0% found this document useful (0 votes)
22 views232 pages

Remaking Chicago - The Political Origins of Urban Industrial - Rast, Joel, 1956 - DeKalb, Ill, 1999 - DeKalb, Ill - Northern Illinois University - 9780875802480 - Anna's Ar

In 'Remaking Chicago', Joel Rast critiques the notion that economic changes necessitate a focus on downtown revitalization for urban fiscal viability, arguing instead that cities have diverse development options influenced by political dynamics. He explores the historical context of Chicago's economic development post-World War II, highlighting the conflict between manufacturing interests and downtown redevelopment efforts. The book emphasizes the significant role of local politics in shaping urban economic trajectories and policy decisions.

Uploaded by

Jmdietr
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
22 views232 pages

Remaking Chicago - The Political Origins of Urban Industrial - Rast, Joel, 1956 - DeKalb, Ill, 1999 - DeKalb, Ill - Northern Illinois University - 9780875802480 - Anna's Ar

In 'Remaking Chicago', Joel Rast critiques the notion that economic changes necessitate a focus on downtown revitalization for urban fiscal viability, arguing instead that cities have diverse development options influenced by political dynamics. He explores the historical context of Chicago's economic development post-World War II, highlighting the conflict between manufacturing interests and downtown redevelopment efforts. The book emphasizes the significant role of local politics in shaping urban economic trajectories and policy decisions.

Uploaded by

Jmdietr
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 232

Change

RemakingChicago

__Ixamining Chicago as
a model for urban economic develop¬
ment in the post-World War II era,
Joel Rast challenges the conventional
belief that structural economic chan ge
has forced cities to concentrate
resources on downtown revitalization
efforts in order to remain fiscally
viable. Rast argues instead that cities
face multiple economic development
choices and that politics play a funda¬
mental role in deciding among them.
During the late 1950s, a coalition
of city officials and downtown busi¬
ness leaders initiated planning efforts
that would help reshape central
Chicago into a modern mecca of
service industries and affluent residen¬
tial neighborhoods, chasing viable
manufacturers from the near down¬
town area in the process. More
recently, however, manufacturers have
sought protection and support from
city government, forming alliances
HC
108
SOCIAL SCIENCES DIVISION ,C4
R37
CHICAGO PUBLIC LIBRARY 1999

400 SOUTH STATE STREET


HWLCHI
CHICAGO, IL 606Q5 Remaking Chicago : the political or

Chicago Public Library

LJ REFERENCE
Form 178 rev. 1-94

SOCIAL SCIENCES SWOON


CHICAGO PUBLIC LIBRARY
400 SOOTH STATE STRfiEI
CHICAGO, IL 60605
.
Remaking Chicago
\

*
The Political Origins of
Urban Industrial Change

si ua?

Remaking

Joel Rast
rffi Northern
Illinois
University
Press
DeKalb
1999
© 1999 by Northern Illinois University Press

Published by the Northern Illinois University Press,

DeKalb, Illinois 60115

Manufactured in tire United States using acid-free paper

All Rights Reserved

Deisgn by Julia Fauci

Library of Congress Cataloging-in-Publication Data

Rast, Joel, 1956-

Remaking Chicago : the political origins of urban

industrial change / Joel Rast.

p. cm.

Includes bibliographical references and index.

ISBN 0-87580-248-6

1. Chicago (Ill.)—Economic conditions. 2. Chicago (Ill.)

—Politics and government—1951. 3. Manufacturing

industries—Illinois—Chicago—History—20th

century. 4. Urban economics—Case studies. I. Title.

HC108.C4 R37 1999

338.9773'll—dc21

98-33177

CIP
SOCIAL SCIENCES VISION
CHICAGO PUBLIC LIBRARY
400 SOUTH STATE STREET
CHICAGO. IL 60005

To my parents

Susanna and Walter Rast


M&i iXk^: '**:
Tii^ STATS >;..
**'•’ Jf JCtoum

*
Contents

Tables and Maps ix

Preface xi

Abbreviations xv

1 The Politics of Urban Economic Development 3

2 Downtown Redevelopment and the Chicago Political Machine

3 Rethinking Industrial Decline:


The Chicago Printing and Apparel Industries 47

4 Community Economic Development and the Crisis


of Machine Politics 83

5 The Battle for the Near North Side 107

6 Toward a Citywide Industrial Policy 132

Conclusion 158

Notes 173

Bibliography 177

Index 195
V
List of Tables and Maps
Tables

1. Budgeted Capital Improvements for Basic Infrastructure in


Three Locations, 1959-1963 36

2. Budgeted Capital Improvements for Basic Infrastructure in


Three Locations, 1977-1981 37

3. Land Values in and near Printing House Row, 1950-1990 54

4. Printing Establishments in the Central Area of Chicago,


1981 and 1993 58

5. Apparel Employment in Cook County, 1970 and 1980 62

6. Concentration of Apparel Firms in the Chicago Garment District,


1951-1981 66

7. Land Values in and near the Chicago Garment District,


1950-1980 67

8. Manufacturing Employment in the City of Chicago, 1958,


1972, and 1983 88

9. Employment in Five Chicago Community Areas, 1972 and 1983 89

10. Employment, Wages, and Taxes Paid by Clybourn


Corridor Business Firms, 1988 128

11. Promised versus Actual Investments in Industrial Streets


and Viaducts, 1990-1994 150

12. Capital Improvements Expenditures in the North Side


Industrial Corridors, 1983-1994 151

13. Land Values in and near Core Zone of Clybourn Corridor PMD,
1975-1994 154

14. Land Values in and near Goose Island PMD, 1975-1994 154

15. Land Values in and near Ravenswood Industrial Corridor,


1975-1994 155

16. Changes in Industrial Firms and Employment in PMDs


and Industrial Corridors, 1988 and 1992 155

Maps

1. Chicago Central Area 23

2. Chicago Downtown Industrial Districts 49


3. North River Industrial Corridor with Clybourn Corridor Sub-Area 112

4. North River Industrial Corridor with Elston Corridor Sub-Area 139


V
Preface

This book began its life as my doctoral thesis in political science at the Uni¬
versity of Oregon. During the summer of 1994, having just completed com¬
prehensive examinations, I was in the position of many new ABDs: euphoria
at finally having all exams and course work behind me was quicldy giving way
to anxiety over the need to identify a suitable dissertation topic. While poring
through the urban politics literature that summer, I stumbled across an arti¬
cle on land-use conflicts in Chicago between centrally located manufacturers
and real estate developers seeking to rehabilitate near-downtown industrial
loft buildings for new commercial and residential uses during the administra¬
tion of progressive mayor Harold Washington (1983-1987). At the time the
article was written, the administration was considering a number of policy
options to protect manufacturers from redevelopment pressures, thereby pre¬
serving well-paying blue-collar jobs for Chicago residents. These policies
were being challenged by a powerful coalition of business interests eager to
see land use in the outlying portions of Chicago’s downtown area give way to
activities that more directly complemented the retail and corporate functions
that dominate the city’s downtown business core.
I had lived in Chicago for several years of Washington’s tenure as mayor,
working as a legal assistant in a large downtown law firm (an experience that
soon had me furiously typing applications for graduate school). However,
most of my information on Chicago politics during those years came from lo¬
cal media sources, which tended to cover redevelopment issues concerning
the central area of the city largely from the perspective of real estate develop¬
ers and their allies within the downtown business community. Like most
Chicagoans at the time, I believed that the decline of city manufacturing was
due largely to changes in locational preferences and decreasing competitive¬
ness on the part of industries themselves. The possibility that, in truth, viable
industrial firms were being uprooted from the downtown area was intriguing
enough that I decided to investigate the issue further. Within several months,
I was convinced I had discovered a worthy subject for my dissertation.
In my efforts to come to grips theoretically with what I was observing in
Chicago, I was drawn to recent scholarship in the field of urban political
economy, particularly the work of Clarence Stone and Stephen Elkin on ur¬
ban regimes. By emphasizing the role of urban governing coalitions rather
than broad structural factors in the formulation of public policy, regime theo¬
rists have sought to introduce a wider appreciation for political choice and
xii Preface

agency within the urban political economy paradigm. The focus on coalition
building could seemingly help account for the neighborhood-oriented poli¬
cies of progressive mayors such as Harold Washington in Chicago, Raymond
Flynn in Boston, and Art Agnos in San Francisco during the 1970s and
1980s, despite powerful structural pressures favoring greater responsiveness
to the interests of property developers and other downtown business elites.
Yet the more I learned about downtown redevelopment in postwar
Chicago, the less convinced I became that existing urban political economy
provided a satisfactory explanatory framework. The problem was this: De¬
spite die emphasis on political variables in much contemporary urban theory,
scholars today continue to privilege the role of economic structure in city
politics. The market, it is generally understood, operates according to its own
logic (a logic of efficiency), which city officials are largely powerless to influ¬
ence. The role of city government is to redistribute and regulate wealth. If
economic forces have determined that cities be refashioned into producers of
advanced corporate services instead of manufactured goods, there is little that
even progressive administrations can do besides taking steps to ensure that
low- and moderate-income neighborhoods receive a share of the benefits of
economic growth.
At a time when basic redevelopment patterns in cities are attributed largely
to the activities of private investors in the market, this book seeks explana¬
tions for urban economic change in an unlikely place: local politics. The ar¬
gument of this study is that two equally efficient possibilities for the redevel¬
opment of Chicago’s central area presented themselves following World War
II. Although a corporate-center strategy favoring commercial and residential
growth proved to be temporarily victorious, it has been successfully contested
more recently by proponents of a viable alternative development strategy
built around locally rooted forms of industrial development. Local politics
have played a decisive role, not simply in redistributing the rewards of eco¬
nomic growth or regulating the activities of private investors but also in es¬
tablishing the very trajectory of redevelopment itself.

In writing this book and the dissertation upon which it is based, I received
valuable support from a variety of sources. First and foremost, I owe a
tremendous debt to my dissertation committee members: Gerald Berk, Dick
Kraus, Dan Goldrich, and Quintard Taylor. My advisor Gerald Berk’s own
work in American political development and his collaborations with Todd
Swanstrom in urban political economy provided the key inspiration for the
questions posed in this study. Although we were not always in complete
agreement, his thoughtful criticisms of my dissertation chapters forced me to
sharpen and, at times, rethink my arguments. In addition, Todd Swanstrom
Preface xiii

and Paul Kleppner both read the entire manuscript and provided constructive
feedback and detailed suggestions for revision. If significant errors remain, it
is only because I was too stubborn to heed the good advice I received.
The University of Oregon provided generous financial support for this ef¬
fort through a university Doctoral Research Fellowship, allowing me to de¬
vote the entire 1995-1996 academic year to full-time research and writing.
My employer, the Center for Neighborhood Technology, graciously agreed
to a leave of absence during the summer of 1998, providing me with the
time I needed to complete revisions to the manuscript necessary for its trans¬
formation into a book. My brothers, Tim and Pete, came to the rescue with
last-minute technical support, straightening out a computer glitch that
threatened to send one of my chapters into oblivion. All maps were prepared
by Tom Willcockson of Mapcraft in Woodstock, Illinois.
Various individuals with firsthand knowledge of events or issues discussed
in the book gave graciously of their time, sitting for interviews or otherwise
sharing key information. I am especially grateful to Margie Gonwa and
Michael Holzer from the Local Employment and Economic Development
(LEED) Council; Donna Ducharme and Greg Longhini from the City of
Chicago; Wim Wiewel, David Ranney, and Lauri Alpern from the University
of Illinois at Chicago; Jody Kretzmann from Northwestern University; Paul
Ginger from the Chicago Association of Neighborhood Development Orga¬
nizations (CANDO); James Lemonides from Greater North Pulaski Devel¬
opment Corporation; Ken Govas from the Industrial Council of Northwest
Chicago; Michael Buccitelli from Jane Addams Resource Corporation; Vickie
Shea from Greater Southwest Development Corporation; Maureen Hellwig
from Erie Neighborhood House; Dorothy Fuller from the Chicago Apparel
Center; Carl Liametz from Chicago Manufacturing Center; Joseph Costigan
from UNITE; James Niesen from the Printing Industry of Illinois and Indi¬
ana Association; and James Maar.
A number of friends and family members provided lodging and valuable
companionship during lengthy stays in Chicago. My uncle and aunt,
Lawrence and Elaine Rast, went above and beyond the call of duty, malting
their condominium on Aldine Street available to me for several extended re¬
search trips. I also spent a number of months at the home of my parents, Su¬
sanna and Walter Rast. Their enthusiasm for this project and good company
helped the time spent away from my partner in California pass far more
quickly than it otherwise would have. Scholarship sometimes yields unantici¬
pated benefits, and the opportunity to return home once again for more than
a short visit was certainly one of them.
Last, but by no means least, I thank my partner, Cliona. Her intellectual
and emotional support, sense of humor, and unwavering confidence in my
abilities kept me going during sometimes trying weeks and months. For that
and much more, I am fortunate beyond words.. .
*
A

V,

'v
List of Abbreviations
CAD computer-aided design
CANDO Chicago Association of Neighborhood Development
Organizations
CAP Citizens Action Program
CCAC Chicago Central Area Committee
CDBG Community Development Block Grant
CDC Community development corporation
CHA Chicago Housing Authority
CMC Chicago Milwaukee Corporation
CTA Chicago Transit Authority
CUED Center for Urban Economic Development
CWED Community Workshop on Economic Development

DED Department of Economic Development


GNPDQ Greater North Pulaski Development Corporation

HUD U.S. Department of Housing and Urban Development

ICNC Industrial Council of Northwest Chicago

IRB Industrial Revenue Bond

JARC Jane Addams Resource Corporation

LEED Local Employment and Economic Development

LIRI Local Industrial Retention Initiative

MCMEA Mayor’s Council of Manpower and Economic Advisors

NCBG Neighborhood Capital Budget Group

NORBIC North Business and Industrial Council

OEM original equipment manufacturer

PMD Planned Manufacturing District (prior to 1988 known


as Protected Manufacturing District)

RERC Real Estate Research Corporation

TIF tax increment financing

UDAG Urban Development Action Grant


.
Remaking Chicago
*

"v
Chapter One

The Politics of
Urban Economic Development

X
A X merican cities have long been symbols of our greatest potential and
deepest failings as a society. Cities like Chicago, New York, Boston, and New
Haven play hosts to prestigious universities where the nation’s rich and pow¬
erful send their children to be educated, yet inner-city elementary schools are
often ill-equipped to teach even basic reading and writing skills. Fifth Avenue
apartments in New York City’s Trump Tower sell for over $10 million, while
sixty thousand to seventy thousand homeless city residents struggle to survive
on the streets below. In Chicago, the notorious Cabrini-Green public¬
housing development stands little more than a stone’s throw from the luxury
high-rises and town houses overlooking Lake Michigan along the city’s Gold
Coast. White, Latino, and African-American office workers in the Loop,
Chicago’s central business district, interact today with a degree of mutual re¬
spect unimaginable thirty years ago, yet at the end of the day they still ride
separate trains home. The juxtaposition of wealth and poverty, tolerance and
intolerance, hope and despair, has bedazzled observers of American cities
since the nineteenth century.
In the decades following World War II, the failings of cities seemed to
grow especially pronounced. During the 1950s, middle-class whites began
what would become a massive exodus from central cities to newly developing
suburban communities, helping to trigger an urban fiscal crisis and causing a
major decline in downtown retailing. Alarmed city officials responded with
the bulldozers of urban renewal, clearing vast tracts of near-downtown land
occupied at the time by small businesses and low- and moderate-income resi¬
dents. The hope was that private developers would build new commercial
and middle- to upper-income housing developments in centrally located ar¬
eas, helping to stabilize downtown property values and creating new markets
4 Remaking Chicago

for downtown retail establishments. Redevelopment, however, was slow to


come. Meanwhile, central area residents displaced by urban renewal projects
were herded into high-rise public-housing projects that became glaring sym¬
bols of a failed urban policy.
During the early 1980s, widespread disillusionment with the future
prospects of American cities gave way in some circles to guarded optimism.
Evidence of a “back-to-the-cities movement” led by middle- and upper-class
young professionals began to receive growing attention in the mass media
(Cicin-Sain 1980). New downtown developments such as Boston’s Quincy
Market and Baltimore’s Harborplace were hailed as signs of an “urban renais¬
sance” (Rosenthal 1980, 9). City officials soon grew mesmerized with the
tourist and convention trades, promoting hotel construction, building con¬
vention centers, and competing in cutthroat fashion for professional sports
franchises. Strong corporate demand for centrally located office space fueled
a downtown construction boom that lasted for much of the 1980s.
Yet many observers soon began to see a downside to the new urban re¬
naissance. Despite successful efforts to resuscitate decaying central business
districts in cities like Boston, Philadelphia, Chicago, and Baltimore, poor
inner-city neighborhoods continued to deteriorate (Abravanel and Mancini
1980; Levine 1989; Bartelt 1989). The growing preoccupation of city offi¬
cials with the glittering trophies produced by downtown revitalization efforts
was creating “divided cities” in which professional workers from gentrified
neighborhoods shared space uncomfortably with low-income and homeless
city residents holding few meaningful economic opportunities (Fainstein,
Gordon, and Harloe 1992). The success of Harborplace notwithstanding,
poverty rates increased in 90 percent of Baltimore’s predominantly black
neighborhoods during the 1970s (Levine 1989, 25). As Mayor Kurt
Schmoke remarked in 1987, “If you were revisiting Baltimore today after a
twenty-year absence, you would find us much prettier and much poorer”
(Judd and Swanstrom 1994, 349).
The roots of the problem, most agreed, lay in the changing structure of
postwar urban economies. Shortly after the close of World War II, manufac¬
turers began moving out of central cities, taking well-paying blue-collar jobs
with them. During the following decades, cities increasingly became produc¬
ers of services rather than goods. Although job growth in the downtown ser¬
vice sectors of many cities helped offset the decline of manufacturing employ¬
ment, the kinds of new jobs being created tended to be either high-paying or
low-paying, fostering a two-tiered distribution of income with a relatively
small middle class (Stanback and Noyelle 1982, 140-142). For many unem¬
ployed factory workers with limited formal education, unskilled positions
opening up at the bottom end of the expanding service sector represented
the only real prospects for reentry into the workforce.
Urban economic restructuring seemed to pose a dilemma for city officials.
The Politics of Urban Economic Development 5

With opportunities for meaningful employment rapidly diminishing in many


inner-city neighborhoods, the pursuit of economic growth was proving less
and less effective as a panacea for the social problems of cities. Yet the alter¬
native—to divert policies and programs from accumulation in favor of efforts
to improve conditions in low- and moderate-income neighborhoods—threat¬
ened to result in a withdrawal of private investment and a loss of resources
for redistribution. Studies of development politics in postwar cities began to
uncover a similar pattern: in most cases, city officials formed close partner¬
ships with business leaders focused around downtown redevelopment efforts
of one form or another, while residents of low- and moderate-income neigh¬
borhoods fought with varying degrees of success for a share of the benefits
(Squires 1989; Cummings 1988; Frieden and Sagalyn 1989). By most ac¬
counts, urban economic restructuring seemed destined to produce an in¬
creasingly adversarial politics of growth versus redistribution, forcing policy¬
makers to straddle a growing divide between downtown business elites and
less privileged neighborhood residents (see Berk and Swanstrom 1994).
This book examines urban redevelopment in Chicago from 1955 to 1997,
focusing in particular on the redevelopment of Chicago’s central area. The
subject matter has been treated elsewhere (see, for example, Bennett 1989;
Squires et al. 1987; Weiss and Metzgar 1989; Ferman 1991; Greer 1983), but
the line- of argument pursued here differs from most other accounts in at least
one key respect. Studies of downtown redevelopment in Chicago and other
central cities typically begin with a discussion of the underlying economic con¬
text, namely, the transformation of the urban economic base from manufac¬
turing to services following World War II. In general, this process is under¬
stood to be driven by forces outside the control of city officials, such as
corporate restructuring efforts and regional variations in land and labor costs.
Urban economic development policy, by most accounts, is highly constrained
by the market. Officials can tax or regulate investment, but they cannot easily
influence the form or trajectory it takes. Because the market has determined
that cities have become efficient locations for service sector growth, but not
for industrial development, city officials must focus economic policies on the
central business district or face the political consequences of economic decline.
There is much at stake in such efforts to come to grips theoretically with
the key structural features of the urban political economy. If it is indeed true
that structural economic change places well-defined limits on the range of vi¬
able urban economic development strategies, then the policy choices open to
public officials are likely to be similarly constrained. Under such circum¬
stances, perhaps the best that residents of low- and moderate-income city
neighborhoods can hope for are urban regimes committed to some form of
reallocation of the economic rewards generated downtown in their favor. Sup¬
pose, however, that theorists have generally misinterpreted the role of private
investment activity in the process of postwar urban economic restructuring,
6 Remaking Chicago

and that public policy has had a significant impact on the trajectory of eco¬
nomic growth, as well as on the distribution of benefits. If so, then we need
to reexamine the course of postwar urban redevelopment to determine why
certain growth strategies were chosen over others, who they favored, and
what the consequences were.
This study represents such an undertaking. Put briefly, I will show that
Chicago’s central area emerged as the focus of a heated political conflict over
the direction of urban economic growth in the postwar era. Within the con¬
text of global economic change, two coherent economic development alter¬
natives presented themselves, each favoring very different segments of the ur¬
ban community. On the first model, the chief beneficiaries would be
downtown property developers, landowners, and other groups with a stake in
the commercial and residential redevelopment of Chicago’s downtown area.
On the second model, those standing to gain were a select group of manu¬
facturers, workers, and residents from working-class communities with an in¬
terest in preserving the city’s blue-collar jobs base. Both models presented
coherent opportunities for economic growth. Neither one could be put into
practice without aggressive support from city government. Contrary to much
popular thinking on the subject of urban economic development, the choice
between them would be made, to a great extent, through politics.

Urban Economic Restructuring and the Corporate-Center Strategy


Contemporary urban scholarship is grounded in a shared narrative of post¬
war urban redevelopment, which portrays key characteristics of the contempo¬
rary urban political economy as the outcome of a far-reaching process of ur¬
ban economic restructuring. During the late nineteenth and early twentieth
centuries, the argument goes, the economic fortunes of cities were deter¬
mined largely by the strength of their manufacturing industries, which tended
to cluster in downtown locations, close to centrally located transportation
nodes. By the end of World War II, however, central city manufacturing
showed signs of growing stagnation, while new industrial development began
to locate increasingly in outlying portions of metropolitan areas (Vernon
1959, 49). By most accounts, the development of the interstate highway sys¬
tem and growth of the trucking industry fueled the exodus of manufacturing
from central cities during the 1950s and the 1960s by making the inexpensive
land and nonunion labor on the urban periphery accessible to industrial firms
(Rain 1968; Moses and Williamson 1967; Hoover and Vernon 1959).
During the following several decades, as innovations in transportation and
communications technologies continued to grow by leaps and bounds, urban
economic restructuring began to assume a global dimension as well (Castells
1985; Froebel, Heinrichs, and Kreye 1977). By the 1970s, a growing num¬
ber of U.S. corporations had relocated their production facilities in Third
The Politics of Urban Economic Development 7

World countries, where wages were a fraction of those earned by American


workers and organizing efforts on the part of labor were typically well con¬
tained (Bluestone and Harrison 1982; Harvey 1986; Beauregard 1989).
With land and labor costs well in excess of those elsewhere, American central
cities were said to be increasingly obsolete as sites for industrial production
(Fainstein and Fainstein 1989, 25; Peet 1987; Massey 1987).
Urban scholars were quick to point out, however, that the decline of man¬
ufacturing jobs within postwar American cities was to some degree offset by
the rapid growth in service sector employment during the same time period.
Corporations shifting their production facilities to the urban periphery and
beyond frequently concentrated decision-making and administrative capacity
within the downtown areas of central cities. Faced with increasingly complex
business environments following World War II, corporate managers came to
rely on a range of advanced corporate services too cosdy and inconvenient to
develop in-house but found in ample supply within the central business dis¬
tricts of large cities (Cohen 1979). Downtowns ultimately became magnets
for legal, financial, accounting, and advertising firms, which grew in propor¬
tion to the rise in corporate headquarters’ activities (Mollenkopf 1983;
Noyelle and Stanback 1983; Stanback and Noyelle 1982).
According to most contemporary urban theorists, structural economic
change, occurring exogenously to municipal politics, left city officials with
relatively few economic development policy options. Most have employed
what sociologist Richard Child Hill has termed the “corporate-center strat¬
egy” for downtown redevelopment (Hill 1983). Recognizing that the heyday
of manufacturing-led urban economic growth is over, development officials
have instead focused their attention on efforts to transform aging industrial
cities into modern corporate service centers. Typically, this involves the pro¬
motion of office building and hotel construction downtown, the develop¬
ment of convention centers and large sports complexes, the fostering of cul¬
tural and entertainment activities in the central business district to promote
tourism and keep downtown streets lively during after-business hours, and
the replacement of industrial and lower-income residential districts in the
near-downtown area with middle- and upper-income housing for downtown
office workers.
The concentration of economic resources on the downtown area and the
disruptive effects of land-use changes have frequendy jeopardized public sup¬
port for the corporate-center strategy. Responding to critics, proponents of
downtown commercial and residential redevelopment insist that their plans
represent the “highest and best use” for central area property, creating job
opportunities, enhancing the city’s tax base, and easing the tax burden for
middle-class homeowners and other city residents. City government, they ar¬
gue, can see to it that the benefits of the corporate-center strategy are truly
widespread by redistributing tax revenues generated downtown in favor of
8 Remaking Chicago

low- and moderate-income city neighborhoods and ensuring that city


residents win a fair share of the jobs and contracts produced by downtown
revitalization efforts.

Urban Political Economy


The narrative of postwar urban economic restructuring provides the key
historical underpinning for the paradigm of urban political economy, which
emerged during the 1970s and the 1980s to unseat pluralism as the model of
choice for studying community power structures. Pluralists had argued that
the most effective way of gauging the true distribution of power within a
given municipality was to determine who prevailed in community decision-
malting, independently of economic structure (Polsby 1963, 4). Case studies
of community power in New Haven, Chicago, and other cities utilizing the
pluralist framework generally found political power to be widely dispersed
among relatively small interest groups, consisting most often of temporary
coalitions formed around particular issues (Dahl 1961; Polsby 1963; Banfield
1961). Pluralists viewed elections as a key mechanism for holding politicians
accountable to their constituents. Because of widespread indifference among
voters on many issues, even small groups could wield considerable power in a
given situation (Dahl 1961, 191).
Pluralists assumed that the government and the economy represented two
distinct, nonoverlapping spheres of activity. As Nelson Polsby wrote, “II a
man’s major life work is banking, the pluralist presumes he will spend his
time at the bank and not in manipulating community decisions” (Polsby
1963, 117). Economic power was of interest to pluralists only to the extent
that it influenced the outcomes of electoral contests. The possibility that
business interests could exercise power in less observable ways, such as im¬
plicit threats to withhold investment, was left unaddressed in pluralist studies
of community power.
With the postwar economic prosperity of the 1950s and the 1960s serving
as a backdrop for the pluralist model of community power, the latter assump¬
tion seemed more plausible than it would during the following decades. By
the 1970s, nationwide economic stagnation and the continuing exodus of
manufacturing firms from older central cities left many urban communities
bending under the weight of impending economic and fiscal collapse. In¬
creasingly, city officials undertook austerity measures, diverting funds from
redistributive social programs into efforts designed to leverage private invest¬
ment and enhance the local tax base. The space for interest group bargaining
seemed to narrow, while businesses profited from a wide array of economic
development incentives, including tax abatements, tax increment financing,
and Urban Development Action Grants (Swanstrom 1985), By the late
1970s, even former pluralists such as Charles Lindblom began to speak of a
The Politics of Urban Economic Development 9

“privileged position” for business among interest groups (Lindblom 1977).


The urban political economy literature contains a wide range of perspec¬
tives, held together by the shared assumption that political power in cities
cannot be understood in isolation from underlying economic structure. In a
fundamental break from pluralism, urban political economists increasingly
emphasized ways in which public officials were systemically predisposed to fa¬
vor the concerns of business over other interest groups (Stone 1980; Cren-
son 1971; Lukes 1974). Interest in coalition-building waned as theorists
sought to explain how economic constraints shaped public policy in ways fa¬
vorable to business, oftentimes with little prodding from business leaders
themselves (Peterson 1981; Goodman 1979; Kantor 1987). According to
structural Marxists, fiscal and economic pressures alone were enough to ex¬
plain the active and enthusiastic participation of city officials in downtown re¬
development, urban renewal, and gentrification schemes, all of which allo¬
cated costs and benefits among different segments of the community in a
highly inequitable fashion (Beauregard 1984; Gordon 1978; Smith 1982;
Hill 1978).
For structural urban theorists, the key feature of the contemporary urban
political economy responsible for the privileged position of business was the
dramatic increase in capital mobility beginning in the 1960s. Innovations in
transportation and communications technologies increasingly freed producers
from the need to locate in close proximity to customers and suppliers (Blue-
stone and Harrison 1982; Peet 1987). No longer tied to dense urban ag¬
glomerations with high land and labor costs, capitalist firms became free to
roam the globe in search of the least expensive production sites (Fainstein
and Fainstein 1989; Castells 1985). According to structural theorists, the
consequences of capital mobility for cities have been far-reaching: interurban
competition generates intensive pressure for more permissive work and envi¬
ronmental regulations (Goodman 1979); the presence of a “reserve army of
places” allows firms to extract costly concessions from workers and communi¬
ties that may exceed the employment and fiscal benefits provided by invest¬
ment (Glickman 1987; Jones and Bachelor 1984; Blair, Fichtenbaum, and
Swaney 1984); internationalization imposes a “global logic” on urban devel¬
opment policy reflecting the interests of powerful multinational corporations,
forcing even progressive urban regimes to fall in line (Castells 1985; Kantor
1987); redistributive policies disappear from the public agenda as the space
for genuine political debate shrinks to a narrow range of nondevelopmental
issues that do not affect the local business climate one way or the other (Pe¬
terson 1981; Kantor 1987; Gottdiener 1987).
Most urban theorists eventually recognized the efforts of urban political
economists to integrate economic structure into studies of community power
as a healthy corrective to the one-dimensional pluralist framework. At the
same time, however, the structural bent ofmuch of the new literature
10 Remaking Chicago

produced a wave of studies remarkably devoid of human content and inca¬


pable of accounting for important differences in the way cities responded to
fiscal and economic pressures. Dissatisfaction with the high level of abstrac¬
tion at which many structural urban theories were being pitched led to two
new theoretical strains within the urban political economy approach, both of
which emphasized the role of business-led coalitions as determinants of polit¬
ical power in cities.
The first of these perspectives, introduced by sociologists John Logan and
Harvey Molotch, holds that community power structures are dominated by
coalitions of land-based interests that stand to profit from the growth of the
municipalities in which they are based (Molotch 1976; Whitt 1982; Feagin
1983; Molotch and Logan 1984; Logan and Molotch 1987). “Growth coali¬
tions” are spearheaded by local property owners (rentiers) who seek to en¬
hance the value of their individual holdings through efforts to intensify land
use. Other coalition partners typically include developers, financiers, utility
companies, the local media, and sympathetic public officials anxious to reap
the increasing tax revenues produced by economic growth. According to Lo¬
gan and Molotch, this coalition has been the driving force behind urban re¬
development schemes in cities across the nation since World War II. As a
consequence, cities have been rendered “growth machines” (Molotch 1976;
Logan and Molotch 1987).
Growth machine theorists do not view economic growth as a universal
good. To the contrary, conflict over economic development policy is said to
form the central political cleavage in contemporary cities, pitting progrowth
activists against community residents who bear the costs of economic growth
in rising rents, displacement, and environmental degradation. In their semi¬
nal work Urban Fortunes, Logan and Molotch (1987) describe this as a con¬
flict between use values and exchange values. Communities struggle to main¬
tain social ties and stability, while the proponents of economic growth seek to
maximize profits and develop property for its highest and best use. Commu¬
nity challenges to the growth machine in the form of growth control move¬
ments have been a popular subject of investigation within the literature on
growth machines (see Molotch and Logan 1984; Vogel and Swanson 1989;
Warner and Molotch 1995; Schneider 1992; Shlay and Giloth 1987; Green
and Schreuder 1991; Tulloss 1995; Calavita 1992; Folz et al. 1993).
A relatively new body of literature focusing on urban regimes represents a
second wave of theorizing emphasizing the role of coalition-building in city
politics within a political economy approach (see Elkin 1985, 1987a; Stone
1989, 1993; Stone and Sanders 1987; Stone, Orr, and Imbroscio 1991).
Regime theorists stress that political power within liberal democratic societies
is divided into two spheres of influence, where economic decision-making re¬
mains largely within the purview of private investors and control of public
policy in the hands of popularly elected government officials. The division of
The Politics of Urban Economic Development 11

labor between state and market means that city officials can accomplish rela¬
tively little through their powers of command. Power is exercised instead
largely through informal partnerships between public officials and private citi¬
zens who control resources necessary for governing. Regimes represent the
informal arrangements through which public and private actors collaborate in
order to produce governing decisions (Stone 1989, 6). Structural constraints,
including the need to generate satisfactory levels of economic activity, mean
that business is almost certain to be well represented in such partnerships
(Elkin 1985; Stone 1987b).
A typical urban regime is a growth coalition, featuring the alliance be¬
tween public officials and place-bound rentiers that serves as the focus of the
growth machine literature. However, other possibilities also exist. In the
words of regime theorist Clarence Stone, “a regime represents an accommo¬
dation between the potentially conflicting principles of the popular control of
government and the private ownership of business enterprises” (Stone
1987b, 269). The need to reconcile these two principles limits the range of
viable regime types. A regime that responded only to the imperative of capital
accumulation would be no more successful than one that responded exclu¬
sively to the demands of the electorate. Nevertheless, the tension between
popular control and economic productivity can be accommodated in more
ways than one. In addition to the growth machine model, regime theorists
have uncovered a variety of regime types, encompassing a range of social, po¬
litical, and economic concerns and objectives (see Elkin 1985; Fainstein,
Fainstein, and Armistead 1983; Stone 1993; Body-Gendrot 1987; Clavel
1986; DeLeon 1992; Orr and Stoker 1994; Nickel 1995; Robinson 1995;
Imbroscio 1995).
Regime theory goes further than any other variant of the urban political
economy literature in asserting the relative autonomy of the political sphere
from structural economic constraints. Yet even regime theorists concede
that progrowth governing coalitions centered around downtown commer¬
cial revitalization largely dominate the contemporary urban landscape. Ac¬
cording to Stephen Elkin, “The battlefield of city politics is not flat but is
tilted toward an alliance of public officials and land interests” (Elkin 1987a,
100). In most contemporary cities, downtown redevelopment is the order
of the day.
In his study of community power in postwar Atlanta, Clarence Stone
(1989) explains the privileged position of Atlanta’s business elite by pointing
to the unequal distribution of resources among potential governing coalition
partners. The downtown redevelopment efforts advanced by Atlanta’s
progrowth coalition produced visible signs of progress for which city officials
could claim partial responsibility. They also generated selective incentives such
as jobs, contracts, and retainer fees that could be used as side payments to
boost support for the regime. The unified outlook and extensive associational
12 Remaking Chicago

networks among the downtown business leadership minimized the difficulty


of coordinating extensive redevelopment efforts involving a wide range of ac¬
tors. According to Stone, attempts to form a progressive, neighborhood-
based regime with minimal business participation failed in Atlanta precisely
because the neighborhoods lacked such economic and associational re¬
sources. For progressive mayor Maynard Jackson, such a regime was not an
empowering one, leading him “inevitably toward accommodation” with the
downtown business community (Stone 1989, 95).

Summing Up
Despite important differences among the various perspectives within the
urban political economy literature, there is widespread agreement on a num¬
ber of crucial points. First, it is widely assumed that the division of labor be¬
tween state and market within liberal democratic societies places most private
investment decisions outside the control of city officials. Fundamental eco¬
nomic changes such as the transformation of urban economies from manu¬
facturing to services are seen as occurring prior to, and outside the sphere of,
municipal politics. Such developments set limits on the range of policy
choices available to city officials. They provide the underlying economic con¬
text for political conflict over redistributive and regulatory questions. Thus,
while contemporary urban theorists disagree over the possibilities for recon¬
ciling capitalism and democracy within the context of global economic re¬
structuring, they generally concur that city officials are largely powerless to
intervene in the process of restructuring itself.
In addition, urban political economists typically subscribe to a model of
firm locational behavior borrowed from neoclassical economics. Under this
view, firms choose production sites through a purely instrumental logic based
on estimates of future production costs in various alternative locations. Re¬
cent technological innovations have helped minimize the “stickiness” caused
by transportation- and communications-related constraints on firm mobility.
As a result, capital today is more footloose than ever, a situation that places
cities in increasingly subservient relationships with mobile investors and in¬
creasingly competitive relationships with one another. As regime theorist
Alan DiGaetano observes,

American cities, as tax-dependent creatures, are forced to compete for foot¬


loose capital in order to sustain themselves fiscally. Competition for capital in¬
vestment within the American system of cities means that cities with the ability
to attract and retain mobile capital will grow, while cities that fail to do so will
sink into economic decline. In short, the unevenness of economic develop¬
ment, a byproduct of capital mobility, creates a hierarchy of cities. (DiGaetano
1989,264)
The Politics of Urban Economic Development 13

The mobility of capital and the division of labor between state and mar¬
ket, in turn, set limits on the range of viable urban economic development
strategies. For many industrial firms, the price of land and labor figures
prominendy in locational decisions. The relatively high cost of urban factory
workers and production space places cities at a disadvantage vis-a-vis subur¬
ban, rural, and Third World locations in the competition for mobile indus¬
trial investment. However, because many high-level corporate services find it
necessary to locate within the central business districts of certain cities, some
municipalities may successfully weather the disruptive effects of capital mo¬
bility through efforts to build on the agglomeration economies provided by
their downtown areas (Hill 1983; Mollenkopf 1983; Cohen 1979; Noyelle
and Stanback 1983). Under this strategy, the importance of redistributive
policies becomes paramount, as the replacement of well-paying manufactur¬
ing jobs with low-level service sector occupations increases disparities in the
social and spatial distribution of wealth and income. Redevelopment efforts
may also clash with residential use values, causing displacement of lower-in-
come residents in near-downtown locations (Molotch 1976; Logan and
Molotch 1987).
Downtown redevelopment and urban economic restructuring thus tend
to magnify the tension between capital and community, which urban politi¬
cal economists today widely recognize as the central political cleavage in
contemporary cities (O’Connor 1973; Mollenkopf 1981; Logan and
Molotch 1987; Stone 1987a). On one side are progrowth coalitions domi¬
nated by downtown business elites; on the other side are neighborhood rep¬
resentatives battling the disruptive effects of accumulation strategies ema¬
nating increasingly from the central business district, imposing substantial
costs while providing little in the way of benefits to neighborhood workers
and residents. Caught in the middle of this divide are city officials, who
must find a way to reconcile the tension between accumulation and legiti¬
mation if the governing arrangements over which they preside are to be suc¬
cessful. Given the division of labor between state and market and the mobil¬
ity of capital, however, their options for doing so are normally seen as quite
limited.

Overview of the Narrative


To a certain point, the redevelopment of Chicago’s central area following
World War II seems consistent with the prevailing narrative of urban eco¬
nomic restructuring. When Richard J. Daley became mayor of Chicago in
1955, the city’s economy was in decline. Under pressure to act, Daley soon
forged an alliance with key representatives from the downtown business com¬
munity focused around rearranging land use in central Chicago in accordance
with the corporate-center strategy for downtown redevelopment. In 1958,
14 Remaking Chicago

the administration announced plans for a major restructuring of the central


area, featuring a new university campus, a convention center, several new
government office buildings, and a network of expressways designed to link
the central business district with oudying portions of the metropolitan area
(Department of City Planning 1958). During the next several decades, the
city would use its powers over land use, infrastructure provision, the distribu¬
tion of federal urban development funds, and other policy tools to help make
these plans a reality.
By most accounts, the Daley administration’s downtown redevelopment
efforts were enormously successful, reversing a thirty-year decline in down¬
town property values and fueling the construction of thirty-two million
square feet of new office space in the Loop between 1962 and 1977 (Weiss
and Metzgar 1989, 135). Yet downtown revitalization also involved certain
less visible economic development trade-offs. When city officials began to im¬
plement their development plans for the central area of Chicago in the late
1950s, the Loop was surrounded by a ring of industrial districts that pro¬
vided nearly 25 percent of Chicago’s manufacturing jobs at the time. By the
1970s, the economic turnaround of the Loop had begun to make turn-of-
the-century industrial loft buildings in near-downtown manufacturing dis¬
tricts such as Printing House Row and River North desirable properties for
alternative uses, as living space for downtown office workers and back office
space for downtown businesses. Both uses commanded rents considerably
higher than manufacturing, luring real estate developers and land speculators
to the area. By the mid-1980s, rising rents, property tax increases, and com¬
plaints from new loft-dwelling “urban pioneers” had driven all but a few in¬
dustrial firms from this section of the city. Missing from most accounts of
downtown redevelopment in Chicago is the story of industrial decline that
accompanied the dramatic resurgence of the central business district follow¬
ing World War II.
The prevailing narrative of urban economic restructuring suggests that
the decline of these near-downtown industrial districts was a foregone con¬
clusion once transportation improvements and technological innovations
made it possible for central city manufacturers to abandon the high-density
environment of the near-downtown area. However, the lure of cheaper land
and labor costs on the urban periphery proved sufficient to attract only cer¬
tain types of central area industrial establishments to new suburban and
rural locations. In general, such firms tended to be relatively large, verti¬
cally integrated, and producing for national or international markets. By
contrast, another class of near-downtown manufacturers—typically small-
to-medium in size, engaged in contracting or subcontracting activities, and
producing for local or regional markets—showed little interest in vacating
their existing central area locations. As Chapters 2 and 3 will illustrate, the
Daley administration’s corporate-center strategy for downtown commercial
The Politics of Urban Economic Development 15

and residential redevelopment would prove disastrous for the latter group
of manufacturers.

Industrial Districts and Regional Development


The notion that some manufacturing firms might actually prefer central
city production sites over suburban, rural, or Third World locations is diffi¬
cult to accept within the dominant paradigm of capital mobility and indus¬
trial decentralization. Increasingly, however, empirical research on firm be¬
havior within certain industrial regions of Europe, North America, and Japan
has begun to call aspects of this paradigm into question. Since the late 1970s,
researchers have identified a growing number of industrial districts in such ar¬
eas as the Emilia-Romagna province of Northern Italy, Baden-Wiirttemberg
in Germany, West Judand in Denmark, the Silicon Valley in California, and
the Route 128 Corridor near Boston, all of which feature spatially rooted ag¬
glomerations of manufacturing firms bound to one another through exten¬
sive subcontracting and networking relationships (Brusco 1982; Pyke, Becat-
tini and Sengenberger 1990; Pyke and Sengenberger 1992; Hirst and Zeitlin
1989; Saxenian 1994). Industrial districts such as these have become the ob¬
ject of considerable scholarly attention in recent years because the regions
within which they are concentrated have, in many instances, experienced su¬
perior economic performance.
The economic success of the new industrial districts comes less from low-
cost factors of production such as cheap land or labor than from the organi¬
zational structures of the districts themselves (Sengenberger and Pyke 1992).
Typically, individual firms specialize in one or several phases of a total pro¬
duction process. Subcontracting and specialization among firms foster
economies of scope, making small-batch production economically feasible
and facilitating rapid response to the quickly changing consumer demands
characteristic of today’s markets.1 In addition, producers can easily increase
or curtail output simply by adding or reducing the number of subcontractors
they employ, relieving them of the risk and expense involved in adding to
their own productive capacities. The most successful districts feature a combi¬
nation of competition and cooperation among subcontractors, who may
share tools, information, and workers with one another while at the same
time competing intensely for each new job opportunity (Harrison 1992; Sen¬
genberger and Pyke 1992). Local and regional governments oftentimes play
supportive roles, providing infrastructure, technical support, and other busi¬
ness services.
According to much of the contemporary literature on industrial districts,
the cement that binds firms in the districts together and accounts for much
of the dynamism of the districts as a whole is trust (Sabel 1989; Lorenz
1989; Zeitlin 1992; Harrison 1992). Subcontractors are willing to pass along
16 Remaking Chicago

design ideas and share information on new production technologies with one
another because they know, based on past experience, that their neighbors
will reciprocate (Sengenberger and Pyke 1992). Each firm owner is acutely
aware that Inis or her individual success is tied to that of the entire district,
encouraging a collective outlook among district-member firms. The line be¬
tween business and community tends to blur. In contrast to the rational, self-
interested behavior attributed to capitalist firms within the neoclassical model
of interfirm relations, economic relations within the industrial districts are so¬
cially embedded (Zeitlin 1992; Lorenz 1989; Granovetter 1985). Producers
and subcontractors will not relocate at the drop of a hat because they know
that these relationships, built through years of repetitive, face-to-face interac¬
tions, cannot be easily reproduced elsewhere.
The new literature on industrial districts is of interest to the story of
downtown redevelopment in Chicago because several of the near-downtown
area’s largest industrial sectors following World War II, namely, printing and
apparel, were organized along lines similar to this. Beginning in the early
twentieth century, manufacturers from both of these sectors clustered in well-
defined printing and garment districts in the Loop area, where they main¬
tained close ties with important downtown customers and contracted and
subcontracted extensively with one another. During the postwar years, even
as many larger, vertically integrated producers abandoned the districts for
lower-cost locations elsewhere, a solid core of small- to medium-sized print¬
ers and apparel manufacturers remained well-rooted in the near-downtown
area. As Chapter 3 illustrates, these manufacturers ultimately fell victim less
to structural economic change than to the Daley administration’s corporate-
center strategy for downtown commercial and residential redevelopment.

The Local-Producer Strategy for Neighborhood Revitalization


During the 1960s and 1970s, proponents of the corporate-center strat¬
egy encountered little resistance either from central area manufacturers ex¬
periencing displacement pressures or from the nearby residential communi¬
ties from which industrial firms drew their employees. In part, this is
undoubtedly because, to the untutored eye, downtown redevelopment
seemed to be driven largely by market forces. What manufacturers saw were
property developers and land speculators seemingly exercising legitimate
property rights by converting industrial loft buildings to higher-rent com¬
mercial and residential space, or otherwise rearranging land use. What they
failed to see, however, were the land-use plans, zoning decisions, capital
improvements programs, building-code revisions, and other public policy
initiatives that attracted property developers and real estate speculators to
the central area in the first place. Convinced that their problems originated
in the market rather than in politics, many central area manufacturers were
The Politics of Urban Economic Development 17

initially discouraged from seeking political solutions.


Reinforcing the quiescence of manufacturers were the crippling effects of
Chicago’s well-entrenched political machine on neighborhood political mo¬
bilization during the postwar years. The overwhelming concentration of pub¬
lic resources on the downtown area that accompanied the corporate-center
strategy was politically feasible, largely because of the well-defined separation
between electoral politics and public policy that characterized Chicago poli¬
tics at the time. Machine leaders used patronage appointments and selective
incentives to assemble winning coalitions of voters from the city’s working-
class neighborhoods, keeping electoral politics on an individual and largely is¬
sueless basis. With potential neighborhood opposition largely diffused, policy
benefits could be channeled into the hands of downtown property developers
and other business elites with little fear of voter repercussions. Under these
circumstances, central area manufacturers seeking protection from displace¬
ment pressures would have found it difficult to identify ready allies.
Nevertheless, by the mid-to-late 1970s this set of conditions was clearly
beginning to change. To begin with, cracks in the machine’s stronghold on
political power in Chicago were becoming visible on the horizon. A federal
court decision declaring the patronage system of political firing unconstitu¬
tional and a reduction of federal funds for urban development and social wel¬
fare programs both jeopardized the machine’s control over key resources it
had long used to reward its supporters in the neighborhoods. In addition,
Chicago’s already well-developed neighborhood movement was becoming
increasingly vocal and united in its outlook by this time, demanding collec¬
tive benefits from city officials who nevertheless continued to respond to
their low- and moderate-income constituents as individuals.
The mobilization of lower-income and working-class communities was
soon accompanied by a strategic shift within certain factions of the neighbor¬
hood movement, away from its consumer-based, social-protest origins to¬
ward a new focus on neighborhood economic development. Prompted by
the precipitous decline of Chicago’s manufacturing base during the 1970s, a
growing number of community-based organizations began to experiment
with various methods of assisting neighborhood manufacturers—establishing
business incubators to nurture small, start-up firms, helping to foster inter¬
firm linkages among local manufacturers, and providing technical assistance
and other business services to industrial companies. As the policy implications
of these new experiments became apparent, new coalitions of neighborhood
economic development organizations, manufacturers, workers, and working-
class residents became forceful advocates of city government support for
neighborhood industrial retention.
Within this context of regime change and neighborhood political mobi¬
lization, proponents of neighborhood economic revitalization began to artic¬
ulate the features of a coherent economic development alternative to the
18 Remaking Chicago

corporate-center strategy. This alternative, which I call the “local-producer


strategy” for neighborhood redevelopment, rested above all on the argument
that preserving Chicago’s base of locally and regionally owned manufacturing
firms was a necessary counterpart to downtown area commercial and residen¬
tial growth. Its proponents rejected the widely shared belief that manufactur¬
ing in Chicago was either dying or dead, ultimately setting forth a detailed
range of policy proposals that they insisted could help revitalize the city’s in¬
dustrial base. These proposals included the targeting of public investments to
industrial areas of the city, the promotion of worker training programs appro¬
priate for the needs of local manufacturers, greater community involvement
in the planning of economic development programs, and policies designed to
make credit available to locally owned businesses. Rejecting the argument by
downtown business leaders that economic development was synonymous
with real estate development, proponents of the local-producer strategy
urged city officials to make the creation and preservation of well-paying jobs
the centerpiece of economic development policy.
Appropriately enough, advocates of the local-producer and corporate-
center strategies clashed first over competing visions of the future of
Chicago’s central area, the section of the city where the two strategies con¬
fronted one another most directly. Clearly, the real estate community’s
postindustrial vision of a ring of affluent residential and commercial neigh¬
borhoods surrounding the Loop could not become a reality as long as manu¬
facturers continued to occupy significant portions of the central area’s loft
space. The struggle between the two groups broke out into the open first
around the issue of zoning policy. A coalition of groups allied with centrally
located manufacturers on Chicago’s Near North Side argued that the piece¬
meal rezoning of central area industrial districts to permit new residential and
commercial uses was forcing otherwise viable manufacturers from the area.
Criticizing city government for its culpability in Chicago’s industrial decline,
the coalition called for legislation to set aside certain portions of the central
area as protected manufacturing districts where further zoning changes to
nonindustrial land uses would be prohibited.
Proponents of the local-producer strategy found key sympathizers within
the administration of progressive mayor Harold Washington, who served as
Chicago’s first nonmachine mayor in the post-Daley era from 1983 to 1987.
When community leaders first questioned the city’s zoning policies in the
central area, Washington officials helped place the issue squarely on the pub¬
lic agenda, ensuring it would receive a full public airing. The ensuing conflict
helped generate a new dialogue over Chicago’s economic future. As people
discovered that certain manufacturers were leaving Chicago only under
duress, they increasingly questioned the necessity or the desirability of the
postindustrial vision of Chicago that downtown business leaders seemed to
embrace. Proponents of the corporate-center strategy were put on the defen-
The Politics of Urban Economic Development 19

sive, no longer able to defend their land-use preferences through simple ideo¬
logical pronouncements such as “highest and best use.” City government’s
role in downtown redevelopment was unmasked and increasingly called into
question. Ultimately, legislation to protect central area industries from fur¬
ther encroachment by incompatible land uses was passed. Since then, city
manufacturers and their allies have won the reluctant support of an adminis¬
tration led by the late Mayor Daley’s eldest son, who, despite strong down¬
town business ties, has taken a number of significant steps toward the imple¬
mentation of a citywide industrial policy.

Revisiting Urban Political Economy


As we saw earlier, the political economy approach to the study of urban
politics provided the important insight over pluralism that public officials
cannot take economic activity for granted, a condition that in turn shapes the
balance of power within cities. Politicians need votes, as pluralists empha¬
sized. However, public confidence in city government depends in no small
part on the extent to which healthy levels of economic activity can be sus¬
tained. Jobs and tax revenues are contingent upon it. Politicians thus need
more than just votes. To be viable over the long term, a regime must include
a strategy for economic development and coalition partners who control the
resources necessary to implement such a strategy. Business is therefore privi¬
leged. It is not simply one among many private-sector interest groups.
Still, the critique of pluralism offered by urban political economy remains
a partial one. Recall that pluralists conceived of politics and markets as dis¬
creet, nonoverlapping spheres of activity. By showing how social and eco¬
nomic factors such as capital mobility, fiscal crisis, and class struggle permeate
and shape the political arena, urban political economists convincingly put to
rest the notion of a self-contained political sphere driven largely by the give
and take of pluralist interest-group bargaining. Political decision-making
could only be understood, they argued, by taking into account the complex
web of social and economic relations within which politics are embedded. To
do any less would simply invite pluralist conclusions.
By the same token, however, urban political economy has largely failed to
extend the same insight into its portrayal of economic relations. Instead, the
market is typically conceptualized as a largely autonomous sphere of activity,
driven in this case by the rational, self-interested behavior of profit-maximiz¬
ing capitalist firms. As a result, developments in the marketplace take on an
almost organic appearance. The economy becomes part of the “natural” ex¬
ternal environment with which political decision makers must learn to cope.
Urban regimes are portrayed as intervening or dependent variables, respond¬
ing to events in the marketplace such as global economic restructuring with
little to no capacity to affect basic economic outcomes.
20 Remaking Chicago

This view of the economy can support only certain kinds of development
politics. By most accounts, the instrumental behavior of mobile, profit-maxi¬
mizing private investors has signaled that cities are becoming efficient loca¬
tions for only a limited range of economic activities, primarily those that cater
to the needs of corporations and financial institutions located in the central
business district. Current scholarly analysis tends to use the corporate-center
strategy as its point of departure, focusing empirical investigation on strug¬
gles over the allocation of costs and benefits associated with downtown rede¬
velopment. The axis of political conflict is typically drawn to pit progrowth
coalitions dominated by downtown business elites against neighborhood resi¬
dents seeking protection from gentrification pressures and demanding a
greater share of the jobs, contracts, and other side payments that downtown
redevelopment makes available. City officials, meanwhile, straddle the divide
between accumulation and legitimation, carefully weighing redistributive
policies against the dampening effects they are likely to have on investment,
since overburdensome taxes and regulations will prompt mobile corporations
to take their investment dollars to some other city where the business climate
is perceived as friendlier.
The above model captures pertain characteristics of postwar urban politics
that are beyond dispute. Without question, struggles between growth coali¬
tions and neighborhood residents over downtown revitalization efforts have
been a defining feature of city politics since the 1950s. Nevertheless, the pit-
falls of viewing urban politics exclusively in these terms become apparent if
we place this model alongside the narrative of postwar redevelopment in
Chicago outlined in the previous section. For example, what if certain indus¬
trial producers representing significant numbers of manufacturing jobs are
not the footloose, atomistic competitors of the urban political economy liter¬
ature but instead rely heavily on extensive networking relationships that root
them firmly in central city industrial districts? And what if public policy, along
with market forces, plays a significant role in determining which kinds of eco¬
nomic activities will survive in such areas of the city? And, finally, what if
manufacturers threatened by displacement pressures caused in part by unsup-
portive public policies should join with neighborhood residents and workers
to demand an alternative set of policies and programs?
To understand the full range of conflicts, development opportunities, and
political possibilities that presented themselves in connection with the rede¬
velopment of Chicago’s central area following World War II, it is necessary to
extend the critique of pluralism offered by urban political economy to its log¬
ical conclusion. The insight that economic and social factors affect political
outcomes in significant ways, while important, is only part of the story. To
complete the picture, we must also take into account the equally significant
external influences that shape outcomes in the economic sphere, leading
profit-seeking investors to behave in sometimes unexpected Ways and pre-
The Politics of Urban Economic Development 21

senting opportunities for local actors to shape basic economic developments


such as urban economic restructuring. Put briefly, we need to move beyond
investigation of the economics of political relations in order to examine the
politics of economic relations as well.2
Contemporary urban political economy, with its emphasis on the econom¬
ics of political relations, invites a particular set of questions. For example, can
mobile investors be compelled to pay their fair share of the city tax burden?
Or, under what circumstances, if any, can low- and moderate-income city
neighborhoods become beneficiaries of economic growth led by downtown
commercial redevelopment? Or, finally, are urban regimes that pursue com¬
munity services over economic growth viable within liberal democratic soci¬
eties? These are not unimportant questions, but they are also not the only
ones worth asking. A model that takes into account the politics of economic
relations generates an additional set of questions. For instance, how can net¬
working relationships among local producers be fostered to root businesses
more firmly in the community? Or, what types of growth strategies appear to
hold the greatest direct benefits for lower-income and working-class neigh¬
borhoods? Or, finally, what kinds of power structures do urban regimes pur¬
suing alternatives to the corporate-center strategy tend to encounter?
The latter questions are the kind that this study principally seeks to ad¬
dress. Unlike contemporary urban political economy, this study does not as¬
sume that capital and community are on opposite sides of a deep divide that
city officials can hope to bridge only imperfectly. By looking seriously at the
politics of economic relations in postwar Chicago, I intend to show how vari¬
ous groups competed to impose their own distinct visions of redevelopment
on the central area of the city. According to one vision, downtown commer¬
cial and residential redevelopment would generate a large share of the wealth
necessary to provide city services and material rewards to lower-income and
working-class neighborhood residents. According to another vision, efforts
to protect and revitalize near-downtown industrial districts would help re¬
build neighborhood productive capacities, reducing the need for a redistribu¬
tion of benefits produced by downtown commercial and residential growth.
Both strategies would require strong support by city government. Both pre¬
sented opportunities for the formation of viable urban regimes. However,
seen through the lens of contemporary urban theory, only one would be
clearly visible.
Chapter Two

Downtown Redevelopment and


the Chicago Political Machine

T ■Iho the casual observer, downtown Chicago today may well seem to epit¬
omize the triumphant postindustrial American city. The Loop, surrounded
just forty years ago by a semicircle of residential slums and industrial loft
buildings threatening the value of downtown real estate, now appears vibrant
and self-assured. The acres of nearby slum dwellings have, for the most part,
long since been cleared away, and many of the old multistory loft buildings
once home to small- and medium-sized manufacturing establishments just
north, south, and west of the central business district have now been con¬
verted to upscale residences for downtown office workers or back office space
for downtown corporations.
Studies of the deindustrialization of American cities have often focused on
the job losses associated with the closure of large plants in major sectors such
as the steel or automotive industries (Bluestone and Harrison 1982). The
Gary-Hammond-East Chicago area, for years the center of a thriving steel in¬
dustry anchored by the Inland and U.S. Steel mills along the lakefront, lost
twenty thousand steelworker jobs in 1979 alone due largely to global overca¬
pacity in steel production and the failure of multinational conglomerates such
as USX (formerly U.S. Steel) to reinvest profits in their Chicago area steel¬
making facilities (Squires et al. 1987, 30-31).
In fact, the loss of manufacturing jobs like these is only part of the story of
Chicago’s industrial decline since the second World War. At roughly the same
time that hundreds of steelworkers were receiving their pink slips on Chicago’s
far South Side, the long-term viability of another set of industries located in
the city’s central area was increasingly being called into question as well. These
were the printers, garment makers, metalworkers, and other manufacturers
who occupied the small, affordable spaces provided by Chicago’s turn-of-the-
century industrial loft buildings, the majority of which are concentrated in the
Map 1. Chicago Central Area
24 Remaking Chicago

near-downtown area of the city. With yearly rents as low as fifty cents per
square foot and strategic downtown locations, loft buildings provided low-cost
production space and convenient access to centrally located customers, suppli¬
ers, and support services (Real Estate Research Corporation 1970, 19;
Ducharme, Giloth, and McCormick 1986). Although much smaller and less
visible than the smokestack industries of the far Southeast Side, these firms still
provided 115,000 jobs for Chicago residents in 1970, 23 percent of the city’s
total manufacturing jobs at the time (McDonald 1984, 25; Mayor’s Council
of Manpower and Economic Advisors 1974).
Chicago’s central area includes the Loop and roughly ten square miles of
additional land.1 As recently as 1970, manufacturing accounted for approxi¬
mately 25 percent of overall land use in this portion of the city (Real Estate
Research Corporation 1970, 2). In the decades following World War II, in¬
dustries clustered in three distinct areas immediately north, south, and west
of the Loop, forming a rough semicircle around the central business district.
The South Loop and Near South Side were home to large numbers of
graphic communications firms. Within this area lies Printing House Row, the
city’s historical center of printing and publishing. As of 1970, the district’s
twelve late-nineteenth-‘and eady-twentieth-century loft buildings still housed
a total of ninety-six printing establishments (Pruska-Carroll 1987). On the
Near West Side, food processing was the leading sector, followed by printing
and apparel manufacturing. Just north of the Loop, in the River North and
Near North Side portions of the central area, the printing and publishing in¬
dustry was dominant, followed closely by food processing and apparel (Real
Estate Research Corporation 1970, 10).
During the twenty-eight-year reign of Chicago’s postwar machine mayors
Richard J. Daley, Michael Bilandic, and Jane Byrne, large portions of these
three areas were transformed from centers of manufacturing into upscale resi¬
dential communities and back office space for downtown corporations. In
many cases, this took place without a fundamental restructuring of the built
environment. Industrial loft buildings were refitted from the inside out for
new, higher-rent uses—as living space for new “urban pioneers” seeking relief
from the tedium of the suburbs, and as office space for service-sector firms
requiring downtown locations but unwilling or unable to pay the more costly
rents commanded in the heart of the central business district. Between 1963
and 1981 alone, manufacturing employment in the central area experienced a
36 percent decline (McDonald 1984, 8-12).
Analysts commonly point to a combination of market forces and trans¬
portation improvements to explain this transformation. The cost of central
area manufacturing space, already high vis-a-vis suburban and rural industrial
sites, grew increasingly so during the 1960s and 1970s as real estate specula¬
tors and commercial and residential property developers bid up the price of
near-downtown industrial land (McDonald 1984; Ducharme, Giloth, and
Downtown Redevelopment and the Chicago Machine 25

McCormick 1986). With growing use of truck transportation following


World War II freeing manufacturers from their dependence on centrally lo¬
cated freight terminals, many chose to relocate in suburban and rural loca¬
tions to escape the congestion and high land costs of the central area
(Mayor’s Committee for Economic and Cultural Development 1966, 8;
Stone 1974, 7). Cheaper land outside the central city enabled firms to build
sprawling, new one-story production facilities, more efficient for rationalized,
assembly-line production techniques than the multistory loft buildings of the
near-downtown area. In addition, the absence of traffic congestion in rural
locations reduced delivery times and difficulties loading and unloading trucks
(Center for Urban Studies 1966, 51-53; Department of City Planning 1964,
58; Greer 1983, 115-116).
The new commercial and residential development in the central area was
ultimately linked to a changing international division of labor, incorporating
cities like New York, London, and Chicago as centers of command and con¬
trol for global corporations (Noyelle and Stanback 1983). As industrial firms
gravitated increasingly to the urban periphery and beyond, Chicago’s Loop
continued to attract a range of corporate and financial activities requiring
downtown locations. Rapidly escalating property values in the central busi¬
ness district eventually led corporations to seek back office space nearby for
less impprtant functions, while the explosion of downtown office workers
during the 1960s and 1970s helped create a market for near-downtown resi¬
dential space (Bennett 1989). By most accounts, the transformation of cen¬
tral area industrial districts to new commercial and residential areas was a
logical by-product of Chicago’s metamorphosis from an industrial into a
postindustrial city (Greer 1983, 115-116; Real Estate Research Corporation
1970).
Lew would question the significance of decision-making by actors in the
marketplace in the shaping of urban land-use patterns. Yet private investment
activity in any particular city takes place according to a set of publicly deter¬
mined rules and incentives in which municipal policy decisions figure promi¬
nently. Lor example, land-use plans prepared by city officials inform develop¬
ers what types of investment activity city government will support within a
given area of the metropolis. Public support for private investment assumes
various guises, such as financial incentives, the placement of capital improve¬
ments, zoning decisions, and the like. Regulations and incentives such as
these help determine whether private investment will be forthcoming, where
it will occur, and what form it will assume. Property zoned exclusively for
manufacturing can be redeveloped for commercial or residential purposes
only if city officials agree to furnish the requisite zoning variances. Industrial
loft; buildings may become viable for higher-rent alternative uses only to the
extent that public subsidies are available for building conversions, to name
just two examples.
26 Remaking Chicago

Local rules and incentives governing private investment do not simply re¬
spond to structural economic changes in the marketplace. Rather, they are
actively shaped by the composition of a city’s governing coalition. As the fol¬
lowing section will show, postwar Chicago was governed by a progrowth
regime consisting of a well-insulated, exclusive alliance between downtown
business leaders and city officials. The key objective of the alliance was the re¬
vitalization of the central business district, which planners believed could not
occur without a fundamental restructuring of near-downtown land use. Ac¬
cording to planning officials, centrally located industrial districts were a prin¬
cipal cause of an unhealthy relationship that existed between the Loop and
surrounding areas, threatening property values and investment activity in the
central business district. Downtown redevelopment ultimately required that
industry and lower-income residents be relocated outside the central area so
that near-downtown property could be reused for new commercial and
middle- to upper-income housing developments that would reinforce the
growth of corporate and financial activities in the Loop.
Whether Chicago’s central area remained a viable location for manufactur¬
ing following World War II is a question we will take up in the following
chapter. For the time being, however, it is enough to show that public policy
was not an intermediate or otherwise derivative force in the redevelopment
of the city’s downtown area during the postwar period. To the contrary,
land-use plans, capital improvements, financial incentives, and zoning- and
building-code regulations were all used aggressively to impose a particular
trajectory of redevelopment on Chicago’s central area. A governing coalition
dominated by downtown corporations, financial institutions, retailers, prop¬
erty developers, and other land-based interests produced a redevelopment
strategy that responded overwhelmingly to the narrow interests of its own
membership. The following section examines the nature of this coalition and
the political and economic strategies that sustained it for nearly thirty years.

Central Area Planning Under the Richard J. Daley Administration


When Richard J. Daley began his first term as mayor of Chicago in 1955,
the city’s manufacturing and service sectors were both in decline. Following a
wartime boom in manufacturing employment, there were already signs that
industry was being lost to the suburbs. Between 1947 and 1954, the city lost
53,209 manufacturing jobs, while manufacturing employment in the rest of
Cook County rose by over 30,000 jobs (McDonald 1984, 10). The central
business district, meanwhile, was suffering falling property values and declin¬
ing retail sales. Between 1930 and 1955, construction of new office buildings
in the Loop came to a virtual standstill (Berry et al. 1976, 76). Stagnation of
downtown employment, white flight to the suburbs, and the growing physical
decay of the residential communities surrounding the Loop all contributed to
Downtown Redevelopment and the Chicago Machine 27

a decrease in downtown pedestrian traffic following World War II. Retail sales
in the central business district fell steadily from 1948 to 1963 (Berry et al.
1976, 73). In 1962, the net profits of the five main Loop department stores
stood at only 30 percent of their 1948 levels (Berry et al. 1976, 75).
Responding to these dire economic circumstances, Daley quickly set out
to forge an alliance with key segments of the downtown business community.
Business leadership had been newly consolidated within the Chicago Central
Area Committee (CCAC), established in 1956 to provide the business com¬
munity with a collective voice to address the problems of the downtown area.
Formed by Holman Pettibone, chairman of the Chicago Title and Trust
Company, the CCAC included representatives from the city’s leading corpo¬
rate, real estate, and financial sectors.2
Business leaders initially kept the new mayor at arm’s length, and under¬
standably so. Daley hailed from the working-class neighborhood of Bridge¬
port on Chicago’s South Side, a world miles apart from the exclusive North
Shore communities where many of the city’s top business executives resided.
He was a loyal machine politician, having risen dutifully through the ranks of
the Democratic party from precinct captain to chairman of the Cook County
Democratic Central Committee, the party’s top position. Such accomplish¬
ments meant nothing to business leaders. In their eyes Daley was a ward
heeler, whose tenure as mayor would likely be marked by the same corrup¬
tion, fiscal irresponsibility, and lack of vision for which previous machine
mayors had won well-deserved reputations (Greer 1983, 121).
However, Daley soon proved he was no party hack. He filled cabinet posi¬
tions and department directorships with respected professionals lacking polit¬
ical ties to the machine, prompting the New York Times to praise the new
mayor as a “reformer at heart” (Biles 1995, 45). Daley quickly made down¬
town revitalization the economic development priority of his administration,
establishing a new Department of City Planning in 1957 to coordinate and
oversee his new building program. By 1964, the department’s budget had
ballooned from $149,500 to $914,500, while staffing increased from twenty-
four to eighty-four employees (Biles 1995, 47). Several new downtown office
buildings, the first to be constructed in over twenty years, were completed
barely two years into Daley’s fledgling mayoralty, followed soon by many
others. These actions rapidly earned Daley the loyalty and respect of the
downtown business elite, allowing him to cement a progrowth alliance. By
paying top wages to city workers and providing thousands of jobs for the
construction trades, Daley also incorporated organized labor as a junior part¬
ner in the coalition. As a result, Chicago experienced little of the labor unrest
that crippled many other cities during the 1960s and 1970s.
Mayor Daley accomplished his planning and development objectives as
successfully as he did due in large part to his skillful use of the machine. By
holding the offices of mayor of Chicago and chairman of the Cook County
28 Remaking Chicago

Democratic Central Committee simultaneously, Daley amassed an unprece¬


dented degree of power, quickly whipping an undisciplined city council into
line.3 As mayor and party chair, Daley himself controlled roughly thirty
thousand patronage jobs and had a key voice in the slating of Democratic
candidates for office (Rakove 1975, 112). Rebellious council members lost
patronage employees and were not slated for reelection by the party. With
the city’s formal and informal political authority concentrated in the
Mayor’s Office, favored redevelopment projects sailed through the city bu¬
reaucracy. In some cases, a brief visit by a developer to the mayor accom¬
plished what under previous administrations would have required months or
even years of painstaking navigation through various layers of city govern¬
ment (O’Connor 1975, 137). Under Daley’s leadership, Chicago earned its
reputation as “the city that works.”
More problematic was Daley’s electoral coalition, consisting of middle-
class white ethnic voters from the city’s Northwest and Southwest Sides and
lower-income black voters from the South and West Sides. Throughout Da¬
ley’s twenty-one-year reign as mayor of Chicago, two key sources of tension
threatening the integrity of this coalition had to be carefully held in check.
The first was caused by the administration’s overwhelming policy emphasis
on redevelopment activities in the central business district. In general, the
residents of Chicago’s lower-income and working-class neighborhoods stood
to gain little from Mayor Daley’s downtown redevelopment efforts. Some, in
fact, had much to lose, particularly those located near the downtown area,
since the city’s plans to surround the Loop with a ring of middle- and upper-
income neighborhoods could not be realized until the homes and industries
occupying these areas were removed.
The second cause of tension stemmed from divisions between the black
and white factions of the machine’s electoral coalition itself. Following World
War II, much of Chicago’s rapidly growing African-American population was
squeezed into a six-square-mile Black Belt on the city’s Near South Side.
Overcrowding caused living conditions in the black ghetto to deteriorate
rapidly during the 1950s and 1960s, creating tremendous pressure for access
by African Americans to housing markets in the surrounding white ethnic
communities (Kleppner 1985, 40-41). Whites, however, motivated by racism
and fear of declining property values, resisted efforts on the part of blacks to
purchase or rent housing in their neighborhoods. School integration, another
point of contention by the early 1960s, was also opposed by many whites
(Biles 1995, 97-101). As racial tensions continued to mount during the
1960s, it became clear that the machine’s biracial coalition could be pre¬
served only to the extent that divisive racial issues could be kept out of the
municipal policy arena (Kleppner 1985, 71).
In managing these two potential threats to his electoral coalition, Mayor
Daley’s skillful handling of the machine once again served him well. Soon af-
Downtown Redevelopment and the Chicago Machine 29

ter taking office, Daley undertook significant steps to increase the ranks of his
patronage army of city employees, hiring an additional 2,500 policemen, 800
firemen, and 500 sanitation workers (Biles 1995, 46). By emphasizing pa¬
tronage jobs, improved city services, and other discretionary favors and re¬
wards in his dealings with the neighborhoods, Daley discouraged group mo¬
bilization around collective demands. Electoral politics became a politics of
the individual, structured around patronage and selective incentives (Ferman
1991). Low-income blacks, with few opportunities for material gain outside
the machine, became Daley’s strongest supporters even though he steadfastly
ignored calls for racial fairness (Kleppner 1985, 71). Likewise, voters from
both white ethnic and African-American neighborhoods were strong machine
supporters despite Daley’s one-dimensional policy emphasis on downtown
redevelopment (Ferman 1991).

Redevelopment Phase One: The 1958 Development Plan


Mayor Daley’s downtown redevelopment goals were set in motion with
the help of several key land-use plans for the central area. The first was the
Development Plan for the Central Area of Chicago issued by the Department
of City Planning in 1958, which proposed to anchor the Loop’s corporate
and retail sectors with a series of public and private improvements in the
downtown and greater downtown areas of the city (Department of City
Planning 1958). Under the plan, 130 acres of underutilized railyards south
of the Loop were to be cleared to make room for a new University of Illinois
campus on the Near South Side. New government buildings were to be con¬
structed on both the north and south ends of the Loop as additional barriers
to surrounding blight. To improve access to the downtown area, the plan
contained proposals for a network of expressways linking the central business
district with outlying portions of the metropolitan area, along with new park¬
ing facilities on the periphery of the Loop. Finally, planners recommended
construction of middle-income housing for up to 50,000 families in areas im¬
mediately north and south of the Loop. Housing was to be privately devel¬
oped on land assembled and cleared by the city under powers of the 1949
U.S. Housing and Redevelopment Act.
There was little mention of central area manufacturers in the 1958 Devel¬
opment Plan. Several square miles west of the central business district from
Halsted Street to Ashland Avenue were set aside under the plan for light in¬
dustrial and commercial uses. However, virtually all central area industrial
land south of the Loop, including Printing House Row, was designated for
“residential re-use” and the future University of Illinois campus. Likewise,
the plan’s long-range projections for extensive residential development north
of the Chicago River did not bode well for the occupants of industrial loft
buildings concentrated in the River North area.
30 Remaking Chicago

The postindustrial thrust of the 1958 Development Plan comes as no real


surprise, given the close working relationship that was evolving between the
downtown business community (as represented by the CCAC) and Mayor
Daley’s planning and development staff.4 A series of reports issued by the
Mayor’s Council of Manpower and Economic Advisors and the Department
of Development and Planning during the 1960s and 1970s contain a similar
theme, remarkably consistent with the economic development preferences of
downtown business leaders (see Department of Development and Planning
1967; Mayor’s Council of Manpower and Economic Advisors 1974, 1977;
Greer 1983, 106). According to these reports, national economic trends
were encouraging the flight of industry from cities like Chicago. The decline
of manufacturing was inevitable, but Chicago’s economic vitality could be re¬
stored if appropriate steps were taken to encourage the growth of high-level
professional, technical, and financial services. For city planning officials and
their allies in the downtown business community, citywide economic devel¬
opment would, in future, necessarily emanate more exhaustively from the
city’s downtown commercial core. Basic economic trends made it increas¬
ingly unnecessary and futile for city government to make painful choices be¬
tween support for industrial or commercial growth. Manufacturing was dying
or dead; economic development would have to become increasingly synony¬
mous with downtown commercial growth.
The coalition forged between the Daley administration and the CCAC was
largely successful in reversing the long decline in downtown real estate activ¬
ity. Between 1962 and 1977, 32 million square feet of new office space were
added to the central business district, servicing a growing demand for cen¬
trally located corporate administrative facilities and professional service firms
(Weiss and Metzgar 1989, 135). A new University of Illinois campus was
built just west of the site proposed in the 1958 Development Plan, despite
protests by Near West Side residents and business owners located in the con¬
struction path (Rosen 1980). By 1964, in addition, the first major effort to
open up the downtown area for large-scale middle- and upper-income resi¬
dential development was in place. Marina City’s two 60-story towers along
the north bank of the Chicago River contained 896 residential units, along
with shopping facilities, restaurants, and a movie theater.

Redevelopment Phase Two: The Chicago 21 Plan


Despite the magnitude and expediency of the Daley administration’s early
redevelopment efforts, by the late 1960s it was becoming painfully clear to
members of Chicago’s growth coalition that their long-term revitalization
plans for the central area were being threatened by a number of unantici¬
pated new developments. For one thing, demolition and rising real estate as¬
sessments were beginning to put pressure on the small retail shops and places
Downtown Redevelopment and the Chicago Machine 31

of entertainment that helped keep downtown streets alive, particularly after


business hours. Between 1968 and 1972, the central business district lost 13
percent of its small retailers, 15 percent of its entertainment spots, and 8 per¬
cent of its restaurants (Berry et al. 1976, 75). For another, the rapidly chang¬
ing demographic balance in the Chicago metropolitan area—suburbanization
of whites accompanied by a growing inner-city African-American
population—was leading to an increase in the ratio of black to white pedes¬
trian traffic downtown. Business leaders typically assumed there was a “tip¬
ping point,” well below 50 percent black, after which whites would begin
staying away altogether (Gapp 1973). As one prominent real estate developer
put it at the time, “I’ll tell you what’s wrong with the Loop. It’s people’s
conception of it. And the conception they have about it is one word: Black.
B-L-A-C-K. Black” (Greer 1983, 140).
These concerns ultimately prompted a sequence of closed monthly meet¬
ings between Daley administration planning officials and key representatives
of the CCAC (Greer 1983, 133). Informing these discussions were a series of
commercial, residential, and industrial land-use studies for the central area
undertaken by two private consulting firms, the Real Estate Research Corpo¬
ration (RERC) and Barton-Aschman Associates, under contract with the City
of Chicago (Real Estate Research Corporation 1970; Barton-Aschman Asso¬
ciates 1^70). According to these reports, the current problems of the down¬
town area stemmed largely from the “unsound” relationship that existed be¬
tween the central business district and its immediate surroundings. Land
given over to low-income housing and industrial uses in neighborhoods bor¬
dering the Loop did nothing to enhance the development of corporate and
retail activity in the central business district and in some cases actively under¬
mined it. The viability of the downtown commercial core could be best pro¬
tected by surrounding the area with middle- and upper-income residential
communities, providing downtown corporations with a nearby pool of work¬
ers to draw from and downtown retail establishments, restaurants, and places
of entertainment with ready markets. This had been an important, though
largely unrealized, goal of the 1958 Development Plan. It was now time to
reposition it to the center of the city’s planning agenda for the central area.5
Fortunately, according to the RERC and Barton-Aschman plans, the costs
associated with such a transformation would be minimal, because land use in
much of the greater downtown area was already in flux. Industry, in particu¬
lar, was said to be in decline due to economic forces largely “exogenous to
the central communities and the city of Chicago” (Real Estate Research Cor¬
poration 1970, 14). Echoing similar arguments by the Mayor’s Council of
Manpower and Economic Advisors and the city’s Department of Develop¬
ment and Planning, RERC warned officials against “futile and wasteful ef¬
forts” to retain industries in the central communities. Policies should be de¬
signed instead to “prepare the city of Chicago and the central communities
32 Remaking Chicago

for their eventual relocation” (Real Estate Research Corporation 1970, 21).
The monthly meetings between administration planning officials and
CCAC representatives continued for several years, culminating in the 1973
publication of Chicago 21: A Plan for the Central Area Communities.
Though published by the CCAC itself, the plan credited twelve City of
Chicago agencies with consulting roles and opened with an introduction by
Mayor Daley, symbolizing the well-integrated public-private partnership that
the Daley administration had fostered around downtown redevelopment.
The findings and recommendations of the Chicago 21 Plan were generally
consistent with the views expressed in the Barton-Aschman and RERC land-
use studies: the upgrading of the central business district was to be the linch¬
pin of citywide economic development, and the physical deterioration and
“economic obsolescence” of industrial and residential areas surrounding the
Loop were the major obstacles to be overcome. Once again, emphasis was
placed on calls for a ring of stable residential communities to surround the
city’s downtown business core.
The centerpiece of the Chicago 21 Plan was a proposal for a “new town
in town” called Dearborn Park that would eventually house up to 120,000
middle-income residents on a 600-acre site south of the central business dis¬
trict. The southern boundary of the Loop had long been a sore spot for city
planners and business leaders alike, due mainly to its proximity to Chicago’s
Near South Side black ghetto. The 1958 Development Plan originally pro¬
posed to anchor this site with the city’s new University of Illinois campus,
but difficulties with land assembly prompted planning officials to choose a
Near West Side location for the university instead, leaving the southern end
of the Loop unprotected. The new South Loop residential development
would, it was hoped, serve two purposes by creating an effective barrier be¬
tween the central business district and low-income communities on the Near
South Side and spurring further transformation of near-downtown areas to
middle- and upper-income residential communities (Emmons 1977; Suttles
1990, 180-186).
Construction of the first phase of the new Dearborn Park housing devel¬
opment began in 1978. Land assembly was financed and promoted by
Chicago 21, a limited dividend corporation whose board members included
some of the city’s most prominent business leaders.6 Investor profits were
capped at a 6.5 percent annual return to keep the cost of Dearborn Park
homes within reach of the middle-income families seen as the backbone of
the new community (Emmons 1977, 15). In spite of the low returns, how¬
ever, the Chicago 21 Corporation had no trouble lining up investors within
the downtown business community. The reason for this, according to a study
of the development undertaken by the Citizens Information Service of Illi¬
nois, is straightforward:
Downtown Redevelopment and the Chicago Machine 33

The appeal is not to profits as such but institutional self-interest. For large firms
headquartered in the Loop with substantial investments in land and buildings,
[Dearborn Park] offers a source of nearby housing for its employees, retail buy¬
ers for its stores, and frequenters for its restaurants and theaters. It assures the
presence of whites at a time when blacks are increasingly users of the Loop. It
offers a stimulus to redevelopment of the Loop’s deteriorated southern fringe.
And it provides what the developers term “a more secure environment to all of
the Loop.” (Emmons 1977, 16)

Although the size of the project was eventually pared down considerably,
Dearborn Park was probably the strongest affirmation yet of the viability of
the downtown area for middle-income residential development. Within one
year after the first phase of the project was completed in 1979, 128 of the de¬
velopment’s 144 town houses were occupied (Enstad 1980). By 1986, the
complex had 1,500 housing units, with few vacancies, and plans to launch a
second phase to include another 1,300 to 1,700 units immediately south of
the existing development site were underway (Suttles 1990, 175).
Following closely on the heels of Dearborn Park were several additional
large-scale downtown residential developments. River City, located several
blocks west of Dearborn Park along the Chicago River, added 2,500 more
dwelling units to the South Loop. In the West Loop, the construction of
Presidential Towers on two square blocks assembled and cleared by the
Chicago Land Clearance Commission created 2,460 additional units of mid¬
dle- and upper-income living space, ready for occupancy by 1985. Conver¬
sion of the industrial loft space south and west of the Loop to residential and
commercial uses closely tracks these three residential developments. By one
estimate, 99 percent of South Loop loft dwellers moved in only after Dear¬
born Park broke ground (Suttles 1990, 175). Of the twenty-six loft conver¬
sions undertaken in the West Loop between 1979 and 1990, only two were
carried out before groundbreaking for Presidential Towers began in August
1983 (Ludgin and Masotti 1985, 1986; Ludgin 1989).
Much of this development took place after Daley’s death in 1976, but it
was the framework created by his administration’s planning efforts that ulti¬
mately made it possible. City planning under the Daley administration was
focused overwhelmingly around one goal—economic growth. It was eco¬
nomic growth that cemented the alliance between the administration and the
CCAC, making it unsurprising that administration officials came to share the
view of business elites that the downtown commercial center should serve as
the primary engine of citywide economic development. Once it became clear
to Daley’s planning staff that the vitality of the downtown business core re¬
quired middle- and upper-income families and office workers to occupy areas
being utilized at the time by small- and medium-sized industries, they were
34 Remaking Chicago

unwavering in their support for commercial and residential redevelopment.


Indeed, they saw little conflict in this. For them, the choice was not between
two viable trajectories of growth but between growth and stagnation.

Capital Improvements and Business Subsidies


Chicago’s postwar downtown revitalization plans were put into practice in
large part through a combination of carefully placed capital improvements
and business subsidies designed to facilitate private commercial and residen¬
tial redevelopment activity in the downtown area. Capital improvements are
public investments in the basic infrastructure of a municipality for projects
such as street and expressway construction and repairs, bridge and viaduct
improvements, mass transit, and water and sewer systems. They also include
public buildings like hospitals, universities, and courthouses. Such projects
can leverage private investment by increasing the value of nearby land parcels,
making such properties more attractive to developers. The reverse, however,
also holds—areas suffering from public disinvestment are likely to experience
private disinvestment as well.7
The Daley administration’s early planning efforts were organized around
an extensive set of public improvements designed to help rekindle private in¬
vestment activity in the city’s downtown commercial core. In the administra¬
tion’s five-year capital improvements budget for 1959-1963, 20 percent of
citywide capital investments were targeted for a handful of projects identified
in the 1958 Development Plan for the Central Area of Chicago (Department of
City Planning 1959). These included the completion of a network of express¬
ways linking the central business district with outlying suburban communities
and the national interstate highway system ($148 million); the construction of
a civil courts building and a new city-county office building in the Loop ($65
million); a new exposition center, McCormick Place, along the lakeffont just
south of the central business district ($33 million); land acquisition and con¬
struction of the initial stage of the Chicago campus of the University of Illi¬
nois ($54 million); and the clearance of 744 acres of land designated as
“blighted” by the Chicago Land Clearance Commission ($79 million).8
The downtown emphasis of public capital investments continued for the
duration of Daley’s tenure as mayor of Chicago, culminating in the city’s
1977-1981 capital budget, which committed nearly $1.1 billion to the
downtown area alone for such projects as the new State Street mall, the con¬
version of Navy Pier to a new recreational and shopping area, and the up¬
grading of the city’s lakefront museums (Department of Planning 1978).9
The budget designated the new Dearborn Park residential development in
the South Loop as one of thirty-nine “community improvement project ar¬
eas” to receive an average of $2.3 million each for street and sidewalk repairs,
landscaping, and other improvements. Between 1979 and 1990, public capi-
Downtown Redevelopment and the Chicago Machine 35

tal investments in Dearborn Park and Printing House Row combined would
ultimately rise to nearly $50 million, leveraging some $700 million in private
investment funds (Ludgin 1989, 17).
Aside from their beneficial effects on downtown development, capital im¬
provements also have a major impact on the viability of central city industrial
districts, where manufacturers are dependent upon city government to pro¬
vide, to maintain, and to upgrade basic infrastructure. For example, low over¬
passes and viaducts may, in some instances, force large trucks to take time-
consuming detours in order to reach local industries, delaying shipments and
increasing freight transportation costs. Restrictive load limits for bridges and
streets present similar difficulties. Narrow inner-city streets create loading and
unloading problems, while traffic congestion increases delivery times. In
some cases, sewer and water lines may be in need of repair. The extent to
which a municipality undertakes basic infrastructure improvements in indus¬
trial areas of the city can thus have an important effect on the prospects for
manufacturing in such locations.
Studies of the locational patterns of Chicago industrial firms carried out by
the Department of City Planning during the early years of the Daley adminis¬
tration show that, already by the late 1950s, the central area was losing man¬
ufacturing firms to outlying areas of the city and suburban communities (De¬
partment of City Planning 1960, 1961). According to these studies, firms
leaving the central area in favor of other city locations were congregating
along two industrial corridors extending northwest and southwest from the
downtown area. City officials blamed economic factors such as high central
area land prices and the obsolescence of multistory production facilities for
these locational shifts. However, an examination of the Daley administra¬
tion’s capital budget for the late 1950s and early 1960s suggests that the City
of Chicago was an active participant in the process.
Table 1 shows the administration’s 1959-1963 budget for bridges and
viaducts, street improvements, and sewers in three areas of the city: the cen¬
tral area (excluding the Loop), the northwest and southwest industrial corri¬
dors combined, and the Loop.10 These are the kinds of public investments
central city manufacturers required if they were to remain competitive with
firms located in modern suburban facilities. Table 1 shows that the central
area was highly underrepresented in two of the three categories of basic infra¬
structure projects—street improvements and sewer construction and repair—
while the northwest and southwest corridors were overrepresented. Although
comprising just under 30 percent of the total geographical area covered by
the three industrial districts, the central area received no funding whatsoever
for sewer improvements, 8 percent of funds allocated for street repairs, and
only 16 percent of total investment dollars. Given this dearth of public in¬
vestments for industrial infrastructure in the near-downtown area, it is not
surprising that manufacturers began to seek new production space elsewhere.
36 Remaking Chicago

By the 1970s, public capital improvements in the central area rose measur¬
ably, but by this time such projects were increasingly designed to leverage
private commercial and residential investment rather than industrial develop¬
ment in this part of the city.11 As Table 2 indicates, capital-investment priori¬
ties shifted in favor of both the Loop and non-Loop portions of the central
area by the late 1970s. Meanwhile, public spending for industrial infrastruc¬
ture in the northwest and southwest industrial corridors, still used predomi¬
nantly for manufacturing by then, fell by substantial amounts during the
1960s and 1970s. Ultimately, the pattern from the late 1950s onward be¬
came one of consistent neglect in the funding of capital improvements for in¬
dustrial areas, while portions of the city perceived to be hubs of commercial
and upscale residential development benefitted from heavy concentrations of
public funds.

Table 1. Budgeted Capital Improvements


for Basic Infrastructure in Three Locations, 1959-1963

Percent of Bridges and


District Total Area Viaducts Streets Sewers Total

Central Area 27 $14,300 $730 $— $15,030


(excluding
Loop)

NWandSW 69 $35,660 $8,490 $32,665 $76,815


Industrial
Corridors

Loop 4 $5,700 $100 $1,020 $6,820

Total 100 $55,660 $9,320 $33,685 $98,665

Source: Department of City Planning (1959). Note: Dollar amounts are in thousands.

Urban Development Action Grants


In addition to the private investment dollars leveraged through its capital
improvements programs, the City of Chicago also influenced the trajectory
of downtown redevelopment through several federally funded grant and loan
programs. Beginning in 1978, the U.S. Department of Housing and Urban
Development (HUD) initiated the Urban Development Action Grant
(UDAG) program to help stimulate economic development in urban areas
experiencing severe economic distress.12 Private investors participating in the
Downtown Redevelopment and the Chicago Machine 37

program received at least $1.00 in UDAG funds for every $2.50 in private
investment committed to a proposed new development. City government
played a key role in determining which projects received UDAG funding,
since program eligibility requirements stipulated that private investors per¬
suade city officials to sponsor their projects and submit applications to HUD
on their behalf. Investors were also required to demonstrate that market
forces alone were insufficient to make their projects workable.
Between 1979 and 1986, the City of Chicago was awarded fifty-eight
UDAGs worth a total of $98 million. An analysis of the UDAG program car¬
ried out by the University of Illinois at Chicago reveals that a large share of
UDAG funds were concentrated in Chicago’s central area, many for the con¬
version of industrial loft buildings to various commercial and residential uses
(Wright 1987). According to the study, 49 percent of Chicago development

Table 2. Budgeted Capital Improvements


for Basic Infrastructure in Three Locations, 1977-1981

Percent of Bridges and


District Total Area Viaducts Streets Sewers Total

Central Area 27 $34,000 $2,554 $4,606 $41,160


(excluding
Loop)

NWandSW 69 $10,895 $5,024 $12,563 $28,482


Industrial
Corridors

Loop 4 $81,819 $57,039 $440 $139,298

Total 100 $126,714 $64,617 $17,609 $208,940

Source: Department of Planning (1978). Note: Dollar amounts are in thousands.

projects relying on UDAG support from 1979 to 1986 were located in the
Loop and surrounding area. In the South Loop, three of Printing House
Row’s twelve industrial loft buildings were converted to residential develop¬
ments with the assistance of $6.8 million in UDAG funds. The nearby Dear¬
born Park and River City residential developments received another $5.1 mil¬
lion through the UDAG program (Wright 1987, 10). On the Near West
Side, seven of nine UDAG-sponsored projects involved the conversion of in¬
dustrial buildings to commercial lofts, with $4.2 million in UDAG funds
leveraging a total of $43 million worth of private investment (Wright 1987,
38 Remaking Chicago

15). According to the study’s principal author, Patricia Wright, the absence
of a coherent city policy for the sponsorship of UDAG-funded economic de¬
velopment projects enabled property developers to dominate the program in
Chicago, causing funds to be concentrated in commercial and residential re¬
development projects in the central area (Wright 1987, 31).

Industrial Revenue Bonds


The preoccupation of city officials with the promotion of commercial and
residential growth in the downtown area of the city did not cause them to
neglect manufacturing entirely. Beginning in 1977 under the administration
of Richard J. Daley’s successor Michael Bilandic, the Department of Eco¬
nomic Development began offering low-interest Industrial Revenue Bonds
(IRBs) to assist companies seeking to relocate, to expand, or to improve their
facilities within the city of Chicago. From 1977 to 1984, when changes in
federal tax laws led to major cutbacks in the program, 102 Chicago firms re¬
ceived a total of $187 million in loans through the IRB program (Clemons,
Giloth, and Tostado 1985, 17). Due to IRB guidelines that weighted the
program in favor of manufacturers, 74 percent of IRB recipients were manu¬
facturing firms (Clemons, Giloth, and Tostado 1985, 2).
The IRB program provided badly needed financial support for certain
Chicago manufacturers, yet it also played a significant role in the deindustri¬
alization of the central area. Program specifications placed a $500,000 mini¬
mum restriction on bond amounts, skewing the distribution of IRBs toward
larger firms that could afford the bonds.13 Smaller manufacturers, meanwhile,
were denied access to credit. Sectors dominated by small- to medium-sized
firms such as printing, apparel, and fabricated metals that made up the bulk
of central area manufacturing establishments were underrepresented in the
program (Clemons, Giloth, and Tostado 1985, 15). Although central area
firms did receive a substantial share of total IRB dollars, bond recipients there
were primarily larger firms from sectors such as primary metals and chemical
and allied products with few strong ties to the near-downtown area. Not sur¬
prisingly, many such firms used the bonds to relocate elsewhere in the city,
freeing up land for new commercial and residential development. Intention¬
ally or not, IRBs thus became another tool in the city’s effort to remove in¬
dustry from the central area.

Zoning for Commercial and Residential Redevelopment


Land-use plans, capital investments, and business subsidies all had an enor¬
mous impact on the trajectory of postwar economic development in
Chicago’s central area. However, the remaking of downtown Chicago could
not have been accomplished without appropriate changes to the city zoning
Downtown Redevelopment and the Chicago Machine 39

ordinance. Zoning laws give development plans their teeth, setting detailed
guidelines for the types of land use permissible in a given area of the city. The
Chicago Zoning Ordinance maps the city into separate districts for residen¬
tial, commercial, and industrial uses, although the separation of categories is,
in practice, rarely so neat. In theory, Chicago’s zoning laws are intended
both “to protect residential, business, and manufacturing areas alike from
harmful encroachment by incompatible uses” and to “conserve and enhance”
the value of private property, goals not always easily reconcilable (Committee
on Buildings and Zoning 1955a, 3-4). When the two principles collide, the
final interpretation of the city zoning ordinance is as likely as not to become
an exercise in local power politics.
Chicago’s first zoning ordinance was passed by city council on April 5,
1923. Since then, the council’s Committee on Buildings and Zoning has
produced two comprehensive amendments to the ordinance, first in 1942
and once again in 1957, two years after Richard J. Daley was elected mayor.
The 1957 Comprehensive Amendment was inspired by many of the same
fears and concerns that prompted the 1958 Development Finn for the Central
Area of Chicago, namely, stagnation and falling property values in the Loop
and growing blight in the surrounding communities.14 Both the tone and
provisions of the amendment closely anticipate the postindustrial thrust of
the 1958 Development Plan. For the central area, zoning was to be a tool to
regulate and encourage the construction of downtown office space (Meltzer
1963, 1). An interim amendment to the zoning ordinance adopted in 1962
extended this emphasis to residential development as well (Meltzer 1963, 3).
Central area manufacturers, by contrast, received little in the way of zon¬
ing protections or incentives under the new amendment because their exodus
from the downtown area was viewed as a foregone conclusion. According to
a 1952 study of industrial land use prepared by the Chicago Plan Commis¬
sion in anticipation of the 1957 rezoning, “such features as obsolete build¬
ings, inadequate space, congested streets, and surrounding blight have placed
the Central Area in a poor position to hold existing industries or attract new
ones” (Chicago Plan Commission 1952, 2). The report’s principal recom¬
mendation was the rezoning for industrial purposes of 1,300 acres of largely
vacant land in outlying areas of the city to foster long-term industrial devel¬
opment in portions of the city deemed more suitable for such activities.
Ultimately, the 1957 Comprehensive Amendment designated roughly half
of the central area property devoted to manufacturing as one of three sub-
types of commercial districts. Commercial districts are the most flexible of the
city’s four primary zoning classifications, allowing various types of office and
residential development, in addition to light manufacturing. The argument by
city planners was that land-use patterns were still unresolved in portions of the
central area. Flexible zoning in such areas would give “market forces” time to
develop before more definitive land-use controls were put into effect (Mayor’s
40 Remaking Chicago

Committee for Economic and Cultural Development 1966, 37-39). In prac¬


tice, this left many central area manufacturers largely unprotected from devel¬
opers hoping to profit from a change in land use, since many conversions
from industrial to both office and residential development could take place in
commercially zoned areas with no zoning changes whatsoever.
For firms engaged in heavy manufacturing, the 1957 rezoning had more
immediate ramifications. All such enterprises unlucky enough to find them¬
selves in areas rezoned as commercial or residential districts now became
“nonconforming uses” under the 1957 Comprehensive Amendment. Busi¬
ness owners were required to cease operation within a limited time period or
face condemnation of their property. This regulation had been included in
the previous comprehensive amendment as well, but generous grace periods
for phasing out nonconforming uses substantially weakened the provision’s
impact.15 By 1957, the city’s tolerance of “industrial nuisances” had waned
considerably. Under the new amendment, nonconforming manufacturing
uses in newly designated commercial or residential districts were given as lit¬
tle as four years to shut down (Committee on Buildings and Zoning 1955b,
48a-50a).16
The 1957 Comprehensive Amendment also introduced important changes
in the methodology used for evaluating the nuisance-creating potential posed
by a given manufacturing establishment. Previously, industries had been as¬
signed to one of two types of manufacturing districts, based exclusively on
the type of product produced. Light industrial activities such as printing and
woodworking were permitted uses in one such district, while noisier, dirtier
operations like chemical production were assigned to another. Under the new
amendment, the type of product produced became irrelevant. Instead, new
“scientific” performance standards were established using such criteria as
noise, vibrations, and emissions as a means of segregating industrial uses.
Three new types of manufacturing districts were established, the highest set
of standards reserved for producers located near the boundaries of residential
or business districts.17
Though hailed by city officials as a way of encouraging manufacturers to
“clean house,” manufacturers themselves protested the changes, arguing that
the proposed new controls would ultimately drive industry out of the city
(Babcock 1972, 100). These fears, as it turned out, were not unfounded.
Higher performance standards faced by industries located in close proximity
to commercially and residentially zoned property acted as a wedge for resi¬
dential and commercial incursion into manufacturing districts. In the central
area, as various borderline properties ultimately became unsuitable for manu¬
facturing under the new ordinance or went on the market for other reasons,
developers often found it relatively easy to acquire them, obtain a zoning
change from the city, and redevelop them for commercial or residential uses
(Giloth and Betancur 1988; Babcock and Larsen 1990). In each case, a new
Downtown Redevelopment and the Chicago Machine 41

set of manufacturers suddenly found themselves on the boundaries of newly


expanded commercial or residential zones, subject to—and quite possibly out
of compliance with—a more stringent set of industrial performance standards
than before. At that point, the process was set to begin all over again.
The new performance standards and reduced time periods for the elimina¬
tion of nonconforming uses might not have posed such a threat to central
area manufacturers if zoning changes were more difficult to procure. Such
changes, however, have long been commonplace. Although the last compre¬
hensive amendment to the Chicago Zoning Ordinance was drafted in 1957,
city council has approved literally hundreds of petitions for minor amend¬
ments to the ordinance since then. Indeed, some of the heaviest concentra¬
tions of industrial loft conversions during the 1970s and 1980s took place in
two areas. River North and the Westgate Mill area of the Near West Side,
which were still zoned exclusively for manufacturing under the 1957 Com¬
prehensive Amendment (Ludgin and Masotti 1985, 1986). By 1980, incre¬
mental changes to the zoning ordinance had removed the last restrictions
against commercial or residential use in both of these districts, along with
most of the remaining portions of the central area.
A petition for a change to the city zoning ordinance in a particular ward is
typically introduced by the alderman of the affected ward, often at the re¬
quest of a ward constituent. Following review and recommendations by the
city zofiing administrator and the Department of Planning and Develop¬
ment, the City Council Committee on Buildings and Zoning schedules pub¬
lic hearings on the proposed change. Final approval of the change requires a
favorable vote of at least two-thirds of city council (Committee on Buildings
and Zoning 1955b, 121a-123a).
Within Chicago city government, there is a longstanding tradition of def¬
erence to aldermen on matters affecting their individual wards. With respect
to zoning issues, this means that an alderman’s recommendation for a zoning
change in his or her own ward will rarely be challenged by the Mayor’s Office
or by fellow council members (Gardiner and Lyman 1978). This tradition
dates back at least to the years of the Daley machine. While Daley generally
centralized decision-making on policies affecting the city as a whole, he rec¬
ognized the importance of aldermanic discretion at the neighborhood level
for purposes of machine-building. Regulatory relief in such matters as zoning
decisions or building inspections, along with patronage appointments and
city service provision, were the currency of politics during the machine years.
Aldermen cemented party loyalties by selectively allocating such favors and
rewards among ward constituents (Rakove 1975). As long as a particular al¬
derman was loyal to the mayor and votes for the machine were being deliv¬
ered in the ward in question, Daley was unlikely to interfere in something as
minor as a zoning variance, particularly if it involved a change to a “higher
and better” use of land.
42 Remaking Chicago

In practice, the combination of aldermanic discretion and standards for


amending the city zoning ordinance, “so vague as to provide no guidance
to developers, the citizenry, or the decisionmakers,” left the direction of
central area zoning after 1957 largely in the hands of ward aldermen and
private developers (Philip Zeitlin Associates 1982, 94). Not until the mid-
1980s would remaining central area manufacturers achieve sufficient organi¬
zational capacity to make a difference in city zoning policy. Until then, the
steady, incremental conversion of central area manufacturing districts to
new commercially and residentially zoned areas made a mockery of at least
one of the city zoning ordinance’s primary stated objectives—to protect res¬
idents, businesses, and manufacturers alike from incursion by incompatible
land uses.

The Chicago Rehabilitation Code


Efforts to convert near-downtown industrial loft buildings to housing or
office space were given a significant boost in 1982, when the new Chicago
Rehabilitation Code was incorporated into the city’s building code. Before
1982, the City of Chicago did pot have a rehabilitation code. Specifications
for new construction, as well as the maintenance, repair, and rehabilitation of
existing buildings, were all contained in one municipal building code. Build¬
ing code regulations stem from concerns over fire protection, general safety,
and hygiene. To the extent that they impose higher standards—and thus
higher costs—for building rehabilitation, they act as a disincentive for the
conversion of industrial loft space to alternative uses. By all accounts, the re¬
habilitation provisions of Chicago’s building code were exceptionally high
before the passage of the city’s new rehabilitation code in 1982 (Metropoli¬
tan Housing and Planning Council 1980).
Ironically, the architect of Chicago’s demanding building rehabilitation
standards was none other than Richard J. Daley himself. Rigorous construc¬
tion standards for building rehabilitation were initiated under the Daley ad¬
ministration and upheld through the mayor’s largely successful efforts to in¬
stitutionalize the policymaking process for updating the building code.18 In
1964, Mayor Daley assembled a group of architects, engineers, building
managers, and labor officials, along with representatives from the Depart¬
ment of Buildings, the Fire Department, and other city agencies to serve on
his new Mayor’s Advisory Committee for revising the building code (Jones
1985, 27). The committee provided a forum for various interested parties to
reach negotiated solutions on code amendments. Its recommendations were
submitted to city council with the endorsement of the mayor, which, during
the Daley years, virtually ensured passage. Through his authority to appoint
and to dismiss committee members, Daley exercised considerable power over
the amendment process. At the same time, his lack of active participation al-
Downtown Redevelopment and the Chicago Machine 43

lowed him to remain aloof from the politically controversial decisions taking
place in the committee (Jones 1985, 34).
To a large extent, rigorous construction standards for building rehabilita¬
tion remained a fixture of building-code amendments under the Daley ad¬
ministration because most of the architects and developers represented on the
Mayor’s Advisory Committee came from large firms involved with the design
and construction of new buildings downtown rather than from smaller firms
focusing on the rehabilitation of existing buildings.19 The large developers on
the committee were, for the most part, indifferent to the rehabilitation provi¬
sions of the code. By contrast, committee members from city agencies such as
the Fire Prevention Bureau, who saw the building code as an important vehi¬
cle for promoting fire safety and upgrading the city’s housing stock, consis¬
tently argued for tough rehabilitation standards. With little opposition voiced
by the private sector participants on the committee, revisions of the code
during the 1960s and 1970s consistently incorporated the city bureaucrats’
demands for stringent rehabilitation provisions (Jones 1985, 35-36).
During the Daley years, there was very little disruption to this exclusive
bargaining system for amending the building code. Daley’s control of pa¬
tronage and the slating of candidates for office guaranteed broad-based sup¬
port for his programs in city council. Disaffected rehabilitation contractors
excluded from the Mayor’s Advisory Committee could always take their
concerns directly to the council, but the strong aldermanic support that Da¬
ley commanded discouraged them from doing so. Moreover, throughout
most of Daley’s tenure as mayor, groups favoring more lenient rehabilitation
standards were largely disorganized and, for the most part, politically non¬
threatening.
Shortly before Daley’s death, however, coalition building among various
rehabilitation forces began to occur. In 1976, the Ad Floe Working Commit¬
tee for Residential Rehabilitation was formed through the joint efforts of the
Chicago chapter of the American Institute of Architects and the Chicago Re¬
hab Network, a community organization involved with the rehabilitation of
housing in low-income city neighborhoods (Metropolitan Housing and Plan¬
ning Council 1980, 2). The committee issued a report in 1978 criticizing the
rehabilitation provisions of the existing building code and calling for more le¬
nient construction standards for residential rehabilitation (Ad Hoc Working
Committee for Residential Rehabilitation 1978). During the next several
years, the committee continued to broaden its membership base, eventually
attracting such politically connected groups as the Committee for Legal
Lofts, a coalition of architects and developers who together controlled five
million square feet of central area loft space, including much of Printing
House Row (Washburn 1981).
By 1979, the voices favoring relaxed rehabilitation standards had grown
sufficiently loud that newly elected Mayor Jane Byrne was compelled to
44 Remaking Chicago

recognize them. Shortly after taking office in April 1979, Byrne assembled a
thirty-member committee chaired by former building commissioner Joseph
Fitzgerald to write what would become the city’s first rehabilitation code.
The committee submitted its final report to the mayor in October 1981.
With Byrne’s approval, the Chicago Rehabilitation Code was passed into law
by city council in March 1982.20
Lacking the authority and centralized power base commanded by Mayor
Daley, Byrne was unable to limit participation in the building-code revision
process to the exclusive group represented in the Mayor’s Advisory Committee
for amending the code (Jones 1985, 39). The new rehabilitation code was a
clear-cut victory for the one-time outsider coalition of rehabilitation groups
and a major defeat for city bureaucrats concerned with fire protection standards
and the upgrading of the city’s housing stock. Previously, any developer wish¬
ing to convert a building to an alternative use had been obliged to bring the
entire building up to modern code standards. Under the new code, only the
remodeled portions of such projects had to meet modern construction stan¬
dards. Fire exit requirements were liberalized as well, and the previous building
height limitation of eighty feet for residential loft conversions was abolished.
Finally, the new code authorized the mayor to appoint a seven-member board
of appeals with the power to grant exceptions to code provisions. Significantly,
no board members were to be drawn from the city bureaucracy.
Along with city bureaucrats, the other major losers in the debate over the
new rehabilitation code were, of course, the small- to medium-sized manu¬
facturers occupying much of the central area loft space in which the real es¬
tate community had recently taken such an interest. These manufacturers had
been the unintended beneficiaries of city bureaucrats’ longstanding insistence
on stringent building rehabilitation standards. Strict code provisions had dis¬
couraged many industrial building owners from undertaking costly building
conversions to higher-rent office or residential uses. By defending building
safety standards, city bureaucrats were thus inadvertent participants in a
deeper struggle over the nature of downtown production space.
However, once the coalition of rehabilitation groups gained access to the
appropriate decision-making arena, forcing a public debate over the issue,
the preoccupation of city bureaucrats with building safety alone meant that
no one spoke publicly for the interests of the manufacturing community.
The industries themselves, at this point still politically unsophisticated and
largely disorganized, were hardly in a position to do so. Ultimately, the de¬
bate was cast as a dispute between meddlesome city bureaucrats and innova¬
tive developers seeking to bring life to a decaying downtown area.21 Not
only was this a confrontation city bureaucrats were unlikely to win, but the
impact that the new rehabilitation code was destined to have on the future
of downtown area industrial space went virtually unnoticed throughout the
entire debate.
Downtown Redevelopment and the Chicago Machine 45

Conclusion

Between 1955 and 1983, the central area of Chicago experienced an enor¬
mous economic upheaval. The mixture of industry, commerce, and low- to
moderate-income residential communities that dominated the area in the
early postwar years gave way to an entirely new agglomeration of corporate
offices, business services, and upscale residential developments. Downtown
area property values rose measurably, while centrally located industrial dis¬
tricts providing well-paying blue-collar jobs for city residents were dismantled
piece by piece, block by block. By the time Chicago’s last machine mayor,
Jane Byrne, was defeated by Harold Washington in 1983, Chicago had as¬
sumed a prominent position in the hierarchy of global cities, its downtown
the headquarters of twenty-six Fortune 500 companies and the location of
the world’s tallest building, the Sears Tower. In international banking, LaSalle
Street was surpassed within the U.S. only by Wall Street, and Chicago’s Board
of Trade and Mercantile Exchange gave the city a decided edge over New
York in commodities and futures trading (Squires et al. 1987, 38).
Urban political economists typically view the redevelopment of older in¬
dustrial cities such as Chicago following World War II as a response to struc¬
tural economic change. Downtown redevelopment and its counterpart, ur¬
ban industrial decline, are said to be linked to a particular phase of capitalist
development, featuring the hypermobility of capital, rapid technological
change, and the persistence of regional and global variations in land and la¬
bor costs. Efforts to remake central business districts arise in response to
structural economic shifts, taking place within the context of a division of la¬
bor between state and market under which private investors are afforded con¬
siderable autonomy. By most accounts, Chicago planning officials perceived a
structural change in the city’s economy from manufacturing to services and
responded in kind with policy changes necessary to support economic
growth on new terms.
Yet the narrative of downtown redevelopment presented in this chapter
calls into question perspectives that draw such fine distinctions between poli¬
tics and markets. The restructuring of Chicago’s central area following World
War II took place through the efforts of a progrowth alliance led by key seg¬
ments of the downtown business community and city planning officials. Al¬
though structural economic changes were beginning to have an impact by
the time this alliance was formed in the late 1950s, the evidence supporting a
fundamental manufacturing-services shift at this early stage was, at best, slim.
The more likely cause of the Daley administration’s decision to embark on a
downtown-oriented economic development strategy was ideological. City
planners ultimately adopted the outlook of their allies within the business
community, leading them to select a real estate-based growth strategy over
potential alternatives.
46 Remaking Chicago

The channeling of economic policy benefits into the hands of downtown


property developers and other business leaders had to be legitimized in the
eyes of neighborhood voters, who typically received few of the benefits of
downtown growth and, in some instances, bore a disproportionate share of
its costs. Part of the solution rested in machine politics. By structuring the
electoral arena around patronage and selective incentives, machine politi¬
cians discouraged their constituents from identifying collectively as low- and
moderate-income neighborhood residents threatened by economic develop¬
ment policies hostile to their interests. The other part rested in an ideology
of privatism, whereby city officials argued forcefully that public policies sup¬
porting downtown redevelopment were formulated in response to well-
defined patterns of private investment activity over which city government
exercised little to no control.
As durable as this set of arrangements would prove to be, it was nonethe¬
less crisis-prone. As we shall see in Chapter 4, growing tensions between the
machine’s governing and electoral coalitions and within the electoral coali¬
tion itself increasingly taxed the ability of machine leaders to hold the contra¬
dictions of growth politics in check. By the 1970s, near-downtown manufac¬
turers, workers, and residents of neighborhoods decimated by industrial
decline began to identify a common enemy in the progrowth alliance of city
planning officials and downtown business elites. Increasingly, the individual¬
istic politics of the machine gave way to coalition-building around a new set
of policies designed to support an economically viable alternative to down¬
town redevelopment. Within this changing political context, possibilities for
the construction of new regimes would begin to emerge.
Chapter Three

Rethinking Industrial Decline


The Chicago Printing and Apparel Industries

^^n the previous chapter, we saw how a progrowth alliance between city of¬
ficials and downtown business leaders encouraged downtown redevelopment
in postwar Chicago to follow a particular trajectory, one from which centrally
located manufacturers were conspicuously absent. As yet, however, little evi¬
dence has been offered to suggest that there was a viable alternative to this.
After all,'city officials and their allies in the downtown business community
justified the procorporate leanings of their redevelopment efforts by arguing
that national and international economic trends marked the death of manu¬
facturing in central cities. This argument rested on a seemingly firm empirical
foundation: manufacturing employment began declining in Chicago and
other major industrial cities shortly after the close of World War II. The pos¬
sibility that the policies and programs described in Chapter 2 were simply
nudging downtown area growth a little faster and more efficientiy down the
path it was destined to take cannot be dismissed out of hand.
This chapter asks two principal questions: First, to what extent did the
central area of Chicago remain a viable site for industrial development in the
years following World War II? Second, how specifically did public-policy deci¬
sions designed to foster commercial and residential redevelopment in the
downtown area affect the prospects of central area manufacturers? The expe¬
riences of two prominent central area industrial sectors, the printing and ap¬
parel industries, will be analyzed to address these questions. These two sec¬
tors have been selected for a number of reasons. To begin with, they have
long been among the top employers in the area. In 1962, central area print¬
ing establishments employed forty-eight thousand workers, over twice as
many employees as any other sector there, while the apparel industry ranked
third among central area employers with seventeen thousand workers (Real
Estate Research Corporation 1970, 4). Taken together, employment in these
48 Remaking Chicago

two sectors alone accounted for 39 percent of all central area manufacturing
employment in 1962.
In addition, these industries share certain features and exhibit certain pat¬
terns of behavior that distinguish them somewhat from many other types of
manufacturers. In general, both printers and apparel manufacturers have
shown more of a tendency to concentrate in the downtown and near¬
downtown areas of Chicago than industries from other sectors. As the fol¬
lowing pages will suggest, this concentration stems in part from the relatively
specialized products that the manufacturers of these two industries typically
produce. In many cases, products are tailored to the specifications of down¬
town customers, and the manufacture of one item may involve a number of
firms. Face-to-face consultation with downtown customers or subcontractors
may be necessary at certain stages of the production process. Partly for this
reason, producers in both sectors congregated at one time in readily identifi¬
able districts. Chicago’s printing and garment districts were, until quite re¬
cently, located within blocks of one another in the near-Loop area.
Put briefly, these two sectors appear to have had stronger ties to the cen¬
tral area of Chicago than any others, both before and after World War II. If
the central area ultimately became an unprofitable or otherwise undesirable
location for these producers independently of the downtown redevelopment
efforts discussed in the previous chapter, it is unlikely that the experiences of
manufacturers from other sectors would have been any different. Conversely,
to the extent that the printing and apparel industries appear to have been vi¬
able industrial sectors undercut by unrepresentative public policies, it be¬
comes possible to imagine a different scenario for downtown redevelopment
than the one presented in Chapter 2. The hypothesis that politics played a
decisive role in the deindustrialization of Chicago’s central area becomes
more plausible.

The Chicago Printing Industry


Chicago’s historical center of printing and publishing lies in an area of the
South Loop bordered by Congress Parkway and Polk Streets to the north
and south, and State and Clark Streets to the east and west (see Map 2). Be¬
tween 1883 and 1912, Chicago’s flourishing community of printing and
publishing firms commissioned the design and construction of twenty-one
loft structures and multifunction office buildings in the district, commonly
known as Printing House Row. Several features of this area made it particu¬
larly attractive to the graphic communications industry. The first was the lo¬
cation of the Dearborn Street Railway Station at the district’s southern
boundary, providing firms with a direct link to producers and consumers na¬
tionwide. The second was the area’s distinctive street layout, consisting of
lengthy, narrow blocks unique to the downtown area. The loft structures
Map 2. Chicago Downtown Industrial Districts
50 Remaking Chicago

built here were, by necessity, long and slablike, a feature that admitted a max¬
imum amount of light into buildings and allowed long lines of presses to be
arranged in an orderly fashion (Commission on Chicago Historical and Ar¬
chitectural Landmarks 1983).
Markets for Chicago’s early printers and publishers varied substantially
from firm to firm. Some companies benefitted from the nationwide growth in
mail-order merchandising around the turn of the century. To help gain access
to America’s potentially huge and largely untapped rural markets, downtown
retailers such as A. Montgomery Ward and Richard W. Sears hired Printing
House Row firms to produce voluminous catalogues with vivid illustrations
and detailed descriptions of store merchandise. Other Chicago printers spe¬
cialized in the printing and publishing of inexpensive books and periodicals for
mass consumption. Still others found work printing material for various social
and political causes, such as the city’s burgeoning labor movement (Commis¬
sion on Chicago Historical and Architectural Landmarks 1983).
In the decades following World War II, as advances in transportation and
communications encouraged certain Chicago manufacturers to relocate to
outlying portions of the metropolitan area, much of the printing industry
continued to exhibit a strong, locational preference for its original South
Loop home. By 1970, when the central area contained just 15 percent of the
Chicago metropolitan area’s total manufacturing employment, over 50 per¬
cent of the region’s printing and publishing jobs were still clustered in this
portion of the city (Real Estate Research Corporation 1970, 4). To a certain
extent, this can be explained by the interrelationships that began to develop
between the printing industry and the city’s expanding corporate-financial
sector following World War II. As the downtown areas of cities like New
York and Chicago became home to growing concentrations of national and
international corporate headquarters, law firms, advertisers, and financial and
cultural institutions, the demand for printed products such as advertising
brochures, lawyers’ briefs, and concert programs increased accordingly. Print¬
ers serving markets like these typically operate under stringent time restric¬
tions, malting a central location near important downtown customers highly
advantageous (Gustafson 1959, 163).
Printed goods also differ in important respects from the typical mass-
produced commodities being manufactured increasingly on the urban pe¬
riphery and beyond. Unlike standardized goods such as ball bearings or radio
transistors, which manufacturers can generally produce in large quantities and
then stockpile, most printed products are uniquely tailored to the detailed
specifications of individual customers. This tends to create an ongoing need
for consultation between printers and their customers, particularly at the pre¬
press stage of production. Layout, typesetting, and preparatory work for
graphics and illustrations must typically be approved by the customer before
a print job goes to press. The close and continual contact between printers
Rethinking Industrial Decline 51

and their customers required under these circumstances is best facilitated by


the location of the two in close proximity to one another (Gustafson 1959,
164; Tobier and Willis 1981).

Technological Change
During the 1960s and 1970s, the printing industry experienced a major
technological upheaval that introduced new complications into the industry’s
locational patterns and preferences.1 Until then, the method for printing let¬
ters and images, known as letterpress, had changed very little during the pre¬
vious five hundred years. Printing was accomplished by inking a raised surface
and pressing it onto paper. The setting of type was done on huge Linotype
machines that cast heavy slugs of type from molten lead. Hot lead was
dropped, one line at a time, into forms, which then had to be lifted and
mounted on presses. For the printing of illustrations and photographs, special
engravings requiring considerable time and skill had to be made. The lengthy
and costly preparation necessary for letterpress printing jobs favored relatively
long production runs with a minimum of graphics. In addition, since typeset¬
ting and printing had to be located in close proximity to one another (due to
the weight of the lead used to set type), they were generally done by the
same firm.
By the 1960s, letterpress was rapidly losing ground to a new method of
printing called lithography (also known as offset printing). Instead of print¬
ing images by inking a raised surface, lithography uses photographic nega¬
tives to produce images on thin photosensitive metal plates chemically
treated to attract ink. The plates, which are easily handled, are bent around
cylinders on lithographic presses to print image after image in a continuous
motion.2 By 1984, letterpress accounted for just 17 percent of commercial
printing revenues nationwide, while lithography’s share had grown to 70 per¬
cent (Center for Urban Economic Development 1987, 18).
The introduction of lithography helped generate a further ripple of tech¬
nological innovations in the printing industry, particularly within the prepress
phase of production.3 First, typesetting was increasingly computerized, a de¬
velopment that replaced hot type with computer-generated typewritten pages
photographed and transferred to chemically treated plates. Computerized
typesetting machines can produce either hard copy or digitized electronic im¬
ages easily transmitted to distant locations via modem or satellite. This meant
that the typesetting and printing of an image no longer had to take place
within the same firm or, for that matter, the same city or country.
In addition, lithography greatly simplified the preparatory work necessary
for the printing of multicolor graphics. Recall that under letterpress, the
preparation of intricate metal engravings was necessary for the printing of
graphic illustrations, a highly labor-intensive undertaking. In lithographic
52 Remaking Chicago

printing, by contrast, graphic images are simply photographed with special


filters to separate them into four basic colors, and the resulting pieces of film
are then realigned for reproduction. Together with new computer controls
on presses that reduced the time required to set up individual press runs, in¬
novations such as this ultimately helped cheapen the cost of small-batch,
multicolor printing jobs considerably. The expansion of niche markets for
multicolor work helped stimulate the growth of new firms specializing in the
preparation of lithographic film and plates, known in the industry as printing
trade shops.
During the early 1960s, the Richard J. Daley administration commis¬
sioned a series of studies examining the impact of technological change on a
number of Chicago industrial sectors, including graphic communications.
The printing industry study, carried out by a private consulting agency, Cor-
plan Associates, concluded that the cost of much of the new, technologically
advanced equipment associated with the growth of lithography would be be¬
yond reach of most small firms (Corplan Associates 1964, 2). The study an¬
ticipated a “weeding out” of smaller, specialized shops and the consolidation
of production in large (160-400 employees), vertically integrated companies
sufficiently capitalized to keep,abreast of the newest technological develop¬
ments in the industry. Because production space in central area loft buildings
was not well suited to the needs of such large firms, the future prospects for
printing in Chicago were said to be grim, unless alternative space could be
located outside the central area (Corplan Associates 1964, 3).
As it turned out, this forecast proved to be largely inaccurate. Technologi¬
cal change in the printing industry did not lead to a reintegration of the pro¬
duction process during the 1970s and 1980s, as Corplan had anticipated. To
the contrary, just 20 percent of Chicago area printing establishments were
fully integrated by the mid-1980s (Ranney and Wiewel 1988, 17). Among
commercial printers, firms doing small-batch, multicolor work, while con¬
tracting out their prepress operations, became the fastest growing part of the
business (Ranney and Wiewel 1987, 93). Within the prepress segments of the
industry, small- and medium-sized typesetting establishments and trade shops
showed the strongest growth, both nationally and in the Chicago area. Be¬
tween 1967 and 1982, typesetting firms and trade shops recorded a 74 per¬
cent gain in total number of establishments nationwide (Ranney and Wiewel
1988, 15). Forecasts by the Illinois Department of Employment Security
showed employment growing in Chicago area typesetting establishments and
trade shops at a rate of 31 percent between 1982 and 1995, the highest
growth rate of any segment of the printing industry (Ranney and Wiewel
1987, 46).
Technological innovations and the new market opportunities that accom¬
panied them had important locational ramifications for the Chicago area
printing industry. As small-batch, multicolor print jobs grew more affordable,
Rethinking Industrial Decline 53

downtown advertising firms became the largest customers for many Chicago
printers. By the mid-1980s, 90 percent of Chicago printing establishments
did at least some advertising work (Ranney and Wiewel 1987, 73). Due to
demands for rapid turnaround and the need for face-to-face communication
between printers and their customers in the production of advertising copy,
printers specializing in these markets generally tended to cluster in the down¬
town area, oftentimes collaborating with one another on individual jobs.
Moreover, because these firms were typically on the small side, with twenty
or fewer employees, the relatively cramped production space available in cen¬
tral area loft buildings did not present the same efficiency problems it did for
larger firms. In a 1987 survey of 320 Chicago printing establishments under¬
taken by researchers from the University of Illinois at Chicago, less than 40
percent of respondents expressed dissatisfaction with the layout and condi¬
tion of their existing production facilities (Ranney and Wiewel 1987, 124).
Other segments of the industry, by contrast, responded to a far greater ex¬
tent to the kinds of structural economic pressures identified in Chapters 1
and 2, relocating production facilities outside the city of Chicago where land
and labor costs were substantially lower. Ironically, the same technological in¬
novations that helped root certain industry segments more firmly in the
downtown area made other segments more footloose. In particular, the abil¬
ity to receive and to transmit digitized images via satellite or modem encour¬
aged many larger commercial printers producing catalogs or periodicals for
national or international markets to set up their press operations in low-cost
suburban, rural, or Third World locations, even if much of their prepress
work continued to be done by more centrally located typesetting firms and
trade shops (Wiewel, Ranney, and Putnam 1990).

Displacement
At roughly the same time that technological advances in lithographic
printing were fueling the growth of centrally located commercial printers,
typesetters, and printing trade shops specializing in small-batch, multicolor
work, these same industry segments began to experience real estate pressures
associated with downtown commercial and residential redevelopment. The
Daley administration’s efforts to revalorize central area property and protect
the investments of downtown retailers and corporations by surrounding the
Loop with stable residential communities helped create a market for residen¬
tial and commercial loft space, the same loft space being used at the time by
printers and apparel manufacturers in the downtown area. Beginning in the
mid-1970s, developers began purchasing industrial buildings in Printing
House Row and elsewhere in the central area, converting them to back office
space or condominiums, and selling or leasing the rehabilitated space for any¬
where from two to six times the going price commanded for industrial use.
54 Remaking Chicago

By 1986, after a flurry of loft conversion activity during the late 1970s and
early 1980s, Printing House Row no longer housed a single graphic commu¬
nications establishment.
In reality, the fate of the printers, typesetters, binders, and other graphic
communications firms on Printing House Row had been largely sealed sev¬
eral decades earlier when the Daley administration, working closely with the
Chicago Central Area Committee (CCAC), issued its Development Plan for
the Central Area of Chicago in 1958. As we saw in the previous chapter, the
plan called for extensive redevelopment in the South Loop and the Near
South Side, including the construction of a new federal-government center,
housing for middle-income families, and a new University of Illinois campus
on a 130-acre site encompassing Printing House Row and underutilized rail¬
road land south of the Loop. When unfruitful negotiations with railroad
company owners caused the city to seek an alternative campus site in 1960,
plans for Printing House Row and the surrounding area were temporarily put
on hold. However, the city had, by that time, already sent a clear message
that it would support efforts to redevelop this area for nonindustrial pur¬
poses. Prices for unimproved loft space in the Printing House Row district
increased sharply during the 1960s. As Table 3 indicates, land values at the
corner of Dearborn and Harrison Streets (the district’s center) were stable
between 1950 and 1960 but jumped from $7.00 to $17.00 per square foot
during the following ten-year period.
For most industrial properties in Chicago, an increase of this magnitude
would be highly unlikely unless the sellers had obtained a zoning change
from the city, assuring purchasers they would be able to develop the property

Table 3. Land Values in and near Printing House Row, 1950-1990

Locadon 1950 1960 1970 1980 1990

Corner of $6.25 $7.00 $17.00 $22.00 $70.00


Harrison and
Dearborn Streets
(Printing
House Row)

Corner of $4.50 $5.00 $10.00 $12.00 $40.00


Harrison and
Wells Streets
(three blocks
west of Printing
House Row)

Sources: Olcott (1950, 1960, 1970, 1980, 1990). Note: Amounts represent dollars per square foot.
Rethinking Industrial Decline 55

for higher-rent, nonindustrial uses. However, industrial property on Printing


House Row, like much of the industrial property elsewhere in the central
area, was zoned C3, which allows a mixture of light industrial, commercial,
and residential uses. No zoning changes were required to convert the indus¬
trial space in this area to residential lofts. This meant that property values in
the Printing House Row district automatically reflected their “highest and
best use,” assuming the city’s willingness to provide infrastructure and other
public support necessary for nonindustrial redevelopment was not in ques¬
tion. Given the Daley administration’s stated preferences for residential and
institutional development in the South Loop area, it is no surprise that by the
1960s land speculators began to bet on a change in land use in Printing
House Row.
Although redevelopment plans for the South Loop and the Near South
Side remained in limbo for well over a decade following publication of the
1958 Development Plan, the CCAC never abandoned its commitment to a
large-scale development south of the Loop. In 1973, plans for the massive
new Dearborn Park middle-income housing development were announced as
part of the CCAC’s Chicago 21 Plan, which was intended once again to pro¬
tect the value of downtown real estate by gentrifying the neighborhoods im¬
mediately surrounding the Loop. The project’s developers, the Chicago 21
Corporation, eventually agreed on a site directly south of Printing House
Row, between Polk Street and Roosevelt Road (see Map 2). The City of
Chicago enthusiastically supported the proposal, committing $7.74 million
in city funds for infrastructure improvements, a new school, and two public
parks for the first phase of Dearborn Park alone (Hollander 1987).
Several years after plans for Dearborn Park were announced, a group of
developers led by Chicago architect Harry Weese began quietly acquiring
property on Printing House Row. By 1979, the group had purchased five in¬
dustrial buildings and spent millions of dollars rehabilitating the loft space for
residential and commercial use. The new conversions would, in the words of
one developer, create “a fresh green ribbon” between the Loop and the new
Dearborn Park housing development, helping to further rejuvenate the
South Loop area (Ziemba 1976). Weese himself would go on to chair the in¬
fluential Committee for Legal Lofts, one of the most prominent groups rep¬
resented in the coalition of civic organizations that successfully pressured the
Jane Byrne administration to relax the city’s building-code standards for loft
conversions in 1982.4 In addition to easing building rehabilitation standards,
city officials agreed to spend $1.5 million for infrastructure improvements
and the construction of a public park in Printing House Row (Hollander
1987). To top it all off, the city agreed to sponsor Urban Development Ac¬
tion Grant applications to finance the conversion of three industrial buildings
in the district, providing Weese’s development group with $6.8 million in in¬
vestment funds (Wright 1987).
56 Remaking Chicago

Although city officials insisted that the buildings were underutilized and
functionally obsolete as sites for manufacturing (see Commission on Chicago
Historical and Architectural Landmarks 1983, 8-9; Department of City Plan¬
ning 1958, 8), a 1987 study of loft-conversion activity on Printing House Row
carried out by the University of Illinois at Chicago suggests otherwise. The
study’s author successfully located and interviewed representatives from forty-
nine of the ninety-six printing firms that had left Printing House Row since
1970.5 Sixty-seven percent of respondents reported they had been pressured to
vacate their existing production space in Printing House Row, through exces¬
sive rent increases, breaking of leases, and harassment (Pruska-Carroll 1987,
21). Half the firms contacted had moved to other locations in the central area
following displacement from Printing House Row. Many emphasized the fi¬
nancial hardship involved in relocation and the difficulty of securing alternative,
suitably located production space. Seventy percent of respondents indicated an
ongoing preference for a near-downtown location (Pruska-Carroll 1987, 24).
Even more important than the time and expense involved in finding alter¬
native production space in the downtown area was the disruption of impor¬
tant subcontracting and networking relationships among firms located in
close proximity to one another. These relationships, built and reinforced
through years of face-to-face contact among manufacturers, helped create a
certain entrepreneurial dynamism relatively unique within agglomerations of
small competitive firms. One commercial printer located in the same Printing
House Row loft building for thirty-seven years described these interrelation¬
ships to me in the following terms:

It was a perfect building for a small printer such as I. We had a very large ink
company [in the building]. Then on the eleventh floor we had Mackin’s
Bindery, and we would send our folding, cutting, and trimming up there. I had
a two-color press, twenty-three inches by thirty-six inches. If I had eight-and-a-
half by eleven work I’d farm it out to Ad Litho next door or Campaign Press
on the seventh floor. We’d follow a line of reciprocity. Campaign Press would
have some work that was twenty-three by thirty-six inches—too large for their
presses—so we would do that work for them. Campaign Press, Ad Litho, and
RLA Press were all both customers and suppliers. So there was reciprocity all
over the building. It was quite advantageous to have everybody in the same
building. (Maar 1996)

Small commercial printers without well-developed typesetting or binding ca¬


pabilities were able to save time and expense by utilizing the services of other
firms in the same building. According to the same printer,

When we got a job off the press, we would just roll it over to the freight eleva¬
tor on a press dolly and send it upstairs to the eleventh floor [for binding]. If I
Rethinking Industrial Decline 57

were sending it out to an outside bindery, we’d have to put it on a skid and
then band it with a steel band. All that takes time. So the building was a great
advantage. (Maar 1996)

Roughly half the printers who relocated from Printing House Row during
the 1970s and 1980s moved into buildings several blocks west or south of
the district (Pruska-Carroll 1987, 26). Redevelopment pressures, however,
were never far behind. In recent years, industrial buildings in many of these
areas have become attractive targets for commercial and residential conver¬
sions as well, and some printers are now being uprooted for a second time.
While I was in Chicago doing research during the winter of 1996, I discov¬
ered one such building in the process of conversion to residential lofts. The
story of its transformation from working to living space provides a vivid illus¬
tration of the way in which downtown redevelopment can destabilize the
market for industrial loft space.
The building, a twelve-story, turn-of-the-century loft structure, is lo¬
cated three blocks west of Printing House Row on West Harrison Street in
the South Loop. During its peak years of industrial use in the 1940s, the
building housed sixty-seven industrial tenants, forty-three of which were
engaged in either printing or related activities (Winters Publishing Co.
1942). Ry the time loft conversion activity began nearby on Printing House
Row in the 1970s, it was still occupied mainly by printers and providers of
ancillary services, many of whom had been there for twenty-five years or
more.
In 1982, the building changed ownership for the first time in several
decades. It is unclear what the buyer’s intentions were, but there are signs
that he purchased the building for speculative purposes. According to ten¬
ants, building maintenance deteriorated rapidly after he assumed ownership
and remained substandard for the duration of his tenure as building owner.
The freight elevator was in a chronic state of disrepair, forcing tenants to rely
on the smaller passenger elevator for deliveries. On several occasions, both el¬
evators were simultaneously out of service. Eventually, firms started to move
out. By 1992, when the building changed hands once again, occupancy had
fallen below 50 percent and many intrabuilding subcontracting and network¬
ing relationships had been destroyed.
In the meantime, property values in this area of the Loop had begun to
accelerate sharply. Like Printing House Row, this area was zoned C3, mean¬
ing that land prices here automatically incorporated the effects of the nearby
loft conversion activity in Printing House Row and the new Dearborn Park
housing development. Table 3 shows that land values at the corner of Harri¬
son and Wells Streets, where the building is located, rose from $12.00 to
$40.00 per square foot between 1980 and 1990. Land speculators here
could more than triple the value of their investments simply by holding onto
58 Remaking Chicago

real estate for eight to ten years and then reselling it at inflated prices. These
prices virtually guaranteed that a change in land use from industrial to nonin¬
dustrial would accompany any real estate transaction here, since rents for in¬
dustrial tenants were far too low to support the property taxes and building
mortgages on such costly real estate.
By winter of 1996, the new owners were busily converting the building to
one and two bedroom residential lofts, at prices ranging from $145,000 to
$320,000. By then, all existing industrial tenants had received notices that
their leases would not be renewed, and only seven printing firms remained in
the building. Three of these firms formerly occupied loft space in Printing
House Row and were now being forced to move for the second time in fif¬
teen years or less. All but one of the seven had successfully located new pro¬
duction space elsewhere in the central area. Most firms were moving several
blocks further west, just across the Chicago River. However, unless strong
steps are taken to curb the pace of redevelopment activity in this part of the
city, it will be only a matter of time before these printers are once again
forced to relocate.

V
Decline
With the reduction of industrial loft space in Printing House Row and
elsewhere in the downtown area, the number of printing establishments in
the central area of Chicago fell sharply during the 1980s and early 1990s.
Table 4 shows the magnitude of this drop: altogether, firms engaged in com¬
mercial printing, typesetting, and printing trade services in the downtown
area decreased in number by nearly 50 percent between 1981 and 1993.
These numbers appear to reflect industry decline within, not movement away
from, the central area of the city. Between 1979 and 1985, Chicago printing
employment located in this part of the city increased from 50 percent to 60
percent (Illinois Department of Employment Security 1979, 1985). Many

Table 4. Printing Establishments in the Central Area of Chicago, 1981 and 1993

Industry Segment 1981 1993 Percent Change

Commercial Printers 312 160 -49


Typesetters 64 36 -44
Trade Shops 25 20 -20

Total 401 216 -46

Sources: Influential Contacts, Ltd. (1981); Inside Contacts (1993).


Note: Includes Standard Industrial Classification (SIC) categories 275, 2791, and 2795.
Rethinking Industrial Decline 59

Chicago printers apparently still preferred a central location but were increas¬
ingly incapable of surviving there.
It would be unfair to blame this decline entirely on real estate pressures
accompanying downtown redevelopment. By the 1980s, Chicago area print¬
ers were wrestling with various other difficulties as well. Most importantly,
the rapid pace of technological change in the industry eventually created fi¬
nancing problems for some firms too small to afford state-of-the-art equip¬
ment (Ranney and Wiewel 1987). In addition, printing employers faced a
growing shortage of skilled labor by this time, caused by declining member¬
ship in the industry’s two principal labor unions, the International Typo¬
graphical Union (ITU) and the Graphic Communications International
Union (GCIU), and by the breakdown of the union apprenticeship programs
(Wiewel, Ranney, and Putnam 1990).
Even though displacement pressures alone cannot fully account for the de¬
cline of centrally located printing establishments in recent years, the loss of
the industry’s spatial compactness is critical. As we saw in Chapter 1, firms lo¬
cated in industrial districts elsewhere have developed collective solutions to
common problems such as labor and equipment shortages, in some cases
sharing tools, workers, and information with nearby establishments (Brusco
1982; Sengenberger and Pyke 1992; Kristensen 1992). These relationships
appear tQ be built, to a great extent, on the mutual trust that eventually de¬
velops among district members as a result of repeated reciprocal exchanges
(Lorenz 1988; Granovetter 1985). Spatial proximity among firms increases
opportunities for these exchanges to occur, thus fostering trust among dis¬
trict-member firms to the extent they perceive such exchanges to be mutually
beneficial (Harrison 1992).
In Chicago, printing firms displaced by the pressures of downtown redevel¬
opment have thus far shown a strong tendency to seek alternative production
space elsewhere in the central area. Yet while the majority of Chicago’s print¬
ing establishments remain in the downtown area, the concentration of firms is
much less pronounced than it was twenty-five years ago when Printing House
Row still served as the nucleus of Chicago’s graphic communications industry.
Opportunities for subcontracting and networking relationships among firms
still exist, but a good deal of the synergy produced by repetitive, face-to-face
interactions among printers, customers, and suppliers has no doubt been lost.
Conditions that might have facilitated cooperative problem-solving among
firms on other issues have been seriously undermined as well.

The Chicago Apparel Industry


The history of the Chicago apparel industry parallels that of the city’s print¬
ing industry in a number of important ways. By 1925, Chicago apparel manu¬
facturers had concentrated in a tightly knit cluster several blocks northwest of
60 Remaking Chicago

Printing House Row along the east bank of the Chicago River (see Map 2).
Like the printing industry, the spatial concentration of apparel firms in the
downtown area was caused principally by uncertainties and rapid fluctuations
in the market for the industry’s output. The importance of fashion to many
segments of the apparel industry, and the inability of producers to predict in
advance which lines of clothing would generate the greatest enthusiasm
among buyers, placed limits on the consolidation of production in large, self-
contained factories. It also encouraged manufacturers to locate near the city’s
large downtown department stores in order to keep themselves abreast of the
continuously changing fashion preferences of retail clothing buyers (De Meir-
leir 1950, 127-128).
While firms in Chicago’s garment district were generally quite small, with
ten or fewer employees, there were also a number of large factories employ¬
ing one hundred or more workers (Magee 1930, 97). At times of peak de¬
mand, the district’s smaller firms could be pressed into service as contractors
for large manufacturers, allowing such firms to respond to seasonal fluctua¬
tions in the market without expanding their own productive capacities
(Carsel 1940, 7-8). At other times, small manufacturers worked either inde¬
pendently or as contractors for “jobbers” in the district, the latter of whom
generally handled the design, purchasing of fabric, and sales of their garments
themselves but contracted out the actual sewing.
This set of arrangements provided the Chicago apparel industry with the
flexibility necessary to respond quickly to extremely volatile, unpredictable
markets. The output of successful lines of clothing could be rapidly increased
without a corresponding expansion in the productive capacity of individual
firms, while less successful lines could be quickly discontinued with minimal
costs. Through contracting and subcontracting relationships, overall produc¬
tion in the district easily shifted in favor of those garments for which demand
was strongest. The spatial concentration of firms facilitated contracting and
subcontracting among manufacturers and jobbers by providing the face-to-
face contact necessary for such relationships to develop and mature. It also
reduced the time and transportation-related cost penalties of subcontracting
and cut down on the distance retail buyers had to cover when visiting the
showrooms of Chicago apparel manufacturers (Magee 1930, 91).
Spatial concentration was important, not only for the synergy of the gar¬
ment district itself but also for the viability of individual firms. By 1910, the
International Garment Workers Union had made significant gains among the
employees of Loop apparel-manufacturing firms. Hoping to escape what
many perceived as an increasingly unfavorable business climate, numerous
firm owners began to set up shop in one of two locations several miles out¬
side the Loop, each featuring large pools of nonunion, immigrant laborers
(Magee 1930, 82). In the long run, however, this strategy proved to be dis¬
advantageous. According to historian Keith McClellan,
Rethinking Industrial Decline 61

Experience showed that to have a central location was far more important in
the garment industry than to achieve small wage differentials. Peripheral firms
found that being isolated from contractors and jobbers reduced “face-to-face”
confrontations and resulted in a loss of business. ... By 1920, the number of
clothing manufacturing firms located outside the Loop had declined markedly,
and by 1928 the industry was more heavily concentrated in the Loop than ever
before. (McClellan 1966, 34)

Postwar Industry Restructuring


Concentration among Chicago apparel manufacturers in the Loop gar¬
ment district continued during the 1930s and early 1940s. Following World
War II, however, developments in clothing production and new market op¬
portunities for some types of apparel firms introduced changes in the loca¬
tional preferences of certain segments of the industry. To begin with, even
though the volatility of fashion and difficulties of mechanizing the produc¬
tion of clothing from soft, limp fabrics have always made the clothing indus¬
try somewhat resistant to mass-production techniques, U.S. apparel manufac¬
turers went further in that direction than ever before (or since) during the
postwar years (Kenyon 1964). Postwar prosperity and greater public accep¬
tance of casual, informal clothing reduced pressure for manufacturers to dif¬
ferentiate themselves on the basis of style (Helfgott 1959, 77). Instead, with
the encouragement of large downtown retailers, many clothing manufactur¬
ers moved to capture scale economies by producing standardized lines of
clothing in long production runs (Zeitlin and Totterdill 1989, 157). Such
firms tended to be large and relatively self-contained. This, together with the
low design content of their product lines, led manufacturers like these to seek
lower-cost production space outside the city center. In Chicago, new concen¬
trations of apparel firms began to emerge on the Near West Side, while man¬
ufacturers producing for more fashion-sensitive segments of the market con¬
tinued to cluster in the city’s original garment district several blocks away in
the West Loop area (De Meirleir 1950, 127-128).
Ultimately, however, apparel manufacturers that had neglected design in
order to achieve scale economies found themselves ill-prepared to respond to
several important changes in U.S. apparel markets during the 1970s. First,
even with standardization, clothing production continued to be highly labor
intensive, with materials handling accounting for up to 80 percent of the to¬
tal time involved in the production of a garment (Floffman 1985, 372).
Once Third World countries acquired the relatively unsophisticated technol¬
ogy and skills for large-batch, standardized garment production, their com¬
paratively lower labor costs enabled them to undercut U.S. producers by
wide margins. By 1980, labor costs represented 35 percent of the total cost
of a typical garment manufactured in the U.S. and only 7 percent of a similar
62 Remaking Chicago

garment produced in an Asian newly industrializing country such as Taiwan


or Hong Kong (Hoffman 1985, 373).
Large domestic clothing producers were also caught off guard by the re¬
cessions accompanying the oil price hikes of the 1970s. The growing instabil¬
ity of domestic demand during the 1970s, together with rising competition
from imports, made it increasingly difficult for many U.S. clothing manufac¬
turers to achieve the sales volumes necessary to amortize the large invest¬
ments they had made in plants and equipment during the 1950s and 1960s.
Table 5 shows the impact of these two new conditions on the eight major
segments of the Cook County apparel industry. Firm closures and relocations
during the 1970s caused declines in employment in virtually every segment
of the industry. Job losses were particularly heavy in standardized industry
segments most susceptible to import penetration, such as women’s and chil¬
dren’s undergarments and girls’, children’s, and infants’ outerwear.

Table 5. Apparel Employment in Cook County, 1970 and 1980

1970 1980
Industry Segment Firms Employees Firms Employees

Men’s, Youths’, and Boys’ 26 6,331 12 2,500-9,999*


Suits, Coats, and Overcoats

Men’s, Youths’, and Boys’ 36 1,378 20 1,078


Furnishings, Work Clothing,
and Allied Garments

Women’s, Misses’, and 88 3,613 59 2,139


Juniors’ Outerwear

Women’s, Misses’, Children’s, 16 1,525 5 719


and Infants’ Undergarments

Hats, Caps, and Millinery 19 445 15 473

Girls’, Children’s, and 9 120 0 0


Infants’ Outerwear

Miscellaneous Apparel and 24 1,172 22 500-599*


Accessories

Miscellaneous Fabricated 246 5,502 173 4,655


Textile Products

Source: Robinson (1985). ‘Exact amounts are withheld to avoid disclosing data for individual companies.
Rethinking Industrial Decline 63

U.S. apparel manufacturers were, however, not the only ones to feel the
pressure of global economic changes in clothing markets. Domestic clothing
retailers eventually discovered that the sourcing of low-cost, standardized
garments from Third World producers could not compensate for the increas¬
ingly sluggish domestic demand for clothing by the 1970s. As a result, retail¬
ers began to make new demands on their suppliers, emphasizing fashion,
quality, and rapid turnaround times as a means of defining and responding
quickly to new market niches. According to economists Jonathan Zeitiin and
Peter Totterdill,

While price remains important, particularly at the lower end of the market, the
struggle for competitive advantage has come to center increasingly on retailers’
and manufacturers’ efforts to target specific groups of consumers defined in
new ways; to seduce customers with attractive, fashionable garments; and to re¬
spond rapidly to short-term trends in the sales of individual product lines.
(Zeitiin and Totterdill 1989, 162)

The growing fragmentation of mass markets for clothing during the 1980s
favored manufacturers capable of producing relatively wide ranges of styles
and flexible enough to switch production between clothing lines in response
to shorty-term sales trends. With bar-coding now providing instant access to
sales-performance data on entire store inventories, retailers were generally
much less willing to place advance orders in large quantities. Retail buyers
who during the 1960s and 1970s finalized most of their orders with suppliers
at the beginning of each fashion season were by the mid-1980s oftentimes
staggering the bulk of their purchases over the course of each season (Costi-
gan 1996). According to Zeitiin and Totterdill, these developments created
new advantages for centrally located domestic clothing producers and disad¬
vantages for Third World manufacturers:

Lead times are too long, minimum production runs too large, quality control
too difficult, and the capital costs of stocks too high to make sourcing in the Far
East worthwhile for many of the more fashion-sensitive types of garment. . . .
Physical proximity is also important for collaboration between retailers and sup¬
pliers in range development and rapid adjustment of production to sales. (Zeitiin
and Totterdill 1989, 167)

In Chicago and other fashion centers, the segment of the apparel industry
in the best position to respond to these new conditions was a small but grow¬
ing community of designer/manufacturers and contractors who typically oc¬
cupied downtown area loft space (Department of Economic Development
1987, 7). Designer/manufacturers generally employ a small staff of cutters
and sewers to produce display garments for their showrooms and to handle
64 Remaking Chicago

small orders, but much of their production work is ordinarily contracted out.
In an industry where uncertainty reigns, this arrangement reduces risk and
increases flexibility for all parties concerned. Designers can temporarily in¬
crease production when orders rise by hiring additional contractors. In turn,
contractors are free to accept orders from more than one designer, when nec¬
essary. To a great extent, designer/manufacturers have taken the place of job¬
bers, who performed a similar role in Chicago’s garment district during the
prewar years. The major difference between the two is that designer firms of¬
ten target a far more specialized, identifiable segment of the market than the
typical jobbers did. According to Dorothy Fuller, vice president of the
Chicago Apparel Center,6

These firms are small, very focused, and probably priced a little higher than the
average but with something rather special to sell. We have, for example, a cou¬
ple of companies that do hand painting—hand painted silk gowns, dresses, cos¬
tumes. They’re expensive and they’re very special, very identifiable, and their
business is probably $3 million. There are also a whole group of designers
whose business is under $1 million. (Fuller 1996)

By the 1980s, some designer firms were beginning to utilize computer-


aided design (CAD) systems as a way of expanding their garment styles and
decreasing turnaround times from concept to delivery. CAD systems reduce
much of the time and effort involved in creating patterns, the basic building
blocks of apparel manufacture. Patterns are used as templates for the cutting
of fabric and must be prepared for each component (such as a collar or a
sleeve) of a particular garment. Before computerization, clothing designers
had to mark out patterns by hand in the full range of sizes being produced.
The patterns for one style alone could easily number in the hundreds, and
the process of grading each pattern up and down to produce multiple sizes
was extremely time-consuming (Hoffman 1985, 375). Computerization
eliminated the need for much of this work. Designer/manufacturers using
CAD systems simply make one set of patterns for each garment, enter the
specifications into the system, and watch while the computer automatically
generates a full set of additional patterns for each size specified by the opera¬
tor. This frees designers to spend more of their time producing new clothing
styles rather than grading patterns to produce multiple sizes of existing styles.
It also allows them to ship orders two to three weeks faster than they other¬
wise could, at considerable cost savings (Liametz 1996).
Despite new market opportunities and technological advances reducing
the cost penalties of small-batch clothing production, the performance of
Chicago’s centrally located designer/manufacturers during the past decade
has not been strong. By the mid-1990s, there were very few of these firms
Rethinking Industrial Decline 65

left in the central area, for reasons to be explored shortly. According to one
comprehensive Chicago area business directory, only 66 apparel manufactur¬
ing establishments were located in the entire central area in 1995, down
from 135 in 1981 (Inside Contacts 1995). To get a better idea of the char¬
acteristics of this segment of the industry and the reasons for its decline, I
selected a random sample of the sixty-six remaining firms for a brief tele¬
phone questionnaire. Although these manufacturers were, for the most part,
considerably less eager to share information than the printers with whom I
spoke, I was still able to interview representatives from twenty of the sixty-
six firms.
In general, the results of the survey suggest that, although designer/man¬
ufacturing firms have declined in numbers, they are still the types of apparel
manufacturers most likely to be found in the near-downtown area. Moreover,
such firms continue to exhibit strong attachments to this part of the city de¬
spite real estate pressures that have forced many firms to relocate one or
more times within the past fifteen years. Thirteen of the twenty firms sur¬
veyed (65 percent) indicated that the manufacturing of designer wear was an
important part of their business. Most of these firms were relatively small,
employing twenty or fewer workers. In addition, fifteen firms (75 percent)
reported that a central location was either moderately important or very im¬
portant jo their companies. The primary reasons given were proximity to
contractors and customers and convenience of transportation. Seventeen of
the twenty firms (85 percent) either contracted out work to, or accepted
work from, other manufacturers. Finally, thirteen firms (65 percent) reported
they had found it necessary to relocate at least once in the past fifteen years
due to rent increases, building demolitions, or building conversions to new
nonindustrial uses.
In many ways, the designer/manufacturer segment of the Chicago apparel
industry resembles the segment of the Chicago printing industry that began
to emerge in the 1970s, when technological innovations in lithographic
printing began to make small-batch, multicolor print jobs economically feasi¬
ble. Both groups have taken advantage of new technologies to reduce the
cost penalties that typically accompany production in short runs, allowing
them to tap into new, previously inaccessible market niches. Both have also
made use of extensive subcontracting relationships to reduce risks and in¬
crease flexibility. This, in turn, helps explain the tendency of firms in both in¬
dustry segments to cluster in industrial districts located near downtown cus¬
tomers. However, both groups have ultimately lost many of the benefits of
spatial proximity through displacement pressures associated with downtown
commercial and residential redevelopment. For the Chicago apparel industry,
these pressures appeared earlier and carried more severe consequences than
they did for the Chicago printing industry.
66 Remaking Chicago

Displacement
The Chicago garment district first began to experience the pressures of
downtown redevelopment shordy after World War II, when planners routed
the new Congress and Dan Ryan Expressways through the West Loop and
Near West Side as part of the city’s efforts to link Chicago’s central business
district with the oudying portions of the metropolitan area. The construction
of the Congress Expressway alone led to the demolition of 250 structures in
the Loop and on the Near West Side, including 157 commercial buildings
(Sturdy 1949). The new freeway cut a path through the southern edge of the
downtown garment district, extending west through the heart of the Near
West Side, where many of the industry’s less fashion-conscious manufacturers
had begun to locate following World War II. Altogether, some 3,000 apparel
industry jobs were lost between 1948 and 1958 due to displacement caused
by freeway construction (Center for Urban Studies 1966, 60). Job losses
were concentrated among Near West Side apparel firms, but five loft build¬
ings in the downtown garment district occupied primarily by apparel manu¬
facturers were also destroyed. The city made no effort to secure alternative
production space for displaced manufacturers elsewhere in the garment dis¬
trict, and many uprooted firms that did not go out of business altogether
moved to new locations in other parts of the metropolitan area (Center for
Urban Studies 1966, 60).
Like Printing House Row, the demise of Chicago’s garment district was
also written into the Daley administration’s 1958 Development Plan for
Chicago’s central area. As late as 1960, over 50 percent of Chicago apparel
manufacturers were still concentrated in an area of the West Loop bordered
by Monroe Street and Congress Parkway to the north and south, and Wells
Street and the Chicago River to the east and west. Table 6 indicates that 116
of Chicago’s 221 apparel manufacturing firms were located within these
boundaries in 1960. Under the 1958 Development Plan, both this area and

Table 6. Concentration of Apparel Firms in the


Chicago Garment District, 1951-1981

Location 1951 1960 1970 1981

Apparel Firms 369 221 148 130


Citywide

Firms Located within 183 (50%) 116 (52%) 57(39%) 26(20%)


Garment District

Sources: Manufacturers News, Inc. (1951, 1960, 1970); Influential Contacts, Ltd. (1981).
Note: Includes SIC categories 231-237.
Rethinking Industrial Decline 67

adjoining properties to the north were to be integrated more extensively into


the central business district (Department of City Planning 1958, 7). Ideally,
the mix of relatively low-intensity commercial and industrial land uses that
made up much of the area between Wells Street and the Chicago River would
be replaced by high-rise office buildings, boosting downtown property tax
revenues and creating the construction jobs that the Daley administration re¬
lied upon to shore up its support from the building trade unions.

Table 7. Land Values in and near the Chicago Garment District, 1950-1980

Location 1950 1960 1965 1970 1980

Corner of $8.50 $8.50 $12.00 $27.00 $60.00


Franklin and
Van Buren Streets
(Garment District)

Corner of $16.00 $20.00 $29.00 $45.00 $70.00


LaSalle and
Van Buren Streets
(Central Lotjp)

Corner of $5.00 $8.00 $10.00 $11.00 $13.00


Jackson and
Desplaines Streets
(Near West Side)

Sources: Olcott (1950, 1960, 1965, 1970, 1980). Note: Amounts represent dollars per square foot.

By the mid-1960s, however, even though redevelopment activity was pro¬


ceeding at a rapid pace east of Wells Street, developers had made fewer in¬
roads into the western portions of the Loop. Table 7 shows that property in
the heart of the garment district was still being priced for low-density indus¬
trial and commercial uses in 1965, at $12.00 per foot, while land values just
two blocks east in the central Loop were more than double that amount.
Wells Street, also serving as the north-south route for the city’s elevated
commuter rail line in the west Loop, appeared to represent a psychological
barrier for the real estate community. Investors seemed reluctant to cross this
line as long as opportunities for new office-tower construction in the heart of
the central business district continued to present themselves. Eventually, city
planners became convinced that a large development project was necessary in
the West Loop to serve as a visual anchor and solidify investor confidence in
the area.
68 Remaking Chicago

In the late 1960s, Sears, Roebuck and Company informed city officials that
it was planning to consolidate its administrative facilities—at the time spread
over various Chicago area locations—in one large downtown high-rise office
building it wanted to construct (O’Connor 1975). Company plans called for
a 110-story, 1,450-foot headquarters building, taller than any other structure
in the world. Sears was interested in a West Loop location and had identified
a site along Franklin Street between Adams and Jackson. The property was
situated within the garment district, and two buildings housing primarily ap¬
parel firms were located on the site. However, assuming the building owners
could be persuaded to sell, there was another more pressing obstacle. The
parcel of land Sears wanted to develop was bisected by Quincy Street, mean¬
ing the company would have to build two towers next to one another instead
of one massive high-rise building. Sears was unwilling to do this.
In 1969, Sears chairman Gordon Metcalf met privately with Mayor Daley
to seek the mayor’s support for the transfer of this block of Quincy Street to
Sears, Roebuck and Company, emphasizing the vast number of construction
jobs the project would generate and the prestige Chicago would enjoy as
home of the world’s largest building (O’Connor 1975, 137). The adminis¬
tration eventually agreed to a deal: it would recommend the sale of Quincy
Street to Sears at a relatively high price of $2.8 million, or $130.00 per
square foot. However, the city would absorb the $1.2 million cost of rerout¬
ing water and sewer lines beneath the street, effectively reducing Sears’ pur¬
chase price to $77.00 per square foot. As Daley biographer Len O’Connor
observed, this was a “bargain basement price” for a public thoroughfare in
downtown Chicago (O’Connor 1975, 137).
City of Chicago Corporation Council Raymond Simon prepared ait ordi¬
nance reflecting the above agreement. With Mayor Daley’s stamp of ap¬
proval, it sailed through the City Council Committee on Streets and Alleys.
According to O’Connor,

Without the imprimatur of His Honor the Mayor, this deal would have been
dumped into the Rules Committee, the bottomless pit into which all anti-ad¬
ministration measures in Chicago City Council are cast. Presented to the full
council as a mighty architectural prize, giving Chicago—with the John Han¬
cock Center built and the Standard Oil Building already under construction—
three of the five tallest structures on earth, there was no dissent when the ordi¬
nance was presented for vote. (O’Connor 1975, 137)

The Chicago Sun-Times reported Mayor Daley’s reaction to the new develop¬
ment as follows:

Daley called [Sears Roebuck and Co.] “one of the most humane [corporations]
because of its interest in people and helping cities find answers Jo social prob¬
lems.” He said the building will constitute a “boost for the area” and expressed
Rethinking Industrial Decline 69

the hope that other firms will move into the West Side area and rehabilitate it.
(Chicago Sun-Times, 6 March 1969)

Mayor Daley was not to be disappointed. Land speculation and redevelop¬


ment activity in the West Loop area began to accelerate even before Sears
broke ground in 1970. Tables 6 and 7 show the impact of these new devel¬
opment pressures on the garment district during the 1970s. As property val¬
ues at the corner of Franklin and Van Buren Streets, just one block south of
the Sears Tower development, increased to $60.00 per square foot by 1980,
the concentration of Chicago apparel firms located in the garment district fell
to just 20 percent. As in Printing House Row, no zoning changes were re¬
quired to redevelop this property for nonindustrial purposes. Most of the
new development here, however, involved land clearance and construction of
massive new high-rise office buildings rather than the adaptive reuse of exist¬
ing buildings favored by developers in nearby Printing House Row. By 1980,
very few of the turn-of-the-century loft buildings that once lined the south
ends of Franklin Street and Wacker Drive (formerly Market Street) were left,
and the area had ceased to function as an identifiable garment district. In
1988, the building housing the last remaining apparel firms in the district
was demolished to build a parking garage for downtown office workers.
The reduction of loft space in the Chicago garment district during the
1970s and 1980s, at roughly the same time that changes in clothing markets
were beginning to create new opportunities for small, flexibly specialized ap¬
parel firms emphasizing design and innovation over economies of scale, came
at an inopportune time for Chicago’s newly developing community of
designer/manufacturers. Although many apparel firms were able to find af¬
fordable space elsewhere in the central area, companies were generally much
more isolated from one another than they had been before. This isolation
carried significant costs in terms of lost networking opportunities, easy access
to contractors and suppliers, informal intraindustry technical assistance, and
other advantages associated with the location of firms in close proximity to
one another. As New York Fashion Council President Gerald Shaw argued in
a 1985 letter to Mayor Edward Koch complaining of the impact of redevel¬
opment pressures on New York City’s garment district,

This industry can only thrive while having in close physical proximity manufac¬
turers (including not just selling activities, but also designing, purchasing, and
shipping), textile vendors and the ancillary, critical trimming makers, machinery
maintained, and freight forwarders. Scattering any of these all over the five bor¬
oughs destroys the synergy unique to the life of our industry. (Lebow 1985)

It is impossible to say how this segment of the Chicago apparel industry


might have fared had city officials responded more evenhandedly to the inter¬
ests of downtown real estate developers and'central area manufacturing
70 Remaking Chicago

establishments. The industry has faced a number of important challenges


apart from real estate pressures, including the financing of expensive CAD
equipment, competition from rising numbers of illegal sweatshops, and in¬
creasingly liberal foreign trade policies for clothing products. An apparel in¬
dustry task force created by the Harold Washington administration during
the 1980s helped focus attention on certain of these issues, with some suc¬
cess. In April 1995, for example, a shared-use CAD facility was opened in the
Chicago Apparel Center to provide small designer firms with access to costly
technologies they cannot afford to purchase themselves. However, another
goal of the task force—the resurrection of a needle trades district in the cen¬
tral area of the city—has thus far proved more elusive. It remains to be seen
whether recent efforts to support the industry can succeed without the less
tangible but still important benefits of spatial clustering.

Conclusion
The redevelopment of postwar Chicago, as we saw in the previous chapter,
took place through the efforts of a progrowth alliance, or “growth machine,”
led by city officials and downtown business leaders. The ideological under¬
pinning of this redevelopment strategy was a narrative of urban economic re¬
structuring and industrial decline embraced by city officials, business elites,
and scholars alike. The central propositions of this narrative, including the
rising mobility of capital, the growing obsolescence of cities as viable loca¬
tions for manufacturing, the division of labor between state and market, and
the equation of economic growth with real estate development and rising
property values, all pointed to one conclusion: that downtown redevelop¬
ment would necessarily become the linchpin of citywide economic growth.
The case studies of the Chicago printing and apparel industries presented
in this chapter and the story of downtown redevelopment told in Chapter 2
both seem to call aspects of this narrative into question. Even more impor¬
tantly, perhaps, they show how it legitimized a particular strategy for down¬
town redevelopment, where viable alternatives existed. As this chapter illus¬
trates, segments of both the printing and apparel industries responded
differently to technological change and industry restructuring following
World War II. Large, vertically integrated firms used assembly production to
achieve scale economies, producing low-priced goods for mass markets and
shifting production to low-cost locations outside city limits. However,
smaller, vertically disintegrated firms achieved economies of scope through
the use of flexibly specialized machinery and subcontracting arrangements.
Interrelationships among manufacturers, suppliers, and subcontractors and
the need to locate in close proximity to downtown customers helped root
these industry segments in Chicago’s downtown area.
The behavior of the latter types of manufacturers is difficult to square with
Rethinking Industrial Decline 71

the assumption of most urban political economists that mobile firms choose
locations chiefly on the basis of factor costs. Printers and apparel manufactur¬
ers in Chicago’s printing and garment districts developed longstanding net¬
working relationships with subcontractors and other manufacturers. Firm
owners passed along job referrals to and exchanged information with nearby
competitors because they knew, based on past experience, that their neigh¬
bors would reciprocate. These relationships, based on nonmarket values like
mutual trust and cooperation, could not be easily reproduced elsewhere. Pro¬
duction was embedded in a social context that was situated, in turn, within a
particular geographical setting.
This set of arrangements helped shelter certain manufacturers from the ef¬
fects of global economic restructuring. By competing more on the basis of
quality, specialization, and rapid turnaround times than on the basis of price,
centrally located printers and apparel manufacturers avoided direct competition
with large, vertically integrated shops using low-cost land and labor to manu¬
facture inexpensive, standardized products. Nevertheless, such competitive
strategies were of little value in fending off real estate speculators acting within
an institutional context in which the principal of “highest and best use” deter¬
mined land-use priorities. Measures to protect central area manufacturers from
real estate pressures would have caused downtown area property values to suf¬
fer. In the eyes of downtown business leaders and their allies in city govern¬
ment, any such efforts would have been viewed as fundamentally antigrowth.
Undergirding the downtown redevelopment plans of Chicago’s
progrowth alliance was an ideology of privatism. City officials insisted
throughout that they were acting within a division of labor between state and
market in which private investment decisions largely determined the course
of central area restructuring. By maintaining that city planning efforts were
formulated in response to, not in conjunction with, market-based activity,
city officials could claim that their activities were essentially nonpolitical. By
all indications, centrally located printing and apparel manufacturers found
such arguments convincing. I asked representatives from twenty firms (thir¬
teen apparel, seven printing) that had been displaced through redevelopment
pressures whether, to the best of their knowledge, public policy had played
any particular role in the events that had forced them to relocate. Nineteen
of the twenty answered this question negatively. As they saw it, their prob¬
lems began and ended with the interests on the part of their former landlords
in obtaining the highest available returns on their property. Not only that,
many seemed to find the question puzzling, as if the answer should have
been obvious to someone with the credentials I claimed to own.
Given these perceptions, it is not surprising that Chicago printers and ap¬
parel manufacturers offered little in the way of resistance when confronted
with real estate pressures caused by commercial and residential redevelopment
activity. Because land speculators and property developers were understood to
72 Remaking Chicago

be exercising legitimate property rights, political mobilization was simply not


viewed as a sensible course of action. As we shall see in Chapters 5 and 6,
however, manufacturers on Chicago’s Near North Side behaved much differ-
endy when confronted with similar pressures during the late 1980s and early
1990s. When piecemeal changes to the city zoning ordinance began to
threaten viable industries by reducing acreage zoned for industrial use in this
portion of the city, manufacturers there forged alliances with neighborhood
residents, community-based organizations, labor unions, and key city officials
in what would prove to be a successful bid for government protection. For
this to occur, a number of critical changes had to take place in Chicago poli¬
tics. These changes are the focus of the following chapter.
Mayor Richard J. Daley leading the NAACP Independence Day parade down State Street, 1963.
African-American voters provided critical support for the Daley machine during the 1950s and
much of the 1960s. (Chicago Historical Society, ICHi-24729)
City officials view an architect’s model of the 1958 Development Plan for the Central Area of
Chicago. The plan called for extensive land-use changes in the central area, including the replace¬
ment of near-downtown industrial districts with middle- and upper-income housing. From left
to right: Clifford Campbell, Deputy Commissioner, Department of City Planning; Mayor
Richard J. Daley; James C. Downs, consultant to the mayor; Ira Bach, Commissioner, Depart¬
ment of City Planning. (Chicago Historical Society, ICHi-24726)
South Dearborn Street in the
Printing House Row district,
1890s. The signage reflects
the diversity of graphic com¬
munications firms and related
business services that clustered
in the district. (Chicago His¬
torical Society, ICHi-20719)

An industrial loft building housing a Chicago apparel manufacturing firm, 1920s. Beginning in
the 1970s, developers began converting buildings like tips to residential and office space.
(Chicago Historical Society, HB-TR-280-8. Photograph by Raymond Trowbridge)
Workers laying patterns at the firm of Hart Schaffner and Marx in the Chicago garment district,
1952. (Chicago Historical Society, ICHi-24111)
Laying of foundation forms for construction of the Sears Tower in the garment district, Novem¬
ber 8, 1970. Mayor Daley played a key role in the site assembly process by endorsing a proposal
authorizing the sale of one block of Quincy Street to Sears, Roebuck and Company. (Chicago
Historical Society, ICHi-26994. Photograph by Eudell H. Greene)
Women from the Harrison-Halsted community on the Near West Side march on City Hall to
protest plans for the demolition of hundreds of neighborhood homes and businesses to make
room for a new University of Illinois campus, April 26, 1961. Although efforts to save the com¬
munity were unsuccessful, Richard J. Daley and future Chicago mayors would be forced to con¬
tend with an increasingly vocal neighborhood movement from the 1960s onward. (Reprinted
with permission, The Chicago Sun-Times ©1996)
i Hi

1 mTCTSt
aCnl
Police shooting at a sniper at the corner of Division Street and Clybourn Avenue during civil un¬
rest following the assassination of Martin Luther King Jr., April 6, 1968. Mayor Daley’s handling
of the disturbance, including orders to Chicago police that suspected looters and arsonists be
shot on sight, fueled the mayor’s growing unpopularity among African-American Chicagoans.
(Chicago Historical Society, ICHi-24421. Photograph by T. Kneebone)

Inauguration of Harold Washington as mayor of Chicago, April 29, 1983. Just behind Washing¬
ton is former mayor Jane Byrne. (The Chicago Public Library, Special Collections Division. Pho¬
tograph by Willy Schmidt)
Mayor Harold Washington shakes hands with workers during a tour of Farley Industries on
Chicago’s Near South Side, July 16, 1987. Washington was a strong proponent of industrial re¬
tention. (The Chicago Public Library, Special Collections Division)
Mayor Richard M. Daley an¬
nouncing his administration’s
support for the establishment
of Planned Manufacturing
Districts (PMDs) in the Goose
Island and Elston Corridor
portions of the North River
Industrial Corridor, June 14,
1990. The districts are in¬
tended to preserve land for in¬
dustrial development by call¬
ing a halt to the rezoning of
industrial property for new
commercial or residential uses.
Daley had been critical of
PMDs during his election
campaign one year earlier.
(Reprinted with permission,
The Chicago Sun-Times
©1996. Photograph by
Amanda Alcock)

N
Chapter Four

Community Economic Development


and the Crisis of Machine Politics

T Jh he growth machine formed by Chicago city officials, downtown busi¬


ness elites, and labor leaders was the rule, not the exception, in postwar ur¬
ban politics. In cities across the nation, similar alliances sought to revitalize
and expand downtown business districts, open up low-income and working-
class neighborhoods to the new middle class, build freeways, sports facilities,
convention centers, and university campuses, and otherwise intensify land use
and increase property values (Molotch 1976; Logan and Molotch 1987;
Stone 1976; Mollenkopf 1983). Higher land values appeared to hold some¬
thing for everyone—greater returns on the investments of landowners, an ex¬
panding tax base for the city, and growth in construction jobs for the build¬
ing trade unions. Realizing these benefits required policies designed to
transfer property into the hands of those willing to develop it to its fullest po¬
tential. Particularly in the downtown areas of large cities, this oftentimes
meant rearranging land-use patterns to encourage new commercial, institu¬
tional, and middle- to upper-class residential development in areas occupied
by industry and lower-income residents (Hill 1983; Mollenkopf 1983; Judd
1983; Fainstein, Fainstein, and Armistead 1983).
If the redevelopment strategies of growth machines were similar from city
to city, so were the ideologies used to legitimize them. The benefits of
growth, according to local business and government leaders, ultimately ex¬
tended to all city residents in one form or another (Elkin 1985; Clavel and
Wiewel 1991, 4). Developing property for its “highest and best use” created
job opportunities and enhanced the city’s fiscal position, easing the tax burden
on middle-class homeowners and other city residents and providing revenues
for redistribution to lower-income neighborhoods. In any case, economic de¬
velopment strategies were highly constrained by the market. Debate over the
substance of economic development policy was thus both unnecessary and
84 Remaking Chicago

counterproductive. In Chicago and other major cities, this message was driven
home by the leading area newspapers, which together with major downtown
banks, department stores, utility companies, and real estate developers,
formed the nucleus of urban growth machines (Molotch 1976; Logan and
Molotch 1987; Feagin 1983; Shlay and Giloth 1987).1
Growth coalitions proved to be politically vulnerable in some cases, how¬
ever, mainly because the costs and benefits of growth were unequally distrib¬
uted, the former falling primarily on lower-income neighborhoods targeted
for urban renewal and the latter accruing principally to downtown develop¬
ers, corporations, and their employees, many of whom lived (and voted) in
the suburbs. In Chicago and elsewhere, resistance to the growth machine
came first in the form of self-preservation efforts by community residents
whose neighborhoods were slated for large-scale clearance and renewal. On
Chicago’s South Side, for example, a neighborhood group called The Wood-
lawn Organization waged a high-profile campaign against the University of
Chicago’s urban renewal plans for the Hyde Park/Woodlawn area in the
early 1960s (Fish 1973). Soon afterward, neighborhood activists from the
Near West Side picketed City Hall and staged a sit-in at Mayor Daley’s office
to protest plans for the destruction of their community in order to make
room for a new University of Illinois campus (Rosen 1980).
Isolated protests such as these were not always successful in blocking re¬
newal projects, but they did force city officials and other members of local
growth coalitions to divert resources into efforts aimed at maintaining social
peace in the neighborhoods (Stone 1976, 1989; Fainstein, Fainstein, and
Armistead 1983). Once neighborhood groups began to act in concert, how¬
ever, these efforts were not always enough. During the early 1970s, for exam¬
ple, community organizations from white ethnic, black, and Latino neighbor¬
hoods in Chicago set aside longstanding antagonism to work together against
the proposed Crosstown Expressway, which included plans for the destruction
of over three thousand units of housing on the city’s Northwest and South¬
west Sides. With the project embroiled in controversy, Illinois Governor Dan
Walker withheld state funding and plans for the expressway ultimately col¬
lapsed (Shearer 1973). A proposal for a 1992 world’s fair in Chicago met with
a similar fate, due once again to the opposition of a broad-based coalition of
neighborhood organizations (Shlay and Giloth 1987; McClory 1986).
Such protest efforts marked the beginnings of a new variant of urban po¬
litical conflict, pitting business and government leaders against community-
based organizations oftentimes acting outside of formal political channels and
institutions (Boyte 1980; Perlman 1976; Karapin 1994; Bennett 1989). In
many cases, projects orchestrated by growth coalitions served as rallying
points for community opposition (Elkin 1985; Fainstein and Fainstein 1983).
The growing sophistication of these opposition efforts turned many of them
into credible political forces. By the 1980s, neighborhood activism in cities
Community Economic Development and Machine Politics 85

across the nation had significantly compromised the ability of urban growth
coalitions to unilaterally dominate local development agendas (Mollenkopf
1983; Abbott 1987; Hartman 1974; DeLeon 1992).
To some extent, the development of Chicago’s neighborhood movement
fits this model of urban conflict, which portrays cities as divided between sup¬
porters and opponents of downtown growth. However, this chapter raises an
additional suggestion as well. Within Chicago’s emerging neighborhood
movement lay the seeds of something more than simple protest. By the early
1980s, a new coalition of neighborhood organizations had begun to form,
not so much in opposition to growth as in favor of an alternative set of eco¬
nomic development priorities to those being advanced by the city’s tradi¬
tional growth coalition of business, labor, and government leaders. Generally
speaking, these priorities included job creation over real estate development,
the promotion and retention of industrial jobs over service-sector jobs, and
neighborhood development over downtown development. In time, the ideas
and practices generated from within this movement formed a coherent alter¬
native to the corporate-center strategy of the growth machine. I call this al¬
ternative the “local-producer strategy” due to its emphasis, above all, on the
preservation of neighborhood productive capacity.
The coalition forming around a local-producer strategy began to mature
as Chicago was entering a period of regime instability and change. From the
late 1960s onward, the political machine led by Mayor Richard J. Daley ap¬
peared increasingly fragile with each passing election. When Daley died in
1976, it was clear that future Chicago mayors would enjoy nowhere near the
level of centralized power and authority that Daley had commanded for the
previous twenty years. As this chapter will show, imperatives to forge new
electoral and governing coalitions on the part of two of Daley’s successors,
Jane Byrne and Harold Washington, created an opening for a new coalition
of neighborhood-based economic development advocates to become a signif¬
icant new voice in Chicago politics.

Building a Neighborhood Economic Development Agenda


The history of Chicago’s neighborhood economic development agenda is
rooted, in part, in the organizing efforts of the late Saul Alinsky and the style
and philosophy of neighborhood organizing he helped to popularize in the
city.2 Alinsky began his organizing career in 1939 in the same Back of the
Yards neighborhood chronicled by Upton Sinclair in The Jungle and eventu¬
ally shifted his attention to other low- and moderate-income black, Latino,
and white ethnic communities across the city during the 1950s and 1960s,
particularly those experiencing racial transition. By the late 1970s, Chicago
had roughly one hundred active neighborhood organizations founded by
Alinsky or one of his followers (Joravsky 1990, 3).
86 Remaking Chicago

Alinsky’s primary organizational goal was to create independent bases of


power in the city’s neighborhoods, an objective no doubt influenced by the
functioning of mainstream political institutions under Chicago’s Democratic
machine. As Chapter 2 showed, the machine was able to preserve its fragile
electoral coalition of white ethnic and African-American neighborhood sup¬
porters while focusing economic development policies on the downtown area
only by keeping substantive policy issues outside the scope of electoral poli¬
tics. The latter objective was accomplished primarily through a quid pro quo
arrangement in which patronage appointments, city services, and other dis¬
cretionary resources were utilized to generate neighborhood support for the
machine. Alinsky realized that only by building autonomous, community-
based institutions that paralleled and ultimately rivaled those of the machine
could the collective needs and concerns of Chicago’s low- and middle-
income neighborhood residents begin to be addressed. As one veteran neigh¬
borhood activist put it, Alinsky’s organizations were essentially “governments
in exile” (Joravsky 1990, 7).
Alinsky’s organizational strategy, which he honed on the Back of the Yards
neighborhood and then applied elsewhere as well, contained four key compo¬
nents.3 First, existing community institutions such as churches, labor unions,
and block clubs served as the building blocks of his organizations. Second,
fusing these institutions into one body with a collective outlook was accom¬
plished by organizing people around concrete, readily identifiable issues, not
“abstract, mountain-top moralizing” or vague ideological pronouncements.
Third, Alinsky organizations were consumer-based, calling for a redirection
of goods and services in favor of low-income and working-class residential
neighborhoods. Finally, organization victories were achieved through con¬
frontational tactics designed to goad reluctant public officials and other tar¬
gets into acting in the interests of the neighborhood.
This approach was effective to a point, but it also contained certain built-
in limitations. The emphasis on immediate, tangible issues identifiable at the
neighborhood level helped build strong organizations but left more funda¬
mental concerns, such as racism and community power relations, unad¬
dressed. As Harry Boyte points out,

Alinsky’s approach, for all its militancy and iconoclasm, could finally fit into the
traditional interest-group form of American politics, in which different power¬
less groups accept the “givens” of income distribution and corporate structure
and compete for scarce but expanding resources through the use of whatever
tactics they can devise. (Boyte 1980, 52)

To some extent, the parochialism fostered by the Alinsky model of neighbor¬


hood organizing helped perpetuate the system of “city trenches” that have
long divided the neighborhoods of cities like Chicago along racial and ethnic
lines (see Katznelson 1981). In white ethnic neighborhoods experiencing racial
Community Economic Development and Machine Politics 87

transition, Alinsky organizations oftentimes played gatekeeper roles, focusing


on community stabilization efforts designed to preserve the ethnic character of
the neighborhoods they served (Squires et al. 1987, 137; Joravsky 1982).
As the 1960s were drawing to a close, these shortcomings grew increas¬
ingly apparent to many Chicago neighborhood activists, and a number of
coalition-building efforts were undertaken to help bridge racial and ethnic
cleavages. Several multiracial organizational networks, including the West
Side Coalition and the Metropolitan Area Housing Alliance, formed in oppo¬
sition to the “block-busting” and redlining of racially changing neighbor¬
hoods by unscrupulous real estate agents and banks (Metzger and Weiss
1988, 8-9).4 Another citywide coalition, the Citizens Action Program, was
established to challenge a utility rate hike requested by Commonwealth Edi¬
son, the local power company, during the autumn of 1969. Following this
successful campaign, Citizens Action Program focused its efforts on such is¬
sues as redlining and the proposed Crosstown Expressway discussed earlier in
the chapter (Shearer 1973; Pavlos 1975). Finally, the Chicago Central Area
Committee’s Chicago 21 Plan for the redevelopment of the central area
served as a rallying point for coordinated opposition by a number of centrally
located neighborhood organizations during the early 1970s (Betancur, Ben¬
nett, and Wright 1991). ,
Each pf these campaigns united organizations from black, Latino, and
white ethnic neighborhoods around a common agenda, demonstrating the
power of coalition-building and the interests shared by all three communities.
The fight against redlining, for example, helped reveal the political and insti¬
tutional roots of neighborhood deterioration, leading many whites to redi¬
rect their anger away from African Americans to banks, real estate brokers,
and machine politicians (Squires et al. 1987, 139). Through collective action,
racial and ethnic divisions among the neighborhoods grew somewhat less
hardened over the years, and the lines of conflict came to reflect a deepening
division between the city’s growth coalition of downtown business and gov¬
ernment leaders and the neighborhood residents, workers, and small business
owners locked out of meaningful opportunities for participation in the city’s
decision-making structures (Gills 1991; Mier and Moe 1991). Yet as long as
these coalition-building efforts represented voices of protest alone, they
could have little impact on the city’s overall planning and development
agenda other than perhaps to forestall certain pet projects advanced by civic
leaders. To do more than that would require the neighborhoods to articulate
an alternative set of practices and ideas to those emanating from downtown.

Economic Crisis and the Rise of the Local-Producer Strategy


The consumer-based, protest-oriented approach of Alinsky-style organization¬
building was developed within a particular economic context, one in which
the availability of well-paying manufacturing-jobs was a reality for many
88 Remaking Chicago

working-class Chicagoans, particularly those from the city’s white ethnic


neighborhoods. Although growth in industrial employment within the city of
Chicago began to taper off following World War II, it was not until the
1970s and early 1980s that manufacturing jobs experienced a sharp decline.
Table 8 lists Chicago manufacturing employment by sector for the years
1958, 1972, and 1983. Citywide manufacturing employment grew from
329,000 to 389,000 between 1958 and 1972, an average increase of just
over 1 percent per year. From 1972 to 1983, however, the situation wors¬
ened considerably, with total manufacturing employment falling by 131,000
jobs, an overall decline of 34 percent.
In many cases, sectoral declines were interrelated, downturns in major end-
use sectors such as steel, for example, leading to widespread layoffs and plant
closures among local supplier industries. Makers of machinery and machine
tools were particularly hard hit by the decline of the Chicago area steel indus¬
try during the 1970s and early 1980s (Midwest Center for Labor Research
1989, 4). Such ripple effects also extended to neighborhood banks, grocery
stores, and small shopping malls, where laid-off workers from steel and steel-

Table 8. Manufacturing Employment in the City of Chicago,


1958, 1972, and 1983

Industries 1958 1972 1983

Lumber, wood, furniture 8,171 15,300 9,821


Stone, clay, and glass 3,670 4,900 3,797
Primary metals 25,033 31,100 14,185
Fabricated metals 34,232 48,000 28,070
Nonelectrical machinery 30,290 36,200 20,760
Electrical machinery 49,625 46,800 25,539
Transportation equipment 5,660 12,100 7,574
Instruments and misc. 22,738 30,500 16,464

Food and kindred products 51,127 50,900 41,795


Apparel and allied products 21,436 18,200 9,063
Paper, printing, publishing 55,148 71,900 54,675
Chemicals, petroleum, coal 11,803 13,000 14,452
Other nondurables 10,099 10,100 11,817

Total 329,032 389,000 258,012

Sources: Department of City Planning (1960); U.S. Bureau of the Census (1976);
Illinois Department of Employment Security (1983).
Community Economic Development and Machine Politics 89

related industries formerly spent and invested their wages. University of Illi¬
nois economist Joseph Persky estimated that the loss of 20,000 Chicago area
steelworker jobs in 1979 ultimately resulted in an additional 10,000 job losses
within the next year (Squires et al. 1987, 30). Area housing markets were af¬
fected as well. Between 1970 and 1980, Chicago lost an estimated 5 percent
of its rental housing stock, or 40,145 units (Weiss and Metzger 1989, 142).
The downward spiral of Chicago’s working-class neighborhoods during the
1970s was cast into sharp relief by the urban renaissance taking place in the
downtown area, led by the record-setting pace of office-building construction
in the Loop. Economic development plans published during the 1970s and
early 1980s under the Richard J. Daley and Jane Byrne administrations antici¬
pated a restructuring of the city’s economic base from manufacturing to ser¬
vices and proposed downtown-oriented policies and programs designed to ac¬
commodate such a shift (Chicago Central Area Committee 1973; Department
of Planning 1982). However, the service sector was proving itself unreliable as
a source of job creation. Although Loop employment posted a modest gain of
11,574 jobs between 1972 and 1983, the number of nonmanufacturing jobs
citywide actually declined by 45,000 during this period (Illinois Department
of Employment Security 1984; Clavel and Wiewel 1991, 19).
Table 9 shows the impact of industrial decline and a lackluster service-
sector performance on employment in five working-class Chicago neighbor¬
hoods during the 1970s and early 1980s. Between 1972 and 1983, employ¬
ment in these five neighborhoods alone fell by over 63,000 jobs, resulting in
a combined average job loss of 35 percent. By 1982, the city’s overall unem¬
ployment rate had risen to a postwar high of 17 percent (Squires et al. 1987,
43). In addition, due to declining unionization rates and the low productivity
of many of the new service-sector jobs, Chicagoans who did have jobs were
oftentimes working for less. As a result, average real incomes fell from
$20,894 in 1969 to $18,776 in 1979, a decrease of 10 percent (Squires et al.
1987,41).

Table 9. Employment in Five Chicago Community Areas, 1972 and 1983

Community Area 1972 1983 Percent Change

Stockyards 46,026 24,963 -46

Pilsen 37,070 24,295 -35

Garfield Park 35,304 20,886 -41

Wicker Park 34,571 24,641 -29

Logan Square 21,253 16,025 -25

Sources: Illinois Department of Employment Security (1972, 1983).


90 Remaking Chicago

In May 1981, as the crisis was drawing to a head, the Chicago Tribune re¬
sponded with a four-part series entitled “Chicago: City on the Brink.” Tri¬
bune economics correspondent R. C. Longworth seemed to identify the
problem accurately: manufacturing employment in the city’s neighborhoods
had gone into a free fall, and service sector job opportunities in the Loop and
elsewhere in the city were not picking up the slack. However, those called
upon by Longworth to propose solutions were primarily business leaders,
academics, and city planning officials, not representatives from the neighbor¬
hoods themselves. The remedies offered were predictable, ranging from
greater public support for downtown redevelopment, to tax holidays and reg¬
ulatory relief as a means of attracting new industries, to University of
Chicago sociologist Gerald Suttles’ suggestion that the unemployed be en¬
couraged to leave the city altogether. According to Suttles,

I don’t think you should give anybody any incentive to stay where there is no
hope of employment. Our incentives ought to be concentrated on getting them
to move elsewhere. . . . We can be human but we have to say, “Look, it doesn’t
make any sense to keep you here.” (Longworth 1981)

The combination of economic decline and dissatisfaction with outsider-de¬


fined solutions such as these ultimately led many community-based organiza¬
tions to broaden their agendas from protest and advocacy around improved
service delivery and community self-preservation to the actual rebuilding of
neighborhood economies (Betancur, Bennett, and Wright 1991; Wright
1992; Clavel and Wiewel 1991, 25-27). A conceptual shift began to occur,
through which neighborhood activists came to perceive their communities
increasingly as sites for both consumption and production (McKnight and
Kretzmann 1984). In African-American neighborhoods, this shift dovetailed
with ideas that had been emerging from the more militant factions of the
civil-rights movement since the 1960s, emphasizing black cultural and politi¬
cal autonomy and community self-reliance (Gills 1991, 43). By the late
1970s, community development corporations (CDCs) from black, Latino,
and white ethnic areas of Chicago were experimenting with a variety of local
development initiatives, including housing rehabilitation, small-business de¬
velopment, and efforts to assist and retain existing neighborhood retail and
industrial establishments.
The funding requirements of these new undertakings encouraged CDCs
to forge new partnerships with municipal government agencies, which typi¬
cally acted as brokers for federal-funding programs designed to promote
community development. Housing development groups received assistance
from the HUD Section 8, Rental Subsidy Program, which supported the
construction or rehabilitation of 8,976 housing units in Chicago between
1975 and 1980 (Metzger and Weiss 1988, 28). Beginning in 1980, housing
Community Economic Development and Machine Politics 91

CDCs could also apply for HUD self-help grants, worth upward of $100,000
each (Slonka 1981; Betancur, Bennett, and Wright 1991, 205). Both hous¬
ing and business development organizations received funding through
HUD’s Community Development Block Grant (CDBG) program, estab¬
lished in 1974 to provide support for economic development projects in low-
and moderate-income communities (Wright 1992; Betancur, Bennett, and
Wright 1991). Finally, many business development CDCs took advantage of
the Small Business Administration Section 503 Certified Development Com¬
pany Program to help build their business loan-packaging capabilities (Met¬
zger and Weiss 1988, 28-29).
Programs such as these boosted the capacity of neighborhood organiza¬
tions to pursue community-based economic development efforts, but they
also forced organizations to struggle with new tensions between community
organizing and community development. The confrontational negotiating
style favored by Alinsky was generally not the most effective way for organi¬
zations to secure a greater share of CDBG funds and other economic devel¬
opment resources from local public officials, yet local discretion in the use of
federal economic development funds meant that community pressure in one
form or another was essential. In many cases, this proved to be a difficult bal¬
ance to strike (Capraro, Ditton, and Giloth 1985, 40-44; Mott 1984).

Although early community economic development efforts in Chicago


were limited primarily to housing rehabilitation and the development of
small-business ventures (Mier and Wiewel 1983; Slonka 1981), by the early
1980s a growing number of organizations also began providing assistance to,
and advocacy on behalf of, small- to mid-sized neighborhood industries
(Mier and Moe 1991; Mier, Wiewel, and Alpern 1992). The latter efforts
were prompted chiefly by the growing realization on the part of community
economic development practitioners that high-paying blue-collar jobs yielded
larger economic multipliers than the employment provided by most other
types of neighborhood development efforts (Lemonides 1984). Some of the
most creative, experimental approaches to neighborhood economic develop¬
ment that blossomed in Chicago during this period involved the efforts of
these industrial development organizations (IDOs). Space does not allow for
an exhaustive review of these efforts here. However, a brief look at a number
of organizations and strategies provides a useful cross section of this segment
of Chicago’s neighborhood economic development community.
Beginning in the early 1980s, the growing shortage of affordable, centrally
located production space for both new and established manufacturers led a
number of Chicago industrial development organizations to initiate efforts to
develop small-business incubators, typically using older industrial loft build¬
ings in near-downtown areas of the city (Auerbach 1980a; Community Work¬
shop on Economic Development 1987; CapratQ, Ditton, and Giloth 1985).
92 Remaking Chicago

Business incubators provide low-cost production space and business services


for small, start-up industries, generally for a limited duration. The most suc¬
cessful of these projects was begun in 1981 by a West Side organization, the
Industrial Council of Northwest Chicago (ICNC), which used a $1.7 million
grant from the Economic Development Administration of the U.S. Depart¬
ment of Commerce to develop a small-business incubator in a largely aban¬
doned multistory industrial building on the city’s Near West Side.
According to ICNC executive director Ken Govas, some of the primary
advantages offered to incubator firms included low rents of between $1.50 and
$1.75 per square foot and access to shared administrative services. In addition,
ICNC support staff acted as intermediaries among incubator companies and
outside institutions, helping to foster cooperative peer relationships among ten¬
ant firms and providing access to other institutions critical to the success of
newly established businesses, such as local banks, technical assistance providers,
and government agencies. By 1985, nearly half the companies located in the in¬
cubator had developed customer/supplier relationships with other incubator
firms. There was also some cooperative bidding on jobs and some sharing of
employees. During the first four years of its operation only four companies
failed, while total employment increased to five hundred workers (Govas 1995).
The fostering of cooperative and collaborative relationships among manu¬
facturing firms located within close proximity to one another became a grow¬
ing priority for other Chicago industrial development organizations as well
during the early and mid-1980s. These efforts sometimes targeted specific in¬
dustrial sectors, such as metalworking or woodworking. In 1985, for exam¬
ple, a North Side organization, the Jane Addams Resource Corporation
(JARC), began operating a consortium of metalworking firms that grew from
eight to twenty-five members during the first three years of its operation.
North Side metalworking establishments faced a number of challenges by the
mid-1980s. Most were small- to medium-sized “job shops,” producing com¬
ponents for nearby original equipment manufacturers (OEMs) such as
Playskool, Sunbeam, and Stewart-Warner. During the late 1970s and early
1980s, many of the area’s larger OEMs began to source elsewhere or to relo¬
cate out of the city altogether, forcing North Side metalworkers to broaden
their customer bases (McCormick 1994). For many, this required technologi¬
cal upgrades, improved quality standards, and much greater attention to mar¬
keting than before, all costly undertakings involving considerable risk for
small, cash-poor manufacturers (Vindasius 1988).
The networking relationships fostered through JARC’s metalworking con¬
sortium enabled member firms both to share the risks and costs associated
with industry restructuring and to expand market opportunities (Vindasius
1988). According to JARC executive director Michael Buccitelli, informal
networking among owners and operators of small job shops is commonplace.
However, a broker like JARC can help expedite and cement such relation-
Community Economic Development and Machine Politics 93

ships, leading in some instances to highly structured cooperative undertak¬


ings such as new product development. As Buccitelli explains,

We do see a lot more sharing of ideas and solutions [among consortium mem¬
ber firms]. Somebody from a small job shop will go over to another job shop or
consortium member that we’ve introduced them to to see how they’ve accli¬
mated. . . . Certain companies network more than others because of their prox¬
imity to one another. A couple of consortium members located on the same
block have been collaborating on new product development ideas, subcontract¬
ing to one another, maybe borrowing materials or purchasing materials from
each other. A lot of these kinds of relationships are certainly developing. (Buc¬
citelli 1995)

Other Chicago industrial development organizations pursued place-


specific rather than sector-specific industrial-advocacy approaches. One such
strategy that proved to be particularly effective was pioneered in 1982 by a
Northwest Side organization, Greater North Pulaski Development Corpora¬
tion (GNPDC). GNPDC focused its efforts on revitalizing a small, visibly de¬
teriorating industrial district in the Humboldt Park area of the Northwest
Side. Following Alinsky, organization staff members formed a broad-based
advisory committee including area manufacturers, neighborhood residents,
representatives from community organizations and nearby churches, acade¬
mics, real estate brokers, and the ward alderman to draft a development plan
for the area (Center for Urban Economic Development 1983; Lemonides
1984). The broad base of community participation helped mobilize alder-
manic support for the plan, leading to commitments on the part of the city
to provide public infrastructure and improve city services in the area during
the following year (Lemonides 1984). The project eventually became the
model for one of the Harold Washington administration’s key industrial pro¬
grams, the Local Industrial Retention Initiative (Lemonides 1995).
Experimentation such as this both required and helped facilitate consider¬
able learning and capacity-building on the part of many Chicago industrial
development organizations. In many cases, as the policy implications of their
industrial outreach efforts began to unfold, organizations broadened their ef¬
forts to include stronger advocacy roles. For example, a Near North Side or¬
ganization, the Local Economic and Employment Development (LEED)
Council, was established in 1982 partly to improve employment linkages be¬
tween residents of the Cabrini-Green public-housing development and
nearby manufacturers in the North River Industrial Corridor (Ducharme
1991). However, organizers soon discovered the area’s manufacturing base
was being threatened by the conversion of industrial buildings and industri¬
ally zoned property to residential and commercial uses. As a result, LEED
Council staff members began to channel their efforts into a major organizing
94 Remaking Chicago

drive among area manufacturers designed to raise public awareness of such


land-use conflicts and to pressure the city to take steps to preserve the area’s
concentration of high-paying manufacturing jobs (Ducharme 1991; Giloth
and Betancur 1988).5

Coalition-Building among Community Development Corporations


In general, both industrial development organizations and other commu¬
nity-based economic development and housing organizations that formed
during the 1970s and early 1980s tended to be much more open to coali¬
tion-building, networking, and the sharing of resources and information than
the older, more turf-conscious, Alinsky-inspired organizations had been
(Gills 1991, 45). From the mid-1970s onward, a considerable amount of for¬
mal and informal networking among such groups began to take place, in
many cases bridging racial, class, and philosophical divides (Mier and Moe
1991; Mier, Wiewel, and Alpern 1992). This made it possible for the new in¬
sights into community development and industrial retention provided by ex¬
periments such as those described in the previous section to be disseminated
among a wide range of neighborhood development groups. It also made the
city’s corporate-center strategy for downtown redevelopment the subject of a
growing challenge from an increasingly vocal chorus of neighborhood orga¬
nizations rapidly coming together around an alternative set of economic de¬
velopment practices and ideas.
Neighborhood housing development organizations were among the first
community development groups to unite formally, establishing the Chicago
Rehab Network in 1976. The coalition’s initial efforts involved providing
technical assistance to local housing groups and acting as a liaison to pro¬
grams offered by the Chicago Department of Housing (Metzger and Weiss
1988, 27-28; Auerbach 1980b). By the late 1970s, the Rehab Network had
also become a vocal policy advocate for community-based housing initiatives,
pressuring the city for greater community participation in public-housing de¬
velopments, use of the city’s own land inventory for low- and moderate-
income housing, and targeting of capital improvements to leverage private in¬
vestment in affordable housing (Auerbach 1980b; Gills 1991).
Economic development groups coalesced in two citywide umbrella organi¬
zations, the Chicago Association of Neighborhood Development Organiza¬
tions (CANDO) and the Community Workshop on Economic Development
(CWED), formed in 1979 and 1982, respectively. CWED, the more policy-
oriented of the two groups, played a key role in Harold Washington’s 1983
mayoral campaign discussed later in the chapter. CANDO began as a coali¬
tion of neighborhood-based business development groups working to pre¬
serve neighborhood retail strips by engaging in policy advocacy and provid¬
ing a variety of services, such as loan packaging and assistance with storefront
Community Economic Development and Machine Politics 95

modernization, to its membership organizations. By the early 1980s, as its


membership base grew increasingly preoccupied with efforts to halt neigh¬
borhood industrial decline, CANDO expanded its focus to include advocacy
and technical assistance for neighborhood industrial development organiza¬
tions (Chicago Association of Neighborhood Development Organizations
1994; Mier, Wiewel, and Alpern 1992).
The advocacy efforts of these new coalitions and the new experimentation
in neighborhood economic development they supported helped broaden the
downtown-versus-the-neighborhoods debate from a conflict focused predom¬
inantly around questions of redistribution and neighborhood preservation to
one in which the substance of economic development policy also became a
subject for legitimate public discussion and debate. According to former
Harold Washington administration officials Robert Giloth and Robert Mier,

The antinomy of neighborhood versus downtown—a long-standing, urban


grass-roots metaphor—was transformed in Chicago and elsewhere in the 1980s
to portray a new set of development choices: manufacturing versus the service
economy; blue-collar jobs versus low-wage Mcjobs; job generation versus real
estate development; industrial expansion versus downtown growth; credit-
starved neighborhoods versus the growth of the finance industry; targeted local
hiring versus regional business climate; and minority/female business versus ef¬
ficiency. (Giloth and Mier 1989, 185)

By the time of the 1983 mayor’s race, the growing strength, sophistica¬
tion, and political mobilization of Chicago’s neighborhood development
community posed a significant challenge to the political power and influence
of the city’s progrowth alliance and its ability to continue to define economic
development in terms beneficial primarily to its own member organizations
and institutions. Increasingly, proponents of downtown development were
forced to defend their economic programs against a coherent local-producer
strategy for neighborhood redevelopment evolving from within the neigh¬
borhoods themselves. Nothing demonstrated this growing vulnerability more
convincingly than Harold Washington’s successful neighborhood-backed
campaign to unseat the growth machine s candidate of choice in the 1983
Democratic primary, incumbent mayor Jane Byrne.

The Jane Byrne Anti-Coalition


Ironically, Byrne herself had initially been a key beneficiary of the grow¬
ing political mobilization occurring within lower- and middle-income
Chicago neighborhoods. Byrne, a Daley protege and former cochair (with
Daley) of the Cook County Democratic Central Committee, had entered
the 1979 mayor’s race against incumbent Maypr Michael Bilandic, selected
96 Remaking Chicago

by Democratic Party insiders to serve the remainder of Mayor Daley’s term


following Daley’s death of a heart attack in 1976. Bilandic’s political strat¬
egy paralleled that of his predecessor. Under the Bilandic administration, the
narrow governing coalition of business and labor leaders, city bureaucrats,
and machine politicians that had dominated policymaking under Daley’s
tenure as mayor remained largely intact, while patronage appointments and
other discretionary favors were deployed to maintain electoral support in the
neighborhoods (Rakove 1982).
For some time now, however, the electoral component of this strategy had
been showing signs of increasing strain. To begin with, a number of groups
excluded from the machine’s governing coalition had begun to demand col¬
lective benefits and a greater voice in the public-policy debate, while the ma¬
chine continued responding to them as individuals, if at all (Ferman 1991).
Blacks, in particular, energized by the civil-rights movement and Martin
Luther King Jr.’s marches and organizing efforts in Chicago during the mid-
1960s, grew more vocal in their demands for equal access to housing and ed¬
ucational opportunities. Mayor Daley, unwilling to jeopardize his support
among white ethnics, responded by appointing anti-integrationists to the
Chicago Housing Authority (CELA) and the Board of Education, defending
the Chicago Police Department against accusations of discriminatory hiring
and promotion practices, and issuing his infamous “shoot to kill” orders to
Chicago police during the riots following King’s assassination in the spring of
1968 (Ferman 1991, 53; Kleppner 1985, 84-85). As a result of actions like
these, voter support for machine candidates in the city’s predominantly
African-American wards declined precipitously from the late 1960s onward
(Kleppner 1985, 75).
Neighborhood organizations, whose constituencies included a racially and
ethnically diverse range of low-income and working-class communities, rep¬
resented an additional base of support for a potential challenge to the ma¬
chine. Although Chicago’s neighborhood movement had yet to coalesce
around a coherent neighborhood agenda by the time of the 1979 mayor’s
race, growing community demands for access to decision-making bodies and
a greater share of municipal service dollars and economic development funds
further undermined the machine’s ability to mobilize neighborhood support
through the provision of particularistic benefits (Ferman 1991).
For that matter, a number of the key resources that the machine had tradi¬
tionally utilized to reward its supporters were by this time in decline anyway.
Most importantly, political firing was declared unconstitutional in 1972 in
the first of the Shakman Decrees, jeopardizing the Cook County Democratic
Central Committee’s control over an estimated 21,500 patronage jobs (Erie
1988, 152-153).6 In addition, federal urban social programs, which machine
politicians had relied upon heavily to reward and to co-opt minorities, experi¬
enced major cutbacks under the Carter administration from 1977 to 1980
Community Economic Development and Machine Politics 97

(Erie 1988, 184-186). The loss of these key resources, combined with grow¬
ing militancy on the part of neighborhood groups and minorities, signifi-
candy reduced party loyalties during the 1970s.
Jane Byrne entered the 1979 mayor’s race with no organization or
precinct workers and little money. However, she well understood the ma¬
chine’s weaknesses and attacked Bilandic where he was most vulnerable, mak¬
ing strong overtures to the neighborhoods and campaigning actively in the
city’s black wards. Throughout the winter of 1979, Byrne appeared regularly
at block-club meetings, church services, and community celebrations in
African-American neighborhoods on the city’s South and West Sides. She
called, in somewhat unspecific terms, for the establishment of neighborhood
planning councils and increased funding for neighborhood-based economic
development and affordable housing programs. She criticized the level of
public support that the Bilandic administration had provided for downtown
redevelopment, promising a more balanced pattern of development between
downtown and the neighborhoods under a future Byrne administration
(Reardon 1990, 54-55; Kleppner 1985, 115).
Byrne’s campaign received a number of unexpected boosts. First, the winter
of 1979 was unusually harsh, the total season snowfall of eighty-seven inches
more than doubling that of an average Chicago winter. For most of the month
of January, many side streets remained impassable, garbage was left uncol¬
lected, and Chicago Transit Authority (CTA) buses and trains ran hours late.
Then, in late January, CTA officials decided to improve service to its suburban
commuters by closing six inner-city stops in predominandy black neighbor¬
hoods during the morning and evening rush hours. After four days of protests
and the threat of a civil-rights suit, the CTA reversed its decision, but the po¬
litical fallout from the incident would continue to dog Bilandic for the remain¬
der of the election campaign (Rakove 1982; Kleppner 1985, 112-113).
In addition, despite slim margins of victory for machine candidates in the
previous two mayoral primaries, Bilandic and his supporters initially saw
Byrne’s candidacy as so frivolous that they did not even bother to recruit a
black or liberal challenger to divide the antimachine vote among several op¬
position candidates. As Byrne’s candidacy gained momentum during the
snowy winter of 1979, the shortsightedness of this decision became obvious,
but by that point it was too late to remedy the situation. The 1979 Democra¬
tic mayoral primary thus shaped up as a two-way contest between a politically
vulnerable incumbent and a challenger whose appeal was based more on the
rising antimachine sentiment shared by growing numbers of Chicagoans than
on a well-articulated agenda for governing the city (Rakove 1982).
On February 27, 1979, the machine’s slated candidate for mayor was de¬
feated at the polls for the first time in a half century, as Byrne edged out
Michael Bilandic in the Democratic mayoral primary by the slim margin of
16,775 votes (Kleppner 1985, 104). Black voters vented their anger at
98 Remaking Chicago

Bilandic by rallying behind Byrne, handing her majorities in fourteen ol the


city’s sixteen predominantly black wards. Byrne also outpolled Bilandic in the
rnosdy white ethnic wards of the Northwest Side and the white middle-class
wards along the lakefront. Several months later, she trounced her Republican
opponent Wallace Johnson in the general election, winning 82 percent of the
popular vote (Holli 1995; Ciccone 1979).

Planning and Development Policy under the Jane Byrne Administration


Following Byrne’s triumph over the machine, a wave of euphoria swept
over the low-income and working-class neighborhood wards that formed the
base of her electoral coalition, as it appeared that the mayor’s office had fallen
into the hands of someone committed to balancing their interests against
those of the city’s downtown business and political leadership. Indeed, a
number of Byrne’s initial acts as mayor gave her supporters cause for opti¬
mism. At Mayor Byrne’s request, Northwestern University Professor Louis
Masotti formed a broad-based transition team of academics, public adminis¬
trators, and municipal reformers to prepare a four-year plan for reforming
city government (Reardon 1990, 52; Holli 1995, 169). Soon after the elec¬
tion, Byrne also took the first steps toward implementing her campaign
promise to establish neighborhood planning councils, awarding the Chicago
consulting firm of John A. Melaniphy Associates a $600,000 contract to con¬
duct a needs assessment of the city’s seventy-seven community areas and pre¬
pare a feasibility study examining various approaches to community planning
(Reardon 1990, 56).
Yet the high hopes of Byrne’s neighborhood supporters were quickly
dashed. The new mayor soon reversed her support for neighborhood plan¬
ning councils with substantive decision-making powers. The release of the
Melaniphy report was suppressed, and Professor Masotti’s transition team
was ultimately given the cold shoulder by the new administration (Holli
1995, 169; Reardon 1990, 56). In the meantime, Byrne undertook steps to
strengthen the administration’s ties with members of the city’s traditional
power structure. She watched quietly in the aftermath of the election as the
Democratic machine’s leadership in city council awarded key committee
chairmanships to its most ardent supporters, taking no action to secure lead¬
ership positions for her own backers in city council. She soon developed close
working relationships with prominent machine aldermen such as Eddie Vr-
dolyak and Edward Burke, individuals she had singled out during her elec¬
tion campaign as part of an “evil cabal of men” unsympathetic to the needs
of the city’s neighborhoods (Kleppner 1985, 117; Reardon 1990, 49-51).
Byrne also moved quickly to mend fences with members of the downtown
business community, some of whom had expressed concern Nover her an¬
tibusiness campaign rhetoric (Gruber 1979)7 Over the opposition of histori¬
cal preservationist groups and neighborhood organizations that had sup-
Community Economic Development and Machine Politics 99

ported her election bid, Byrne sought approximately $120 million in direct
subsidies and tax abatements for a major redevelopment project in the North
Loop she had criticized during her election campaign (Bennett et al. 1988).
She supported the developers of Presidential Towers, a $200 million, 2,346-
unit upscale housing development on the Near West Side, by helping to
arrange an insured, low-interest $158 million federal mortgage and a low-in¬
terest $L80 million construction loan financed through the sale of tax-ex¬
empt bonds (Ferman and Grimshaw 1992).8 Finally, she gave the administra¬
tion’s backing to plans for a 1992 Chicago world’s fair, spearheaded by
former Commonwealth Edison utility executive Thomas Ayers with the sup¬
port of other key members of Chicago’s growth coalition. The proposal,
which sparked the mobilization of a citywide coalition of opposition groups,
involved an estimated cost of between $885 million and $1 billion and called
for the razing of viable community areas and businesses on the city’s Near
South Side (Shlay and Giloth 1987).
Efforts such as these were consistent with the policies articulated in the
major planning document issued by the Byrne administration, the Chicago
1992 Comprehensive Plan, which committed the administration to using its fi¬
nancing powers and controls over land use and public infrastructure to sup¬
port the plans and objectives of the downtown business community (Depart¬
ment of Planning 1982). The plan represented a ten-year redevelopment
program for the city, culminating in the 1992 Chicago World’s Fair. Its vision
of Chicago in the year 1992, like earlier economic development and land-use
plans published during the Richard J. Daley administration, was that of the
postindustrial city, driven by the growth of its finance, trade, and service in¬
dustries downtown.
While expressing a commitment to the retention of manufacturing jobs in
the neighborhoods, Mayor Byrne supplied few supportive resources or pro¬
grams. Of a total of $163 million budgeted for citywide economic develop¬
ment expenditures in the Byrne administration’s 1981-1985 Five Year Capi¬
tal Improvements Program, $70 million (43 percent) alone was targeted lor
redevelopment projects in the Loop (City of Chicago 1981). Meanwhile, in
an apparent attempt to shore up her electoral standing in low- and moderate-
income communities, Byrne channeled CDBG funds into traditional services
such as street cleaning and snow removal instead of economic development
projects, continuing the machine’s tradition of currying favor with its lower-
income and working-class constituencies through service provision and other
localized benefits (Rich 1993, 200).

Interpretations of the Byrne Administration


Mayor Byrne’s alliance with the downtown business community and ma¬
chine leadership in city council following an election campaign in which she
promised a more equitable distribution of resources and policies seems
100 Remaking Chicago

consistent with the views of most contemporary urban political economists,


who, as we saw in Chapter 1, tend to be largely pessimistic about the long¬
term prospects of neighborhood-based governing regimes. City officials lack
the authority to rule by themselves, a situation that leads them to seek part¬
nerships with nongovernmental actors in possession of resources necessary
for governing (Stone 1989; Elkin 1985; Stone, Orr, and Imbroscio 1991).
Although many resources and activities are essential for the well-being of
cities, the need for economic growth to provide jobs and to maintain tax rev¬
enues means that economic resources are particularly valuable (Stone 1993;
Fainstein 1990). According to political scientist Clarence Stone, the concen¬
tration of economic resources in the hands of the downtown business com¬
munity produces a “strong tug” toward some form of business/government
alliance and gives populist regimes a “built-in tendency towards instability”
(Stone 1987b, 286-287). Campaign promises for fairness and equity in city
planning and policymaking tend to dissipate quickly once the election is over
and the need for policies to support downtown development becomes evi¬
dent (Stone 1987b; Elkin 1985; Logan and Molotch 1987).
A number of explanations for Byrne’s behavior in the aftermath of her
election victory have been advanced, some of which are consistent with this
land of reasoning. For example, Holli (1995) and Reardon (1990) have both
emphasized the city’s deteriorating fiscal position in the late 1970s. Shortly
after Byrne assumed office, her financial advisors discovered a previously
undisclosed $100 million deficit in the city budget. Public revelation of the
deficit caused Standard and Poor’s to lower Chicago’s general obligation
bond rating from AA to A-, and Byrne is said to have been forced away from
a redistributive agenda under pressure from the city’s creditors to demon¬
strate her administration’s commitment to restoring the city’s fiscal integrity.
It has also been suggested that Byrne’s electoral coalition of blacks, lake-
front liberals, and Northwest Side white ethnics was too diverse to refashion
into a governing coalition, leaving Byrne with little alternative but to seek al¬
liances with the very elements of Chicago’s power structure she had run
against during her election campaign (Rakove 1982; Ferman 1991; Kleppner
1985, 118). According to political scientist Milton Rakove,

[Byrne’s electoral coalition] could only come together for winning an election
or, rather, defeating an incumbent. It was not a coalition with which Byrne (or
anyone else) could govern the city, unless she was prepared to undertake drastic
changes in die thrust of public policies. And if she did that, she would immedi¬
ately alienate segments of the coalidon, since they had contradictory ideas of
what should be done and inherendy conflicting interests. (Rakove 1982, 230)

Each of these arguments would be more convincing if it were not for the
fact that Harold Washington, faced with similar pressures and constraints af-
Community Economic Development and Machine Politics 101

ter assuming the mayor’s office in 1983, behaved much differently than
Byrne. Washington’s electoral coalition of blacks, Latinos, and lakefront lib¬
erals was similar in its diversity to that of his predecessor, yet, unlike Byrne,
Washington forged a governing coalition that included each of these groups.
Moreover, he did so despite inheriting a $121 million budget deficit from the
Byrne administration and facing a disappointing B+++ Standard and Poor’s
debt rating for much of his tenure as mayor.9
The key explanation for Byrne’s behavior lies not so much with the city’s
budgetary position or with “irreconcilable differences” among members of
her electoral coalition, although each of these pressures was certainly a con¬
tributing factor. More importantly, Byrne and her supporters had simply made
no attempt to develop a positive program for governing the city once she was
in office. Her electoral coalition was an “anti-coalition,” united around its dis¬
satisfaction with the Bilandic administration’s downtown-oriented policies and
neglect of neighborhood concerns but lacking a shared consensus around an
alternative set of governing principles.10 Such a consensus was not an impossi¬
ble one to achieve, as the following discussion of Harold Washington’s may-
oral campaign illustrates. It would, however, require both a stronger commit¬
ment to the neighborhoods than Byrne appears to have held and a
well-orchestrated effort by the neighborhoods themselves to advance a coher¬
ent economic development policy agenda of their own. Not until the 1983
mayor’s race would such an agenda become clearly recognizable.

CWED and the Harold Washington Mayoral Campaign


Jane Byrne’s insensitivities to the needs and interests of the lower-income
and working-class communities largely responsible for her 1979 election vic¬
tory fueled a growing sense of outrage among these constituencies, as it be¬
came increasingly evident that the new mayor had no intention of delivering
on her campaign promises for more fairness and equity in city government.
This frustration was compounded by newly elected President Ronald Rea¬
gan’s policy of “New Federalism,” which featured cutbacks in federal funds
for community development and social services, and a market-based ap
proach centered around the creation of “enterprise zones” to address the
problems of disinvestment and deindustrialization in urban communities.
Neighborhood activists protested the budget cuts and voiced concerns that
the tax breaks and regulatory relief offered to firms within the new enterprise
zones would encourage destructive competition among inner-city communi¬
ties (Giloth and Mier 1989).
In early 1982, a church-based civic organization called the Community
Renewal Society invited several hundred representatives from Chicago-based
neighborhood organizations to a conference whose purpose was to develop a
citywide neighborhood response to the enterprise zone legislation then under
102 Remaking Chicago

consideration in the Illinois state legislature. At the conference, it became


clear that, in addition to the critique of enterprise zones, there was an interest
in developing a positive set of policy alternatives based on the new commu¬
nity development efforts with which neighborhood organizations from across
the city had been experimenting (Mier, Wiewel, and Alpern 1992). At this
point, no such agenda had yet been advanced in Chicago on a citywide basis.
As one conference participant put it,

There’s a sense in which this enterprise zone idea made us aware, for the first
time, of what a policy gap there was. Didn’t we have a set of ideas about how to
develop communities and create jobs that would put this to shame? So we sort
of got backed into understanding what a policy void there was. At that point in
the early 1980s, the first CDCs had only been in existence for probably ten
years, so it wasn’t like there was a huge, decades-long set of experiences. The
idea that you could knit together practices that were being invented in commu¬
nity development, kick it up to another level and say, “What policy implications
might there be here?” was a new idea. (Kretzmann 1996)

During the next eight to ten months, organization leaders continued to


meet on a regular basis, eventually uniting formally as the Community Work¬
shop on Economic Development (CWED). A key purpose of the meetings
was to produce a policy statement that incorporated the experiences and pol¬
icy concerns of both organization leaders and their constituencies. By late
summer of 1982, the main principles and recommendations of the CWED
policy statement had been put into writing (Community Workshop on Eco¬
nomic Development 1982).11 The document identified seven primary goals
and seventeen specific policy recommendations through which these goals
were to be implemented. Included were calls for greater community owner¬
ship and control over economic resources, more community involvement in
the planning of economic development programs, policies designed to pro¬
mote full employment, and a shift of public resources in favor of neighbor¬
hood development projects. “Top-down” solutions to community develop¬
ment were explicitly rejected in the statement’s preamble, which stated:

During the past two decades, community-based organizations throughout


Chicago have accumulated a vast store of knowledge and experience about eco¬
nomic development. Their understanding of programs that have been successful
or unsuccessful, combined with direct information about community needs and
strengths, provides them with the expertise and perspective necessary for the
design of appropriate policies that can foster self-sustaining economic develop¬
ment in existing Chicago communities. This point also leads to the conclusion
that communities must participate actively in locally initiated and controlled
planning for community economic development. (Community Workshop on
Economic Development 1982)
Community Economic Development and Machine Politics 103

At the same time that CWED-member organizations were building a con¬


sensus around the principles contained in the CWED policy statement, many
black Chicagoans were beginning to mobilize around an additional set of
concerns more specific to African Americans. In July 1982, Mayor Byrne
capped a series of high-profile snubs of the black community by replacing
two African-American members of the Chicago Housing Authority with
whites, tipping the balance of the authority from a black majority to a white
majority (Kleppner 1985, 136-146). In response, a wide range of commu¬
nity organizations led by Jesse Jackson’s Operation PUSH formed the Peo¬
ple’s Coalition to Boycott ChicagoFest, a week-long outdoor music festival
sponsored by the Mayor’s Office of Special Events. The boycott seemed both
to energize and to galvanize the black community, cutting across longstand¬
ing divisions and ultimately fueling a voter registration drive launched earlier
in the summer (Rivlin 1992). By November 1982, black voter registration
since the 1979 mayoral election had increased by 127,000, while voter regis¬
tration among white Chicagoans rose by only 1,656 during the same period
(Kleppner 1985, 148).
The success of the drive convinced many African-American leaders that the
mayor’s office was within their grasp (Mier and Moe 1991; Rivlin 1992).
The favored candidate among black Chicagoans was Harold Washington, at
the time a U.S. Congressman from the 1st Congressional District on
Chicago’s South Side. Washington had strong ties to many of the city’s black
neighborhood organizations, but he was also well-known and well-liked by
progressive whites. On November 10, 1982, after considerable prodding by
neighborhood activists and African-American political leaders, he formally
announced his entry as a candidate in the upcoming Democratic mayoral pri¬
mary scheduled for the following February.
Washington quickly appointed two individuals active in CWED, Hal
Baron and Kari Moe, to leadership positions on the campaign’s Research and
Issues Committee, where the central policy proposals of the campaign plat¬
form were to be developed. At Baron’s suggestion, Washington agreed to
forego the conventional approach to issues development, which was to as¬
semble a small group of “experts” compatible with the candidate’s politics to
write position papers. Instead, roughly fifteen issues teams staffed by ten or
more individuals each were designated to conduct research and write papers
on specific topics such as energy, housing, transportation, neighborhoods,
and economic development. The papers drafted by each team were published
collectively as the “Washington Papers,” the campaign’s central policy docu¬
ment (Mier and Moe 1991).
The economic development issues team included a number of additional
CWED participants and ultimately served as a gateway for certain key ideas
and proposals voiced in the CWED policy statement to find their way into the
Washington Papers.12 One of the most important of these ideas was the argu¬
ment that job creation, not real estate deyelopment, should be the long-term
104 Remaking Chicago

goal of economic development policy. Public policy had operated for years un¬
der the assumption that these two goals were synonymous—that employment
opportunities for Chicago residents were an obvious corollary of policies that
encouraged the development of property for its “highest and best use.” In
severing the connection between the two and identifying job creation as the
long-term policy objective, the Washington Papers set the stage for a more
open public debate over the shape of the city’s economic future. In cases
where the land-use preferences of commercial or residential property develop¬
ers and those of manufacturers conflicted, for example, property values would
no longer represent the sole standard against which public decisions would be
determined and evaluated. This was a radical departure from the reigning ide¬
ology of Chicago’s progrowth alliance. Other priorities voiced in the Wash¬
ington Papers—including full employment, an emphasis on neighborhood de¬
velopment over downtown development, and a commitment to retaining
traditional manufacturing industries—promised a similar departure from the
economic development policies, practices, and ideological claims of previous
governing coalitions.
The Washington Papers were important as well because they provided the
campaign with a positive set of governing principles. The broad-based, par¬
ticipatory framework through which the campaign issues papers were devel¬
oped helped bring together the diverse strands of Washington’s electoral
coalition around a common agenda for governing the city. It meant that
Washington’s electoral coalition, unlike that of Jane Byrne, would be more
than an anti-coalition, united in its dissatisfaction with the status quo but
lacking a coherent set of positive policy alternatives. Ultimately, many of the
key ideas, policies, and programs appearing in the Washington Papers—
including the emphasis on jobs over real estate development—were repro¬
duced in the Washington administration’s economic development plan,
Chicago Works Together: 1984 Chicago Development Plan (City of Chicago
1984). Policies articulated in the Washington Papers and the 1984 Chicago
Development Plan not only helped to establish the administration’s eco¬
nomic development agenda but also could be conveniently held over the ad¬
ministration’s head by members of the Washington coalition when the pres¬
sures of governing began to push the administration in less progressive
directions (Ducharme 1991).
Not surprisingly, the Washington Papers did not play well within the city’s
downtown business community, which divided its support between Washing¬
ton’s two rivals in the Democratic mayoral primary, Jane Byrne and Richard
M. Daley, the eldest son of the late mayor. It was a campaign marked by deep
divisions along class and racial lines. In an editorial entitled “White Business
and Black Mayors,” Crain’s Chicago Business openly questioned whether the
city’s white business establishment could work with an African-American
mayor.
Community Economic Development and Machine Politics 105

[Washington] has to win the support of a basically suspicious business commu¬


nity, which in the final analysis, is as it should be. Rep. Washington comes from
a neighborhood—culturally and politically—that is foreign and forbidding to
white business people. They harbor a very real fear that they will be unable to
deal with Rep. Washington’s administration because they don’t know the terri¬
tory, and they fear also that they’ll be unwelcome. {Crain’s Chicago Business, 13
December 1982)

Speaking on behalf of the Byrne campaign to an assembly of two hundred


Northwest Side precinct captains, Democratic Party Chairman Eddie Vr-
dolyak was considerably less subde, urging his troops on with the following
words: “It’s a racial thing. Don’t kid yourself. I’m calling on you to save your
city, to save your precinct. We’re fighting to keep this city the way it is”
(Rivlin 1992, 155).
In the end, Washington’s narrow victories in the Democratic primary and
general election several months later were due to the mobilization of many of
the same forces responsible for Jane Byrne’s upset over Michael Bilandic in
the 1979 mayor’s race. Voter turnout among blacks in the 1983 general elec¬
tion was a remarkable 85 percent, with 99 percent of those votes cast for
Washington over his white Republican challenger, Bernard Epton. Washing¬
ton als9 received 82 percent of the Latino vote and up to one-third of the
white vote in the north lakefront wards (Rivlin 1992, 196-197).
With Harold Washington’s election victory in the spring of 1983, the
community development agenda seemed to have come full circle. What had
begun as a set of practices and ideas within neighborhood development orga¬
nizations, whose policy implications were collectively debated and hammered
out in the forum provided by CWED, had now penetrated the mayor’s office
on the wings of the Washington campaign. However, few neighborhood ac¬
tivists were naive enough to believe that Washington’s victory marked an end
to the longstanding political influence held by the various members of
Chicago’s growth machine. For most, it was clear that Washington’s govern¬
ing agenda would ultimately involve some form of compromise between the
downtown business leadership, the administration, the machine faction of
city council, and community leaders. As Washington was sworn in as mayor
of Chicago on April 29, 1983, neighborhood activists waited anxiously to see
what form that compromise would take.

Conclusion
Chicago’s governing arrangements, remarkably stable for decades follow¬
ing the election of Richard J. Daley in 1955, fell increasingly into crisis by
the early 1980s. It was a crisis at once economic, social, political, and ideo¬
logical. Court rulings that declared political patronage unconstitutional and
106 Remaking Chicago

cutbacks in federal aid to cities undermined the ability of the machine to re¬
ward its supporters in the neighborhoods and discourage group mobilization
around collective demands. In addition, the contrast between the deindustrial¬
ization and soaring unemployment rates in the neighborhoods and the row-
upon-row of glistening office towers rising in the Loop threatened to open
the city’s role in downtown redevelopment and economic restructuring to
greater public scrutiny and debate. Finally, the failure of the downtown service
sector to provide jobs for displaced manufacturing workers undermined the
credibility of arguments that all Chicagoans benefitted from economic devel¬
opment policies focused primarily on revitalizing the central business district.
These developments combined to fuel a neighborhood movement whose
base was, by this time, already well-established. They also helped expand the
residential base of the movement to include growing numbers of manufac¬
turers and other community business establishments that, like neighborhood
residents, were shut out of the closed decision-making process fostered by
the politics of downtown growth. The outreach efforts of industrial devel¬
opment organizations like Jane Addams Resource Corporation, the Indus¬
trial Council of Northwest Chicago, and Greater North Pulaski Develop¬
ment Corporation represented a turning point for the neighborhood
movement in Chicago, helping to redefine the relationships of neighbor¬
hood manufacturers to one another, to the community, and to city govern¬
ment. As a result, the downtown-versus-the-neighborhoods debate shifted
onto new terrain. The significance of neighborhoods as production sites was
incorporated into the neighborhood critique of the growth machine, and
the existing residential/consumer base of the movement was expanded to
include a strong producer component as well. Calls for redistribution be¬
came liberally supplemented with new demands for balanced growth, as a
local-producer strategy for neighborhood redevelopment emerged to chal¬
lenge the practices and ideologies of the progrowth alliance.
The Jane Byrne victory of 1979 provided an indication of the new political
possibilities. However, it was not until four years later, when Harold Wash¬
ington was elected mayor, that Chicago would experience a fundamental
regime change. Under Washington, the alliance between business and gov¬
ernment leaders was replaced by a new governing coalition in which neigh¬
borhood organizations figured prominently. “Balanced growth” would be¬
come the centerpiece of Washington’s economic development platform, and
downtown business elites and their allies would be forced into a debate with
neighborhood activists over how that concept should be operationalized. In
the context of this debate, the beginnings of a citywide industrial policy
would emerge. The following chapter explores these beginnings.
Chapter Five

The Battle for the Near North Side

A
wL JLs we saw in Chapter 1, urban political economists provide com¬
pelling reasons for the dominance of land-based interests in city politics. Key
among these is the argument that city officials and landed interests both share
an overriding concern with the intensification of land use (Molotch 1976; Lo¬
gan and Molotch 1987; Domhoff 1983, 166-173). Property owners want to
attract “higher and better” uses to enhance the value of their real estate hold¬
ings, while city officials require healthy rates of economic growth to boost tax
revenues and provide jobs for city residents. Efforts to rearrange and intensify
land use accomplish each of these objectives by increasing the city property-
tax base and creating temporary construction jobs and new forms of perma¬
nent employment. Capital mobility cements the alliance between city officials
and landed interests, since both groups share an interest in attracting mobile
investors to maintain high-intensity land uses and neither can achieve this goal
without the cooperation of the other (Elkin 1985, 1994).
According to most urban theorists, alliances of this nature are the perva¬
sive type of governing coalition in contemporary cities not simply because
they are mutually beneficial but also because few viable alternatives exist.
Nonbusiness groups like neighborhood associations may hold useful organi¬
zational resources, but these are no match for the substantial economic hold¬
ings of landed interests and their allies in the downtown business community
(Stone 1989, 1993). City officials must promote economic growth, and the
pressures of global economic restructuring and deindustrialization have, by
most accounts, made the growth machine’s corporate-center strategy for
downtown redevelopment the only game in town (or, at least, the only
winnable game). Under these circumstances, even administrations with a pro¬
gressive bent can do little more than redistribute some of the benefits of
downtown growth in favor of low- and moderate-income neighborhoods
(Dreier and Erlich 1991; Turner 1992; Wong 1988).
108 Remaking Chicago

For decades following World War II, Chicago’s governing arrangements


conformed well to this model. Yet the election victory of Harold Washing¬
ton in 1983 seemed to represent a potential turning point. Washington had
incorporated themes from community economic development circles—like
balanced growth, jobs over real estate, and manufacturing over services—
into his election campaign. Such themes set up a potential conflict with the
real estate-based growth coalition’s goal of rising property values, since the
preservation of well-paying manufacturing jobs required that land values
and property taxes remain affordable for industrial users. Indeed, as the case
studies of the printing and apparel industries in Chapter 3 illustrate, land
speculation was perhaps the central cause of industrial displacement in
Printing House Row and the Chicago garment district. By drawing a dis¬
tinction between jobs and real estate development and prioritizing the for¬
mer, Washington set his administration on a new course in which land-
based interests suddenly found themselves in the unfamiliar position of
political outsiders.
If the arguments and assumptions of contemporary urban political econ¬
omy are correct, however, property developers and their progrowth allies
should not have remained outsiders for long. Faced with difficult choices be¬
tween growth and redistribution, Washington would have found himself
drawn increasingly into alliance with members of the business community
holding the economic resources necessary for governing. Fortunately, certain
events that transpired during the Washington years provide useful empirical
evidence with which to test this proposition. Shortly after Washington as¬
sumed office, significant conflict erupted on the city’s Near North Side be¬
tween property developers seeking to redevelop industrial space for new
commercial and residential uses and manufacturers attempting to maintain a
foothold in this area of the city in the face of intensive real estate speculation.
The city’s zoning policy became an arena of political conflict, and the admin¬
istration was forced to make hard choices between a real estate-based devel¬
opment strategy backed by powerful members of the city’s downtown
growth coalition and a jobs-based development strategy supported by a coali¬
tion of manufacturers, workers, and neighborhood organizations.
This chapter questions whether the dynamics of that conflict, the adminis¬
tration’s response, and the alternative political and economic possibilities to
which it pointed can be grasped within the framework provided by contem¬
porary urban political economy. Contrary to the scenario painted in much of
the literature today, the options that policymakers faced forced them to
choose not between growth and redistribution but between two growth
strategies benefitting different segments of the community. Moreover, the
choice between these growth strategies was made politically through an ex¬
pansion of the city’s controls over land use, not through Darwinian competi¬
tion in the marketplace. The following narrative portrays urban economic re-
The Battle for the Near North Side 109

structuring as open-ended and politically contested, suggesting that opportu¬


nities for meaningful development choices continue to exist within the con¬
text of global economic change.

The Harold Washington Administration:


Broadening the Economic Development Debate
Studies of community power have often shown how economic elites domi¬
nate the public agenda at the expense of low- and moderate-income city resi¬
dents or workers by limiting the scope of legitimate public debate over redis¬
tributive or regulatory questions or by preventing conflict from arising in the
first place (Crenson 1971; Bachrach and Baratz 1962; Gaventa 1980). How¬
ever, power relations may also be a factor in determining the range of public
debate over economic development strategies themselves. Chicago mayors
Richard J. Daley, Michael Bilandic, and Jane Byrne each presided over nar¬
row governing coalitions dominated by a select group of land-based interests
drawn from the downtown business community. The use of public policy to
support the commercial and residential downtown redevelopment agenda of
this segment of the business community was possible, in part, because of a
narrow public decision-making process in which the proponents of alterna¬
tive development strategies did not participate in any significant form.
As we saw in the previous chapter, Harold Washington’s electoral coalition
included a diverse group of community-based organizations advocating,
among other things, a local-producer strategy for neighborhood redevelop¬
ment. An important part of this agenda, as articulated in the Community
Workshop on Economic Development (CWED) policy statement and later in
the “Washington Papers” themselves, was the argument that traditional
Chicago industries could survive under the appropriate political conditions
and that city government should expand its one-dimensional focus on down¬
town growth to include greater attention to the needs of city manufacturers.
While mindful of the need to maintain the confidence and support of the
downtown business community, Washington nevertheless undertook a num¬
ber of steps during his first term in office that helped place these concerns
within the parameters of legitimate public discourse. Taken together, these
steps helped create a fertile political climate for manufacturers threatened by
nearby commercial and residential redevelopment to demand and receive
protection from city government.
To begin with, Washington placed a number of community leaders in
prominent positions within the city bureaucracy. The key appointment was
that of Robert Mier as Commissioner of the Department of Economic De¬
velopment (DED). In 1978, Mier had helped establish the Center for Urban
Economic Development at the University of Illinois at Chicago, a think tank
and technical-assistance provider for community-based organizations. Mier
110 Remaking Chicago

was also a founding member of CWED and helped draft the economic devel¬
opment platform of the “Washington Papers.” Mier’s appointment, together
with those of other community-minded academics and neighborhood advo¬
cates such as Robert Giloth, Arturo Vazquez, and Rari Moe, created outposts
in the bureaucracy where elements of an industrial policy could be articulated
and advanced.
In May 1984, the Washington administration released its economic devel¬
opment plan, Chicago Works Together: 1984 Chicago Development Plan (City
of Chicago 1984). As we have seen earlier, the development plans issued by
the Richard J. Daley and Jane Byrne administrations had focused overwhelm¬
ingly on real estate development, with particular emphasis placed on the
downtown and near-downtown areas of the city (Department of City Plan¬
ning 1958; Department of Planning 1982). In keeping with the policy objec¬
tives identified in the “Washington Papers,” the Washington administration’s
plan prioritized job creation over real estate development and emphasized the
need to retain traditional manufacturing industries in the city’s neighbor¬
hoods. The plan played an important role in framing the economic develop¬
ment policy debate in Chicago during the Washington years. For example,
one of the major downtown civic organizations, the Commercial Club of
Chicago, subtitled its own development plan released in December 1984
“Jobs for Metropolitan Chicago” (Commercial Club of Chicago 1984).
One aspect of the Washington administration’s strategy for retaining man¬
ufacturing jobs involved the creation of task forces for two prominent
Chicago industries, steel and apparel, and the funding of a printing industry
task force convened by the Center for Urban Economic Development at the
University of Illinois. While the efforts of each task force culminated in a
number of significant policy innovations, one of their most important func¬
tions was to help defeat the common perception that all manufacturing in
Chicago was either dead or dying (Giloth and Mier 1989). Each task force
produced a substantial research paper and a smaller document that summa¬
rized the research findings and outlined policy recommendations. In each
case, researchers successfully identified key industry segments seemingly unaf¬
fected by the decentralizing pressures of global economic restructuring (see
Markusen 1985; Robinson 1985; Ranney and Wiewel 1987). For example,
the printing industry task-force report pointed out how technological inno¬
vations had opened up new markets for small, centrally located graphic com¬
munications firms for whom face-to-face communication with downtown
customers was essential, making this one of the strongest industry segments
(Center for Urban Economic Development 1988). The administration
helped generate publicity and support for task-force findings by issuing press
releases and holding briefings for editors from the major Chicago newspa¬
pers, local labor leaders, and city council members in conjunction with the
release of individual task-force reports (Mier 1986).
The Battle for the Near North Side ill

Finally, the Washington administration significantly boosted the organiza¬


tional capacity of a number of industrial development organizations through
the Local Industrial Retention Initiative (LIRI), introduced in 1983 by DED
Commissioner Mier. The LIRI program, modeled after Greater North Pu¬
laski Development Corporation’s target area industrial improvement project
described in Chapter 4, funded industrial development organizations to orga¬
nize manufacturers around policy issues of concern to neighborhood indus¬
tries. Previously, city economic development officials had attempted to ad¬
dress the concerns of manufacturers by contacting them individually and then
developing solutions to industry problems within the bureaucracy. By con¬
trast, the Washington administration’s LIRI program created a central plan¬
ning role for manufacturers themselves, providing them with both a strong
sense of ownership of the process and an organizational presence with which
to pressure the city to make good on commitments for funding, infrastruc¬
ture, and other assistance (Reardon 1990, 165-170). In 1983, LIRI’s first
year, four groups received a total of $76,000 in CDBG funds. By 1987, the
number of LIRI groups had grown to eight, while the program’s annual
budget had increased to $387,500 (Alter, Wiewel, and Alpern 1992, 90).

Industrial Displacement:
Identifying the Problem and Developing a Solution
One of the LIRI organizations that the Washington administration began
funding early in its first term was a Near North Side industrial development
organization called the Local Employment and Economic Development
(LEED) Council. One of the most diverse areas of the city, the Near North
Side is home to the affluent residential community of Lincoln Park near the
lakeshore, the Cabrini-Green public housing development, and the largest in¬
dustrial corridor on Chicago’s North Side. Founded in 1982 as a unit of the
New City YA1CA, LEED Council was established in part to help neighbor¬
hood residents and businesses design a community economic development
strategy that would link the low-income residents of Cabrini-Green with em¬
ployment opportunities in the nearby North River Industrial Corridor
(Ducharme 1991, 225).
The North River Industrial Corridor encompasses 566 acres of land,
roughly half of which lie within the northern portion of Chicago s central area,
the rest extending for another several miles along the Chicago River to the
northwest. The boundaries of the corridor are, roughly, Wellington and
Chicago Avenues to the north and south, and Clybourn Avenue and the
Kennedy Expressway to the east and west (see Map 3). Twenty-seven industrial
sectors are represented in the corridor, with major employers including piimary
and fabricated metal products, printing and publishing, rubber and plastics,
and food products (Department of Planning and Development 1992, 15).
112 Remaking Chicago

Map 3. North River Industrial Corridor with Clybourn Corridor Sub-Area


The Battle for the Near North Side 113

In 1986, industrial firms located within the North River Corridor em¬
ployed roughly 16,000 manufacturing workers, representing somewhere be¬
tween 8 and 10 percent of Chicago’s total manufacturing jobs at the time
(Ducharme, Giloth, and McCormick 1986, 18). Until 1983, the zoning clas¬
sifications for the entire corridor were either Ml, M2, or M3, dividing the
corridor into light, medium, and heavy manufacturing districts. Apart from
its industrial zoning, a number of other characteristics had long attracted
manufacturers to this area of the city, including its proximity to the central
business district, waterway access to the Great Lakes and the Mississippi River
via the Chicago River, and a rail spur from the Soo Line Railroad (Depart¬
ment of Planning and Development 1992, 11-14).
Yet the North River Corridor also had features that would eventually make
it attractive to nonindustrial users as well. In particular, portions of the corri¬
dor contained concentrations of the multistory industrial loft buildings that
real estate developers in areas nearby had recently begun to target for conver¬
sion to more lucrative commercial and residential uses. Although such activi¬
ties were still generally concentrated closer to the Loop by the time LEED
Council was founded in 1982, a recent flurry of loft conversions just south of
the North River Corridor in the River North area suggested that redevelop¬
ment was, by this time, poised to strike. Between 1977 and 1984, over seven
thousand manufacturing and durable wholesaling jobs were lost in River
North, many the result of adaptive reuse of the area’s numerous industrial loft
buildings for nonmanufacturing purposes (Giloth and Betancur 1988, 284).
In 1983, developers Tern Horowitz and Robert Matthews approached
forty-third ward alderman Martin Oberman with the first request for a re¬
zoning of industrial property in the North River Corridor. The developers
wanted to convert a vacant piano factory on Clybourn Avenue to residential
lofts and needed the property rezoned from manufacturing to residential in
order to proceed with their plans. LEED Council staff members quickly or¬
ganized companies near the proposed Clybourn Lofts development, most of
whom were opposed to the zoning change due mainly to fears that noise and
odors produced by nearby industries would provoke complaints from the
new loft dwellers. In response, the developers agreed that all loft purchase
agreements would include a clause indicating the buyers understood they
were purchasing property in a manufacturing area. LEED Council ultimately
withdrew its opposition to the development, and Alderman Oberman agreed
to support the zoning change.
During the next two years, however, development hit Clybourn Avenue
and its immediate surroundings at a brisk pace. Oberman’s office received no
less than seven additional requests for zoning changes on sizeable parcels ol
land, and the remaining industrial buildings on Clybourn Avenue soon be¬
came interspersed with a growing number of upscale stores, restaurants,
movie theaters, and several additional residential developments (Joravsky
114 Remaking Chicago

1988). Property values in the area soared, and before long nearly every in¬
dustrial building on the market was being priced for conversion to residential
or commercial uses, at values two or more times the going industrial rate
(Ducharme 1991, 228).
The rapid pace of redevelopment soon began to compromise the industrial
integrity of the area along Clybourn Avenue. Rising property values priced
existing manufacturers out of the market for expansion space and discour¬
aged outside manufacturing firms from moving into the area. They also led
to sizeable property-tax increases; in 1984, tax assessments in the North
River Corridor rose by an average of 100 percent, boosting industry operat¬
ing costs (Ducharme, Giloth, and McCormick 1986, 24). In addition, the
doubts manufacturers harbored about the likely tolerance of their new resi¬
dential neighbors for the sounds and smells of industrial development were
quickly confirmed. Oberman’s office was soon bombarded with complaints
from the new loft dwellers, and companies feared lengthy and expensive
court battles would inevitably follow (Joravsky 1988). Ultimately, redevelop¬
ment began to have a chilling effect on industrial investment activity along
Clybourn Avenue, as uncertainties about the future of the area led more and
more firms to defer expenditures for plant and equipment (Ducharme 1991,
228-229).
Despite the claims of real estate developers to the contrary, the redevelop¬
ment activity that hit Clybourn Avenue during the mid-1980s was not a sim¬
ple expression of market forces. Rather, the city’s zoning laws and practices
favored the real estate community. To begin with, the Chicago Zoning Ordi¬
nance offered little protection to manufacturers because zoning changes
were, in most cases, relatively easy to procure. In principle, changes to the
zoning ordinance required the approval of at least two-thirds of city council.
In practice, however, aldermen held sway over such decisions within their
own wards. This practice, as we saw in Chapter 2, has its origins in Chicago’s
machine-style political tradition. Zoning variances were one of a wide variety
of selective incentives used by ward politicians to cement party loyalties, and
aldermen learned to respect one another’s boundaries where such matters
were concerned to avoid possible retaliation and threats to their own political
fiefdoms. The relative absence of bureaucratic hurdles to overcome helped
fuel real estate speculation along Clybourn Avenue. Because investors were
confident that their requests for zoning changes would be processed with few
hassles, M-zoned property was soon being priced at commercial or residential
rates even before the variances had actually been granted (Ducharme, Giloth,
and McCormick 1986, 22-23).
Piecemeal changes to the city zoning ordinance did more than simply
destabilize property values along Clybourn Avenue, however. Since the last
comprehensive revision of the Chicago Zoning Ordinance in 1957, manufac¬
turers located near commercial or residential districts have faced more restric-
The Battle for the Near North Side 115

tive operating standards for noise, vibrations, and odors than manufacturers
with no commercial or residential neighbors.1 The rezoning of a nearby M-
zoned land parcel to allow residential or other nonindustrial uses could thus
place a given manufacturing establishment out of compliance with the zoning
ordinance, even if it had been there long before any nonindustrial types of
land use were established nearby.
This provision of the zoning ordinance was introduced following a wartime
boom in manufacturing, during which time citywide manufacturing jobs in¬
creased from 403,000 in 1939 to 615,000 in 1954 (McDonald 1984, 10).
Concerned about the nuisance-creating potential of the rapid industrial expan¬
sion, legislators established industry performance standards as a way of stem¬
ming further incursion of heavy industry into residential and commercial areas
of the city (Babcock 1972). Now that the tables were turned, however, and
the problem was residential and commercial encroachment upon industrial ar¬
eas, the performance standards became a weapon that real estate developers
could utilize in their efforts to rearrange land use along Clybourn Avenue.
In late 1984, staff members from LEED Council and the Research and
Development Division of the city’s Department of Economic Development
began a joint research project on the issue of industrial displacement on
Chicago’s Near North Side. At this point, the notion that industry needed
or desired protection from city government had won few converts within
the Washington administration. However, Robert Giloth, Deputy Commis¬
sioner of the Department of Economic Development under Washington and
eventual head of the department’s Research and Development Division, had
had firsthand experience with the issue during his tenure as executive direc¬
tor of the Eighteenth Street Development Corporation, a Near South Side
community development corporation, during the 1970s. The efforts of
Giloth and other community activists had helped persuade officials in the
Jane Byrne administration to reject a developer’s plan to convert a cluster of
nineteen industrial buildings in the Pilsen community of the Near South
Side to a mixed-use commercial and residential development (Giloth and
Menashe 1981).
The research team, led by Giloth and LEED Council Executive Director
Donna Ducharme, published its findings in a 1986 report, Business Loss or
Balanced Growth: Industrial Displacement in Chicago (Ducharme, Giloth,
and McCormick 1986). The study documented job losses resulting from
loft-conversion activity in River North from 1978 to 1985 and argued that a
similar fate awaited blue-collar workers in the North River Industrial Corri¬
dor if existing redevelopment trends were allowed to continue. Using the
economic development priorities voiced in the “Washington Papers” and the
administration’s 1984 Chicago Development Plan as leverage for their argu¬
ments, the authors of the study charged that city government support for the
rezoning of industrial property was inconsistent with the administration’s
116 Remaking Chicago

commitment to balanced growth. The report concluded with a recommenda¬


tion that areas of the city experiencing industrial displacement be designated
as “protected industrial districts” in which further zoning changes from man¬
ufacturing to commercial or residential designations would be prohibited.
The study proved to be significant for a number of reasons. First, it helped
sensitize administration officials to the issue of industrial displacement, both
within the Department of Economic Development and elsewhere in the bu¬
reaucracy. Even more importantly, it made a major impression on forty-third
ward alderman Oberman, whose approval of rezoning requests until that
point had helped feed the real estate speculation occurring along Clybourn
Avenue. After reading a draft of the report in late 1985, Oberman called a
moratorium on further zoning changes in the forty-third ward portion of the
North River Industrial Corridor and appointed a task force composed of rep¬
resentatives from LEED Council, the Departments of Planning and Eco¬
nomic Development, and the alderman’s office to develop a comprehensive
strategy for achieving a balance among the different sets of interests compet¬
ing for space in the area.
Understanding Oberman’s sudden concerns for the plight of Clybourn
Avenue manufacturers requires a brief digression on the somewhat unique
characteristics of forty-third ward politics. Once a machine stronghold, the
forty-third ward eventually became one of the city’s handful of independent
wards as the area’s working-class German ethnics were displaced by an influx
of middle-class professionals in the late 1960s and early 1970s (Fremon
1988, 281-288). The new ward residents were typically employed in the cor¬
porate and financial sectors downtown and, unlike their predecessors, did not
generally depend on ward politicians for patronage jobs and other individual
favors. Since the early 1970s, this has left aldermen representing the forty-
third ward more free than their counterparts in the city’s machine wards to
focus their attention on policymaking, which is ultimately what their middle-
class constituents want and expect. All this does not mean that Oberman was
oblivious to the political opportunities that his support for rezoning requests
along Clybourn Avenue provided. It does mean, however, that once a con¬
flict arose, he was more willing than most Chicago aldermen to consider the
issue from a policy rather than a machine-building perspective.2
In early 1986, Oberman’s task force on industrial displacement began to
hold monthly meetings with Near North Side manufacturers as part of its ef¬
forts to help resolve the land-use conflicts occurring in the North River Cor¬
ridor. The meetings served two key purposes. First, they gave die issue of in¬
dustrial displacement further credibility within city government by providing
administration officials with firsthand accounts of the pressures manufacturers
were experiencing in the area. Second, they helped build a sense of commit¬
ment among manufacturers to the process of devising a solution, increasing
the likelihood that they would weather the time-consuming organizing work
The Battle for the Near North Side 117

that would necessarily be a part of any long-term political settlement


(Ducharme 1991,230-231).
In July 1986, the task force recommended that a Protected Manufacturing
District (PMD) be established in the eastern portion of the North River In¬
dustrial Corridor between Clybourn Avenue and the Chicago River (the
“Clybourn Corridor”), where most of the real estate speculation that had
buffeted the area in the previous several years had been concentrated (see
Map 3). The new district would consist of a core zone, where manufacturing
would be the only permitted land use, and an exterior buffer zone, where a
mixture of commercial and industrial uses would be allowed in order to
buffer the heavier manufacturers in the district’s core from the residential de¬
velopment taking place outside the boundaries of the PMD. No further resi¬
dential development whatsoever would be allowed within the new district.
Nearly two years later, Oberman’s successor as alderman of the forty-third
ward, Edwin Eisendrath, would use the recommendations of the task force as
the basis for a Protected Manufacturing District ordinance introduced into
city council under his name.

Building the Coalition


During the two-year period between the issuance of the task-force report
and the introduction of the PMD ordinance into city council, LEED Council
engaged in a monumental organizing drive designed to identify and secure
bases of support for the initiative within the community in order to provide
sympathetic public officials with a sufficiently broad political base from which
to act. Part of this time was spent crafting and gaining city council approval
of a PMD enabling ordinance, which provided the legislative guidelines for
the establishment of protected manufacturing districts in specific areas of the
city (Ducharme 1991, 232). While adding extra time to the process, passage
of the enabling ordinance meant that future campaigns to create additional
PMDs in areas besides the Clybourn Corridor could be developed within a
clear legislative framework, simplifying the process considerably.
Support for the Clybourn Corridor PMD involved considerable risk for
public officials. Real estate developers active in the area made a persuasive
case, arguing that their proposals brought life to declining and obsolescent
industrial areas by refurbishing underutilized and abandoned buildings. Ac¬
cording to developers, loft rehabilitations and other commercial and residen¬
tial redevelopment projects represented the highest and best use for the
property, while PMDs were politically misguided efforts to restrict property
rights and interfere with market forces.3 As Tern Horowitz, developer of
Clybourn Lofts, argued in 1986, “These buildings are basically fossils. Cities
live. If the economics push the industries out, it is for the good of the city”
(Barry 1986).
118 Remaking Chicago

The position of the real estate community was reinforced by a series of ar¬
ticles and editorials published in the Chicago Tribune from 1986 to 1988
praising the efforts of property developers in the Clybourn Corridor and
characterizing LEED Council and supportive city officials as antigrowth and
antidevelopment. The Tribune's high-profile campaign against the PMD
made LEED Council’s organizing work all the more critical to the success of
the initiative, as media criticism left public officials increasingly reluctant to
carry the initiative forward before strong support among a broad and diverse
segment of the community had been mobilized. According to LEED Coun¬
cil’s Donna Ducharme, “What we had to do was to cast the issue widely
enough that lots of different types of people could find a way to hang their
hats on it” (Ducharme 1995).

The Manufacturers
At the center of the coalition backing the Clybourn Corridor PMD were,
naturally enough, the Clybourn area manufacturers themselves. Companies
supported the PMD for different reasons. Some businesses, like segments of
the printing industry described in Chapter 3, required near-downtown loca¬
tions for proximity to customers, suppliers, and subcontractors and were not
enticed by the cash incentives being dangled by commercial or residential de¬
velopers interested in acquiring their property. Others had major investments
in heavy machinery and equipment, making relocation costs prohibitive. One
specialty steel producer whose equipment included a 6,000-ton forging press
calculated it would need to sell its land for at least $90.00 per square foot in
order to relocate (Goff 1988). Even the priciest real estate on Clybourn Av¬
enue was only going for half that amount in the mid-1980s.
Organizing the companies and building a commitment to the process on
the part of firm owners and representatives proved to be difficult and time-
consuming work. Most had never been involved in a grassroots political cam¬
paign such as this, and many were initially skeptical it would succeed. Ulti¬
mately, however, a number of company representatives emerged as forceful
and articulate spokespersons on behalf of the PMD, testifying at hearings,
writing letters to city officials, and presenting their cases in the local and na¬
tional media. One firm even hired a publicist. For many of these manufactur¬
ers, the campaign marked a significant shift in the way they perceived and in¬
teracted with city government. As one Washington administration official
observed at the time,

What you’re seeing in tire Clybourn area for the first time in Chicago is the en¬
ergizing of the manufacturing sector and its entry into the public policy debate.
They’ve had their collective consciousness raised about the importance of pub¬
lic policy to their continued existence—something the real estate developers
who opposed the PMD have always known. (Goff 1988)
The Battle for the Near North Side 119

Still, support for the initiative among Clybourn Corridor manufacturers


was not unanimous. As property values continued to escalate during the mid-
1980s, some firms with weaker ties to the area wanted the freedom to cash in
their real estate holdings at commercial or residential rates, if they so chose.
A number of companies that had supported the PMD when it was first pro¬
posed in 1986 had become vocal critics of the initiative by 1988.4 Although
such firms were clearly in the minority, the Tribune chose to emphasize the
split between those firms supporting and those opposing the PMD in its cov¬
erage of the controversy.5

The Community
Neighborhood organizations of various kinds were a second key compo¬
nent of the PMD coalition. Along with LEED Council, three additional
North Side industrial development organizations—Jane Addams Resource
Corporation, Greater North Pulaski Development Corporation, and North
Business and Industrial Council—became strong supporters of the initiative.
In addition, both of Chicago’s citywide community economic development
umbrella organizations—the Community Workshop on Economic Develop¬
ment (CWED) and the Chicago Association of Neighborhood Development
Organizations (CANDO)—provided critical support for the campaign in var¬
ious ways.6 A number of resident-based organizations from low- and moder¬
ate-income neighborhoods bordering the North River Industrial Corridor
also joined forces with the coalition. Their concerns were twofold: if redevel¬
opment continued to move westward, gentrification would threaten both the
livelihoods of community residents employed in nearby industries and the in¬
tegrity of lower- and middle-income residential neighborhoods.
Support sometimes came from unexpected places. Once the new commer¬
cial development along Clybourn Avenue began to cause growing traffic con¬
gestion and parking problems, some of the middle- and upper-income resi¬
dents living east of Clybourn Avenue in Lincoln Park began to see their
industrial neighbors in a more favorable light. Sensing an opportunity to
broaden the coalition still fhrther, LEED Council held a meeting with several
resident-based neighborhood organizations from the Lincoln Park area, in¬
cluding the Concerned Allied Neighbors, Sheffield Neighbors, and the
RANCH Triangle Association, in September 1986. The proposed ordinance
was revised to accommodate a number of concerns raised by organization
members, and each of these groups ultimately chose to back the initiative
(Department of Economic Development 1988, 11).
Along with the Clybourn Corridor companies and neighborhood organiza¬
tions supporting the PMD, labor also played a significant role in the coalition.
Because LEED Council relied heavily on the companies themselves to educate
their workers and mobilize them on behalf of the PMD, support from labor
was strongest among workers whose employers backed the initiative, while
120 Remaking Chicago

workers employed by companies opposed to the measure were typically less


enthusiastic. Ultimately, the Chicago Federation of Labor became a strong
proponent of the initiative, allowing LEED Council to reach some workers in
companies that were not members of the PMD coalition. Altogether, roughly
ten union locals eventually identified themselves as supporters of the PMD
(Ducharme 1995).

The Administration
The final major component of the PMD coalition was the Washington ad¬
ministration itself, whose endorsement of the initiative was critical for a num¬
ber of reasons. For one thing, the enabling legislation made city council ap¬
proval of the PMD contingent upon the Plan Commission’s
recommendations. More importantly still, support for the initiative on the
administration’s part would send a strong signal to both developers and man¬
ufacturers that the city was serious about protecting industry on Chicago’s
Near North Side. Without a strong, unambiguous message from the adminis¬
tration, the ordinance alone would probably not be enough to stop real es¬
tate speculation and convince' manufacturers to make a commitment to the
area. As one dubious Clybourn area manufacturer noted, “I don’t know if
[the PMD] can provide meaningful protection in light of the way ordinances
get changed so easily” (Goozner 1986).
For much of the controversy, however, Washington himself remained offi¬
cially silent on the issue, refusing to take a public stance either for or against
the PMD. This comes as no real surprise, since the controversy was clearly
pulling the administration in two different directions. On the one hand, the
initiative was consistent with a number of themes Washington had articulated
during the previous several years, prioritizing jobs over real estate develop¬
ment, neighborhoods over downtown, and traditional blue-collar industries
over service sector and high-tech employment strategies. Also, a number of
organizations backing the proposed ordinance were key members of the
Washington coalition. On the other hand, support for the initiative risked
upsetting the administration’s already tenuous relationship with the down¬
town business community and left Washington susceptible to the kinds of
antigrowth charges the Tribune tended to level whenever the focus of the ad¬
ministration’s economic development policies began to stray too far from
downtown. As Chicago’s first African-American mayor, Washington was par¬
ticularly sensitive to the antidevelopment charge.
For a long time, the administration’s response to this dilemma was to play
both sides of the fence, approving zoning changes from industrial to nonin¬
dustrial designations while taking certain steps to support manufacturers, in¬
cluding funding LEED Council’s antidisplacement work A While couched in
the rhetoric of balanced growth, this strategy simply produced confusion
The Battle for the Near North Side 121

over the city’s policy stance on the question of industrial displacement, as the
following memorandum from DED Commissioner Robert Mier to Commu¬
nity Relations Director Jane Ramsey appears to suggest:

The Protected Manufacturing District (PMD) ordinance and policy is coming


under increasing criticism from the Tribune and development community. If it
is passed, the Tribune and other media could portray the Mayor as anti-growth
or anti-development. To counter this, we need a coordinated strategy to pre¬
vent this from happening while, at the same time, placing the Administration
out front on the PMD issue. One thing that could help is for the Mayor to be
involved with some type of manufacturing investment in the Goose
Island/Clybourn Corridor. . . . The announcement of some type of industrial
expansion will help convince the public that our PMD policy makes sense. In
addition, it would be helpful if we could identify the Mayor with some type of
major “yuppie” development. For instance, we could create a lot of fanfare
around a project representing the 5 billionth dollar of commercial and residen¬
tial development in the central area. ... By having the Mayor identified with
both of these processes—industrial expansion and boom time we can perhaps
begin to deal with the difficult public perception that balanced growth is not
only possible but is happening, now. (Mier 1987a)
/

Ultimately, the administration’s willingness to take a strong stand against in¬


dustrial displacement proved to be contingent on the strength and diversity
of the political base that had been mobilized around the issue in a given situ¬
ation. As LEED Council’s Donna Ducharme recalls, “The administration
people talk about [the PMD] now like it was their idea and they always
wanted to do it, but for a long time they were very hesitant to move forward
on it. Every step of the way, I would have to organize a broader base for
them to stand on to get them to take the next step (Ducharme 1995).
Two high-profile zoning controversies that took place in the North River
Industrial Corridor during 1986 and 1987 when public debate over the
PMD was beginning to gather momentum help illustrate Ducharme’s point.
The first involved the proposed conversion of a former scrap smelting plant
on Goose Island to a mixed-use, commercial/residential development. Plans
for the development, to be called River Lofts, were officially unveiled in Sep¬
tember 1986. Goose Island, a 160-acre island in the city’s forty-second ward
formed by a split in the Chicago River immediately south of the proposed
Clybourn Corridor PMD, was zoned exclusively for heavy industry at the
time. To proceed with their plans, the developers would need a zoning
change from manufacturing to a business or commercial designation. Due to
the parcel’s riverfront location, Plan Commission approval was required.
Both the administration and forty-second ward alderman Burton Natarus
initially gave their blessings to the project. However, LEED Council soon
122 Remaking Chicago

discovered there was significant opposition to the development on the part of


Goose Island manufacturers. A rapid organizing drive of area industries con¬
ducted by LEED Council staff members led to a series of meetings between
city officials and Goose Island manufacturers, during which manufacturers
expressed their reservations about the River Lofts project and reiterated their
opposition to the developers’ request for a zoning change. Because of the
manufacturers’ protests, the rezoning application was continued before the
Plan Commission for nine successive months. During this time, administra¬
tion officials also began to express reservations about the development, draw¬
ing harsh editorial criticism from the Tribune.8
In July 1987, an agreement was finally reached. The Plan Commission de¬
cided to support the zoning change, but only after the project’s developers
agreed to numerous concessions designed to mollify the concerns of Goose
Island manufacturers. Most importantly, the development company promised
to limit its activities on Goose Island to the River Lofts project alone and
agreed to support the passage of a future Goose Island Protected Manufac¬
turing District ordinance to ensure that the rest of the island would remain
industrial for years to come. Due largely to effective community organizing
on the part of LEED Council, the administration had taken its firmest stance
yet in support of North Side manufacturers threatened by nearby commercial
and residential development.
Just one month before the River Lofts agreement was finalized, another
development company announced plans for what would soon become a sec¬
ond high-profile zoning controversy in the North River Industrial Corridor.
In this case, the development firm of Centrum Properties wanted to demol¬
ish a complex of vacant industrial buildings on a seventeen-acre parcel at the
northern end of the corridor and redevelop the site as a shopping center and
retail strip, to be called Riverpoint Center. Again, both the site itself and the
surrounding land parcels were zoned for heavy manufacturing only. To carry
out its plans, Centrum needed the area rezoned for commercial use. The
large size of the parcel and riverfront location meant, once again, that Plan
Commission approval was necessary.
While outside the proposed Clybourn Corridor PMD, the size of the pro¬
ject and its location within a contested area of the North River Industrial
Corridor made this case a litmus test of the administration’s policy stance on
North Side industrial/commercial/residential land-use conflicts. In a Sep¬
tember 1987 memorandum to DED Commissioner Robert Mier and Plan¬
ning Commissioner Elizabeth Hollander, Deputy Planning Commissioner
David Mosena wrote,

It’s a great site, good for industrial, but prime land for retail. In fact, it’s be¬
cause it’s such a great retail site—near a growing yuppie market—that whatever
we choose to do, it will really give off a clear message that will have an effect on
The Battle for the Near North Side 123

the marketplace in that area. . . . The stakes are high—real money stands to be
lost. Everyone is watching to see what happens. Whichever message comes out
of this decision, it will be the most powerful message yet from the City.
(Mosena 1987a)

The negotiations over the River Lofts development on Goose Island had
helped strengthen the administration’s resolve to protect the integrity of
North Side industrial districts threatened by redevelopment pressures. While
still publicly silent on the PMD question, city officials initially expressed
reservations over Centrum’s proposal. One administration official quoted in
Crain’s Chicago Business insisted, “We’re ready to go to war over this. It
would send a bad signal. People would wonder if it’s possible for the city to
stop rezoning of industrial land in the future if it can’t stop this” (Goff
1987). In response, the Tribune stepped up its editorial pressure, unleashing
one of its harshest attacks yet against the administration’s efforts to curb in¬
dustrial displacement on the city’s Near North Side:

Mr. Washington has let his economic planners embark on a zany crusade to
snuff out commercial and residential growth in areas that they these insulated
City Hall planners—have decreed should be reserved for manufacturing. In¬
vestors who want to convert abandoned old factory buildings into job-produc¬
ing, tax-producing commercial complexes are told no, take your money to
some other city. And don’t think they won’t, if Chicago continues this perverse
ideological nonsense. (Chicago Tribune, 28 September 1987, sec. 1, p. 14)

In September 1987, the city began negotiating with Centrum over the fu¬
ture of the proposed retail center. Centrum insisted that the highest and best
use for the site was commercial and offered to contribute to a linked develop¬
ment program in exchange for the city’s approval of the zoning change
(Mosena 1987b). Already by this point, the administration’s commitment to
retaining the site for future industrial users appeared to be wavering, al
though officials continued to voice reservations about the project publicly.
From the city’s standpoint, the problem with opposing this particular project
outright was the lack of organized opposition to the proposed development
among nearby manufacturers and neighborhood groups. Unlike Goose Is¬
land, neither LEED Council nor any other North Side industrial develop¬
ment organization had created a strong organizational presence around the
issue of industrial displacement in this area of the North River Corridor.
Without the solid backing of nearby manufacturers, the city’s opposition to
the development could be easily characterized as antigrowth. Conceding that
a denial of the rezoning would represent a sizeable risk for the administra¬
tion, Deputy Planning Commissioner Mosena wrote in a late September
memorandum to Commissioner Hollander,
124 Remaking Chicago

The retail market of west Lincoln Park is real, neighborhood organizations to


the east support the shopping center, and the depth of our industrial support
against the shopping center is weaker than it was in the Clybourn PMD or
Goose Island. On the other hand, support of this rezoning without qualifica¬
tion puts this industrial area totally up for grabs and amounts to no policy from
the City for this area. (Mosena 1987b)

By November 1987, the administration’s strategy had changed completely.


Opposition to the zoning change had been dropped from consideration, and
the question now was how to announce approval of the project while inflict¬
ing as little damage as possible on the proposed PMD ordinance. Here once
again, the lack of strong community organizing around the issue of industrial
displacement in this area of the city proved to be the decisive factor. Initially,
there was some discussion of making the announcement within the context
of a moratorium on further zoning changes in the thirty-second ward por¬
tion of the North River Industrial Corridor, where the Centrum develop¬
ment was to be located. Like the settlement reached with the developers of
the River Lofts complex on Goose Island, this would help quell speculation
among area manufacturers ahd property developers that this particular zon¬
ing change was simply the first of many to follow. However, Greg Longhini,
Manager of Industrial Land-Use Policy in the Department of Planning,
counseled against this approach. In a November 1987 memorandum to
Deputy Commissioner Mosena, Longhini argued,

Inconsistency is difficult for all of us to proceed with. Case-by-case analysis is


tough. However, to map out an area for a moratorium, without a study or com¬
munity planning process, could lead to difficulties, such as our rationale. . . .
Donna [Ducharme] has never concentrated west of the river. We don’t know
what the industrialists would say. In a nutshell, we don’t know our political base
there very well. (Longhini 1987)

Two days later, the Department of Planning issued a press release announc¬
ing its recommendation that Centrum Properties’ petition for a zoning change
be approved. There was no mention of a moratorium on future zoning
changes in the thirty-second ward, although weaker language expressing the
city’s commitment to preserving the overall industrial “character” of the area
did appear. For its part, Centrum promised to require all tenants of its new
Riverpoint Center to make use of the city’s First Source jobs program, target¬
ing low- and moderate-income Chicagoans, in filling their new retail positions,
and agreed to develop an industrial project elsewhere on the North Side.
Ironically, it was the city’s decision to drop its opposition to Riverpoint
Center that finally pushed the administration off the fence on the issue of the
proposed PMD ordinance. By this time, the coalition of manufacturers,
The Battle for the Near North Side 125

neighborhood organizations, and labor unions supporting the initiative was


largely in place. Fearful that the Centrum announcement would send the
wrong message to both the manufacturing and real estate communities, the
administration acted quickly during the following weeks to counter any per¬
ceptions that the city’s commitment to safeguarding the interests of North
Side manufacturers was wavering. A bus tour of the Clybourn Corridor—to
include Mayor Washington and top administration officials, labor and com¬
munity representatives, and members of the local news media—was hastily
arranged. During the tour Washington finally made it official: his administra¬
tion would support the passage of a Protected Manufacturing District ordi¬
nance to help shelter Near North Side manufacturers from the pressures of
nearby commercial and residential redevelopment. To counter charges by de¬
velopers that the Near North Side was no longer a viable area for manufac¬
turing, Washington also used the opportunity announce plans for a new in¬
dustrial park on Goose Island.
The cases of River Lofts and Riverpoint Center illustrate the importance
of coalition-building between manufacturers and neighborhood organiza¬
tions around the issue of industrial displacement. Both sites were situated in
the midst of viable industrial areas, surrounded by manufacturers threatened
by the proposed conversion of nearby land parcels to new uses incompatible
with industrial development. However, only in the case of the River Lofts de¬
velopment on Goose Island did neighborhood activists step in to organize
area manufacturers around a collective response to redevelopment pressures.
A strong, united manufacturing presence on Goose Island gave sympathetic
Washington administration officials the ammunition they needed to deflect
charges by real estate developers that the area was no longer desirable for in¬
dustrial use.
Manufacturers near Centrum Properties’ proposed Riverpoint Center, by
contrast, were located in an area of the city where community organizations
had not yet allied themselves with neighboring industrial firms. Resistance to
the developers’ plans on the part of nearby manufacturers was, accordingly,
weak and sporadic. Because profitable manufacturers failed to stand up in suf¬
ficient numbers to demand protection from the city, opposition to the devel¬
opment on the part of Washington officials seemed to give credence to the
antigrowth charges being leveled at the administration by real estate develop¬
ers and their allies. Lacking sufficient evidence in this case to refute such
claims, administration officials quickly fell in line with Centrum’s demands.

The PMD Controversy after Washington


Mayor Washington’s decision to end his long public silence on the PMD
and to announce his support for the initiative appeared to represent the last
and, in some ways, most significant link in thy. PMD coalition. However,
126 Remaking Chicago

LEED Council and its allies on the Near North Side quickly suffered a major
setback. Just three weeks after visiting the Clybourn Corridor and endorsing
the PMD proposal, Washington collapsed suddenly and died at his desk in
City Hall, the victim of a massive heart attack.
Washington’s death ignited a fierce power struggle between the Washing¬
ton and machine factions of city council over the city’s future political leader¬
ship. The acting mayor was, by law, to be chosen from among the council’s
fifty aldermen. Eugene Sawyer, a soft-spoken black alderman with former ties
to the machine, emerged as the compromise choice for mayor. Sawyer imme¬
diately pledged to remain faithful to Washington’s neighborhoods-oriented
policy agenda, but his history as a loyal machine foot soldier dating back to
the 1950s caused many to question his sincerity. Media commentaries on the
likely fate of Washington’s policies called the future of the Clybourn Corridor
PMD initiative an open question.
Although acting mayor Sawyer quickly announced his support for the
PMD, the city’s commitment to providing meaningful protection for North
Side manufacturers continued to be called into question in the following
months. In the spring of 1988, developer Duncan Henderson approached
the city with plans to convert a‘ former Ludwig Drum factory in an industrial
area just west of the North River Industrial Corridor to a mixed-use,
work/live development. The residential component of the project meant the
site would have to be rezoned from its present manufacturing designation to
allow Henderson to proceed with his plans. Ward Alderman Terry Gabinski,
a machine loyalist, supported the development. Although several nearby la¬
bor and resident-based community organizations were critical of Hender¬
son’s plans, the area was not officially served by any of the city’s LIRI groups
or any other industrial development organization. Opposition to the project
by area manufacturers was consequently weak and ineffective, much as it had
been in the case of Centrum Properties’ Riverpoint Center development sev¬
eral months earlier. In July 1988, the Plan Commission somewhat reluctantly
approved the zoning change.
Also that spring, a North Side automobile dealer, Hanley Dawson, began
negotiations to purchase a six-acre site within the core zone of the proposed
Clybourn Corridor PMD. The company wanted to relocate its automobile
dealership to the site and asked that the parcel’s existing M3 zoning designa¬
tion be changed to allow the new commercial development to go forward as
planned. No rezonings within the core zone of the proposed PMD had been
entertained since the PMD map had entered the public domain over two
years earlier. Fearful of the message that an exception for Hanley Dawson,
one of the city’s largest automobile dealers, would convey, Planning Com¬
missioner Elizabeth Hollander strongly urged Sawyer to oppose the develop¬
ment. In an April 1988 memorandum to the mayor, Hollander argued,
The Battle for the Near North Side 127

The entire development community—commercial, industrial, and neighbor¬


hood organizations—are familiar with the PMD concept, and many with the
map. In other words, re-zoning to allow in Hanley Dawson could create a
firestorm within the community that is actively pursuing a PMD policy to pre¬
serve and create meaningful jobs. ... If Hanley Dawson does not receive firm
word from the City and from Alderman Eisendrath that their proposal is not
sufficient to justify a re-zoning, serious damage could be done to the proposed
PMD ordinance coming up before the Council. (Hollander 1988)

Hollander’s recommendations seemingly fell on deaf ears. According to me¬


dia accounts, Hanley Dawson was able to persuade a top Sawyer advisor to
intervene on its behalf (Goozner 1988a). By the time forty-third ward aider-
man Eisendrath introduced the PMD ordinance into city council early that
summer, the boundaries of the core zone had been redrawn to accommodate
the auto dealer’s wishes.
During this time, however, the base of community support around the
PMD continued to expand. In the months following Washington’s death, the
proposed ordinance was the subject of numerous articles and commentaries
in both the local and national media. A number of local publications, includ¬
ing Crain's Chicago Business, the Reader, and, on occasion, the Chicago Sun-
Times, played important educational roles, helping to offset the Tribune's
one-dimensional coverage of the issue. Feature stories on Clybourn Corridor
companies such as Finkl Steel, a highly profitable, 109-year-old steel foundry
employing more than 400 workers, presented Chicagoans with a more nu-
anced portrait of industrial decline than many had previously been exposed
to. People began to learn that not all manufacturers were leaving Chicago for
the same reasons. As Charles Finkl, chairman of the steel company that bears
his name, argued in a New York Times cover story in December 1987, “It’s
one thing to lose basic industries to international forces beyond our control.
But it’s another thing to force healthy industries out of the city through un¬
wise and piecemeal zoning policies” (Schmidt 1987).
The educational role played by the media was complemented by studies of
the Clybourn Corridor carried out during the summer of 1988 by the De¬
partment of Economic Development, the Mayor’s Office of Employment and
Training, and an independent research team led by Peter Creticos and
Northwestern University business professor Louis Masotti. The studies con¬
tained similar findings: each concluded that the economic contribution to
Chicago’s economy made by Clybourn Corridor manufacturers far out¬
weighed that made by the retail and other service-sector establishments in¬
creasingly populating the area. Two of the studies focused on employment
multipliers, concluding that the higher value added by manufacturing work¬
ers, along with their relatively higher wages, allowed such individuals to
128 Remaking Chicago

support up to twice as many additional jobs as did retail employees (Sandro


1988; Creticos and Masotti 1988). The third study included a direct compari¬
son of employment, wages, and taxes paid by the industrial and nonindustrial
firms in the Clybourn Corridor, at the time more or less evenly divided be¬
tween the two. The results, reproduced in Table 10, show that for each vari¬
able considered, the industrial firms outperformed their nonindustrial coun¬
terparts by sizeable margins (Department of Economic Development 1988).

Table 10. Employment, Wages, and Taxes Paid


by Clybourn Corridor Business Firms, 1988

Industrial Nonindustrial

Number of firms 31 29
Total employment 1,701 451
Total payroll (millions) $38.9 $4.8
Average wage $22,900 $10,700
City taxes paid \ $1,367,000 $404,000

Source: Department of Economic Development (1988).

The research and publicity over industrial displacement on the Near North
Side eventually helped begin to unmask some of the ideological claims being
made by the real estate community to generate support for its goals and am¬
bitions in the Clybourn Corridor. In particular, the familiar litany of the
“highest and best use,” which once held such weight in land-use conflicts of
all kinds, had itself now become the subject of growing discussion and de¬
bate. Especially with the current focus of economic policy squarely on jobs
instead of real estate development, property developers were hard-pressed to
explain why their proposals were better for Chicago than preserving land for
manufacturing, even if they did lead to higher property values. Furthermore,
growing publicity over the role of zoning policy in determining land-use
trends in contested areas of the Near North Side weakened developers’ argu¬
ments that displaced manufacturers were victims of market forces alone. This
change in the terms of discourse over industrial decline further legitimized
LEED Council’s organizing work and helped reduce the Sawyer administra¬
tion’s risks in supporting the PMD initiative.
By the summer of 1988, the debate over the PMD was rapidly approach¬
ing a climax. Once forty-third ward alderman Eisendrath introduced the or¬
dinance into city council in late June, only two major hurdles remained: a
community hearing to be held by the Department of Planning to explain the
proposal and solicit comments, and a hearing by the Plan Commission to de-
The Battle for the Near North Side 129

termine the industrial viability of the district and the need for PMD status.
Both were scheduled for later that summer. At the close of the second hear¬
ing, the Plan Commission would issue its recommendations. Although final
approval of the ordinance still rested with the city’s fifty aldermen, a positive
vote by the Plan Commission was not expected to be challenged in city
council.
Ultimately, both hearings proved to be dominated heavily by PMD sup¬
porters. Of twenty-eight individuals testifying at the Department of Plan¬
ning’s hearing on the PMD in July, four were opposed to the PMD, while
twenty-four supported the initiative. Those arguing against the ordinance in¬
cluded two Lincoln Park residents, a realtor, and one area manufacturer. Tes¬
tifying in favor of the PMD were representatives from eight nearby industrial
firms, two labor unions, and six community organizations, along with seven
area residents and one industrial worker (Department of Planning 1988).
The following month, after a lengthy session that included five hours of addi¬
tional testimony, the Plan Commission voted 7—0 in favor of the PMD.
In October 1988, over two years after former forty-third ward alderman
Martin Oberman’s task force on industrial displacement originally proposed
the idea, city council voted to approve the Plan Commission s recommenda
tion to create the city’s first Planned Manufacturing District in the Clybourn
Corridoj portion of the forty-third ward.9 The basic features of the ordinance
remained largely as originally conceived by the task force. The new district
would consist of two parts: a forty-one acre core zone surrounded by a sev¬
enty-four acre buffer zone. Residential conversions would be prohibited in
both portions of the district, but certain commercial and retail development
would be allowed in the buffer zone at the discretion of the zoning board.
The Plan Commission was given the authority to review the PMD periodi¬
cally and recommend changes to, or repeal of, the ordinance should it be
deemed ineffective in fulfilling its intended objectives.10
Passage of the Clybourn Corridor PMD ordinance represented, without a
doubt, a key legislative milestone in Chicago. Never before had city officials
acted so aggressively to protect central area manufacturers threatened by the
pressures of nearby commercial and residential redevelopment. Yet both sup¬
porters and opponents of the measure alike were well aware that important
questions still lay ahead. The threat of a lawsuit by Clybourn Corridor manu¬
facturers opposed to the initiative had already been raised, and the city’s ten¬
dency to vacillate between support for the industrial and real estate communi¬
ties throughout the duration of the conflict raised questions about whether
area manufacturers and property developers would accept the decision as final.
The debate over the PMD brought a number of other issues to the surface
as well. Protecting manufacturers from real estate speculation was one thing,
but PMDs alone would clearly not be enough to halt the decline of
Chicago’s manufacturing base, much less rejuvenate industry. In addition to
130 Remaking Chicago

preserving the integrity of the city’s industrial districts, city government


would also need to proactively support industrial development in the city—
through planning, infrastructure improvements, financial assistance, and
other such efforts. In short, the city would need to begin furnishing Chicago
manufacturers with the same kinds of public support the downtown business
community had enjoyed for the past half-century. The following chapter fo¬
cuses on the aftermath of the Clybourn Corridor PMD debate, examining
how elements of a citywide industrial policy have begun to take root in
Chicago despite the return of the downtown business leadership to a central
position within the city’s governing coalition.

Conclusion
The events in this chapter appear to depart in a number of ways from the
expectations of most contemporary urban political economists. Most impor¬
tantly, the alliance between land-based interests and city officials that urban
theorists today view as a nearly inescapable component of present-day urban
regimes failed to materialize around an economic development strategy for
the intensification of land use on Chicago’s Near North Side. Instead, offi¬
cials from the Harold Washington and Eugene Sawyer administrations sided
with neighborhood organizations and their allies around a policy initiative
designed to preserve well-paying blue-collar jobs by maintaining Near North
Side property values at below-market levels.
How were Washington and Sawyer administration officials able to resist
the pull of the real estate community and its allies? Consider, first of all, the
alliance between Near North Side community organizations and manufactur¬
ers, which provided the former with a set of economic resources and the lat¬
ter with the organizational capacity to influence public policy. In essence, this
coalition allowed city officials to circumvent the accumulation/legitimation
divide within which urban political economists today tend to frame virtually
all conflict between downtown and the neighborhoods. Relieved in this case
of the necessity to balance legitimate neighborhood concerns with the need
for economic growth, both Washington and Sawder officials were more free
than usual to make decisions consistent with the interests of their administra¬
tions’ neighborhoods political base.
Significant as well is the role of public policy in this story. The growing
transformation of the Clybourn Corridor from a manufacturing district to a
residential, shopping, and entertainment area for downtown office workers
was not the result of simple economic pressures, global or otherwise, that ur¬
ban political economists typically single out as the driving force behind urban
economic change. As this chapter illustrates, real estate speculators and devel¬
opers took advantage of an easy-to-manipulate zoning ordinance to inflate
property values in the Clybourn Corridor above the level that could support
The Battle for the Near North Side 131

manufacturing. As commercial and residential incursion into the area in¬


creased, industry performance standards for noise, odors, and vibrations cre¬
ated an additional wedge for the real estate community by placing profitable
manufacturers out of compliance with the city zoning ordinance. Ultimately,
the zoning ordinance itself became the object of struggle in a dispute that
would determine, politically, which of these two equally viable economic
strategies would prevail.
To be sure, real estate developers and their allies represented formidable
adversaries, continually testing the mettle of Washington and Sawyer admin¬
istration officials. Yet the power of the development community rested less
on objective economic conditions than on its ability to define growth in
terms most beneficial to its own membership. This proved increasingly diffi¬
cult once Washington and his jobs-based economic development agenda ar¬
rived on the scene. Despite the Tribune’s attempts to portray supporters of
the PMD initiative as antigrowth and antidevelopment, it ultimately became
clear that the real estate-based growth strategy of rising property values and
property tax revenues was simply one of two contested economic develop¬
ment strategies in the Clybourn Corridor, the other being the preseivation of
the area’s industrial base along with the well-paying manufacturing jobs it
supported. Based on simple economic criteria alone, there was no rational ba¬
sis for choosing commercial or residential redevelopment over manufactur¬
ing.11 Like Printing House Row and the Chicago garment district years ear¬
lier, politics would play a leading role in determining the trajectory of
redevelopment. In this case, however, the outcome would be shaped through
a vision shared by a coalition of neighborhood organizations, manufacturers,
workers, and their allies in city government instead of real estate developers,
downtown business leaders, and sympathetic city officials.
Chapter Six

Toward a Citywide Industrial Policy

widespread evidence supporting the argument by urban politi-


cal economists that city officials today find themselves irresistibly drawn into
alliances with land-based interests around downtown revitalization efforts,
cases do emerge from time to time that fit this model either imperfectiy or
not at all. The Harold Washington and Eugene Sawyer administrations’ sup¬
port for the Clybourn Corridor Planned Manufacturing District initiative dis¬
cussed in the previous chapter represents one such case, but there have been
various others as well. In the late 1970s, for example, Cleveland’s populist
mayor Dennis Kucinich engaged in a series of head-to-head confrontations
with the city’s downtown business community, abolishing tax abatements for
downtown construction projects and at one point returning a $41 million
federal grant for a downtown rail system, calling it a “contemptuous substi¬
tute” for the real transit needs of Clevelanders (Swanstrom 1985, 8). In
Hartford, progressive city council leader Nicholas Carbone tested the bound¬
ary between public and private sectors by helping the city gain part owner¬
ship of a number of downtown development projects, which officials then
used to win commitments for minority hiring in the construction and opera¬
tion of the developments (Clavel 1986).
Contemporary urban political economists readily acknowledge such cases
but deny they represent departures from existing theory. If anything, the ex¬
periences of progressives such as Kucinich in Cleveland and Carbone in Hart¬
ford are said to confirm their views, because both regimes were ultimately
toppled by progrowth coalitions dominated by downtown business elites.
During Kucinich’s first term in office, downtown bankers angry with the new
mayor’s confrontational negotiating style sent the City of Cleveland into de¬
fault by refusing to refinance $14 million in short-term notes. Playing on
fears that Kucinich was damaging the city’s investment climate, probusiness
candidate George Voinovich easily beat Kucinich in the next mayoral elec¬
tion. Likewise, Nicholas Carbone’s unsuccessful bid for mayor of Hartford in
Toward a Citywide Industrial Policy 133

1979 was followed by the rapid dismantling of the neighborhood-oriented


programs he had pushed through city council during previous years.
In Chicago, when Richard M. Daley emerged as the winner of a 1989 spe¬
cial election called to fill the late Mayor Washington’s unexpired term, many
observers anticipated a similar retrenchment. Daley’s campaign had been
heavily bankrolled by downtown attorneys, developers, and business leaders
upset with their limited access to City Hall during the previous six years. In
his campaign speeches and literature, Daley presented himself as a staunch
advocate of economic growth committed to solidifying Chicago’s status as a
world-class city by building on its strengths as a corporate-financial center.
He criticized the recently established Clybourn Corridor PMD and appeared
to display little understanding of, or interest in, other types of neighborhood
economic development and revitalization efforts.
However, while Daley officials have in fact eliminated or curtailed certain
of Harold Washington’s neighborhood-oriented policies and programs, such
reversals have been fewer than anticipated, and in some instances the admin¬
istration has even expanded upon programs introduced by Washington. In¬
dustrial policy is a case in point. Daley eventually reversed his campaign
stance against planned manufacturing districts and, despite strong opposition
from key members of his downtown business-dominated governing coalition,
pushed''successfully for the passage of two additional PMDs in the North
River Industrial Corridor in 1991. Efforts by Daley officials to curb industrial
displacement have been supplemented more recently with capital improve¬
ments, tax increment financing, and other policies and programs designed to
help rejuvenate Chicago’s industrial base.
By all appearances, the Richard M. Daley administration began as an effort
to recreate the progrowth alliance of downtown business leaders and city of¬
ficials that governed Chicago from 1955 to 1983. Indeed, downtown rede¬
velopment efforts have played a central role in the Daley administration’s
overall economic development policy agenda. Yet as we saw in Chapter 4, the
political arrangements that sustained Chicago’s postwar progrowth regime
fell increasingly into crisis by the late 1970s, paving the way for Harold
Washington’s election victory in 1983. As this chapter will show, efforts to
reconstruct the old regime on new political terrain have been only partially
successful. Meanwhile, proponents of a local-producer strategy for neighbor¬
hood redevelopment have continued to seize opportunities to move their
agenda forward in important ways.

The Richard M. Daley Administration: Between Machine and Reform


The April 1989 special election for mayor once again divided die city of
Chicago along racial lines. The race shaped up as a three-way contest among
acting Mayor Eugene Sawyer, Richard M. Daley, and Timothy Evans, an
134 Remaking Chicago

African-American alderman who had served as the late Mayor Washington’s


floor leader in city council. Sawyer and Evans each sought to build support
among Washington’s former coalition of blacks, Latinos, and lakefront liber¬
als, but neither candidate was able to generate any enthusiasm to speak of
outside the black community. With Sawyer and Evans dividing the African-
American vote, Daley coasted to victory with strong support in the city’s
white ethnic, Latino, and lakefront wards (Grimshaw 1992).
The election campaign proved to be a referendum of sorts on the Wash¬
ington administration’s balanced growth economic development agenda. The
Chicago Tribune helped set the terms of debate with a seven-part series enti¬
tled “Chicago on Hold,” published late in the summer of 1988 shortly be¬
fore the campaign got underway. The series was a direct attack on commu¬
nity activists, progressive politicians, and former Washington administration
officials working to resist gentrification and displacement pressures in inner-
city neighborhoods. According to series author John McCarron, such indi¬
viduals were little more than self-serving opportunists, drumming up political
support by using downtown area redevelopment projects to play on the fears
of neighborhood residents, workers, and business owners. McCarron also
criticized the proposed PMD ordinance, ridiculing the notion that preserving
Chicago’s industrial base was either possible or desirable and arguing that
growth in service-sector employment would offset the decline in factory jobs
(McCarron 1988).
Candidate Daley presented himself as the alternative to what McCarron
scornfully referred to as the “anti-development crusade,” distancing himself
from controversial programs and policies such as PMDs being advanced from
within the neighborhoods. Daley’s economic development platform was
structured around the corporate-center strategy for downtown redevelop¬
ment. His campaign speeches reinforced this theme, generally containing
only token references to traditional manufacturing while emphasizing the im¬
portance of the downtown area to the city’s overall economic prosperity. Ad¬
dressing the Chicago Kiwanas Club several weeks before the Democratic pri¬
mary, Daley argued:

We should strive to preserve and expand manufacturing. But we must also look
to areas that promise greater job growth. We can’t afford to focus on only one
economic sector and ignore the growth industries of the 1990s. Chicago is one
of the nation’s three largest financial centers. Financial services provided
330,000 Chicago jobs in 1986—about ten percent of total metropolitan em¬
ployment. These financial jobs are growing at twice the rate of other employ¬
ment. . . . Chicago is home to world-class universities and cultural institutions.
We have trained professionals, a vibrant downtown, and a leading convention
center. The next mayor should use these qualities to attract service and financial
industries. (Daley 1989)
Toward a Citywide Industrial Policy 135

Relying heavily on large donations from corporate officials and leading


downtown business establishments, Daley amassed a record $7.1 million in
campaign contributions (Gibson 1989a). By contrast, Eugene Sawyer man¬
aged to raise just $3 million, while Timothy Evans’s severely underfunded
campaign raised less than $500,000 (Goozner and Ziemba 1989; Gibson
1989b). In March 1989, the Evans campaign released a report contending
that 30 percent of Daley’s donations came from just one percent of his con¬
tributors, and that 40 percent came from lawyers, large law firms, and real es¬
tate developers (Hardy and Casuso 1989). Developer J. Paul Bietler, whose
firm accounted for one of Daley’s four donations of $100,000 or more, ex¬
plained Daley’s strong support in the real estate community as follows:

We’re tired of tire anti-development posture that has been promulgated at City
Hall in the last six years. The extent of the contributions to Daley’s campaign
by the real estate community, I think, is clearly making that message known. In
Daley we see someone whom we believe is capable of seeing the bigger picture;
someone who is capable of solving problems in a more logical way; someone
who won’t seek taxes [on development] without first establishing a dialogue
with the real estate community. (Goozner and Ziemba 1989)

Daley’s own politics have been characterized as a curious and often contra¬
dictory blend of machine and reform (Grimshaw 1992; Ferman 1991). His
roots in the machine are deep and personal, yet during a six-year stint as
Cook County state’s attorney from 1983 to 1989, he made high-quality,
professional appointments and resisted any temptation to reserve lower-level
jobs for Democratic party loyalists (Grimshaw 1992, 212-213). In the 1989
mayoral campaign, he presented himself as a classic managerial reformer,
pledging to put city government on a more businesslike footing by privatiz¬
ing certain government services, making city workers more productive, and
eliminating the city’s $150 million budget deficit. In general, Daley has made
good on these promises, yet his upbringing within the machine has made its
mark on his administration as well.
For example, in his acceptance speech on April 4, 1989, Daley promised
his administration would maintain the participatory governing style intro¬
duced by the late Mayor Washington, a clear attempt to distance himself
from the centralized model of governing embraced by his father (Dold and
Hardy 1989). However, within a short period of time, the administration s
decision-making powers had been concentrated in the hands of an inner cir¬
cle made up of three long-term Daley aides, none of whom held reform cre¬
dentials. Many of the administration’s top bureaucrats were left virtually
powerless, unable to make even the smallest decisions without first receiving
approval from the mayor’s top echelon of advisors (Hornung 1990).
Daley’s efforts to privatize certain government functions revealed similar
136 Remaking Chicago

tensions between machine and reform. During the first several years of his
administration, one-time bastions of patronage, such as city-run janitorial and
towing services, were farmed out to private contractors, helping Daley bring
the city’s budget deficit under control. At the same time, however, reports
began to surface that many of the mayor’s key political contributors were re¬
ceiving lucrative city contracts for legal work and other services. One particu¬
larly generous law firm did $445,000 worth of business with the city during
Daley’s first year in office, nearly three times more business than any other
firm (Gibson 1990). Critics soon began to charge that “pinstripe patronage”
was filling the void left by the Shakman Decree’s restrictions on political hir¬
ing and firing, breathing new life into a crippled machine.
In the economic development policy arena, Daley quickly undertook steps
to establish himself as a builder in the same tradition as his father, emphasiz¬
ing flashy downtown area development projects. Within months of his elec¬
tion in 1989, Daley convinced the Illinois state legislature to finance a $150
million renovation of Navy Pier, a rusting, underutilized municipal dock near
the Loop that the city wanted to convert to an entertainment complex. Dur¬
ing the first two years of his administration, Daley also helped engineer a
$987 million expansion of the McCormick Place convention complex and
was influential in the planning and construction of the $175 million United
Center, a new basketball/hockey stadium on the Near West Side (Reardon
1992a). In addition, plans were announced for the construction of a third
major airport in Chicago; a $2 billion, privately financed casino and enter¬
tainment complex; and a $775 million trolley system to transport tourists and
office workers through Chicago’s expanding downtown area (Reardon
1992a; Washburn 1993).
Yet despite his centralized governing style and preoccupation with
grandiose development projects, Daley displayed flashes of sensitivity during
the first years of his administration to the concerns of neighborhood resi¬
dents, workers, and business owners located outside the Loop and affluent
lakefront wards. A number of progressives received top-level appointments in
the mayor’s new cabinet, including David Mosena, a strong proponent of in¬
dustrial retention within the Washington and Sawyer administrations, and
Joseph James, former director of economic development and international
trade for the city of Austin, Texas. Daley also sponsored a city ordinance re¬
quiring that 25 percent of all city contracts be awarded to minority-owned
firms and established the Capital Improvements Advisory Committee in
1991 to provide a forum for public input into and oversight of the allocation
of funds for citywide capital improvements projects (Kass 1990; Ferman and
Grimshaw 1992, 122).
Overtures to the neighborhoods such as these were an indication of the
degree of change that Chicago’s political landscape had undergone during
the previous several decades. As we saw in earlier chapters, Chicago politics
Toward a Citywide Industrial Policy 137

during the machine era were marked by a separation between electoral politics
and public policy, whereby patronage appointments and selective incentives
were utilized to assemble winning electoral coalitions of neighborhood voters,
while policy benefits were channeled into the hands of downtown landowners
and other business elites. By the time Richard M. Daley was elected mayor in
1989, however, any attempt to recreate this set of arrangements would have
certainly been futile. Not only were key machine resources like patronage and
federal urban programs in short supply, but Harold Washington had also cre¬
ated a climate of high expectations within the city’s low- and moderate-in¬
come neighborhoods by binding the organizing efforts of community-based
organizations, providing them with access to key decision-making arenas, and
helping to create a new sense of legitimacy for community-based political ac¬
tivity. As political scientist William Grimshaw has argued,

“Son of Boss” may well be bent on reinstating a centralized and closed style of
machine government; however, he is entangled in the web of reform spun by
Washington. Blacks, women, lakefront liberals, Hispanics, and other dispos¬
sessed groups who had been empowered by Washington’s reforms are disin¬
clined to give up their gains. They want to maintain the access, decision-making
influence, and material and symbolic benefits they acquired under reform gov¬
ernment. (Grimshaw 1992, 208)

By all indications, Richard M. Daley’s governing coalition is not unlike


that of his father, dominated by real estate developers and other downtown
business elites, lawyers and large law firms, and some party regulars (Ferman
and Grimshaw 1992, 122). By and large, public policy under the new Daley
administration has been consistent with the interests of this coalition, as Da¬
ley’s penchant for large-scale, downtown redevelopment projects would ap¬
pear to suggest. Yet due to the mobilization of low- and moderate-income
neighborhoods and the absence of resources to help dampen neighborhood
expectations, Daley has been compelled to expand the circle of public-policy
beneficiaries in Chicago, selectively broadening his governing coalition to in¬
clude some degree of representation for certain neighborhood groups and
organizations.
The Richard M. Daley administration thus straddles a number of contradic¬
tions, occupying the middle ground between machine and reform, between
centralization and decentralization, and between downtown and the neigh¬
borhoods. Despite his personal upbringing within the machine and his strong
ties to the downtown business community, Daley himself has shown a willing¬
ness to embrace the policy concerns of neighborhood organizations and their
allies on a selective basis, even when doing so places him at odds with promi¬
nent members of his governing coalition. These tensions and contradictions
are evident in the Daley administration’s approach to industrial policy.
138 Remaking Chicago

Industrial Land Use

During the time the Clybourn Corridor PMD proposal was being debated
in the mid-to-late 1980s, land-use conflicts between manufacturers and com¬
mercial and residential property developers were beginning to emerge in
other portions of the North River Industrial Corridor as well. As we saw in
Chapter 5, developers of the River Lofts commercial/residential complex on
Goose Island successfully lobbied the city for a zoning variance in 1987 de¬
spite strong opposition from Goose Island manufacturers. By 1988, land
speculators betting that gentrification would ultimately sweep the island had
purchased much of the available industrial property there, driving property
values well above industrial rates and making expansion space for Goose Is¬
land manufacturers increasingly unaffordable (Goozner 1988b). Similar real
estate pressures were also developing in the nearby Elston Corridor sub-area
of the North River Industrial Corridor (see Map 4). By the time the
Clybourn Corridor PMD was established in October 1988, planning was un¬
derway to create additional PMDs in each of these two areas.
The Local Employment and Economic Development (LEED) Council,
the major organizational force behind the Clybourn Corridor PMD initiative,
also led the effort to set aside Goose Island and the Elston Corridor as the
city’s next two PMDs. During 1988 and 1989, LEED Council staff members
met with representatives from all companies in each area to explain the PMD
concept and to build support for the initiatives. Once a core group of sup¬
porters had been identified and mobilized around the proposals, staff mem¬
bers worked with the companies to achieve a consensus regarding the PMD
boundaries and specific land-use guidelines. As in the Clybourn Corridor, the
support base among area manufacturers was eventually broadened with ap¬
peals to labor and nearby community organizations.
On Goose Island, the PMD planning process was dominated by the
Chicago Milwaukee Corporation (CMC), a railroad company that owned a
twenty-five acre unused railyard on the island, which made it Goose Island’s
largest landowner. In 1987, CMC had agreed to develop an industrial park on
the property following guarantees provided by Washington administration of¬
ficials for tax concessions and public infrastructure improvements. However,
with land speculators driving the value of Goose Island property progressively
higher during 1987 and 1988, CMC began to rethink its plans. In July 1988,
the company voiced its opposition to the proposal for a PMD on Goose Is¬
land and announced that its plans for an industrial park there had been put on
hold (Goozner 1988b). The following month, CMC President Edwin Jacob¬
son insisted the company’s Goose Island property would be developed for
“the highest and best use that would be most beneficial for the company’s
shareholders” (Hornung 1988). According to CMC Vice President Wayne
Delfino, “retail, residential, office or even a hotel—or any combination
Map 4. North River Industrial Corridor with Elston Corridor Sub-Area
140 Remaking Chicago

thereof—would all be better uses than industrial” (Fitzgerald 1988).


Despite the opposition of certain large landowners like CMC, a broad base
of support around the Goose Island and Elston Corridor PMD proposals had
emerged by the time Richard M. Daley succeeded Eugene Sawyer as mayor
of Chicago in the spring of 1989. Daley’s election victory, however, appeared
to represent a major stumbling block for the coalition of PMD supporters. In
his election campaign, Daley had expressed deep reservations about the value
of PMDs in the North River Industrial Corridor, insisting that residential and
commercial developments of the sort CMC wished to construct were more
desirable uses than manufacturing for riverfront property. In a January 1989
interview with Crain’s Chicago Business, Daley argued,

Everybody talks about bringing manufacturing back. There aren’t going to be


any more soap factories on Clybourn Avenue. . . . This city is changing. You’re
not going to bring manufacturing back. . . . The [Chicago] river is too impor¬
tant. The river should not be a dumping ground for industries. The river is go¬
ing to be for townhouses, homes, apartments, and offices. (Crain’s Chicago
Business, 30 January 1989)
1
To complicate matters further, neither of the two aldermen in whose
wards the Goose Island and Elston Corridor PMDs were to be located were
strong supporters of the initiatives. Both were machine aldermen reluctant to
back legislation that promised to undermine their authority over land use.
Furthermore, their half-hearted support for the process thus far had been
prompted largely by pressure from the Washington and Sawyer administra¬
tions. With the election of a mayoral candidate on record as opposing more
PMDs, it appeared that, if anything, pressure from the Mayor’s Office would
now be applied in the opposite direction. To virtually no one’s surprise,
forty-second ward alderman Burton Natarus withdrew his support for the
Goose Island PMD several months after Daley was elected, arguing the pro¬
posal would “interfere with the natural marketplace” and reduce land values.1
Lacking both mayoral and aldermanic support, the PMD proposals for Goose
Island and the Elston Corridor faced an extremely uncertain future by the
summer of 1989.
During the early stages of his administration, however, Daley withheld
judgment on the proposal for additional PMDs in the North River Industrial
Corridor. Instead, he asked the Departments of Planning and Economic De¬
velopment to conduct a thorough analysis of the industrial viability of Goose
Island and the Elston Corridor and make recommendations on the need for
protective legislation. The selection of these two departments was significant.
Both were headed by progressive administrators who had chosen to retain a
number of Washington and Sawyer appointees to staff key. positions within
their respective agencies. As a result, each department contained a solid core
Toward a Citywide Industrial Policy 141

of bureaucrats who were familiar with the PMD concept and committed to
moving the process forward.
Not surprisingly, departmental staff built a strong case for the industrial vi¬
ability of both Goose Island and the Elston Corridor, arguing that displace¬
ment pressures posed a serious threat to profitable manufacturers in each
area. According to reports issued joindy by the Departments of Planning and
Economic Development, both locations were ideal sites for manufacturing
because of their proximity to downtown, convenient access to transportation
of various kinds, industrial infrastructure, and diverse industrial bases (De¬
partment of Planning 1990a, 1990b). Industrial real estate markets were
found to be strong, with demand for suitable industrial space well exceeding
the limited supply in each area. Speculative pressures were blamed for inflat¬
ing property values above industrial rates and holding industrial expansion in
both locations to less than its full potential.
At roughly the same time, an additional study of Goose Island was carried
out by a task force convened by LEED Council in response to Mayor Daley’s
initial lack of enthusiasm for the Goose Island PMD proposal. The study, which
received technical assistance from the Departments of Planning and Economic
Development, included an economic and revenue impact analysis of two differ¬
ent land-use scenarios for Goose Island: a commercial/residential scenario and
an industrial scenario. Researchers projected that preserving Goose Island as an
industrial district would contribute $337 million more to the Chicago area s
Gross Regional Product by the year 2001 than a commercial/residential sce¬
nario, due largely to the comparatively higher economic multipliers associ¬
ated with industrial development. In addition, although property tax rev¬
enues were projected to be higher under a commercial/residential scenario,
when revenues from additional sources were taken into account, industry
came out well ahead of commercial or residential alternatives (LEED Council
1990).
Mayor Daley’s opposition to the two new PMD proposals continued to
soften during the winter and spring of 1990. Then in June of that year he
made a startling announcement: he had reconsidered his campaign position
against planned manufacturing districts and was now prepared to support the
planning process for the Goose Island and Elston Corridor PMDs.2 Caving
into pressure from the administration, the aldermen representing both areas
soon agreed to back the initiatives as well. Later that month, despite strong
editorial criticism from the Tribune and appeals from property developers urg¬
ing him to rethink his decision, Daley personally introduced both ordinances
into city council.3 With the backing of the administration and both area aider-
men, the ordinances easily survived a full city council vote later that fall.
Daley’s decision to reverse his campaign stance against PMDs and to take a
position opposed by many of his allies in the real estate community was
prompted by a number of considerations. First, Daley’s new planning
142 Remaking Chicago

commissioner, David Mosena, a supporter of the PMD proposals, had quickly


gained the confidence and respect of the mayor, so much so that Daley went
on to appoint him chief of staff after his reelection in 1991. Mosena’s clout
with the mayor allowed him effectively to end-run Daley’s inner circle of ad¬
visors, none of whom appeared to share Mosena’s enthusiasm for industrial
protection (Hornung 1991). Once his staff had marshaled sufficient evidence
in support of the new PMD initiatives, Mosena was able to present the infor¬
mation directly to the mayor, ensuring that his department’s findings and rec¬
ommendations would get a fair hearing from Daley himself. Mosena’s per¬
sonal lobbying effort on behalf of the PMDs is generally singled out as the
key reason for Daley’s change of heart on the issue (Goozner 1990).
Yet Mayor Daley also faced strong community pressure to continue the
Washington and Sawyer administrations’ policy of industrial retention. Un¬
like a number of other neighborhood-oriented initiatives introduced by
Washington and Sawyer, the PMD proposals originated within the commu¬
nity, not the administration, and were thus more difficult to kill. Supporting
the PMDs was a way for Daley to respond to the policy demands of a politi¬
cally mobilized community without compromising his strong progrowth,
prodevelopment policy agenda. The combination of key bureaucratic sup¬
port, community pressure, and research demonstrating the value of North
Side manufacturers to Chicago’s economy all help explain the mayor’s will¬
ingness to reconsider his position on PMDs.

Industrial Land-Use Plans


In addition to its support for the Goose Island and Elston Corridor PMD
initiatives, the Daley administration also released a series of industrial land-
use plans for the North, South, and West Sides of Chicago between 1991
and 1994, the first such plans issued by the City of Chicago since 1970 (De¬
partment of Planning and Development 1991, 1992, 1994). Like PMDs, the
new plans were intended to head off conflicts between manufacturers and
commercial or residential property developers by clearly identifying areas of
the city the administration was committed to retaining for industrial use. Un¬
like PMDs, however, they contained no legal mechanisms to adjudicate such
disputes. With the publication of the third and final plan of the series in
1994, Department of Planning and Development staff had identified twenty-
two separate industrial corridors on the North, South, and West Sides of the
city, representing a total of 13,761 acres of land.
Although published by the Daley administration, the new industrial land-
use plans were the culmination of a planning process that began during the
Washington administration. Acting on the recommendations of a task force
on industrial and commercial land use established by Mayor Washington in
1986, the Eugene Sawyer administration’s Department of Planfling began an
Toward a Citywide Industrial Policy 143

inventory of industrial land several years later on Chicago’s North Side, the
area of the city where land-use conflicts between manufacturers and real es¬
tate developers had become most pronounced (City of Chicago 1987, 12). A
broad segment of the North Side community—including industrial develop¬
ment organizations, real estate brokers, bankers, and industry representa¬
tives—participated in the planning process. With the assistance of these
groups and individuals, city planners mapped out the boundaries of four
North Side industrial corridors, releasing their findings in a draft report pub¬
lished in January 1989 (Department of Planning 1989). In most cases, the
plan called for “complete protection of all M-zoned land” within a particular
industrial corridor. Besides identifying the boundaries of individual corridors,
the plan included specific recommendations for infrastructure improvements,
marketing assistance, and other business services for area manufacturing firms.
The Daley administration’s industrial land-use plans were an extension of
this planning process, with a number of key differences. First, the base of
community participation in the planning process was far narrower than be¬
fore, limited to the four or five industrial development organizations active in
each of the three study areas. Second, the new plans focused exclusively on
land use and infrastructure improvements; provisions for marketing and other
business assistance present in the Sawyer administration’s earlier plan were
dropped;. Finally, the recommendation for complete protection of manufac¬
turing-zoned land that had figured prominently in the Sawyer administra¬
tion’s plan was replaced by a weaker proposal recommending that rezoning
requests inside the twenty-two designated industrial corridors be subject to a
more stringent review process than petitions for zoning changes elsewhere.
Although the latter provision was never implemented, the Daley adminis¬
tration has shown some commitment to maintaining the industrial integrity
of the corridors. In the summer of 1995, for example, developer Ron Shipka
approached the city with plans to build a six-hundred-unit residential devel¬
opment on a recently vacated twenty-acre industrial site within the North
River Industrial Corridor, several blocks north of the Clybourn Corridor
PMD. Zoned for heavy manufacturing at the time, the site could be devel¬
oped for residential use only if city officials would agree to Shipka s request
for a zoning variance. Sold as industrial property, the parcel was worth
roughly $4 million. With a zoning change, Shipka was willing to pay $17
million (Driscoll 1996).
Shipka’s plans quickly drew criticism from nearby manufacturers, who an¬
ticipated a barrage of complaints and possible legal action from residents of
the new development upset with the noise and odors produced by neighbor
ing industrial firms. The most prominent critic was the Vienna Sausage Man¬
ufacturing Company, which employed 480 workers in a 100,000-square-foot
food processing plant directly across the Chicago River from the proposed
development. Vienna insisted that an approval of Shipka’s rezoning request
144 Remaking Chicago

would ultimately force the company to relocate. According to Vienna Chair¬


man James Bodman, “Our first choice is to stay in Chicago. But we smoke
wood for our sausage, we clean grease out of the water and, on hot days, it
smells here. Add our diesel trucks and noisy rooftop refrigeration systems,
and it will be only a matter of time before nearby residents drive us out”
(Murphy 1996).
Ward alderman Terry Gabinsld was initially in favor of rezoning the site.
However, Daley officials were unwilling to support the proposal without first
testing the market for industrial buyers. With the assistance of LEED Coun¬
cil, Department of Planning and Development staff began marketing the
property to North Side manufacturers in need of expansion space. When a
number of buyers surfaced, Daley came out firmly against the rezoning re¬
quest, arguing that individuals who had spent several hundred thousand dol¬
lars for a condominium would soon pressure the city to close down nearby
factories.4 Gabinski, a close ally of the mayor in city council, soon relented,
and by spring of 1996 the proposal was considered dead.
Despite cases such as this one, the weak provisions governing rezoning re¬
quests in the city’s new industrial land-use plans have left the administration
relatively free to accommodate the interests of commercial and residential
property developers, when it so chooses. In 1994, for instance, developers in
the nearby Addison Industrial Corridor approached area alderman Richard
Mell with a plan to build 330 townhouses on an eleven-acre, manufacturing-
zoned site occupied until recendy by a shipbuilding company. The property
was worth roughly $5 per square foot as industrial space, but the owners
were asking nearly three times that price in anticipation of a zoning change
allowing residential redevelopment of the site (Reardon 1995).
Mell, a prominent machine alderman whose name had briefly surfaced as a
possible choice for acting mayor following Harold Washington’s death in
1987, wanted the development approved, arguing the property would other¬
wise lie vacant. However, opposition to the project soon began to materialize.
Grain’s Chicago Business published a number of editorials in 1995 and 1996
urging Mell to withhold support for the rezoning request. Several neighbor¬
hood organizations active in the area, including the North Business and In¬
dustrial Council and the Lake View Citizens’ Council, also became strong vo¬
cal critics of the proposal. Finally, a number of neighboring industrial firms,
including an electronic games manufacturer employing one thousand work¬
ers, insisted that complaints from residents of the proposed new development
would ultimately force nearby industries to relocate (Snyder 1995).
Although planning officials initially asked Mell to withhold his endorse¬
ment of the proposal pending the outcome of an effort by the city to market
the property to an industrial user, both the administration and Alderman
Mell continued to leave the door open to a zoning change. A 1996 proposal
that would have divided the site between two manufacturing firms eventually
Toward a Citywide Industrial Policy 145

collapsed when the property owners refused to lower their asking price for
the land to an industrial rate. The following year a compromise plan was an¬
nounced: four acres of the site would be retained for industrial use, while
seven additional acres would be rezoned and developed as a residential com¬
plex of townhomes and condominiums. Such high-profile rezonings within
the city’s officially designated industrial corridors have been relatively infre¬
quent, but they have damaged the city’s policy on industrial land use by
sending mixed signals to industries and property developers, undermining
the stability necessary for both groups to undertake long-range plans.

Resources for Industrial Development


The Daley administration’s industrial land-use planning efforts, while sig¬
nificant, represented only one piece of a comprehensive industrial policy for
the city of Chicago. By the 1990s, many of the city’s industrial corridors
showed the effects of long-standing neglect. Decades of street resurfacings
had made viaducts in many industrial areas impassable to large trucks, forcing
drivers to take time-consuming circuitous routes in order to reach local in¬
dustries. Potholes eighteen inches deep or more were a common sight on
many industrial streets. In addition to public infrastructure concerns, many
firms required assistance with job training, marketing, achieving compliance
with environmental regulations, and securing conveniently located expansion
space within the city.
During the early 1990s, the Daley administration began to take a more
aggressive stance toward industrial retention, supplementing its efforts to
protect the integrity of the city’s industrial districts with a number of pro¬
grams and policies designed to support industrial development in Chicago
more actively than in the past. The change in policy was prompted by a num¬
ber of considerations. First, the administration’s attempts to safeguard manu¬
facturing-zoned property through PMDs and industrial land-use plans con¬
tinued to draw heavy criticism from real estate developers, who pointed to
abandoned industrial buildings and undeveloped property in places like
Goose Island as evidence the policy was not working. In a somewhat ironic
twist, the real estate community’s growing demands that the city reverse its
policy on industrial land use pushed officials to begin taking the necessary
steps to produce tangible results in the PMDs and other officially recognized
industrial corridors.
In addition, Mayor Daley experienced a major embarrassment in the sum¬
mer of 1992 when city officials were unable to locate a suitable new site for
the catalog operation of Spiegel, Inc., which had outgrown its existing quar¬
ters on Chicago’s Southwest Side. Spiegel, its 2,200-employee payroll mak¬
ing it the largest employer in Mayor Daley’s own neighborhood of Bridge¬
port, announced in late September that it had selected a suburb oi
146 Remaking Chicago

Columbus, Ohio, as the location for its new warehousing and distribution
center. Spiegel’s departure, coming at a time when the city was in the midst
of negotiations over a planned third Chicago airport and a $2 billion casino
and entertainment complex, highlighted the mayor’s preoccupation with
flashy, headline-grabbing megaprojects, and the failure of his administration
to develop a comprehensive set of policies to retain well-paying manufactur¬
ing jobs for Chicago residents.
To top it all off, Daley’s megaprojects were themselves beginning to en¬
counter considerable hostility in the Illinois state legislature, which proved to
be far less forthcoming with funding and approval for the mayor’s pet pro¬
jects than for those of his father several decades earlier. In 1991, Daley suc¬
cessfully pushed a proposal for a $987 million expansion of the McCormick
Place convention center through the legislature, but only after plans for a
$398 million domed stadium had been dropped from the project (Reardon
and Pearson 1991). Plans for a new Chicago airport and casino and enter¬
tainment complex did not fare as well. By the end of 1992, both projects had
collapsed due to the unwillingness of downstate and suburban lawmakers to
support the construction of high-profile developments that appeared to hold
few benefits for their noncity constituents.
Legislative roadblocks like these were rare during the peak years of the
Richard J. Daley machine, when state assembly approval of large develop¬
ment projects for Chicago oftentimes required little more than a telephone
call from the mayor. During the 1950s and 1960s, much of the population,
industry, and wealth of the state of Illinois was concentrated in Chicago, giv¬
ing Mayor Daley and the Cook County Democratic machine considerable
clout in the Illinois state legislature. According to political scientist Milton
Rakove, Daley worked out a quid pro quo arrangement with downstate Re¬
publican legislators under which they agreed to give him what he wanted for
Chicago as long as he agreed not to interfere with their control over state
government patronage and other perquisites of office (Rakove 1975, 209).
The power relationships that undergirded this set of arrangements have
since changed considerably, due chiefly to the effects of suburbanization. Be¬
tween 1952 and 1992, Chicago’s share of the Illinois vote dropped from 41
percent to 22 percent, and a growing proportion of the state’s wealth and
productive capacity is now concentrated in portions of the metropolitan area
outside the city of Chicago (Judd and Swanstrom 1994, 304). As a result,
Chicago has far less power than it once did in the state capitol. The former
alliance between Chicago Democratic machine politicians and downstate Re¬
publican legislators has given way to a new alliance between suburban and
downstate lawmakers increasingly hesitant to provide funds or approval for
what many of their constituents view as pork-barrel projects for Chicago
politicians. Still stinging from the legislative defeats of his airport and casino
proposals in late 1992, Daley announced in January 1993 that his new eco-
Toward a Citywide Industrial Policy 147

nomic development priorities would consist of relatively unglamorous objec¬


tives such as maintaining infrastructure, improving city services, and retaining
business in the city (Reardon 1993).

The Model Industrial Corridors Initiative


One of the Daley administration’s early efforts to expand its industrial pol¬
icy beyond simple land-use controls was a program established in 1994 called
the Model Industrial Corridors Initiative. By the summer of 1993, the ad¬
ministration was nearing completion of its industrial land-use plans for the
North, South, and West Sides of the city and was now ready to take a more
proactive stance toward retaining manufacturing jobs in the twenty-two in¬
dustrial corridors identified in the plans. In a strong demonstration of his
newfound sympathies for Chicago manufacturers, Daley authorized the hir¬
ing of LEED Council executive director Donna Ducharme as the city’s top
industrial development officer in July 1993, placing her in charge of program
development within the newly designated industrial corridors.
Soon after her appointment in the Department of Planning and Develop¬
ment, Ducharme convened a task force to help define the ideal features of a
viable, inner-city industrial corridor and to develop a program to help indi¬
vidual corridors better achieve these characteristics. The task force a diverse
group including manufacturers, real estate brokers, developers, representa¬
tives from industrial development organizations, and city planning staff
agreed on a set of five basic objectives: industrial corridors should be safe, ac¬
cessible and functional, competitive and marketable, well-managed, and
attractive. The Model Industrial Corridors Initiative was established to meet
these objectives by funding industrial development organizations participat¬
ing in the city’s Local Industrial Retention Initiative (LIRI) program to write
strategic plans for their service areas addressing these five goals. Each group
was also promised anywhere from $1 million to $1.5 million to seed the im¬
plementation of their plans. By 1997, funding proposals submitted by twelve
LIRI organizations from the North, South, and West Sides of Chicago had
been approved by the city.
One of the first organizations funded under the new program was Greater
Southwest Development Corporation, a Southwest Side industrial develop¬
ment organization whose service area consists of a nine-hundred-acre indus¬
trial corridor anchored by a number of large food-products manufacturers,
including Tootsie Roll, Nabisco, and Kool-Aid. Greater Southwest’s strategic
plan, issued in March 1995, is representative of the types of plans that orga¬
nizations participating in the Model Industrial Corridors Initiative have pro¬
duced. Its primary focus is land use and infrastructure. Among other things,
the plan identifies locations where the need for street repairs, viaduct up¬
grades, and other physical improvements is most urgent. It also contains a
148 Remaking Chicago

survey of vacant land parcels in the corridor with the best prospects for in¬
dustrial redevelopment, along with proposals for marketing the property
(Greater Southwest Development Corporation 1995).
In addition to the above recommendations, the plan contains proposals
for increasing intracorridor networking relationships among area firms and
for improving the skills of the local workforce. In a survey of corridor firms
carried out as part of the Model Corridors planning process, nearly 50 per¬
cent of companies expressed an interest in common warehousing and cooper¬
ative purchasing of raw materials or finished goods. Opportunities for the
area’s larger manufacturers to use the services of local rather than outside
suppliers are identified as well. In addition, firm owners singled out the need
for job training as a top priority. Under the plan, employers would work with
nearby educational institutions to develop new programs designed to meet
the specific training needs of local employers. Both these efforts, job training
and the fostering of intracorridor networking relationships, would be over¬
seen by a steering committee made up of local business and community rep¬
resentatives.
Under Ducharme’s leadership, the Model Industrial Corridors Initiative
was developed as a decentralized planning process. By providing LIRI orga¬
nizations with the resources to write and implement their own strategic
plans, the administration has helped create a strong organizational presence
in the neighborhoods that is committed to seeing the process through to
completion. LIRI organizations participating in the program work closely
with area manufacturers to prepare their strategic plans; as a result, the manu¬
facturers are themselves generally strong supporters of the process, willing to
devote time and resources to ensure that the plans are fully implemented.
Aside from improving the program’s chances of success, the decentralized
planning process has helped the administration broaden its base of support in
the neighborhoods by offering policy benefits to LIRI organizations and
their members and allies.

Capital Improvements for Industrial Infrastructure


A certain number of the industrial infrastructure projects identified as pri¬
orities in the strategic plans written by industrial development organizations
participating in the Model Industrial Corridors Initiative can be financed by
the seed money that the city provides for the implementation of each plan.
However, the considerable costs of such projects as bridge and viaduct up¬
grades mean that additional funding sources are oftentimes necessary. The
City of Chicago allocates hinds for industrial infrastructure through its capital
budget. As we have seen in previous chapters, city officials have historically
concentrated capital improvements dollars disproportionately in the Loop
and nearby areas. The Richard M. Daley administration’s infrastructure prior-
149
Toward a Citywide Industrial Policy

ities are not inconsistent with this pattern, but in recent years the share of the
city’s capital improvements funds targeted to industrial areas has increased.
In large part, this increase has come about through the efforts of a city¬
wide coalition of seventy-five neighborhood organizations called the Neigh¬
borhood Capital Budget Group (NCBG), which was founded in 1987 to
provide Chicago citizens with a greater voice in the capital budgeting
process. The City of Chicago’s procedure for planning and funding capital
projects centers around the Capital Improvements Program, a comprehensive
list of citywide capital improvements scheduled to be implemented within
five years. Until recently, the general public was excluded from the capital
budgeting process. Under pressure from NCBG, however, the Daley admin¬
istration established the Capital Improvements Advisory Committee in 1991,
which provides formal channels for public input into the drafting of the city s
Capital Improvements Programs.
While democratizing the planning process somewhat, community partici¬
pation in the formulation of the Capital Improvements Program does not en¬
sure that Chicago neighborhoods will actually receive a growing share of cap¬
ital improvements funds. This is because the Capital Improvements Progiam
is only a plan; it cannot be fully implemented until revenue sources have been
identified for each project. Typically, projects listed in the first year of the
plan haye received funding commitments from the city and are likely to be
completed within a reasonable time period. The status of projects furthei
down the line in the third, fourth, and fifth years of the plan is far less cer¬
tain. Many are never completed at all. In a 1982 study, the Metropolitan
Housing and Planning Council determined the city had implemented no
more than 70 percent of its planned capital improvements projects during the
previous decade (Nagel 1982).
In February 1992, NCBG published a study of the Daley administration’s
1992-1996 Capital Improvements Program which revealed that $350 mil¬
lion in neighborhood infrastructure improvements listed in the plan had no
identifiable funding sources (Reardon 1992b). Included in this amount were
over half the industrial infrastructure projects recommended in the adminis¬
tration’s recently published industrial land-use plan for the city’s West Side.
The study, which received significant media coverage, estimated the adminis¬
tration could leverage an additional $880 million in private investment fbnds
by fully financing the city’s 1992-1996 Capital Improvements Program.
NCBG urged the administration to sell revenue bonds to cover the $350 mil¬
lion shortfall.
The timing of the study was fortuitous. Less than two months later, the
Chicago River burst through a hole in the city’s century-old system of undei-
ground freight tunnels, flooding scores of downtown buildings and bringing
business in the Loop to a virtual standstill for several days. The flood high¬
lighted the poor condition of much of the city’s aging infrastructure and put
150 Remaking Chicago

additional pressure on the Daley administration to secure the revenues neces¬


sary to finance long-overdue repairs. In May 1992, Mayor Daley announced
plans to sell $160 million in General Obligation Bonds for infrastructure im¬
provements, over 90 percent of which were to be located in neighborhood
residential and industrial areas (Davis 1992). Since then, the administration
has issued bonds for neighborhood infrastructure on an annual basis.
The bond offerings, largely the result of pressure by NCBG, have reduced
the disparity between neighborhood industrial infrastructure projects identi¬
fied in the city’s Capital Improvements Program and those that are actually
completed. Table 11 evaluates the degree to which street and viaduct im¬
provements identified in four Capital Improvements Programs issued by the
Daley administration were implemented by the city. The administration’s per¬
formance from 1990 to 1992 was abysmal. Of a total of $37 million
promised, only $8 million were actually invested. Bond issues for neighbor¬
hood infrastructure from 1992 onward helped narrow this gap considerably.
During 1993 and 1994, the city funded nearly 70 percent of industrial street
and viaduct improvements fisted in the Capital Improvements Programs, a
less than stellar performance but clearly an improvement.
The Daley administration’s, record on industrial infrastructure compares
favorably with that of the Washington and Sawyer administrations. Table 12
shows the dollar amounts budgeted by the Washington/Sawyer and Daley
administrations for industrial infrastructure improvements within the seven
industrial corridors identified in the city’s North Side industrial land-use plan
issued in 1992.5 During the six-year period from 1983 to 1988, the Wash¬
ington and Sawyer administrations committed just over $22 million for in¬
dustrial infrastructure within the North Side industrial corridors, while the
Daley administration spent roughly $28.5 million during the five-year period

Table 11. Promised versus Actual Investments in


Industrial Streets and Viaducts, 1990-1994

Year Amount Promised Amount Invested

1990-1991 $18,638 $2,689


1992 $18,376 $5,329
1993 $19,041 $19,228
1994 $43,103 $22,919

Total $99,158 $50,165

Source: Neighborhood Capital Budget Group (1995). Note: Dollar amounts are in thousands. Calculations
are based on the 1990-1994 through 1994-1998 Capital Improvements Programs for the city of Chicago.
151
Toward a Citywide Industrial Policy

between 1990 and 1994. From 1990 to 1994, the Daley administration
completed infrastructure investments worth a total of $80 million in the
city’s twenty-two officially recognized industrial corridors, while seeming
funding for an additional $90 million worth of improvements (Department
of Planning and Development 1996).

Table 12. Capital Improvements Expenditures in the


North Side Industrial Corridors, 1983-1994

Improvement 1983-1988 1990-1994


Project Washington/Sawyer Daley

Street $10,525 $12,465

Intersection/Signalization $528 $125

Bridge $— $9,400

Viaduct $11,204 $6,636

$22,257 $28,626
Total

Sources: Department of Planning (1989); Department of Planning and Development (1996). Note: Dollar
amounts are in thousands. Table includes projects completed and projects fhnded for construction.

Effective community pressure has caused the Daley administration to


reprioritize its capital budgeting process in order to increase binding for in¬
dustrial infrastructure. Yet the city’s capital budget remains, above all, highly
skewed in favor of the downtown area. An analysis of the city’s Capital Im¬
provements Programs covering the years 1990 through 1999 reveals that the
downtown area is slated to receive an annual average of $19.4 million in eco¬
nomic development infrastructure funds during the 1990s. Meanwhile, forty-
two of the city’s other forty-nine wards will receive a yearly average of just
$1.5 million or less in development funds (Neighborhood Capital Budget
Group 1996, 14). This gap is beginning to narrow somewhat, but there can
be no doubt that the Daley administration’s capital improvements priorities
continue to reflect the strong position that downtown business interests oc¬
cupy in the city’s present governing coalition.

Tax Increment Financing


In recent years, the City of Chicago has come to rely increasingly on a rel¬
atively new policy innovation called tax increment financing (TIF) to provide
public improvements and other economic development incentives for
Chicago industries. Tax increment financing authorizes the creation of special
152 Remaking Chicago

districts where taxes paid on any increase in land values are set aside for rede¬
velopment projects within the district boundaries. The key requirement for
the establishment of a TIF district is the finding that, absent significant public
investment, property values within the targeted area are likely to remain at or
near existing levels for the foreseeable future. Once a district has been created,
the assessed value of the land is frozen, and all incremental tax revenues from
this point forward are placed in a TIF fund to be used for redevelopment pur¬
poses. Eligible redevelopment activities include land acquisition, site prepara¬
tion, building rehabilitation, public capital improvements, and job training
and other related educational programs. State legislation requires that all Illi¬
nois TIF districts be dissolved after a maximum of twenty-three years.
By 1997, Chicago had seventeen industrial TIF districts covering well over
3,000 acres of land. According to data compiled by the Department of Plan¬
ning and Development, industrial TIF districts in Chicago had, by that point,
created roughly 2,660 new jobs and retained approximately 6,750 others, of¬
tentimes by financing costly predevelopment work such as environmental re¬
mediation of brownfield sites (Department of Planning and Development
1998, 7). On Goose Island, for example, a recently established TIF district
played a key role in securing commitments from several major investors, lead¬
ing to the construction of a 120,000-square-foot Federal Express distribution
center and a 375,000-square-foot window-manufacturing facility. Both com¬
panies were persuaded to build there when the city agreed to use TIF funds
for environmental cleanup and job-training purposes (Fingeret 1998).
Although tax increment financing has been used successfully as an indus¬
trial development tool in Chicago, the program is not without its detractors.
In particular, critics charge that the Daley administration has used the policy
too liberally, establishing TIF districts in areas like the central business district
where, it is argued, private investment dollars would have been forthcoming
without such aggressive city intervention. In such cases, tax increment financ¬
ing may divert revenue from local taxing authorities, taking funds that would
otherwise go to the school board, Park District, and other units of local gov¬
ernment and utilizing them for economic development purposes alone
(Washburn and Martin 1998).

Evaluating Industrial Policy in Chicago


It is still far too early to say with any degree of certainty whether planned
manufacturing districts, tax increment financing, the Model Industrial Corri¬
dors Initiative, and other programs and policies introduced by the Washing¬
ton, Sawyer, and Daley administrations to protect and support industrial de¬
velopment in Chicago have been successful. Major investments on the part of
manufacturing firms require long-range planning. Expansion plans made by
companies within the city’s three PMDs in the years following passage of the
Toward a Citywide Industrial Policy 153

initiatives in some cases have yet to be fully implemented. Current data on


job and firm retention and growth within the PMDs may thus not be reliable
indicators of future trends. Programs like tax increment financing and the
Model Industrial Corridors Initiative are even more difficult to evaluate,
since many industrial TIF districts are less than two years old and few LIRI
organizations participating in the city’s Model Industrial Corridors Initiative
have moved beyond the planning phase of the program.
Despite these limitations, it may still be possible to make some preliminary
judgments on the effectiveness of Chicago’s newly emerging industrial policy.
One way of gauging the extent to which the three PMDs and the city’s new
industrial land-use plans are likely to curb the displacement of manufacturers
within Chicago industrial districts is to examine recent trends in property val¬
ues within the PMDs and officially recognized industrial corridors. If the real
estate community is convinced that the city is serious about preserving these
areas for industrial development alone, we would expect to see a decline in
real estate speculation and a stabilization of land values in such locations.
Faced with the likelihood that rezoning requests there would be denied,
speculators would have little incentive to bid up the price of industrial land to
commercial or residential rates. As property values level off, nearby expansion
space should become increasingly affordable to area manufacturers, while
property tax increases associated with rising land values should become less
pronounced and burdensome to industrial firms.
Tables 13, 14, and 15 illustrate trends in land values from 1975 to 1994
for three locations: the Clybourn Corridor PMD; the Goose Island PMD;
and the Ravenswood Corridor, a North Side industrial corridor in which loft
conversions to residential use began to occur during the late 1980s. Property
values within the Goose Island PMD and the Ravenswood Industrial Corri¬
dor rose substantially during the 1980s but leveled off after 1990. In the
Clybourn Corridor, land values stabilized only within the core zone of the
PMD, while values in the PMD buffer zone more than kept pace with the
price of commercially zoned property nearby. This comes as no particular
surprise, since industrial and commercial development are both permitted uses
within the PMD buffer zone.6 If the Clybourn buffer zone is excluded from
consideration, however, all three areas exhibited considerable price stability
from 1990 to 1994, following sizeable increases during the 1980s. This, cou¬
pled with the fact that property values in the vicinity of each area continued
to increase measurably during the early 1990s, is a strong indication that land
speculators and property developers are taking the city’s new industrial reten¬
tion policies seriously.
Of course, the ultimate test of any city’s industrial policy is its effect on job
and firm retention and growth. Again, it is still much too soon to make defini¬
tive judgements, but early data are of interest. Table 16 shows job and firm re¬
tention rates in the three PMDs and four of the city’s officially recognized
154 Remaking Chicago

industrial corridors between 1988 and 1992.7 During this time, the city of
Chicago lost a total of 534 manufacturing firms and 33,700 manufacturing
jobs, declines of 12 and 15 percent, respectively. As Table 16 indicates, the
performance of industrial firms within the city’s PMDs was considerably better
than this. In the three PMDs combined, the number of industrial firms rose
by 3 percent, while manufacturing employment experienced a 7 percent de¬
cline. By once again eliminating the Clybourn Corridor buffer zone from the
picture, the 7 percent decline in manufacturing jobs becomes a 4 percent in¬
crease. With the possible exception of the Ravenswood Corridor, the four in¬
dustrial corridors listed in Table 16 also showed considerably greater stability
than the city of Chicago as a whole.
These preliminary findings suggest that Chicago’s industrial policy is al¬
ready having an impact. Considering that much of the private investment ac¬
tivity that recent policies and programs like planned manufacturing districts

Table 13. Land Values in and near Core Zone


of Clybourn Corridor PMD, 1975-1994

Location 1975 4980 1985 1990 1994

PMD Core Zone $2.00 $2.50 $3.00 $6-12.00* $6-12.00*

PMD Buffer Zone $2.50 $2.50 $4.00 $40.00 $50.00

Corner of North $2.00 $2.50 $15.00 $30.00 $40.00


and Clybourn
(Opposite SE
Corner of PMD)

Sources: Olcott (1975, 1980, 1985, 1990, 1994). Note: Amounts for PMD core and buffer zones represent
maximum land values. *Includes areas within southern portion of PMD buffer zone.

Table 14. Land Values in and near Goose Island PMD, 1975--1994

Location 1975 1980 1985 1990 1994

Goose Island PMD $1.75 $2.00 $2.00 $9.00 $12.00

Corner of Halsted $2.00 $2.00 $2.50 $12.00 $15.00


and Blackhawk
(two blocks east of PMD)

Sources: Olcott (1975, 1980, 1985, 1990, 1994). Note: Amounts represent dollars per square foot.
Amounts for Goose Island PMD are maximum land values.
Toward a Citywide Industrial Policy 155

and tax increment financing seem likely to generate has not yet been fully re¬
alized, the modest job and firm growth within the city’s three PMDs is cause
for optimism. Still, if Chicago is to compete successfully with suburban in¬
dustrial parks and other noncity locations, administration officials will need to
concentrate further resources and planning on industrial areas of the city.
Programs such as the Model Industrial Corridors Initiative, while promising,
are highly underfunded. As with land use, efforts to reprioritize capital im¬
provements, public financing, and other economic development incentives to

Table 15. Land Values in and near Ravenswood Industrial Corridor,


1975-1994

Location 1975 1980 1985 1990 1994

Ravenswood $2.25 $2.50 $3.00 $6.00 $6.00


Industrial
Corridor

Corner of Ashland $4.00 $3.75 $4.50 $9.00 $11.00


and Wilson
(two blocks tast of corridor)

Sources: Olcott (1975, 1980, 1985, 1990, 1994). Note: Amounts represent dollars per square foot.
Amounts for Ravenswood Industrial Corridor are maximum land values for M-zoned property.

Table 16. Changes in Industrial Firms and Employment


in PMDs and Industrial Corridors, 1988 and 1992

1988 1992 Percent Change


Firms Jobs Firms Jobs Firms Jobs
Location

75 1,484 72 1,582 -4 7
Pilsen Corridor
122 6,565 110 6,871 -10 5
North River Corridor
46 2,372 51 2,325 11 -2
Addison Corridor
101 2,688 99 1,991 -2 -26
Ravenswood Corridor

154 5,221 159 4,832 3 -7


PMDs
112 4,111 119 4,261 6 4
PMDs minus Clybourn
Buffer Zone

4,377 220,600 3,843 186,900 -12 -15


City of Chicago

Sources: LEED Council (1996); U.S. Bureau of the Census (1987, 1992).
Note: Jobs represent manufacturing employment.
156 Remaking Chicago

reduce disparities between downtown and the neighborhoods will continue


to pit neighborhood residents, workers, and manufacturers against powerful
land-based interests concerned chiefly with generating public support for re¬
development projects in the downtown area. Organizational pressure from
the neighborhoods will no doubt continue to play a crucial role in determin¬
ing the city’s response.

Conclusion
By all appearances, Richard M. Daley’s victory in Chicago’s 1989 mayoral
contest represented the final nail in the coffin of the Harold Washington re¬
form movement. Daley’s campaign speeches emphasizing the importance of a
vibrant downtown commercial core to Chicago’s economy and his strong fi¬
nancial support within the downtown business community left few doubts
about where his economic development priorities would lie. Surely, efforts
undertaken by Harold Washington to broaden the base of participation in
city government, to reallocate resources in favor of low- and moderate-in-
come neighborhoods, and to resurrect Chicago’s declining manufacturing
sector would quickly become a thing of the past. While few anticipated a re¬
turn of the machine in its classic harm, most expected Daley’s leadership style
and policy choices at least to approximate those of his father.
Such an agenda may well have been utmost in the mind of the younger
Daley as well. However, Richard M. Daley has found himself on much differ¬
ent political terrain than his father was decades earlier. Patronage jobs, city
contracts, and other material incentives that propelled the elder Daley to six
election victories between 1955 and 1976 even as he concentrated policy
benefits in the hands of a select group of downtown business leaders became
increasingly scarce by the late 1970s. Meanwhile, community-based organi¬
zations previously excluded from decision-making arenas became legitimate
and well-respected participants in policy discussions and debates under the
Harold Washington administration, in some cases forming alliances with
neighborhood manufacturers around economic policies and programs hold¬
ing direct rewards for community residents and workers. These organizations
and their coalition partners were not anxious to relinquish titeir gains follow¬
ing the death of Harold Washington in 1987 and Eugene Sawyer’s loss to
Richard M. Daley several years later. Responding to new political realities,
Daley has selectively broadened his governing coalition to include certain
representatives from Chicago’s working-class and industrial neighborhoods.
Neighborhood manufacturers seeking protection and support from city
government have fared particularly well under the Daley administration. Due
in no small part to the publicity surrounding the Clybourn Corridor PMD
debate during the mid-to-late 1980s, policies designed to protect inner-city
manufacturing districts from encroachment by commercial or residential de-
Toward a Citywide Industrial Policy 157

velopment now enjoy broad-based support within Chicago’s working-class


neighborhoods. Following the example set by Washington and Sawyer, Daley
has used city government support for neighborhood manufacturers as a
means of bridging the accumulation/legitimation divide that frequently jeop¬
ardizes the long-term success of redistributive programs and policies in con¬
temporary cities. By supporting PMDs, neighborhood infrastructure pro¬
grams, and the like, Daley has expanded his base of support in the
neighborhoods without compromising his administration’s strong progrowth,
prodevelopment economic-policy stance. In doing so, he has clearly ruffled
some feathers in certain segments of the real estate community, but his other¬
wise strong support for downtown development has left his standing within
much of the downtown business community largely untarnished.
Seen through the lens of urban political economy, the administration of
Richard M. Daley resembles that of his father in key respects. Downtown re¬
development has once again become the centerpiece of economic develop¬
ment policy. The alliance between business leaders and city officials has been
reestablished, and the divide between governing and electoral coalitions has,
to some degree, been recreated. In reality, however, Chicago’s current gov¬
erning arrangements are less a return to the old regime than a partially suc¬
cessful attempt to reconstruct the old regime under new political and eco¬
nomic conditions. Business leaders today must share the stage with new
actors, ideas, and economic development strategies that fit uneasily into the
public/private, accumulation/legitimation divides of contemporary urban
theory. The corporate-center strategy may once again be the economic devel¬
opment strategy of choice, but Chicago also has the makings of a citywide in¬
dustrial policy. Assuming that neighborhood coalition-building around in¬
dustrial retention and redevelopment remains as effective as it has been in the
recent past, Chicago’s newly emerging industrial policy is likely to gather mo¬
mentum, even within an administration whose economic development priori¬
ties appear to lie elsewhere.
Conclusion

T ■Jkhe redevelopment of postwar Chicago was shaped, in significant part,


through politics. Soon after World War II, proponents of a corporate-center
strategy for downtown commercial and residential redevelopment won the
necessary political backing to impose their vision of a postindustrial city on the
central area of Chicago. In time, however, their plans were challenged and ul¬
timately compromised by another set of interests committed to preserving the
manufacturing jobs base of the near-downtown area. Proponents of a local-
producer strategy for neighborhood redevelopment formed an alliance with
the Harold Washington administration, helping to effect a regime change in
which local economic development priorities, in addition to the distribution
of economic rewards, were rearranged to benefit new segments of the urban
community. More recently, manufacturers and their allies have successfully ex¬
tended their local-producer strategy, pressuring the Richard M. Daley admin¬
istration to assemble additional components of a citywide industrial policy.

Rethinking Urban Political Economy


The political contest between advocates of the corporate-center and local-
producer strategies in Chicago has gone largely unnoticed by contemporary
urban scholars because existing theory teaches that urban political conflict,
where economic development questions are concerned, centers on redistribu¬
tive and regulatory matters—not the trajectory of economic growth itself.
Within the urban political economy paradigm, the key struggles are those
waged by low- and moderate-income neighborhood residents seeking a
greater share of the benefits of economic growth, along with protection from
displacement caused by major land-use changes. When urban political econo¬
mists speak of the “business community” and its role in city politics, what
they are really talking about is a particular segment of the business commu¬
nity, chiefly those land-based interests that form the nucleus of urban growth
machines. Business establishments located outside the central business district
Conclusion 159

count for relatively little in contemporary urban theory, as do coalition-build¬


ing efforts between such enterprises and community-based organizations.
It is worth considering for a moment why this is so. Urban political econ¬
omy, as we saw in Chapter 1, begins with a series of propositions about the
government-economy relationship and the behavior ol present-day capitalist
firms. Chief among these assumptions is the division of labor between state
and market, which relegates policymaking to popularly elected officials and
economic decision-making to holders of private investment capital. According
to political scientist Stephen Elkin, social well-being depends to a great extent
on market transactions because major social decisions are made through the
activities of private business firms afforded considerable discretion (Elkin
1987a, 130). Given the limited powers of city government, investment must
be induced rather than commanded, creating a privileged position for business
in city politics. Decisions by private investors acting in the marketplace create
problems, such as deindustrialization and urban economic restructuring,
which city officials must wrestle with. While officials may exercise some degree
of choice in responding to such developments, the marketplace ultimately sets
the parameters for political intervention. As Alan DiGaetano argues,

[T]he relationship between the economy and the local state is not a direct one,
but is mediated by city governing coalitions. That is, while the economic con¬
text sets the boundaries for what is possible in urban development politics, a
city’s governing coalition interprets what can and should be done within that
context. (DiGaetano 1989, 279)

Urban political economists argue that the balance of power between city
officials and private investors in the post-World War II era has shifted notice¬
ably in favor of the latter, due chiefly to the effects of a second key feature of
the urban political economy: the rising mobility of capital. The transportation
and communications revolution has made investors increasingly footloose,
capable of moving plants and equipment to distant locations on relatively
short notice. Capital mobility has set up a bidding war among cities for pri¬
vate investment, forcing city officials to offer an array of incentives to capital¬
ist firms in order to sustain adequate levels of investment activity. Although
place-based assets—such as the agglomeration economies often found in cen¬
tral business districts—remain indispensable to certain investors, they are not
a constraint on mobility. Since the value of all such assets is said to be re¬
flected in land prices, investors can always sell their holdings and move else¬
where without being penalized by the market.
Finally, urban theorists generally maintain that mobile investors make loca¬
tion decisions based on a narrow economic logic. Specifically, firm owners
and managers estimate the future costs of doing business in various places,
locating new facilities where they are anticipated to generate the highest prof¬
its for the enterprise. This logic of decision-making, combined with capital
160 Remaking Chicago

mobility, accounts for the current trend of industrial decentralization and


concentration of corporate and financial activity downtown. Manufacturers
seek low-cost land and labor in suburban, rural, and Third World settings,
while high-level corporate and financial institutions take advantage of myriad
business support services that the central business districts of large cities offer.

These three propositions—the division of labor between state and market,


the mobility of capital, and the instrumental rationality of capitalist firm be¬
havior—lead, in turn, to a number of conclusions about urban politics that
are central to the urban political economy paradigm.

An alliance exists between city officials and landed interests. Public officials,
held accountable along with private investors for the city’s economic perfor¬
mance, must forge a working relationship with the business community.
While the specifics of this relationship may vary over time and space, it is al¬
most certain to involve a close alliance between city officials and major
landowners, both of whom share an interest in promoting high-intensity land
uses. According to Stephen Elkin, this alliance has been sealed in the postwar
era by the mobility of capital.

Among other things, local politicians are drawn into alliance with those whose
interests are tied to preserving the value of local land uses because both are
deeply concerned to hold and attract mobile capital. Those who hold fixed cap¬
ital in land need to attract high-intensity use to maintain and enhance its value.
And local politicians wish to have the political benefits that a vibrant local econ¬
omy will send their way. (Elkin 1994, 136)

Economic development is synonymous with real estate development and rising


land values. For urban political economists, economic growth means foster¬
ing higher and better uses for property by attracting mobile investors. To use
the terminology employed by Logan and Molotch (1987), the goal of devel¬
opment is to enhance the exchange value of land. The benefits of growth—
including jobs and other material rewards, rising property-tax revenues, and
favorable publicity for city officials—all follow from this objective.

Development politics divide along a growth/antigrowth axis. Economic


growth, despite its advantages, is not without controversy. Efforts to boost
exchange values of land through rearranging and intensifying land use may
disrupt residential use values by driving up rents or otherwise causing dis¬
placement. Moreover, because the market has generated a pattern of invest¬
ment activity that increasingly concentrates wealth downtown, redistributive
Conclusion 161

policies are necessary if economic disparities between downtown and the


neighborhoods are to be reduced. Officials from some cities have sought to
address such concerns by taxing downtown growth and using the proceeds
for neighborhood development or by establishing quotas for minority partici¬
pation in city-sponsored redevelopment projects. However, claims that such
efforts damage the local business climate and scare away mobile investors
have made public officials extremely cautious about pursuing policies and
programs like these.

The above propositions and the conclusions to which they seem to lead
may indeed successfully capture certain rough characteristics of the urban po¬
litical economy. Still, the portrait that emerges is a relatively featureless one,
largely devoid of the kinds of actors, alliances, conflicts, and political oppor¬
tunities that the story told in this book focuses upon. Urban theorists today
frequently lament the dominance of growth machines in city politics and the
plight of low- and moderate-income city neighborhoods, yet they are often at
a loss to identify possibilities for change. By reexamining the central proposi¬
tions of the urban political economy paradigm in view of the evidence pre¬
sented in this study, we can begin to see why. Simply put, urban political
economy defines opportunities for genuine political agency in exceptionally
narrow^terms, far more so than the evidence here seems to support.
Consider, first of all, the division of labor between state and market. Ur¬
ban political economists interpret the corporate-center strategy for down¬
town redevelopment embarked upon by cities like Chicago, Baltimore, and
New York following World War II as a response to urban economic restruc¬
turing. According to the prevailing viewpoint, structural economic change
left policymakers with a limited range of options. They could downplay mat¬
ters of economic development and focus attention instead on service provi¬
sion to neighborhood constituents. However, given a commitment to eco¬
nomic growth, officials were drawn inevitably toward a development strategy
that required major land-use changes in the downtown area. The market, in
other words, set the parameters for political response.
Yet what precisely are the market forces in the story of downtown redevel¬
opment in Chicago? Clearly, private investors made decisions that had certain
consequences, but efforts to untangle their activities from those of city offi¬
cials committed to a particular program of redevelopment are problematic.
For example, rising land values in Chicago’s central area were an important
cause of industrial decline there, but land speculators were guided in their ac¬
tions by land-use plans furnished by city officials and a pattern of public capi¬
tal investments and other incentives designed to foster commercial and resi¬
dential development in this part of the city. Had the city’s priorities been
otherwise, there is no reason to assume that land prices would have escalated
to the extent they did. The notion that structural economic change mandated
162 Remaking Chicago

a particular course of action by city officials becomes even less plausible when
we consider the origins of the Daley administration’s downtown redevelop¬
ment program in the late 1950s, well before officials could have discerned any
long-term trend of industrial decentralization.
The corporate-center strategy in Chicago was the product of an alliance
between landed interests and city officials, not a political response to struc¬
tural economic change. Actors from the public and private sectors worked
hand-in-hand to produce land-use changes that would reshape both the
physical appearance of the city and its underlying economic structure. There
were, of course, good reasons for public officials to enter into such an al¬
liance. The growth coalition of land-based interests was well-organized, po¬
litically astute, and controlled substantial economic resources. In addition,
major land-use projects provided politicians with visibility and a steady stream
of material benefits that could be used to build and maintain public support.
The point, however, is that the business-government alliance was not an in¬
tervening variable responding to structural economic change but an au¬
tonomous force responsible for basic economic outcomes.
Consider, as well, the other central propositions of the urban political
economy paradigm: capital mobility and the instrumental rationality of firm
decision-making. To be sure, technological innovations relaxed locational
constraints on certain types of business establishments in the postwar era, en¬
couraging many industrial firms in particular to seek low-cost land and labor
outside city limits. However, the pattern of industrial decentralization was
uneven across industries and industry segments. Large, vertically integrated
firms utilizing mass-production techniques to capture scale economies fled
willingly in many cases, while smaller, vertically disintegrated firms utilizing
flexibly specialized production methods to achieve economies of scope left
only when redevelopment pressures presented them with few alternatives.
In Chicago, segments of key central area industries like printing and ap¬
parel were deeply dependent upon the economies of place provided by near¬
downtown locations. Part of this advantage came through agglomeration
economies—the reduction of individual firm costs that result from the loca¬
tion of a variety of business services nearby. Since agglomeration economies
are realized through simple exchange relations between and among firms,
they are reflected in land values. That is, landowners are free to trade such as¬
sets in the marketplace because other firms can presumably realize the same
benefit from the land.
Yet there were more than simple agglomeration economies at work in
Chicago’s central area industrial districts. As we saw in Chapter 3, interfirm
relations in the printing and apparel industries were permeated by noncapital¬
ist values like trust and cooperation, which led in turn to information-shar¬
ing, job referrals, and other reciprocal arrangements among firms. Place-
based assets like these were fostered through spatial proximity, yet, unlike
Conclusion 163

agglomeration economies, they were the product of individual social rela¬


tionships, which made them unavailable for purchase and sale in the market¬
place. As such, their value was not incorporated into the price of land. When
central area industrial land prices did finally rise significantly, it was due to an
entirely different phenomenon: real estate speculation by commercial and
residential property developers. Ultimately, government intervention de¬
signed to curb property value increases was selectively applied, not to inhibit
economic growth but to preserve a particular form of capitalist production
dependent upon rooted relations of place.

Having examined the key assumptions of the urban political economy


framework in view of the evidence provided in this study, we are now in a po¬
sition to set forth an alternative set of conclusions about the urban political
economy that takes into account the possibilities for meaningful political
agency identified in the story told here.

The nature of the business-government alliance is historically specific. Given


the importance placed by voters on healthy rates of economic growth and the
dependence of city tax revenues on private investment activity, politicians
who ignore the concerns of business are unlikely to enjoy lengthy and pros¬
perous political careers. However, the alliance between city officials and
landed interests that so preoccupies urban political economists is only one of
multiple possible scenarios for business-government cooperation. Since
higher land values may be detrimental to certain types of capitalist produc¬
tion, such as geographically rooted forms of industrial capital, city officials
may' take steps to curb real estate speculation or otherwise act in ways con¬
trary to the interests of the real estate community without compromising
their commitment to economic growth. The nature of the business-govern
ment alliance is thus politically contingent, not structurally determined.

The definition of economic development is politically contested. Recall that ur¬


ban political economists define economic growth as the intensification of land
use. Efforts to rearrange and intensify land use increase the exchange value of
land, generating profits for developers and enhancing the city’s property-tax
base. Yet, as we have just seen, capitalist firms depend on place-based assets
that are not always reflected in the price of land. Indeed, the development of
property for higher and better uses may well be a hindrance to certain types
of capitalist growth, should land speculation break up viable producer net¬
works. By interpreting growth as a measure of the exchange value of land
alone, urban political economy has taken sides with landed interests in an ide¬
ological conflict over how economic development is to be defined.
164 Remaking Chicago

Chicago’s planned manufacturing district initiatives came about, in no small


part, through a successful struggle to sever the identification of economic
growth with real estate development alone. Proponents of PMDs voiced a
theme first articulated during the Harold Washington mayoral campaign—that
job creation, not real estate development, should be the ultimate goal of eco¬
nomic development policy. Once the definition of growth became the object
of ongoing political struggle, developers’ claims that their proposals repre¬
sented the highest and best uses for city land rang increasingly hollow, particu¬
larly since redevelopment often meant replacing high-paying manufacturing
jobs with low-level service-sector positions. When research focusing on con¬
tested areas of the city such as Goose Island projected that, despite lower
property values, the returns to the city would be higher for manufacturing
than for residential or commercial scenarios, even formerly skeptical city offi¬
cials became increasingly forceful proponents of industrial retention.

Development politics include conflict over different growth strategies. Al¬


though political cleavages between downtown and the neighborhoods re¬
main a central fixture of urban development politics, this divide does not al¬
ways involve struggles between neighborhood residents and business elites
over the costs and benefits of growth. Contrary to dualities often posed in
the urban political economy literature, city officials are not necessarily forced
into balancing acts between accumulation and legitimation, or between the
exchange values and use values of land held dear by land-based interests and
neighborhood residents, respectively. Rather, because neighborhoods are
themselves repositories of key productive assets, conflict may involve compe¬
tition for the political resources necessary to implement growth strategies be¬
ing advanced by different groups. Should alliances develop among neighbor¬
hood manufacturers, workers, and community-based organizations, as they
did in Chicago, city officials may be supportive of both neighborhood inter¬
ests and viable forms of economic growth.

The alternative urban political economy outlined in the last few pages has
important implications for urban regime formation. Scholars utilizing die rel¬
atively new paradigm of urban regime theory have built on the assumptions
of urban political economy to identify the various types of urban governance
structures that are viable in liberal democratic societies such as ours (see, for
example, Stone 1987b, 1993; Stone, Orr, and Imbroscio 1991; Elkin 1985).
Unsurprisingly, the typologies set forth fall within a relatively narrow range.
City officials, their choices limited by capital mobility, fiscal pressures, and
other structural constraints, find themselves drawn by necessity into alliances
with downtown business elites around redevelopment efforts that hold few
Conclusion 165

benefits for low- and moderate-income neighborhoods. Progressive regimes


representing neighborhood interests are unsustainable in all but a few situa¬
tions, since the need to focus resources on economic growth overwhelms all
other considerations.
If the argument of this book is correct, however, and the structural factors
at work in city politics are less pervasive than they are generally imagined to
be, then.the possibilities for regime formation are likewise more open-ended
than scholars generally hold. Governing coalitions committed to fostering
economic development need not rule at the expense of neighborhoods.
Rather, alliances among locally rooted producers, workers, and community-
based organizations could serve as the cornerstone for urban regimes com¬
mitted to building on place-based productive assets that exist in city neigh¬
borhoods. Instead of balancing the interests of downtown business elites in
economic growth against those of low- and moderate-income city residents
in consumption, such a regime would tap the potential of neighborhoods to
become centers of flexibly specialized, locally rooted production. The extent
to which Chicago’s local-producer strategy for neighborhood redevelopment
has carried over into the Richard M. Daley administration, despite the ad¬
ministration’s close ties with downtown business, is powerful evidence of the
feasibility of urban regimes built around such alliances.

Chicago Exceptionalism?
Although this study focuses on the historical experiences of one city, I
have suggested that the story told here contains lessons that extend well be¬
yond Chicago. Such an argument is convincing, however, only to the extent
that similar political and economic development choices and conflicts have
emerged in other cities as well. If the political culture or economic conditions
of postwar Chicago are in important respects unique or exceptional, there are
limits to the value of this case for more comprehensive theory building.
The presence of viable networks of small- and medium-sized producers in
other postwar central cities is, in fact, well documented. In cities such as New
York, Boston, San Francisco, and Los Angeles, industries from sectors like
printing, apparel, food processing, medical supplies, and electronics have clus¬
tered in well-defined districts, oftentimes subcontracting with neighboring
firms and taking advantage of nearby business services (see, for example, Hall
1959; Rapkin 1963, 91-114; Kenyon 1964; Scott 1988; Fernandez Kelly
1989- King 1987; Angwin 1996). As in Chicago, key industry segments have
shown considerable resilience in the face of industry restructuring, meeting
the competition of lower-cost producers outside city limits with high-quality,
specialized products and rapid turnaround times. In many cases industries
have been forced to contend with commercial and residential redevelopment
pressures as well. Documented case studies of industrial displacement include
166 Remaking Chicago

New York City, Baltimore, San Francisco, New Orleans, Philadelphia, Boston,
Syracuse, and Portland, in addition to Chicago (Center for Urban Economic
Development 1985; King 1988a; Zukin 1982; City of New York 1981).
A number of cities besides Chicago have also implemented policies de¬
signed to protect centrally located industrial districts from competing land
uses. In New York City, for example, real estate pressures developing during
the 1970s began to threaten the integrity of Manhattan’s garment district,
where commercial zoning allowed the conversion of industrial loft buildings
to office space “as of right” (Zukin 1982; City of New York 1981). Respond¬
ing to pressure from industry groups, apparel-related labor unions, and com¬
munity representatives, the city planning commission established new zoning
restrictions for the garment district in April 1987. Under the new regula¬
tions, developers are allowed to convert industrial space to office use only if
they guarantee that an equal amount of space will be preserved for manufac¬
turing (King 1988a; Kennedy 1988). The measure protects over three mil¬
lion square feet of manufacturing space within the district.
In Boston, growing demand for downtown office space began to threaten
centrally located industrial districts like Fort Point Channel and the Newmar¬
ket area during the mid-1980s (Malone 1987; King 1988a). At that time, the
city zoning ordinance allowed property zoned for manufacturing to be
reused for commercial purposes without a zoning change or any other formal
review process. As in Chicago, rents and property values in the near-down¬
town area experienced strong upward pressures during the 1980s, threaten¬
ing profitable manufacturers from sectors requiring central locations, such as
printing and publishing. In response, the Boston Redevelopment Authority
and the Economic Development and Industrial Corporation developed a
proposal to establish light manufacturing districts where the conversion of
manufacturing space to nonindustrial uses would be tightly regulated. Under
the proposal, approved by the Boston Zoning Commission in December
1988, developers may convert no more than 40 percent of an industrial
building within an officially recognized manufacturing district to office space
(King 1988c). Permitted industrial uses in the twelve new districts estab¬
lished since 1988 include printing and publishing, computer and related in¬
dustries, medical research and development, and a number of other light in¬
dustries.
Finally, city officials in Portland, Oregon, have established sixteen indus¬
trial sanctuaries throughout the city since 1981, each designed to discourage
gentrification or conversion and to promote future industrial development
(King 1988a). Within such areas, most office and retail development is either
forbidden or carefully regulated, while residential uses are limited to caretak¬
ers’ living quarters and artists’ living and work space. In addition, Portland
planning officials have used infrastructure improvements, low-interest loans
for manufacturers, and tax increment financing to promote industrial devel-
Conclusion 167

opment within the sanctuaries. From 1981 to 1988, manufacturing jobs


within the sanctuaries increased by 30 percent even though citywide blue-
collar employment experienced a 10 percent decline (King 1988a).
On the surface, these policies appear to resemble the planned manufactur¬
ing districts established on Chicago’s Near North Side during the late 1980s
and early 1990s, which protected the integrity of North Side manufacturing
districts by calling a halt to the issuing of zoning variances for nonmanufac¬
turing uses. Yet there is an important difference. In New York City, Boston,
and Portland, policies designed to protect manufacturers were developed
within the city bureaucracy for the benefit of manufacturers and their allies.
In each case, city officials perceived a need for protective legislation and acted
accordingly. Chicago’s planned manufacturing districts, by contrast, origi¬
nated within the neighborhoods themselves. A coalition of neighborhood or¬
ganizations, manufacturers, and labor unions developed the legislation with
little formal participation by the city and then pressured often reluctant offi¬
cials to act. Only after a strong, independent political base had been orga¬
nized in favor of the districts did government officials agree to endorse the
proposals.
To understand why manufacturers and community organizations rather
than bureaucrats were the driving force behind protective legislation for man¬
ufacturing in Chicago, it is necessary to consider for a moment the impact of
the Democratic machine on community politics in postwar Chicago. Recall
that machine politicians maintained support through the provision of partic¬
ularistic benefits like political patronage and material incentives and rewards,
a practice that made them unlikely champions of divisive legislation such as
planned manufacturing districts aimed at addressing the collective aspirations
of neighborhood constituents. In cities like New York and Boston, political
machines were undermined by the steady erosion of their Irish and German
immigrant electoral bases before and after World War II, a trend that paved
the way for new progressive and reform regimes in which neighborhood ac¬
tivists sometimes gained a foothold (DiGaetano 1991). Chicago, by contrast,
was a different story. Responding to similar demographic changes, Richard J.
Daley reinvigorated the Chicago machine by reaching out to non-Irish white
ethnics and African Americans (Rakove 1975). A massive infusion of federal
funds during the 1960s and 1970s helped Daley and his successors, Michael
Bilandic and Jane Byrne, keep the machine alive until Harold Washington’s
election in 1983, long after machine governments elsewhere had gone into
decline.
Neighborhood politics in Chicago thus developed by necessity on fiercely
independent political footing. During the 1960s and 1970s, when neighbor¬
hood groups elsewhere began to forge partnerships with progressive city offi¬
cials community organizations in Chicago found access to formal political
channels largely closed off to them by an intransigent machine. Chicago
168 Remaking Chicago

neighborhoods decimated by economic crisis and deindustrialization were


forced to take matters into their own hands. Only through protest, community-
based planning, and extensive coalition-building could they expect to gain the
attention of largely indifferent machine politicians. Chicago’s planned manufac¬
turing districts developed within this rich tradition of community organizing
and community-based planning, fostered by the long-standing presence of a
highly unresponsive political machine.
Does the resilience of the machine make Chicago an exceptional case? In
some respects, perhaps, but the likely success of policies that challenge
growth machines ultimately turns on the basis of community support that
undergirds them, not on whether they originate inside or outside city hall.
Chicago’s planned manufacturing district initiatives were able to withstand
the death of Mayor Washington and the initial skepticism of Richard M. Da¬
ley because of the diverse coalition of supporters that organizations like the
Local Economic and Employment Development (LEED) Council had
painstakingly assembled on behalf of the proposals. By contrast, Boston’s
plan to create light manufacturing districts nearly collapsed in its infancy
when progressive city officials from the Raymond Flynn administration tried
to move the plan forward before a strong, independent political base had
been mobilized around the proposal (King 1988b). The legacy of the ma¬
chine helped produce a strong, community-based challenge to the corporate-
center strategy in Chicago. Whether similar challenges elsewhere enjoy the
necessary level of community support to survive the opposition of property
developers and their allies over the long term remains to be seen.

The Shape of Things to Come


Like many American cities today, Chicago stands at a crossroads. The
corporate-center strategy has produced a dazzling array of impressive new de¬
velopments in the downtown area. With the completion of the $987 million
McCormick Place expansion in January 1997, Chicago became home to the
largest exhibition and meeting facility in North America. A spectacular new
museum and entertainment complex on Navy Pier has proved to be a major
tourist draw. Yet the spillover from new developments such as these has led to
few visible improvements in the city’s low- and moderate-income neighbor¬
hoods. New jobs are being created in the tourist and convention trades, but
the majority of them are low-paying and unskilled. Meanwhile, suburban
commuters and residents of wealthier lakeffont wards hold most of the better¬
paying, knowledge-intensive positions opening up within the downtown ser¬
vice sector. Urban redevelopment in Chicago has become a “tale of two
cities.” Pockets of prosperity and renewal are surrounded by depressed areas
largely untouched by the revitalization occurring in other parts of the city
(Levine 1989, 25).
Conclusion 169

As in other cities, Chicago’s downtown renaissance has been spearheaded


by a progrowth coalition of land-based interests and other business elites. Al¬
though occupying a central position within the governing coalition of cur¬
rent mayor Richard M. Daley, Chicago’s growth machine has nonetheless
struggled in recent years. The emergence of a broad-based, politically sophis¬
ticated neighborhood movement prevents business elites today from unilater¬
ally dictating the terms of redevelopment as they once did. In addition, with
population shifts increasingly tilting the balance of power in the Illinois state
legislature in favor of suburban and rural areas, state lawmakers have grown
reluctant to approve large-scale development projects in Chicago that
promise few benefits for the rest of the state. Since 1985, plans for a 1992
world’s fair in Chicago, a third Chicago airport, a casino and entertainment
complex, and a downtown light rail system have all collapsed due to commu¬
nity opposition, unsupportive state legislators, or some combination of both.
Recent setbacks to the corporate-center strategy, together with pressure
from within the neighborhoods, are creating space for a new dialogue over
Chicago’s economic future. “Balanced growth” has entered the vocabulary
of many Chicagoans. Interest in the preservation of Chicago’s manufacturing
base has been substantially reawakened, and the contribution of Chicago
manufacturers to the city’s overall economy is more widely appreciated now
than perhaps at any time since the 1960s. Even Mayor Daley, initially no
friend of industry, has become a convert, supporting planned manufacturing
districts and other policies to retain industrial firms in the city. Proponents of
a local-producer strategy for neighborhood revitalization continue to make
inroads into the Daley administration’s downtown business-dominated gov¬
erning coalition, carrying with them possibilities for the formation of alterna¬
tive regimes.
Even so, the movement from planned manufacturing districts to the con¬
solidation of a new regime in which neighborhood manufacturers would re¬
place downtown corporations, financial institutions, and real estate develop¬
ers as the key beneficiaries of economic development policy represents a
substantial leap. Attempts to form such a local-producer regime face signifi¬
cant hurdles. To begin with, global economic pressures continue to chip
away at the city’s manufacturing base. Recent tax giveaways necessary to re¬
tain large Chicago manufacturers like Nabisco and Tootsie Roll underscore
the mobility of certain types of producers and the ease with which they are
able to play cities off against one another to win the most generous package
of benefits (see Strahler 1996). Unless firms like these can be more deeply in¬
tegrated into the local economy, through the fostering of local supplier rela¬
tionships, for example, the costs to the city of retaining them may well ex¬
ceed the benefits they provide.
A local-producer regime faces other serious obstacles as well. The activities
of neighborhood-based industrial development organizations have increased
170 Remaking Chicago

the organizational capacity of Chicago manufacturers, creating a structure for


programs like the Model Industrial Corridors Initiative to develop within. Yet
the emerging associational ties among neighborhood producers are still no
match for those of the downtown business elite, whose well-oiled network of
civic cooperation offers opportunities for the execution of highly visible,
complex economic development projects involving substantial economic re¬
sources. The bottom line is that, under a local-producer regime, city officials
would be forced to work harder to bring together the resources necessary to
complete large numbers of relatively low-profile economic development pro¬
jects, most of which would likely receive little attention or publicity outside
of the neighborhoods in which they were located.
Finally, despite effective community organizing and coalition-building
during the past few decades, troublesome divisions continue to separate the
natural popular constituencies of a local-producer regime. Industrial develop¬
ment organizations have done an effective job of organizing local manufac¬
turers, while resident-based community organizations are well-established in
many lower-income and working-class communities. Yet all too often indus¬
trial- and resident-based groups from identical community areas work in iso¬
lation from one another. Furthermore, when coalition-building does take
place, as it did in connection with planned manufacturing district initiatives
on the Near North Side, labor is oftentimes only a weak link in the alliance.
With unionization rates currently hovering at 20 percent of the labor force or
less, many industrial workers lack the necessary forum to develop and articu¬
late collective demands, making outreach efforts involving labor difficult and
time-consuming. Last, but by no means least, racial and ethnic cleavages con¬
tinue to impede coalition-building across working-class neighborhoods. For
many working-class whites, in particular, resentment of downtown business
leaders is well surpassed by racial animosity toward black workers living
nearby. City trenches are perhaps less pronounced and impenetrable than in
years past, but they continue to divide Chicago neighborhoods.
Chicago’s present governing coalition is, in many respects, an unstable
one, balancing a weakened growth machine against politically mobilized
communities frustrated by the failure of downtown redevelopment to uplift
their neighborhoods. While a return to the exclusive business-government
partnership fostered by the late Richard J. Daley is not out of the question,
several alternative scenarios are conceivable as well. Under the first of these, a
less arrogant, more conciliatory growth machine might emerge in reener¬
gized fashion. Members of this coalition might be willing to compromise
their plans in order to win financial backing from downstate legislators and,
at the very least, pay lip service to the redistributive demands of neighbor¬
hood organizations. More likely than not, the gulf between downtown and
the neighborhoods would continue to widen under such a scenario, the
trickle of benefits from downtown growth creating few new opportunities for
Conclusion 171

the redevelopment of low- and moderate-income neighborhoods. By neces¬


sity, economic development policy would remain highly centralized. Business
elites would continue to dominate important policy and planning decisions,
although input from neighborhood groups might be solicited on a somewhat
more regular basis.
There is another possibility. Chicago producers, workers, and community
residents might begin to build more directly on the economies of place
achievable through the wise and efficient use of local resources. Drawing on
the legitimate manufacturing expertise that exists in the city’s working-class
and industrial neighborhoods, Chicago communities could better shield
themselves from the effects of global economic restructuring by developing
flexible, innovative networks of production. The alliance between neighbor¬
hood producers, workers, and community residents could, in turn, serve as
the constituency for a new regime supported by an alternative set of eco¬
nomic resources to those held by the downtown growth machine.
The seeds of such a regime are clearly visible in the activities of industrial
development organizations and in programs and policies such as planned
manufacturing districts and the recent Model Industrial Corridors Initiative.
Weaving experiments such as these into more broad-based political and eco¬
nomic alliances will require both strong political leadership and a greater
awareness among neighborhood producers, workers, and community resi¬
dents of the powers they can achieve when relations of mutual trust and re¬
spect supplant those of indifference and hostility. It is a path strewn with
countless obstacles, unlikely to reward its supporters with highly visible signs
of progress, but one that nonetheless offers considerable promise for the revi¬
talization of Chicago neighborhoods left in the shadows of downtown rede¬
velopment.
\
"N
Notes
1: The Politics of Urban Economic Development
1. Economies of scope occur when one firm is able to produce a variety of
products more efficiendy than separate firms could. Economies of scale refer to pro¬
ductivity gains, or decreases in average production costs, that result from increasing
the scale of production.
2. For a parallel argument, see Swanstrom (1993).

2: Downtown Redevelopment and the Chicago Political Machine


1. For purposes of this study, the central area is bounded by Lake Michigan to
the east, North Avenue to the north, Ashland Avenue to the west, and Cermak Road
to the south.
2. Business establishments represented in the CCAC included Sears, Roebuck
and Co.; First Federal Savings and Loan; Carson Pirie Scott; Commonwealth Edison
Co.; Peoples Gas Co.; First National Bank of Chicago; Hart Schaffner & Marx; Stan¬
dard Oil Co.; Skidmore, Owings & Merrill; Hilton Hotels Corp.; Illinois Central In¬
dustries; Chicago Title and Trust Co.; Harris Trust and Savings Bank; CNA Financial
Corp.; Continental Illinois National Bank; Inland Steel Co.; Scribner & Co.; Marshall
Field & Co.; Northern Trust Co.; and Real Estate Research Corporation.
3. /After his election, Daley announced his intention to resign as chair of the
Cook County Democratic Central Committee, but it was a promise he failed to keep.
He eventually claimed the party refused to accept his resignation (Biles 1995, 55).
4. The CCAC is listed first among private sector groups consulted by the city in
preparation of the 1958 Development Plan.
5. The downtown emphasis of the RERC plan is not surprising, considering
that RERC attorney James C. Downs, Jr. was president of the CCAC during the time
the plan was prepared.
6. The key actors in the Chicago 21 Corporation were Thomas Ayers (chair¬
man and president, Commonwealth Edison), John H. Perkins (president, Continental
Illinois National Bank), Raymond C. Wieboldt, Jr. (president. Dearborn Park Corpo¬
ration), Ferd Kramer (Draper and Kramer), Donald Erikson (semor partner, Arthur
Andersen and Co.), Warren G. Skoning (vice-president. Sears, Roebuck and Co.), and
Philip Klutznick (chairman, Urban Investment and Development Company).
7. In a study of the City of Chicago’s capital budgets issued during the 1970s
and 1980s, Greer (1986) found a strong correlation between the spatial distribution
of public and private investment dollars.
8. Of 744 total acres slated for clearance by the Commission, 646 tell within a
two-mile radius of the Loop (Department of City Planning 1959, 120-121).
9 The 1977-1981 Capital Improvements Program was published during the
Michael Bilandic administration but largely prepared in the last years ot the Daley ad¬
ministration (Greer 1986, 34). x, , .
10. For purposes of this study, the northwest corridor extends trom North Av¬
enue to Irving Park Road along the Kennedy Expressway (1.5 miles on either side)
for a total of nine square miles. The southwest corridor extends westward from
174 Notes to Pages 36-51

Halsted Street to the city limits along the Stevenson Expressway (1.5 miles on either
side) for a total of fifteen square miles. The central area is roughly eleven square miles,
including 1.5 square miles occupied by the Loop.
11. For example, the reconstruction of bridges over the Chicago River to im¬
prove transportation between the Loop and the near-downtown area became a key
public investment priority during this period.
12. HUD’s definition of economic distress made the entire city of Chicago eli¬
gible for UDAG funds.
13. Firms receiving IRBs averaged 172 employees in size (Clemons, Giloth,
and Tostado 1985, 2).
14. See “Council Votes Far Reaching Zoning Laws,” Chicago Tribune, 30 May
1957, sec. 1, p. 8.
15. The 1942 Comprehensive Amendment allowed nonconforming industrial
concerns to continue operating for the duration of the useful lives of the buildings in
which they were housed, generally seventy-five to one hundred years (see Municipal
Code of Chicago, Chapter 194A).
16. Amortization schedules for elimination of nonconforming uses were based
on the assessed value and type of construction of the building housing the offending
use. Buildings assessed at $5,000 or less were given two to four years to be brought
into conformance. More valuable structures had anywhere from ten to forty years to
comply, depending on whether they were made of brick, concrete, or wood.
17. Noise generated by manufacturing facilities on the boundaries of residen¬
tial or business districts was not to exceed 79 decibels. Any use creating intense
earth-shaking vibrations had to be set back at least 250 feet from the boundaries of
residential, business, or commercial districts (Committee on Buildings and Zoning
1955b, 98a).
18. Because of rapid technological advances in building construction, periodic
revisions are essential if a municipal building code is to incorporate modern construc¬
tion standards on a consistent basis.
19. The composition of the committee reflected Daley’s belief that public dis¬
putes were most efficiently resolved by limiting debates to the most powerful interests
concerned. When asked why no community organizations were represented on the
advisory committee, a Daley administration staff member responded, “We don’t feel
that is conducive to a rational analysis of the situation” (Jones 1985, 30).
20. The Rehabilitation Code became Chapter 78.1 of the Municipal Code of
Chicago.
21. See Gary Washburn, “'Restrictive’ Codes Thwart a Loft Revival,” Chicago
Tribune, 18 January 1981, sec. N14, p. 1.

3: Rethinking Industrial Decline


1. The discussion in this section draws heavily on Ranney and Wiewel (1987,
1988) and Wiewel, Ranney, and Putnam (1990).
2. See Wiewel, Ranney, and Putnam (1990, 373) and Ranney and Wiewel
(1987, 65) for a helpful technical explanation of both letterpress and offset printing.
3. The printing production process encompasses three primary phases: (1) pre¬
press, which includes layout, typesetting, and other activities necessary to prepare an im-
Notes to Pages 51-117 175

age for printing; (2) press, where the ink is actually applied to paper; and (3) binding.
4. See Chapter 2. Not so incidentally, 1982 was also the year in which loft con¬
version activity in Printing House Row reached its peak (Pruska-Carroll 1987, 37).
5. The other forty-seven firms presumably either failed to survive displacement
from their original Printing House Row locations or went out of business for other
reasons.
6. The Chicago Apparel Center is a large exhibition center and buying facility
for apparel manufacturers and designers, one of four regional marts in the country. It
contains 850 showrooms, includes a hotel, and employs approximately 3,000 people.

4: Community Economic Development and the Crisis of Machine Politics


1. See, for example, “Downtown Growth Helps Taxpayers,” Chicago Tribune,
30 October 1986, sec. 1, p. 22.
2. See Finks (1984) and Horowitt (1989) for comprehensive discussions of
Alinsky’s organizing career.
3. The Alinsky model of neighborhood organizing is detailed in Alinsky’s two
organizing manifestos. See Alinsky (1969, 1971).
4. “Block-busting” was a practice through which real estate agents reaped
windfall profits by inducing panic selling among white homeowners in neighborhoods
experiencing racial transition. Homes were typically resold to African-American fami¬
nes at inflated prices.
5. LEED Council’s organizing efforts around the issue of industrial displace¬
ment are described at length in Chapter 5.
6. See Shakman v. the Democratic Organization of Cook County et al., no. 69 C
2145 (U.S. District Court for the Northern District of Illinois, Eastern Division).
7. See also “Her Honor Flirts with Business,” Business Week, 10 December
1979,pp.61-66.
8. In 1990, the federal government was forced to assume payments on the
$158 million mortgage when the project’s developers defaulted on the loan. See J.
Linn Allen, “Blazing a Trail West,” Chicago Tribune, 7 July 1996, sec. S, p. 1.
9. See, for example, “Chicago’s Budget and Its Bond Rating,” Crain’s Chicago
Business, 20 October 1986, p. 12.
10. See DeLeon (1992) for a similar argument based on the Art Agnos admin¬
istration in San Francisco.
11. A final version of the CWED Policy Statement was never published, al¬
though several additional drafts of the document were issued in late 1982 and early

1983. . . , ,
12. CWED participants on the economic development issues team included
Robert Mier, John Kretzmann, and Donna Ducharme.

5: The Battle for the Near North Side


1. See Chapter 2.
2. According to one account, Oberman’s refusal to engage in backroom deals
made him the butt of jokes among machine aldermen (Rivlin 1992, 118).
3. See, for example, William E. Schmidt, “Chicago Plan Aims to Curb Factory
176 Notes to Pages 117-54

Loss,” New York Times, 10 December 1987, sec. 1, p. 1; and Merrill Goozner, “City
Out to Forge Compromise in Industrial-Residential Clash,” Crain’s Chicago Business,
3 November 1986, p. 1.
4. See “Pro-PMD? That was then, this is now,” Crain’s Chicago Business, 22
August 1988, p. 10.
5. See, for example, Merrill Goozner, “Clybourn Zoning Fight Breaks All the
Rules,” Chicago Tribune, 27 July 1988, sec. 1, p. 1.
6. For example, CANDO developed the Securing Older Buildings Project to
identify ways to continue using older industrial buildings for manufacturing purposes.
7. Plan Commission approval is required for zoning changes on land parcels
two acres or larger in size and for all parcels along the Chicago River. From 1983 to
1987, Washington’s Plan Commission supported 139 requests for rezonings of indus¬
trial property and opposed only one (Mier 1987b).
8. See “Chicago’s Sham Industrial Policy,” Chicago Tribune, 20 July 1987, sec.
l,p.8.
9. In early 1988, the name of the ordinance was changed from Protected Man¬
ufacturing District to Planned Manufacturing District.
10. See Municipal Code of Chicago, Chapter 194D.
11. In fact, the studies of the Clybourn Corridor by the Department of Eco¬
nomic Development, the Mayor’s Office of Employment and Training, and Creticos
and Masotti cited earlier in the chapter suggest that, if anything, retaining the corri¬
dor as an industrial district was the more lucrative development strategy of the two.

6: Toward a Citywide Industrial Policy


1. See “Let Goose Island Have Its Golden Eggs,” Chicago Tribune, 8 June
1989, sec. 1, p. 24.
2. See Ray Hanania, “Daley Unveils Plan To Protect Two Areas,” Chicago Sun-
Times, 14 June 1990, p. 63.
3. See “Don’t Stop the Clock on Goose Island,” Chicago Tribune, 18 June
1990, sec. 1, p. 10.
4. See “Daley Sides Frankly with Sausage Plant,” Chicago Tribune, 2 March
1996, sec. 1, p. 5.
5. Comparable data for industrial corridors located on the city’s South and
West Sides are not currently available.
6. The Clybourn Corridor PMD is the only one of tire city’s three PMDs di¬
vided into core and buffer zones.
7. The North River, Addison, Ravenswood, and Pilsen Industrial Corridors are
the only industrial corridors for which data are presendy available.
Bibliography
Abbott, Carl. 1987. The new urban America: Growth and politics in Sunbelt cities.
Chapel Hill: University of North Carolina Press.
Abravanel, Martin D., and Paul K. Mancini. 1980. Attitudinal and demographic con¬
straints. In Urban revitalization, ed. Donald B. Rosenthal. Beverly Hills. Sage.
Ad Hoc Working Committee for Residential Rehabilitation. 1978. Progress report of
findings to date regarding residential rehabilitation and the Chicago Building
Code. Chicago: Ad Hoc Working Committee for Residential Rehabilitation.
Alinsky, Saul D. 1969. Reveille for radicals. New York: Vintage Books.
__ 1971. Rules for radicals: A practical primer for realistic radicals. New York:
Vintage Books.
Angwin, Julia. 1996. Price wars unraveling local clothing makers. San Francisco
Chronicle, 20 March, 1(C).
Auerbach, Jesse. 1980a. Cooperation stabilizes community. The Neighborhood Works,
12 September, 1.
__ 1980b. Development without displacement. The Neighborhood Works, 8 Feb¬
ruary, 1.
Babcock Richard F. 1972. The new Chicago Zoning Ordinance. In Urban land use
policy The central city, ed. Richard B. Andrews. New York: The Free Press.
Babcock,'Richard F., and Wendy U. Larsen. 1990. Industrial-strength zoning. In Spe¬
cial districts: The ultimate in neighborhood zoning, ed. Richard Babcock. Cam¬
bridge: Lincoln Institute of Land Policy.
Bachrach, Peter, and Morton S. Baratz. 1962. Two faces of power. American Politi¬
cal Science Review 56: 947-952.
Banfield, Edward C. 1961. Political influence. Glencoe, IL: Free Press.
Barry, Patrick. 1986. Downtown growth threatens industry. The Neighborhood Works
9 (March): 1, 14-15. ,
Bartelt David W. 1989. Renewing center city Philadelphia: Whose city? Which pub¬
lic’s interests11 In Unequal partnerships: The political economy of urban redevelop¬
ment in postwar America, ed. Gregory D. Squires. New Brunswick, NJ: Rutgers

University Press. . . .
Barton-Aschman Associates. 1970. Commercial land-use policies and locattonal princi¬
ples. Chittgo:Jiaiton-Asch.rnan. Associates.
Beauregard, Robert A. 1984. Structure, agency, and urban redevelopment. In Cities
in transformation, ed. Michael Peter Smith. Beverly Hills: Sage.
_ ed. 1989. Economic restructuring and political response. London: Sage.
Bennett Larry 1989. Postwar redevelopment in Chicago: The declining politics ot
party and the rise of neighborhood politics. In Unequal partnerships: The politi¬
cal economy of urban redevelopment in postwar America, ed. Gregory D. Squires.
New Brunswick, NJ: Rutgers University Press.
Bennett, Larry, Kathleen McCourt, Philip W. Nyden, and Gregory D. Squires. 1988.
Chicago’s North Loop Redevelopment Project: A growth machine on ho d n
Business elites and urban development, ed. Scott Cummings. Albany: State Uni¬
versity of New York Press.
Bennett. Larry, Gregory D. Squires, Kathleen McCourt, and Philip Nyden. 1987.
178 Bibliography

Challenging Chicago’s growth machine: A preliminary report on the Washington


administration. International Journal of Urban and Regional Research 11:
351-362.
Berk, Gerald, and Todd Swanstrom. 1994. Expanding the agenda of regime theory:
Republican cities in the age of industry. Paper delivered at the 1994 annual con¬
vention of the American Political Science Association (31 August—4 September),
New York, NY.
-. 1995. The power of place: Capital (im)mobility, pluralism, and regime theory.
Paper delivered at the 1995 annual convention of the Urban Affairs Association,
(3-6 May), Portland, OR.
Berry, Brian, Irving Cutler, Edwin Draine, Y. Kiang, Thomas Tocalis, and Pierre de
Vise. 1976. Chicago: Transformations of an urban system. Cambridge: Ballinger.
Best, Michael H. 1989. Sector strategies and industrial policy: The furniture industry
and the Greater London Enterprise Board. In Reversing industrial decline? ed.
Paul Hirst and Jonathan Zeitlin. Oxford: Berg.
Betancur, John J., Deborah E. Bennett, and Patricia A. Wright. 1991. Effective strate¬
gies for community economic development. In Challenging uneven development:
An urban agenda for the 1990s, ed. Philip W. Nyden and Wim Wiewel. New
Brunswick: Rutgers University Press.
Biles, Roger. 1995. Richard J. Daley: Politics, race, and the governing of Chicago.
DeKalb: Northern Illinois University Press.
Blair, John P, Rudy H. Fichtenbaum, and James A. Swaney. 1984. The market for
jobs: Locational decisions and the competition for economic development. Ur¬
ban Affairs Quarterly 20: 64-77.
Bluestone, Barry, and Bennett Harrison. 1982. The deindustrialization of America.
New York: Basic Books.
Body-Gendrot, Sophie N. 1987. Grass-roots mobilization in the Thirteenth Ar-
rondissement of Paris: A cross-national view. In The politics of urban development,
ed. Clarence Stone and Heywood Sanders. Lawrence: University Press of Kansas.
Boyte, Harry C. 1980. The backyard revolution. Philadelphia: Temple University Press.
Brusco, Sebastiano. 1982. The Emilian model: Productive decentralization and social
integration. Cambridge Journal of Economics 6: 167-184.
Buccitelli, Michael. 1995. Interview by author, 19 January, Chicago.
Cafferty, Pastora San Juan, and William C. McCready. 1982. The Chicago public-pri¬
vate partnership experience: A heritage of involvement. In Public-private part¬
nerships in American cities, ed. R. Scott Fosler and Renee A. Berger. Lexington,
MA: Lexington Books.
Calavita, Nico. 1992. Growth machines and ballot box planning: The San Diego case.
Journal of Urban Affairs 14: 1-24.
Capraro, James, Andrew Ditton, and Robert Giloth. 1985. Neighborhood economic de¬
velopment: Working together for Chicago’s future. Chicago: Department of Eco¬
nomic Development.
Carroll, Margaret. 1980. Saving old buildings: Stumbling blocks go with the territory.
Chicago Tribune, 15 March, 13(N1).
Carsel, Wilfred. 1940. A history of the Chicago Ladies’ Garment Workers’ Union.
Chicago: Normandie House. \
Castells, Manuel. 1985. High technology, economic restructuring, and the urban-
179
Bibliography

regional process in the United States. In High technology, space and society, ed.
Manuel Castells. London: Sage.
Center for Urban Economic Development. 1983. Strategies for target area industrial
development. Chicago: Center for Urban Economic Development, University ol
Illinois at Chicago. ..
_ 1985. Industrial displacement in major American cities and related policy op¬
tions. Chicago: Center for Urban Economic Development, University of Illinois

at Chicago. „
--.170
1987.
/ . J-Printing
llJLILlllg, employment
w-- in Chicago: Trends and_ prospects. . Chicago. Ce
ter for Urban Economic Development, University of Illinois at Chicago.
_. 1988. Printing in Chicago: The final report of the printing industry task force.
Chicago- Center for Urban Economic Development, University of Illinois at Chicago.
Center for Urban Studies. 1966. Mid-Chicago economic development study. Vol. 2.
Chicago: Mayor’s Committee for Economic and Cultural Development.
Chicago Association of Neighborhood Development Organizations. 1994. CANDU
1994 annual report. Chicago: CANDO.
Chicago Central Area Committee. 1973. Chicago 21: A plan for the central are. com-
munities. Chicago: Chicago Central Area Committee.
Chicago Plan Commission. 1952. Chicago industrial study summary report. Chicago.
Chicago Plan Commission.
Ciccone, F. Richard. 1979. Byrne beat Bilandic: Black vote, CTA woes, snow cite .
Chicago Tribune, 1 March, 1(1).
Cicin-Sain Biliana. 1980. The costs and benefits of neighborhood revitalization.
Urban revitalization, ed. Donald B. Rosenthal. Beverly ffills: Sage.
City of Chicago. 1981. Five year capital improvements program, 1981 198b. unicag .

_^-ltyi984^Chicago works together”: 1984 Chicago development plan. Chicago:

Cltyi98^Report of the Mayor’s Task Force on Neighborhood Land Use: Industrial


and commercial land use. Chicago: City of Chicago. .
City of New York. 1981. Lofts: Balancing the equities. New York: City Planning Commi .
Clavel Pierre. 1986. The progressive city. New Brunswick, NJ: Rutgers nivers'ty r«s-
Clavel’ Pierre and Wim Wiewel, eds. 1991. Harold Washington and the neighborhoods.

^^^"^rdo Tcudo 1985. Monitoring


Chicago’s industrial revenue bond performance: 1977-1984. Chicago. Departmen

of Economic Development. „ . . D dis_


Cohen, Robert B. 1979. The internationalization of capital and U.S. cities. 1 . .
sertation New School for Social Research.
Commercial Club of Chicago. 1984. Make no little plans: Jobs for metropolitan
Chicago. Chicago: Commercial Club of Chicago. Printing
Commission on Chicago Historical and Architectural Landmarks. 1983. Printing
House Row district: Preliminary summary of information. Chicago: Commission
on Chicago Historical and Architectural Landmarks.
Committee on Buddings and Zoning. 1955a. A report on the proposed
amendment to the Chicago Zoning Ordinance. Chicago: City Councd Committee

on Buildings and Zoning.


180 Bibliography

Committee on Buildings and Zoning. 1955b. Proposed comprehensive amendment to


the Chicago Zoning Ordinance. Chicago: City Council Committee on Buildings
and Zoning.
Community Workshop on Economic Development. 1982. Draft platform of the
Community Workshop on Economic Development. Chicago: CWED.
-. 1987. Digging in: Community-grown business ventures. Chicago: CWED.
Corplan Associates. 1964. Technological change: Its impact on industry in metropolitan
Chicago, printing and publishing industries. Chicago: Illinois Institute of Tech¬
nology Research Institute.
Costigan, Joseph. 1996. Interview by author, 20 February, Chicago.
Crenson, Matthew A. 1971. The un-politics of air pollution: A study of non-decision¬
making in the cities. Baltimore: Johns Hopkins University Press.
Creticos, Peter A., and Louis H. Masotti. 1988. The economic effect of Chicago in¬
dustrial businesses: An examination of Clybourn Avenue manufacturers. Chicago:
Creticos and Associates.
Cummings, Scott, ed. 1988. Business elites and urban development: Case studies and
critical perspectives. Albany: State University of New York Press.
Dahl, Robert A. 1961. Who governs? New Haven: Yale University Press.
Daley, Richard M. 1989. City government and Chicago’s economic future. Speech to
members of the Chicago Kiwanis Club, 14 February.
Davis, Robert. 1992. Daley plans “bond sale to fix infrastructure. Chicago Tribune, 22
May, 3(2C).
DeLeon, Richard E. 1992. The urban antiregime: Progressive politics in San Fran¬
cisco. Urban Affairs Quarterly 27: 555-579.
De Meirleir, Marcel J. 1950. Manufactural occupance in the west central area of
Chicago. Ph.D. dissertation, University of Chicago.
Department of City Planning. 1958. Development plan for the central area of Chicago.
Chicago: Department of City Planning.
-. 1959. Five year capital improvements program 1959-1963. Chicago: Depart¬
ment of City Planning.
-. 1960. Locational patterns of major manufacturing industries in the city of
Chicago. Chicago: Department of City Planning.
-. 1961. Industrial movements and expansion, 1947-1957. Chicago: Department
of City Planning.
. 1964. Basic policies for the comprehensive plan of Chicago. Chicago: Depart¬
ment of City Planning.
Department of Development and Planning. 1967. The comprehensive plan of Chicago,
conditions and trends: Population, economy, land. Chicago: Department of Devel¬
opment and Planning.
Department of Economic Development. 1987. Cooperation for survival and growth:
New designs for apparel manufacturing in Chicago. Chicago: Department of Eco¬
nomic Development.
-. 1988. Clybourn Corridor Planned Manufacturing District. Chicago: Depart¬
ment of Economic Development.
Department of Planning. 1978. 19/7—1981 capital improvements program. Chicago:
Department of Planning. \
-. 1982. Chicago 1992 comprehensive plan. Chicago: Department of Planning.
181
Bibliography

_. 1988. Report of proceedings of a public hearing re: Planned Manufacturing


District (PMD) application, Clybourn Corridor, 43rd Ward. Department of Plan¬
ning, 27 July. ,
__ 1989. Corridors of industrial opportunity: A plan for industry on Chicago
North Side (draft). Chicago: Department of Planning.
_. 1990a. Staff report to the Chicago Plan Commission: Goose Island Planned
Manufacturing District. Chicago: Department of Planning.
_ 1990b. Staff report to the Chicago Plan Commission: Elston Corridor
Planned Manufacturing District. Chicago: Department of Planning.
Department of Planning and Development. 1991. Corridors of industrial opportune
A flan for industry m Chicago’s West Sid,. Chieago: Department of Planning and
Development. . . y
_. 1992. Corridors of industrial opportunity: A plan for industry in Chicago s
North Side. Chicago: Department of Planning and Development.
_. 1994. Corridors of industrial opportunity: A plan for industry m Chicago s
South Side. Chicago: Department of Planning and Development.
_ 1996 Industrial corridor capital investment 1995: A guide to industrial im
provement projects. Chicago: Department of Planning and Development.
__ 1998. Proposed industrial redevelopment project areas: Tax increment financing
program. Chicago: Department of Planning and Development.
DiGaetano, Alan. 1989. Urban political regime formation: A study in contrast. Jou
nal of Urban Affairs 11: 261-281. , CTT ,
_., 1991. Urban political reform: Did it kill the machine? Journal of Urban His-

Dold^Bruct^and Thomas Hardy. 1989. Chicago elects second Mayor Daley.


Chicago Tribune, 5 April, 1(1). _ , , c ,
Domhoff G William. 1983. Who rules America now? New York: Simon and Schuste .
Dreier, Peter, and Bruce Ehrlich. 1991. Downtown development and urban reform.
The politics of Boston’s linkage policy. Urban Affairs Quarterly 26. 3 •
Driscoll Paul A 1996. Wieners or losers: Chicago sausage maker says either it or big
residential proposal must go. Chicago Tribune, 3 March, 3(16).
DucCme DoL 1991. Planned Manufacturing Districts: How a community initia¬
tive became city policy. In Harold Washington and the neighborhoods, ed. Pierre
Clavel and Wim Wiewel. New Brunswick: Rutgers University Press.
_ 1995 Interview by author, 26 January, Chicago.
Ducharme, Donna, Robert Giloth, and Lynn McCormick. 1986 .Business loss or ba
ancedgrowth: Industrial displacement in Chicago. Chicago: Department o

EMnTep^rXTwendeth century urban regimes. Journal of Urban Affairs 7:

_f 1987a. City and regime in the American republic. Chicago: University of

ChT9S87brStSate and market in city politics: Or. the “teal” Dallas. In The politic,
-fifuZ delToplnt, ed. Clarence Stone and He^ood Sanders. Lawrence:

Umi994yB>usTnessf Stt relations in the commercial republic. Journal of Political

Philosophy 2:115-139.
182 Bibliography

Emmons, David. 1977. Dearborn Park/South Loop New Town: A project in the
Chicago 21 plan. Chicago: Citizens Information Service of Illinois.
Enstad, Robert. 1980. Even without shops, Dearborn Park project is drawing “pio¬
neers.” Chicago Tribune, 14 August, 1(N6).
Erie, Steven P. 1988. Rainbow’s end. Berkeley: University of California Press.
Fainstein, Susan S. 1990. Economics, politics, and development policy: The conver¬
gence of New York and London. In Beyond the city limits, ed. John R. Logan and
Todd Swanstrom. Philadelphia: Temple University Press.
Fainstein, Susan S., and Norman I. Fainstein. 1983. Regime strategies, communal re¬
sistance, and economic forces. In Restructuring the city, ed. Susan Fainstein, Nor¬
man Fainstein, Richard Hill, Dennis Judd, and Michael Smith. New York: Long¬
man.
-. 1989. Technology, the new international division of labor, and location: Con¬
tinuities and disjunctures. In Economic restructuring and political response, ed.
Robert A. Beauregard. London: Sage.
Fainstein, Susan S., Norman I. Fainstein, and P. Jefferson Armistead. 1983. San Fran¬
cisco: Urban transformation and the local state. In Restructuring the city, ed. Su¬
san Fainstein, Norman Fainstein, Richard Hill, Dennis Judd, and Michael Smith.
New York: Longman.
Fainstein, Susan S., Ian Gordon, and Michael Harloe, eds. 1992. Divided cities: New
York and London in the contemporary world. Cambridge: Blackwell.
Feagin, Joe R. 1983. The urban real estate game. Englewood Cliffs, NJ: Spectrum.
Ferman, Barbara. 1991. Chicago: Power, race, and reform. In Big city politics in tran¬
sition, ed. H. V. Savitch and John Clayton Thomas. London: Sage.
Ferman, Barbara, and William J. Grimshaw. 1992. The politics of housing policy. In
Politics of policy innovation in metropolitan Chicago, ed. Kenneth Wong and ferry
Nicholas Clark. Greenwich, CT: JAI Press.
Fernandez Kelly, Patricia. 1989. International development and industrial restructur¬
ing: The case of garment and electronics industries in Southern California. In In¬
stability and change in the world economy, ed. Arthur MacEwan and William K.
Tabb. New York: Monthly Review Press.
Fingeret, Lisa. 1998. Chicago approves 1,500-acre TIF chunk. Real Estate Journal,
16 February, 1.
Finks, P. David. 1984. The radical vision of Saul Almsky. New York: Paulist Press
Fish, John Hall. 1973. Black power/white control: The struggle of the Woodlawn Orga¬
nization in Chicago. Princeton: Princeton University Press.
Fitzgerald, Michael. 1988. Developer opposes Goose Island PMD. Inside Lincoln
Park, 3 August.
Folz, David H., Linda Gaddis, William Lyons, and John M. Scheb. 1993. Saturn
comes to Tennessee: Citizen perceptions of project impacts. Social Science Quar¬
terly 74: 793-803.
Fremon, David K. 1988. Chicago politics ward by ward. Bloomington: Indiana Uni¬
versity Press.
Frieden, Bernard J., and Lynne B. Sagalyn. 1989. Downtown, Inc.: How America re¬
builds cities. Cambridge: MIT Press.
Froebel, Folker, Jurgen Heinrichs, and Otto Kreye. 1977. The tendency towards a
new international division of labor. Review 1: 73-88.
183
Bibliography

Fuller, Dorothy. 1996. Interview by author, 2 February, Chicago.


GaDD Paul 1973 Power elite drafts city master plan. Chicago Tribune, 21 May, 1( )■
Gardiner, John A„ and Theodore R. Lyman. 1978. Decisions for sale: Corruption and
reform in land-use and building regulation. New York: Praeger.
Gaventa, John. 1980. Power and powerlessness: Quiescence and rebellion in an p-
palachian valley. Urbana: University of Illinois Press. . ,
Gibson, Ray. 1989a. Daley campaign reaps $7.1 million, sets ftind-raising record.
Chicago Tribune, 4 July, 2(2C). , ., , im
_ 1989b. Evans borrows as contributions lag. Chicago Tribune, 21 Marc , ( )■
_ 1990. Daley backers reap lucrative rewards via city contracts. Chicago Tn-

Gills&Doug9.199L Chicago politics and community development: A social movement


perspective. In Harold Washington and the neighborhoods, ed. Pierre Clavel and
Wim Wiewel. New Brunswick: Rutgers University Press. , ,
Giloth Robert, and John Betancur. 1988. Where downtown meets neighborhood:
Industrial displacement in Chicago, 1978-1987. Journal of the American Plan-

Gilo Menashe. 1981. Rehab the brewery for more jobs? The

GilotE^Rob^rt^^d^Rob^t^^en 1989. Spatial change and social justice; ^™ative


economic development in Chicago. In Economic restructuring and political re¬
sponse, ed. Robert A. Beauregard. London: Sage.
Glickman Norman J. 1987. Cities and the international division of labor In The cap
italist city: Global restructuring and community politics, ed. Mic ae mit an

Goff!^isa^l987^ Qybomri1 rmd^bd^lder to woo city with linkage,

_1988 2FinkPrsmrPWhy industry needs protection. Crain’s Chicago Business,

8 August, 1.
Gonwa Margaret. 1996. Interview by author, 17 May, Chicago.
Goodman, Robert. 1979. The last entrepreneurs: America’s regional war for jo s an
dollars New York: Simon and Schuster. . . , ,
Goozner, Merrill. 1986. City out to forge compromise in industrial-residential c as .
Crain’s Chicago Business, 3 November, 1. . T i
_. 1988a. Clybourn zoning fight breaks all the rules. Chicago Tribune, 27 Juy,

_ii1 1988b. Industry may give up fight for Goose Island. Chicago Tribum, 20

Collision course: City planners, largest landholder differ on what to do


with Goose Island property. Chicago Tribune, 26 September, 1(8). .
Goozner, Merrfll^and StanTey Ziemba. 1989. Big gifts to Daley reflect city's Mansion,

Chicago 2//'b;S;t32elopment and the h,story of Amencan cities.


G°1ed. William K. Tabb and Larry Sawers. New York:

Gottdiener! Mark. 1987. The decline of urban politics: Pohtual theory uni the crisis of
the local state. London: Sage.
184 Bibliography

Govas, Ken. 1995. Interview by author, 18 January, Chicago.


Granovetter, Mark. 1985. Economic action and social structure: The problem of em¬
beddedness. American Journal of Sociology 91: 481-510.
Greater Southwest Development Corporation. 1995. Strategic plan for the Greater
Southwest Model Industrial Corridor. Chicago: Greater Southwest Development
Corporation.
Green, Brian M., and Yda Schreuder. 1991. Growth, zoning, and neighborhood or¬
ganizations: Land use conflict in Wilmington, Delaware. Journal of Urban Af¬
fairs 13: 97-110.
Greer, James L. 1983. The politics of declining growth: Planning, economic transfor¬
mation, and the structuring of urban futures in American cities. Ph.D. disserta¬
tion, University of Chicago.
. 1986. Capital investment in Chicago: Fragmented processes, unequal outcomes.
Chicago: Center of Urban Research and Policy Studies, University of Chicago.
Grimshaw, William J. 1992. Bitter fruit: Black politics and the Chicago machine,
1931-1991. Chicago: University of Chicago Press.
Gruber, William. 1979. Business, finance execs see little vote impact. Chicago Tri¬
bune, 1 March, 6(6).
Gustafson, W. Eric. 1959. Printing and publishing. In Made in New Fork: Case studies
in metropolitan manufacturing, ed. Max Hall. Cambridge: Harvard University
Press.
Hall, Max. 1959. Made in New Fork: Case studies in metropolitan manufacturing.
Cambridge: Harvard University Press.
Hardy, Thomas, and Jorge Casuso. 1989. Evans assails Daley’s contributions. Chicago
Tribune, 17 March, 3(2C).
Harloe, Michael, Peter Marcuse, and Neil Smith. 1992. Housing for people, housing
for profits. In Divided cities: New Fork and London in the contemporary world, ed.
Susan S. Fainstein, Ian Gordon, and Michael Harloe. Cambridge: Blackwell.
Harrison, Bennett. 1992. Industrial districts: Old wine in new bottles? Regional Stud¬
ies 26: 469-483.
Hartman, Chester W. 1974. Ferba Buena: Land grab and community resistance in San
Francisco. San Francisco: Glide.
Harvey, David. 1978. The urban process under capitalism: A framework for analysis.
International Journal of Urban and Regional Research 2: 101-131.
-. 1986. The urbanization of capital. Baltimore: Johns Hopkins University
Press.
Helfgott, Roy B. 1959. Women’s and children’s apparel. In Made in New Fork: Case
studies in metropolitan manufacturing, ed. Max Hall. Cambridge: Harvard Uni¬
versity Press.
Hill, Richard C. 1978. Fiscal collapse and political struggle in decaying central cities
in the United States. In Marxism and the metropolis, ed. William K. Tabb and
Larry Sawers. New York: Oxford University Press.
. 1283. Crisis in the motor city: The politics of economic development in De¬
troit. In Restructuring the city, ed. Susan Fainstein, Norman Fainstein, Richard
Hill, Dennis Judd, and Michael Smith. New York: Longman.
Hirst, Paul, and Jonathan Zeitlin, eds. 1989. Reversing industrial decline? Industrial
structure and policy in Britain and her competitors. Oxford: Berg.
185
Bibliography

Hoffman, Kurt. 1985. Clothing, chips and competitive advantage: The impact of mi¬
croelectronics on trade and production in the garment industry. World Develop¬
ment 13: 371-392.
Hollander, Elizabeth. 1987. Memorandum to George Munoz, 14 May.
-. 1988. Memorandum to Eugene Sawyer, 7 April.
Holli, Melvin G. 1995. Jane M. Byrne: To think the unthinkable and do the un-
doable. In The mayors: The Chicago political tradition, ed. Paul M. Green and
Melvin G. Holli. Carbondale: Southern Illinois University Press.
Hoover, Edgar, and Raymond Vernon. 1959. Anatomy of a metropolis. Cambridge.
Harvard University Press.
Hornung, Mark. 1988. CMC unit trying to duck Goose Island deal? Crain’s Chicago
Business, 8 August, 2.
_ 1990. Daley inner circle: Pols, kin, fat cats. Cvain’s Chicago Business, 23 April, 4.
_. 1991. He wants government to work: Why pragmatic Mosena was Daley’s
pick to run city staff. Crain’s Chicago Business, 10 June, 3.
Horowitt, Sanford. 1989. Let them call me rebel: Saul Alinsky, his life and legacy. New
York: Knopf.
Illinois Department of Employment Security. 1972. Where workers work tn the
Chicago standard metropolitan statistical area. Springfield: State of Illinois.
_ 1979. Where workers work in the Chicago standard metropolitan statistical
area. Springfield: State of Illinois. .
_ 1983. Where workers work in the Chicago standard metropolitan statistical
area. Springfield: State of Illinois. . .
_ 1984. Where workers work in the Chicago standard metropolitan statistical
area. Springfield: State of Illinois. . .
_ 1985. Where workers work in the Chicago standard metropolitan statistical
area. Springfield: State of Illinois.
Illinois Institute of Technology. 1963. City of Chicago industrial renewal study.
Chicago: Illinois Institute of Technology.
Imbroscio David L. 1995. An alternative approach to urban economic development:
Exploring the dimensions and prospects of a “self-reliance” strategy. Urban Af¬
fairs Review 30: 840-867.
Influential Contacts, Ltd. 1981. Contacts influential: Commerce and industry direc¬
tory Chicago: Influential Contacts, Ltd.
Inside Contacts. 1993. Inside contacts USA: Marketing information directory.
Chicago: Inside Contacts.
_. 1995. inside contacts USA: Marketing information directory. Chicago: inside
Contacts.
Jones, Bryan D. 1985. Governing buildings and building government: A new perspec¬
tive on the old party. University, AL: University of Alabama Press.
Jones, Bryan D„ and Lynn W. Bachelor. 1984. Local policy discretion and die corpo¬
rate surplus. In Urban economic development, ed. Richard D. Bingham and John
P. Blair. London: Sage.
Joravsky, Ben. 1982. Organizers look beyond Alinsky for power. The Neighborhood
Works 5 (May): 10-12. ., .
_ 1988. No entry, gentry: New rules for the Clybourn Corridor. Chicago
Reader, 2 September.
186 Bibliography

-. 1990. Alinsky’s legacy. In After Alinsky: Community organizing in Illinois, ed.


Peg Knoepfle. Springfield, IL: Illinois Issues.
Judd, Dennis R. 1983. From cowtown to Sunbelt city: Boosterism and economic
growth in Denver. In Restructuring the city, ed. Susan Fainstein, Norman
Fainstein, Richard Hill, Dennis Judd, and Michael Smith. New York: Longman.
Judd, Dennis R., and Todd Swanstrom. 1994. City politics: Private power and public
policy. New York: Harper Collins.
Rain, John F. 1968. The distribution and movement of jobs and industry. In The met¬
ropolitan enigma, ed. James Q. Wilson. Cambridge: Harvard University Press.
Kantor, Paul. 1987. The dependent city: The changing political economy of urban
economic development in the United States. Urban Affairs Quarterly 22:
493-520.
Karapin, Roger. 1994. Community organizations and low-income citizen participa¬
tion in the U.S.: Strategies, organization, and power since the 1960s. Paper de¬
livered at the 1994 annual convention of the American Political Science Associa¬
tion (31 August-4 September), New York, NY.
Kass, John. 1990. Minorities get a cut of city business. Chicago Tribune, 1 August,
1(2C)-
Katznelson, Ira. 1981. City trenches: Urban politics and the patterning of class in the
United States. New York: Pantheon.
Kennedy, Shawn G. 1988. Side streets getting fashion showrooms. New York Times, 8
May, 7(X).
Kenyon, James B. 1964. The industrial structure of the New York Garment Center.
In Focus on geographic activity: A collection of original studies, ed. Richard S.
Thoman and Donald J. Patten. New York: McGraw-FIill.
King, John. 1987. BRA director’s proposed rule changes have developers up in arms.
Boston Globe, 13 December, 37(A).
-. 1988a. Protecting industry from yuppies and other invaders. Planning 54: 4—8.
—-. 1988b. Boston zone change for light industry hits opposition. Boston Globe,
22 April, 66.
-. 1988c. New plan for light manufacturing offered. Boston Globe, 5 November, 41.
Kleppner, Paul. 1985. Chicago divided: The making of a black mayor. DeKalb, IL:
Northern Illinois University Press.
Kretzmann, John. 1996. Interview by author, 29 May, Chicago.
Kristensen, Peer Hull. 1992. Industrial districts in West Jutland, Denmark. In Indus¬
trial districts and local economic regeneration, ed. Frank Pyke and Werner Sen-
genberger. Geneva: International Institute for Labour Studies.
Lebow, Joan. 1985. SA’s plea to Koch: Help keep us on SA. Women’s Wear Daily, 16
May.
LEED Council. 1990. Keeping jobs for Chicago’s future: A development impact as¬
sessment of Goose Island. Chicago: LEED Council.
-. 1996. Draft report on Chicago Planned Manufacturing Districts. Chicago:
LEED Council.
Lemonides, James. 1984. Stoking the smokestacks. The Neighborhood Works 7
(March): 1, 11-17.
——-. 1995. Interview by author, 17 January, Chicago.
Levine, Marc V. 1989. The politics of partnership: Urban redevelopment since 1945.
187
Bibliography

In Unequal partnerships: The political economy of urban redevelopment in postwar


America, ed. Gregory D. Squires. New Brunswick, NJ: Rutgers University Press.
Liametz, Carl. 1996. Interview by author, 20 February, Chicago.
Lindblom, Charles E. 1977. Politics and markets. New York: Basic Books.
Logan, John R., and Harvey L. Molotch. 1987. Urban fortunes: The political economy
of place. Berkeley: University of California Press.
Logan, John R, and Todd Swanstrom. 1990. Urban restructuring: A critical view. In Be¬
yond the city limits: Urban policy and economic restructuring in comparative perspec¬
tive, ed. John R. Logan and Todd Swanstrom. Philadelphia: Temple University Press.
Longhini, Greg. 1987. Memorandum to David Mosena, 10 November.
Longworth, R. C. 1981. Chicago: City on the brink. Chicago Tribune, 13 May, 1(1).
Lorenz, Edward H. 1988. Neither friends nor strangers: Informal networks of sub¬
contracting in Lrench industry. In Trust: Making and breaking co-operative rela¬
tions, ed. D. Gambetta. Oxford: Basil Blackwell.
_. 1989. The search for flexibility: Subcontracting networks in British and
French engineering. In Reversing industrial decline? ed. Paul Hirst and Jonathan
Zeitlin. Oxford: Berg.
Ludgin, Mary. 1989. Downtown development: Chicago, 1987-1990. Chicago: Depart¬
ment of Planning.
Ludgin, Mary, and Louis H. Masotti. 1985. Downtown development: Chicago,
1979-1984. Chicago: Center for Urban Affairs and Policy Research, Northwest¬
ern University.
_< 1986. Downtown development: Chicago, 1985-1986, Chicago: Center tor Ur¬
ban Affairs and Policy Research, Northwestern University.
Lugar, Michael I. 1984. Federal tax incentives as industrial and urban policy. In Sun¬
belt/snowbelt: Urban development and regional restructuring, ed. William IC. Tabb
and Larry Sawyers. New York: Oxford University Press.
Lukes, Steven. 1974. Power: A radical view. London: MacMillan.
Maar, James. 1996. Interview by author, 7 February, Chicago.
Magee, Mabel A. 1930. Trends in location of the women’s clothing industry. Chicago:
University of Chicago Press.
Malone, M. 1987. Fort Point Channel rezoning plan offered. Boston Globe, 2 Decem-
ber, 30. ~
Manufacturers News, Inc. 1951. Illinois manufacturers directory. Chicago: Manulac-
turers News.
_. 1960. Illinois manufacturers directory. Chicago: Manufacturers News.
__ 1970. Illinois manufacturers directory. Chicago: Manufacturers News.
Markusen Ann. 1985. Steel and southeast Chicago: Reasons and opportunities for in¬
dustrial renewal. Evanston: Center for Urban Affairs and Policy Research,
Northwestern University.
Martin, Cathie J. 1991. Shifting the burden: The struggle overgrowth and corporate
taxation. Chicago: University of Chicago Press.
Massey Doreen 1987. The shape of things to come. In International capitalism and
industrial restructuring, ed. Richard Peet. Boston: Allen and Unwin
Mayor’s Committee for Economic and Cultural Development. 1966 Mid-Chicago
' economic development study. Vol. 1. Chicago: Mayor’s Committee for Economic
and Cultural Development.
188 Bibliography

Mayor’s Council of Manpower and Economic Advisors. 1974. Chicago’s economy.


Chicago: Mayor’s Council of Manpower and Economic Advisors.
-. 1977. Chicago’s economy on the move. Chicago: Mayor’s Council of Manpower
and Economic Advisors.
McCarron, John. 1988. Chicago on hold: Politics of poverty. Chicago Tribune, 28
August, 1(1).
McClellan, Keith. 1966. A history of Chicago’s industrial development. In Mid-
Chicago economic development study. Vol. 3, ed. Center for Urban Studies, Uni¬
versity of Chicago. Chicago: Center for Urban Studies, University of Chicago.
McClory, Robert. 1986. The fall of the fair. Chicago: Chicago 1992 Committee.
McCormick, Lynn. 1994. The rise and fall of collaborative labor relations in
Chicago’s small-firm metalworking network. Paper delivered at the 1994 annual
convention of the Industrial Relations Research Association, (3-5 January),
Boston, MA.
McDonald, John F. 1984. Employment location and industrial land use in metropoli¬
tan Chicago. Champaign: Stipes.
McKnight, John, and John Kretzmann. 1984. Community organizing in the 80s: To¬
ward a post-Alinsky agenda. Social Policy 14 (winter): 15-17.
Meltzer, Jack. 1963. Zoning for residential development in the central area. Chicago:
Jack Meltzer Associates.
Metropolitan Housing and Planning Council. 1980. Chicago Building Code and resi¬
dential rehabilitation: A working paper. Chicago: Metropolitan Housing and
Planning Council.
Metzger, John T., and Marc A. Weiss. 1988. The role of private lending in neighbor¬
hood development: The Chicago experience. Evanston: Center for Urban Affairs and
Policy Research, Northwestern University.
Midwest Center for Labor Research. 1989. The deindustrialization of Chicago.
Chicago: Midwest Center for Labor Research.
Mier, Robert. 1986. Memorandum to A1 Miller, 18 July.
— -. 1987a. Memorandum to Jane Ramsey, 10 August.
— -. 1987b. Memorandum to Ernest Barefield, 2 October.
Mier, Robert, and Kari J. Moe. 1991. Decentralized development: From theory to
practice. In Harold Washington and the neighborhoods, ed. Pierre Clavel and Wim
Wiewel. New Brunswick: Rutgers University Press.
Mier, Robert, and Wim Wiewel. 1983. Business activities of not for profit organiza¬
tions: Surviving the New Federalism? Chicago: Center for Urban Economic De¬
velopment, University of Illinois at Chicago.
Mier, Robert, Wim Wiewel, and Lauri Alpern. 1992. Decentralization of policy mak¬
ing under Mayor Harold Washington. In Politics of policy innovation in metropol¬
itan Chicago, ed. Kenneth Wong and Terry Nicholas Clark. Greenwich, CT: JAI
Press.
Mollenkopf, John H. 1981. Community and accumulation. In Urbanization and ur¬
ban planning in capitalist society, ed. Michael Dear and Allen J. Scott. London:
Methuen.
-. 1983. The contested city. Princeton: Princeton University Press.
Molotch, Harvey. 1976. The city as a growth machine: Toward a political economy of
place. American Journal of Sociology 82: 309-332.
Bibliography 189

Molotch, Harvey, and John Logan. 1984. Tensions in the growth machine: Over¬
coming resistance to value-free development. Social Problems 31: 483^499.
Mosena, David R. 1987a. Memorandum to Robert Mier and Elizabeth Hollander, 11
September.
-. 1987b. Memorandum to Elizabeth Hollander, 29 September.
Moses, Leon, and Harold F. Williamson. 1967. The location of economic activity in
cities. American Economic Review 57: 211-222.
Mott, Andrew. 1984. Mixing organizing with development. The Neighborhood Works
7 (January): 1, 24-27.
Murphy, H. Lee. 1996. New plan for Cotter and Co. site intensifies fight with indus¬
try. Crain’s Chicago Business, 5 February, 1.
Nagel, Laura O. 1982. Portfolio for the future: Chicago’s long range infrastructure
planning needs. Chicago: Metropolitan Housing and Planning Council.
Neighborhood Capital Budget Group. 1995. A matter of delivery: A report on the
City of Chicago’s industrial street and viaduct improvement programs, 1990
through 1999. Chicago: Neighborhood Capital Budget Group.
-. 1996. Moving beyond the basics: Building Chicago for the next century.
Chicago: Neighborhood Capital Budget Group.
Nickel, Denise R. 1995. The progressive city? Urban redevelopment in Minneapolis.
Urban Affairs Review 30: 355-377.
Noyelle, Thierry J., and Thomas M. Stanback, Jr. 1983. The economic transformation
of American cities. Totowa, NJ: Rowman and Allanheld.
O’Connor, James. 1973. The fiscal crisis of the state. New York: St. Martin’s Press.
O’Comior, Len. 1975. Clout: Mayor Daley and his city. Chicago: Henry Regnery Co.
Olcott, George C. 1950. Olcott’s land values blue book of Chicago. Chicago: George C.
Olcott and Co.
_ 1960. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
__ 1965. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
_. 1970. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
_. 1975. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
__ 1980. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co. „
_ 1985. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
-. 1988. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
__ 1990. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott
and Co.
-. 1994. Olcott’s land values blue book of Chicago. Chicago: George C. Olcott

Orr, Marion E., and Gerry Stoker. 1994. Urban regimes and leadership in Detroit.
Urban Affairs Quarterly 30: 48-73.
Pavlos, Elliott Arthur. 1975. Chicago’s Crosstown: A case study in urban expressways.
190 Bibliography

In The manipulated city, ed. Stephen Gale and Eric C. Moore. Chicago: Maaro-
ufa Press.
Peet, Richard. 1987. Industrial restructuring and the crisis of international capitalism.
In International capitalism and industrial restructuring, ed. Richard Peet.
Boston: Allen and Unwin.
Perlman, Janice E. 1976. Grassrooting the system. Social Policy 7 (Sept./Oct.): 4—20.
Peterson, Paul E. 1981. City limits. Chicago: University of Chicago Press.
Philip Zeitlin Associates. 1982. Chicago Zoning Ordinance review. Chicago: Philip
Zeitlin Associates.
Piore, Michael, and Charles Sabel. 1984. The second industrial divide: Possibilities for
prosperity. New York: Basic Books.
Polsby, Nelson W. 1963. Community power and political theory. New Haven: Yale
University Press.
Pruska-Carroll, Marika. 1987. The Printers Row: A case of industrial displacement.
Masters Project, University of Illinois at Chicago.
Pyke, Frank, Giacomo Becattini, and Werner Sengenberger, eds. 1990. Industrial dis¬
tricts and inter-firm cooperation in Italy. Geneva: International Institute for
Labour Studies.
Pyke, Frank, and Werner Sengenberger, eds. 1992. Industrial districts and local eco¬
nomic regeneration. Geneva: International Institute for Labour Studies.
Rakove, Milton L. 1975. Don’t make no waves, don’t back no losers. Bloomington: In¬
diana University Press.
-1982. Jane Byrne and the new Chicago politics. In After Daley: Chicago poli¬
tics in transition, ed. Samuel K. Gove and Louis Masotti. Chicago: University of
Illinois Press.
Ranney, David C., and Wim Wiewel. 1987. The graphic communications industry in
the Chicago metropolitan area. Chicago: Center for Urban Economic Develop¬
ment.
-. 1988. Technological change and industrial restructuring: The graphic commu¬
nications industry. Chicago: Center for Urban Economic Development.
Rapkin, Chester. 1963. The South Houston industrial area: A study of the economic sig¬
nificance of firms, the physical quality of buildings, and the real estate market in an
old loft section of Lower Manhattan. New York: Department of City Planning.
Rast, Joel S. 1993. The social output of the post-Fordist firm: Lessons from the auto¬
mobile industry. Unpublished manuscript.
Real Estate Research Corporation. 1970. Industrial functions in the Chicago central
communities. Chicago: Real Estate Research Corporation.
Reardon, Kenneth M. 1990. Local economic development in Chicago 1983-1987.
Ph.D. dissertation, Cornell University.
Reardon, Patrick T. 1992a. Just like dad, Daley building a legacy. Chicago Tribune, 29
March, 1(2C).
-. 1992b. Infrastructure work could pump $880 million into city, study says'.
Chicago Tribune, 25 February, 4(2C).
-. 1993. Daley had big plans, but now just big problems. Chicago Tribune, 10
January, 1(2C).
-. 1995. Mayor worries about keeping factories afloat. Chicago Tribune, 28 Au¬
gust, 1(1).
Bibliography 191

-. 1996. Mell may keep site industrial. Chicago Tribune, 17 April, 3(2).
Reardon, Patrick T., and Rick Pearson. 1991. McCormick goes McLean: No dome in
new plan. Chicago Tribune, 22 May, 1(1).
Rich, Michael J. 1993. Federal policymaking and the poor. Princeton: Princeton Uni¬
versity Press.
Rivlin, Gary. 1992. Fire on the prairie: Chicago’s Harold Washington and the politics of
race. New York: Henry Holt and Co.
Robinson, Carla J. 1985. Analysis of the Chicago apparel and fashion industry.
Chicago: Department of Economic Development.
Robinson, Tony. 1995. Gentrification and grassroots resistance in San Francisco’s
Tenderloin. Urban Affairs Review 30: 483—513.
Rosen, George. 1980. Decision-making Chicago style: The genesis of a University of Illi¬
nois campus. Urbana: University of Illinois Press.
Rosenthal, Donald B., ed. 1980. Urban revitalization. Beverly Hills: Sage.
Sabel, Charles F. 1989. Flexible specialization and the re-emergence of regional
economies. In Feversing industrial declineP ed. Paul Hirst and Jonathan Zeitlin.
Oxford: Berg.
Sandro, Phillip H. 1988. Job impact analysis program economic impact statement.
Chicago: Mayor’s Office of Employment and Training.
■ Saxenian, AnnaLee. 1994. Regional advantage: Culture and competition in Silicon
Valley and Route 128. Cambridge: Harvard University Press.
Schattschneider, E. E. 1960. The semisovereign people. New York: Holt, Rinehart and
Winston.
Scherrer, Christoph. 1991. Seeking a way out of Fordism: The US steel and auto in
dustries. Capital and Class 44:93—120.
Schmidt, William E. 1987. Chicago plan aims to curb factory loss. New York Times,
10 December, 1(1).
Schneider, Mark. 1992. Undermining the growth machine: The missing link between
local economic development and fiscal payoffs. Journal of Politics 54: 214-230.
Scott, Allen J. 1988. Metropolis: From the division of labor to urban form. Berkeley:
University of California Press.
Sengenberger, Werner, and Frank Pyke. 1992. Industrial districts and local economic
regeneration: Research and policy issues. In Industrial districts and local eco¬
nomic regeneration, ed. Frank Pyke and Werner Sengenberger. Geneva: Interna¬
tional Institute for Labour Studies.
Shea, Vickie. 1996. Interview by author, 25 May, Chicago.
Shearer, Derek. 1973. CAP: New breeze in the Windy City. Ramparts 12 (October):
12 16
Shefter, Martin. 1985. Political crisis/fiscal crisis: The collapse and revival of New York
City. New York: Basic Books.
Slilay Anne B., and Robert P. Giloth. 1987. The social organization of a land-based
elite: The case of the failed Chicago 1992 World’s Fair. Journal of Urban Affairs
9.305—324.
Slonka, Tom. 1981. Housing subsidies put to work. The Neighborhood Works, 9 Janu-

Smith^Neil. 1982. Gentrification and uneven development. Economic Geography 58:


139-155.
192 Bibliography

Snyder, David. 1995. Dick Mell, alderman vs. Dick Mell, industrialist. Crain’s
Chicago Business, 3 April, 10.
Squires, Gregory D., ed. 1989. Unequal partnerships: The political economy of urban
redevelopment in postwar America. New Brunswick, NJ: Rutgers University Press.
Squires, Gregory D., Larry Bennett, Kathleen McCourt, and Philip Nyden. 1987.
Chicago: Race, class, and the response to urban decline. Philadelphia: Temple Uni¬
versity Press.
Stanbaclc, Thomas M., and Thierry J. Noyelle. 1982. Cities in transition: Changing
job structures in Atlanta, Denver, Buffalo, Phoenix, Columbus (Ohio), Nashville,
Charlotte. Totowa, NJ: Allanheld, Osmun and Co.
Stone, Clarence N. 1976. Economic growth and neighborhood discontent. Chapel Hill:
University of North Carolina Press.
--. 1980. Systemic power in community decisionmaking: A restatement of strati¬
fication theory. American Political Science Review 74: 978-990.
-. 1987a. The study of the politics of urban development. In The politics of ur¬
ban development, ed. Clarence Stone and Heywood Sanders. Lawrence: Univer¬
sity Press of Kansas.
-. 1987b. Summing up: Urban regimes, development policy, and political
arrangements. In The politics of urban development, ed. Clarence Stone and Hey¬
wood Sanders. Lawrence: University Press of Kansas.
-. 1989. Regime politics: Governing Atlanta, 1946-1988. Lawrence: University
Press of Kansas.
-. 1993. Urban regimes and the capacity to govern: A political economy ap¬
proach. Journal of Urban Affairs 15: 1-28.
Stone, Clarence N., Marion E. Orr, and David Imbroscio. 1991. The reshaping of ur¬
ban leadership in U.S. cities: A regime analysis in urban life and transition. In
Urban life in transition, ed. Mark Gottdiener and Chris G. Pickvance. Newberry
Park, CA: Sage.
Stone, Clarence N., and Heywood Sanders. 1987. The politics of urban development.
Lawrence: University Press of Kansas.
Stone, Donald. 1974. Industrial location in metropolitan areas. New York: Praeger.
Strahler, Steven R. 1995. “Can’t” has become the cant in Daley’s city that worked.
Crain’s Chicago Business, 12 June, 3.
-. 1996. How sweet the subsidy. Crain’s Chicago Business, 6 May, 15.
Strong, James. 1989. Two outsiders picked for Daley cabinet. Chicago Tribune, 9
May, 1(3).
Sturdy, Frank. 1949. Old buildings topple in path of expressway. Chicago Daily Tri¬
bune, 30 November, 4(1).
Suttles, Gerald D. 1990. The man-made city: The land-use confidence game in Chicago.
Chicago: University of Chicago Press.
Swanstrom, Todd. 1985. The crisis of growth politics: Cleveland, Kucinich, and the
challenge of urban populism. Philadelphia: Temple University Press.
-. 1988. Urban populism, uneven development, and the space for reform. In
Business elites and urban development: Case studies and critical perspectives, ed.
Scott Cummings. Albany: State University of New York Press.
-. 1993. Beyond economism: Urban political economy arid the postmodern
challenge. Journal of Urban Affairs 15: 55-78.
Bibliography 193

Tobier, Emanual, and Mark A. Willis. 1981. The commercial printing industry: Is it
still made in New York? In New York City’s changing economic base, ed. Benjamin
J. Klebaner. New York: Pica Press.
Tolliday, Steven, and Jonathan Zeitlin, eds. 1987. The automobile industry and its
workers: Between Fordism and flexibility. New York: St. Martin’s Press.
Tulloss, Janice K. 1995. Citizen participation in Boston’s development policy: The
political economy of participation. Urban Affairs Review 30: 514-537.
Turner,’ Robyne S. 1992. Growth politics and downtown development: The eco¬
nomic imperatives in Sunbelt cities. Urban Affairs Quarterly 28. 3—21.
U.S. Bureau of the Census. 1976. Census of manufactures, 1972. Washington, D.C.:
U.S. Government Printing Office.
-. 1987. Census of manufactures, 1987. Washington, D.C.: U.S. Government
Printing Office.
_. 1992. Census of manufactures, 1992. Washington, D.C.: U.S. Government
Printing Office.
Vernon, Raymond. 1959. The changing economic function of the central city. New
York: Committee for Economic Development.
Vindasius, Julia. 1988. Industry-led community development: A report on the tar¬
geted development project. Unpublished manuscript.
Vogel, Ronald K., and Bert E. Swanson. 1989. The growth machine versus the anti¬
growth coalition: The battle for our communities. Urban Affairs Quarterly 25:
63-85.
Warner, Kee, and Harvey Molotch. 1995. Power to build: How development persists
despite local controls. Urban Affairs Review 30: 378-406.
Washburn, Gary. 1981. “Restrictive” codes thwart a loft revival. Chicago Tribune, 18
January, 1(N14).
_ 1993. Loop trolley estimate hits $775 million. Chicago Tribune, 29 July,
11(2C).
Washburn, Gary, and Andrew Martin. 1998. City OKs Pilsen redevelopment plan,
Chicago Tribune, 11 June, 1(2).
Weiss, Marc A., and John T. Metzgar. 1989. Planning for Chicago: The changing
politics of metropolitan growth and neighborhood development. In Atop the ur¬
ban hierarchy, ed. Robert A. Beauregard. Totowa, NJ: Rowman and Littlefield.
Whitt, J. Allen. 1982. Urban elites and mass transportation: The dialectics of power.
Princeton: Princeton University Press.
Wiewel, Wim. 1990. Economic development in Chicago: The growth machine meets
the neighborhood movement. Local Economy 4: 307-316.
Wiewel, Wim, David Ranney, and George W. Putnam. 1990. Technological change in
the graphic communications industry: Implications for economic development
planning. Economic Development Quarterly 4: 371-382.
Wiewel, Wim, and Nicholas C. Reiser. 1989. The limits of progressive municipal eco¬
nomic development: Job creation in Chicago, 1983-1987. Community Develop¬
ment Journal 24: 111-119.
Winters Publishing Co. 1942. Chicago central business and office building directory,
1942-1943. Chicago: Winters Publishing Co.
Wong, Kenneth K. 1988. Economic constraint and political choice in urban policy¬
making. American Journal of Political Science_32. 1—18.
194 Bibliography

Wright, Patricia A. 1987. The impact of federal and state Urban Development Action
Grants on Chicago, 1979-1986. Chicago: Center for Urban Economic Develop¬
ment, University of Illinois.
-. 1992. Choices ahead: CDCs and real estate production in Chicago. Chicago:
Nathalie P. Vorhees Center for Neighborhood and Community Improvement.
Zeitlin, Jonathan. 1992. Industrial districts and local economic regeneration:
Overview and comment. In Industrial districts and local economic regeneration,
ed. Frank Pyke and Werner Sengenberger. Geneva: International Institute for
Labour Studies.
Zeitlin, Jonathan, and Peter Totterdill. 1989. Markets, technology, and local inter¬
vention: The case of clothing. In Reversing industrial decline? ed. Paul Hirst and
Jonathan Zeidin. Oxford: Berg.
Ziemba, Stanley. 1976. “Village” development downtown under study. Chicago Tri¬
bune, 4 February, 3(1).
Zukin, Sharon. 1982. Loft living: Culture and capital in urban change. Baltimore:
John Hopkins University Press.
Index
Accumulation vs. legitimation, 13, 20, coalition, 85, 98, 100-101; govern¬
130,157, 164 ing coalition, 85, 100-101, 109; in¬
Ad Hoc Working Committee for Resi¬ terpretations of probusiness policies
dential Rehabilitation, 43 of, 99-101; relations with African
Addison Industrial Corridor, 144, 176n Americans of, 103
Agglomeration economies, 13, 159,
162-163 Cabrini-Green housing development, 3,
Agnos, Art, xii, 175n 111
Aldermen, 41-43, 93, 98, 114, 116, Capital improvements, 25-26, 34-36,
129,140-141, 144, 175n 133,148-152, 155
Alinsky, Saul, 85-87, 93, 175n Capital Improvements Advisory Com¬
American Institute of Architects, 43 mittee, 136,149
Arthur Andersen and Co., 173n Capital Improvements Program, 16, 99,
Atlanta, Ga., 11-12 149-150, 173n
Austin, Tex., 136 Capital mobility, 9, 12-13, 15, 19, 45,
Ayers, Thomas, 99, 173n 70-71, 107,159-160, 162, 164
Carbone, Nicholas, 132
Back of the Yards, 85-86 Carson Pirie Scott, 173n
Balanced growth, 106, 108, 116, Centrum Properties, 122-126
120-121, 134, 169 Chicago 1992 World’s Fair; plans for, 99;
Baltimore, Md., 4, 161, 166 opposition to, 84, 99, 169
Baron, Hal, 103 Chicago 21 Corporation, 32, 55
Barton-Aschman Associates, 31-32 Chicago Apparel Center, 64, 70, 175n
Bietler, J. Paul, 135 Chicago apparel industry, 16, 38, 47-48,
Bilandic, Michael, 24, 38, 101, 109, 59- 72; characteristics of, 47-48,
167, 173n; 1979 election campaign, 60- 65, 70-71; computer-aided de¬
95-98, 105 sign (CAD), 64, 70; contracting and
Block-busting, 87, 175n subcontracting, 48, 60, 63-64,
Bodman, James, 144 69-70; decline of, 62, 65;
Boston, Mass., xii, 3-4, 15, 165-168 designer/manufacturers, 63-65,
69-70; displacement of firms, 65,
Bridgeport, 145
69, 108; flexible specialization,
Brownfields, 152
Buccitelli, Michael, 92-93 63-65, 69-70; jobbers, 60; labor,
Building code, 16, 26, 42M3, 55, 174n 60-61; locational patterns of, 24,
59, 60-61, 70-71; overseas compe¬
Burke, Edward, 98
Business incubators, 17, 91-92 tition, 61-63, 70; postwar industry
Business, privileged position of, 9, 19, restructuring, 61-65, 70; sweat¬
shops, 70; task force, 70, 110. See
159
Byrne, Jane, 24, 45, 106, 115, 167; also Chicago garment district
1979 election campaign, 95-98; Chicago Association of Neighborhood
1983 election campaign, 104-105; Development Organizations
and Chicago Rehabilitation Code, (CANDO), 94-95, 119; Securing
43M4, 55; economic development Older Buildings Project, 176n
planning by, 89, 99, 110; electoral Chicago Central Area Committee
196 Index

(CCAC), 27, 30-33, 54-55, 87, cremental changes to, 41-42; per¬
173n formance standards, 40-41,
Chicago garment district, 59-61, 64-66, 114— 115, 131. See also Zoning
108; and 1958 Development Plan, Citizens Action Program, 87
66; land values in, 67, 69; and Sears City Council, 28, 42^13, 98, 105, 110,
Tower, 68-69 114, 117, 120, 126, 128-129, 134,
Chicago Housing Authority (CHA), 96, 141, 144; Committee on Buildings
103 and Zoning, 39, 41; Committee on
Chicago Milwaukee Corp. (CMC), 138, Streets and Alleys, 68
140 City of Chicago: Board of Education, 96;
Chicago political machine, 41, 96, Chicago Land Clearance Commis¬
98-99, 105, 114, 116, 126,140, sion, 33-34, 173n; Chicago Transit
144, 175n; decline of, 96-97, Authority (CTA), 97; Department
105-106; effects on neighborhood of Buildings, 42; Department of
political mobilization, 17, 46, 86, City Planning, 27, 29, 35; Depart¬
137, 167-168. See also Patronage; ment of Development and Planning,
Selective incentives 30-31; Department of Economic
Chicago printing industry, 16, 38, Development, 38, 109, 111,
47-60; characteristics of, 47-53, 65, 115- 116, 127, 140-141, 176n; De¬
70-71, 162; decline of, 58-59; dis¬ partment of Housing, 94; Depart¬
placement of firms, 53-58, 65, 108; ment of Planning, 116, 124,
effects of technological change on, 128-129, 140-142; Department of
51- 53, 59, 65, 70; labor shortages, Planning and Development, 41,
59; labor unions, 59; letterpress, 51, 142, 144, 147, 152; Mayor’s Coun¬
174n; lithography, 51; locational cil of Manpower and Economic Ad¬
patterns of, 24, 48-53, 58-59, visors, 30-31; Mayor’s Office of
70-71, 118, 162; and real estate Employment and Training, 127,
speculation, 53-59, 71-72; subcon¬ 176n; Mayor’s Office of Special
tracting, 48, 56-57, 59, 65, 70, Events, 103; Plan Commission, 39,
118; task force, 110; trade shops, 120-122, 126, 128-129, 176n
52- 53, 58. See also Printing House Cleveland, Ohio, 132
Row Clybourn Corridor Planned Manufactur¬
Chicago Rehab Network, 43, 94 ing District (PMD), 117-130, 138,
Chicago Rehabilitation Code, 42M-4, 143, 153-154, 156, 176n;and Eu¬
174n; Mayor’s Advisory Commit¬ gene Sawyer administration,
tee, 42M:4 126- 128; and Harold Washington
Chicago River, 29-30, 33, 60, 67, 111, administration, 120-125; media
113, 117, 121, 143, 149, 174n, coverage of, 118-119, 127; opposi¬
176n tion to, 119, 129; passage of, 129;
Chicago Sun-Times, 68-69, 127 recommendation for, 117; studies,
Chicago Tide and Trust Co., 27, 173n 127- 128; support for, 117-119
Chicago Tribune, 90, 118, 120-123, Clybourn Corridor, 117, 176n
127, 134,141 Clybourn Lofts, 113, 117
Chicago Zoning Ordinance, 38-42, 114, CNA Financial Corp., 173n
130; 1942 Comprehensive Amend¬ Columbus, Ohio, 146
ment of, 39, 174n; 1957 Compre¬ Commercial Club of Chicago, 110
hensive Amendment of, 39^11; in¬ Committee for Legal Lofts, 43, 55
197
Index

Commonwealth Edison Co., 87, 99, 148-151; and the Chicago machine,
173n 135-137; and downtown redevelop¬
Community Development Block Grants ment, 133-134, 136-137, 146,
(CDBGs), 99, 111 157; governing coalition, 137, 151,
Community Renewal Society, 101 156-157, 169-170; and industrial
Community Workshop on Economic De¬ land-use plans, 142-147; and
velopment (CWED), 94, 102-103, planned manufacturing districts,
105,109-110, 119, 175n 133, 140-142, 145, 168; and tax
Community-based organizations, increment financing, 151-152; and
102-103, 109, 137, 156; alliances zoning changes, 143-145
with manufacturers, 17, 91-94, Dearborn Park, 32-35, 37, 55, 57
106, 125, 159, 164-165; coalition Dearborn Park Corp., 173n
building by, 84, 87, 94-95; and lo¬ Deindustrialization, 22, 26, 38, 87-90,
cal-producer strategy, 17, 90-95, 101,106-107, 159
109; opposition to growth coali¬ Delfino, Wayne, 138
tions, 84-85. See also Neighborhood Department of Housing and Urban De¬
economic development velopment (HUD), 36, 90-91,
Concerned Allied Neighbors, 119 174n
Continental Illinois National Bank, 173n Development plans: 1958 Development
Cook County Democratic Central Com¬ Plan, 29-31, 34, 39, 54-55, 66,
mittee, 27-28, 95-96, 173n 173n; Chicago 1992 Comprehen¬
Corplan Associates, 52 sive Plan, 99; Chicago 21 Plan,
Corporate-center strategy, xii, 7, 13-14, 30-33, 55, 87; Chicago Works To¬
16-18,20-21,85, 107, 134, gether: 1984 Development Plan,
157-158, 161-162, 168-169 104,110, 115
Crain’s Chicago Business, 104—105, 123, DiGaetano, Alan, 159
127, 140,144 Division of labor between state and mar¬
Creticos, Peter, 127 ket, 10-13,45,70-71, 159-161
Crosstown Expressway, 84, 87 Downs, James C., Jr., 173n
Downtown redevelopment, xi, 3—4,
Daley, Richard J., 52, 84, 95-96, 105, 20^8, 132, 158; and Chicago gar¬
173n; African-American support for, ment district, 65-70; effects on in¬
29, 96; and capital improvements, come distribution, 4; and growth
34-36; and the Chicago machine, coalitions, 170; ideological claims
27-29, 41, 85, 146,156,167;and regarding, 46, 70-71, 83; and Print¬
the Chicago Rehabilitation Code, ing House Row, 53-59, 66, 70; role
42-43, 174n; and downtown rede¬ of public policy in, xii, 5-6, 16, 19,
velopment, 13-14, 24,26-34, 25-46,48,71,106,109,161-162;
53-55,66-69,89,99,110,162; structural influences on, 5, 7, 9,
electoral coalition, 28-29; govern¬ 11-13,45, 100,161-162,
ing coalition, 109, 170; relations 164-165. See also Industrial dis¬
with downtown business commu¬ placement
nity, 27; and zoning, 39 Draper and Kramer, 173n
Daley, Richard M., 145-146, 157-158, Ducharme, Donna, 115, 118, 121, 124,
165; 1983 election campaign, 104; 147-148, 175n
1989 election campaign, 133-135,
156; and capital improvements, Economic restructuring: global, 6-7, 12,
198 Index

19, 71, 107, 110, 171; urban, 4-6, Growth machines, 70, 105-107, 161,
8,13-14, 70, 108-109, 159, 161 168-171; objectives of, 10, 83; op¬
Economies of scale, 61, 70, 162, 173n position to, 10, 84-85; participants
Economies of scope, 15, 70, 162, 173n in, 10-11, 84, 158. See also Growth
Eighteenth Street Development Corpo¬ coalitions
ration, 115
Eisendrath, Edwin, 117, 127-128 Hanley Dawson, 126-127
Electronics industry, 165 Harris Trust and Savings Bank, 173n
Elkin, Stephen, xi, 159-160 Hart Schaffner & Marx, 173n
Elston Corridor Planned Manufacturing Hartford, Conn., 132
District (PMD), 138-142; studies, Henderson, Duncan, 126
140-141 “Highest and best use,” 7, 19, 41, 55,
Enterprise zones, 101 71, 83, 104, 107, 117, 123, 128,
Epton, Bernard, 105 138,163-164
Erikson, Donald, 173n Hilton Hotels Corp., 173n
Evans, Timothy, 133-135 Hollander, Elizabeth, 122-123,
126-127
Federal Express, 152 Horowitz, Tern, 113, 117
Finkl, Charles, 127
Finkl Steel, 127 Illinois Central Industries, 173n
First Federal Savings and Loan, 173n Industrial Council of Northwest Chicago
First National Bank of Chicago, 173n (ICNC), 92, 106
Fitzgerald, Joseph, 44 Industrial displacement, 14, 20, 53-58,
Flexible specialization, 63-65, 162, 165 65-72, 108, 111-131, 133,
Flynn, Raymond, xii, 168 138-145, 153, 165-168; commu¬
Food processing industry, 24, 165 nity opposition to, 18-19, 123;
Fuller, Dorothy, 64 studies, 115-116
Industrial districts, 14-16, 18, 25-26,
Gabinski, Terry, 126, 144 35,45, 123, 130, 141, 145, 153,
Giloth, Robert, 95, 110, 115 162, 166; networking among firms
Goose Island, 145, 152; and River Lofts within, 15-16, 20-21, 56-57, 59,
development, 121-125, 138. See 65, 148; studies of, 15-16. See also
also Goose Island Planned Manufac¬ Flexible specialization; Subcontract¬
turing District ing
Goose Island Planned Manufacturing Industrial Revenue Bonds (IRBs), 38,
District (PMD), 122, 138-142, 174n
153-154; studies, 140-141, 164 Inland Steel Co., 173n
Govas, Ken, 92
Governing coalitions, xi, 26, 46, 107, Jackson, Jesse, 103
130,165 Jackson, Maynard, 12
Greater North Pulaski Development Jacobson, Edwin, 138
Corporation (GNPDC), 93, 106, James, Joseph, 136
111,119 Jane Addams Resource Corporation
Greater Southwest Development Corpo¬ (JARC), 92, 106, 119
ration, 147-148 John A. Melaniphy Associates, 98
Growth coalitions, 10-11, 20, 30, 84, John Hancock Center, 68 \
87, 99, 108 Johnson, Wallace, 98
199
Index

King, Martin Luther, Jr., 96 McCormick Place, 34, 136, 146, 168
Klutznick, Philip, 173n Medical supplies industry, 165
Koch, Edward, 69 Mell, Richard, 144
Kramer, Ferd, 173n Metcalf, Gordon, 68
Kretzmann, John, 102, 175n Metropolitan Area Housing Alliance, 87
Kucinich, Dennis, 132 Metropolitan Housing and Planning
Council, 149
Labor unions, 119-120, 125, 129, 166, Mier, Robert, 95, 109-111, 121-122,
170; support for machine, 27 175n
Lake View Citizens’ Council, 144 Model Industrial Corridors Initiative,
Lincoln Park, 111, 119, 129 147_148,152-153,155,170-171
Lindblom, Charles, 8 Moe, Kari, 103, 110
Local Employment and Economic Devel¬ Molotch, Harvey, 10, 160
opment (LEED) Council, 93-94, Mosena, David, 122-124, 136, 142
111, 113,115-116, 123, 144, 147,
168, 175n; and Clybourn Corridor Natarus, Burton, 121, 140
PMD campaign, 117-119, 121, Navy Pier, 34, 136, 168
126, 128; and Clybourn Lofts devel¬ Neighborhood Capital Budget Group
opment, 113; and Elston Corridor (NCBG), 149-150
PMD campaign, 138; and Goose Is¬ Neighborhood economic development:
land PMD campaign, 138, 141; and and community development corpo¬
River Lofts development, 121-122. rations (CDCs), 90-91; and
See also Ducharme, Donna corporate-center strategy, 94; and
Local Industrial Retention Initiative deindustrialization, 87-90; govern¬
(LIRI), 93, 111, 126, 147-148, 153 ment funding for, 90-91; history of,
Local-producer strategy, 16-19, 85, 87, 85-87; and industrial development
95, 106, 109, 133, 158,165, organizations (IDOs), 91-94, 106,
169-170 111; and local-producer strategy,
Loft buildings, 24-25, 91; and Chicago 17,90-95, 106. See also
garment district, 63, 66, 69; conver¬ Community-based organizations
sion to nonindustrial uses, 14, 16, New Haven, Conn., 3, 8
22,25,33,37-38,41-42,44, New Orleans, La., 166
53-58, 65, 113, 115,117, 153, New York, N.Y., 3, 45, 161, 165-167;
166, 175n; and Printing House garment district, 69, 166
Row, 48, 50, 53-58, 175n; and New York Times, 27, 127
North Business and Industrial Council,
River North, 29
Logan, John, 10, 160 119,144
North River Industrial Corridor, 93,
Longhini, Greg, 124
111-117, 119, 121-122, 124, 126,
Longworth, R. C., 90
Los Angeles, Calif., 165 133,138-140, 143, 176n
Northern Trust Co., 173n
Marina City, 30
Oberman, Martin, 113-114, 116-117,
Marshall Field 8c Co., 173n
Marxism, 9 129,175n
Masotti, Louis, 98, 127
Patronage, 17, 28-29, 41, 43, 46, 96,
Matthews, Robert, 113
105, 116, 136-137, 156, 167. See
McCarron, John, 134
200 Index

also Chicago political machine Selective incentives, 11, 17, 29, 46, 114,
Peoples Gas Co., 173n 137
Perkins, John H., 173n Shakman Decree, 96, 136, 175n
Pettibone, Holman, 27 Shaw, Gerald, 69
Philadelphia, Pa., 4, 166 Sheffield Neighbors, 119
Planned/protected manufacturing dis¬ Shipka, Ron, 143
tricts (PMDs), 18, 117, 133, 141, Simon, Raymond, 68
164, 167-171; effectiveness of, Sinclair, Upton, 85
152-155. See also Clybourn Corri¬ Skidmore, Owings & Merrill, 173n
dor Planned Manufacturing District; Skoning, Warren G., 173n
Elston Corridor Planned Manufac¬ Spiegel, Inc., 145-146
turing District; Goose Island Standard Oil Building, 68
Planned Manufacturing District Standard Oil Co., 173n
Pluralism, 8-9, 19-20 Steel industry task force, 110
Portland, Ore., 166-167 Stone, Clarence, xi, 11-12, 100
Presidential Towers, 33, 99 Subcontracting, 14-16, 56-57, 59, 70,
Printing House Row, 14, 24, 35, 37, 43, 165
48-50, 53, 56, 58, 60, 108, 175n; Suburbanization, 3, 26, 31, 146
and 1958 Development Plan, 29, Suttles, Gerald, 90
54-55; and Dearborn Park, 55, 57; Syracuse, N.Y., 166
land values in, 54-55, 57. See also
Chicago printing industry Tax increment financing (TIF), 8, 133,
151-153, 155, 166
Ramsey, Jane, 121
RANCH Triangle Assn., 119 United Center, 136
Ravenswood Corridor, 153-155, 176n University of Illinois at Chicago, 29-30,
Reagan, Ronald, 101 32, 34, 37, 54, 56, 84; Center for
Real Estate Research Corp., 31-32, 173n Urban Economic Development,
Redistribution, xii, 5, 7-9, 12-13, 109-110
20-21, 83,95, 100, 106-109, Urban Development Action Grants
157-158, 160-161, 170 (UDAGs), 8, 36-38, 55, 174n
Redlining, 87 Urban Investment and Development
River City, 33, 37 Co., 173n
River Lofts, 121-125, 138 Urban political economy, xi-xii, 8-13,
Riverpoint Center, 122-126 19-21, 45, 71, 100,107-108, 130,
132,157-165
San Francisco, Calif., xii, 165-166, 175n Urban regime theory, xi, 10-11, 164
Sawyer, Eugene, 136, 140, 142, 152, Urban regimes, xi, 10-11, 19, 21, 130,
156-157; 1989 election campaign, 165
133-135; and capital improvements, Urban renewal, 3, 9
150-151; and Clybourn Corridor
PMD, 126-128, 130-132; and in¬ Vazquez, Arturo, 110
dustrial land-use plans, 142-143 Vertical integration, 52, 71
Schmoke, Kurt, 4 Vienna Sausage Manufacturing Co.,
Scribner & Co., 173n 143-144
Sears Tower, 45, 68-69 Voinovich, George, 132 \
Sears, Roebuck, and Co., 68-69, 173n Vrdolyak, Eddie, 98, 105
Index 201

Walker, Dan, 84
Washington, Harold, xii, 45, 93, 127,
135-138, 144, 152, 156-158,
167-168; 1983 election campaign,
95, 103-105; and capital improve¬
ments, 150-151; and Clybourn
Corridor PMD, 120-126, 130-132;
economic development platform,
104, 108-111, 134, 164; electoral
coalition, 85, 100-101, 105, 109,
134; governing coalition, 100-101,
106; and industrial displacement, xi,
18, 115,120-125,130-131,140,
142; and industry task forces, 70;
relations with business community,
120, 130-131. See also Local Indus¬
trial Retention Initiative
Washington Papers, 103-104, 109-110,
115
Weese, Harry, 55
West Side Coalition, 87
Westgate Mill, 41
Wiebolt,"Raymond C., 173n
Woodlawn Organization, 84

Zoning: and industrial displacement, 16,


18,25-26, 93-94, 108,113-116,
153, 166; controversies, 113,
121-124, 138; rezoning of indus¬
trial land, 18, 40-41, 54-55, 69,
93-94, 113-116, 120-127,138,
143-145, 153, 166, 176n. See also
Chicago Zoning Ordinance
A

\
"N
*

\
>
\
/

* ✓
%

\
K

\
N
with labor and community organiza¬
tions concerned with the decline of
well-paying industrial job opportuni¬
ties. Responding to these pressures,
city officials from the Harold Wash¬
ington, Eugene Sawyer, and Richard
M. Daley administrations have taken
steps to implement a citywide indus¬
trial policy.
Remaking Chicago portrays urban
economic development as open-ended
and politically contested. It demon¬
strates that who governs matters and
shows how opportunities exist for
creative local responses to urban eco¬
nomic restructuring. Based on exten¬
sive research, this well-written case
study will appeal to those interested in
urban planning and politics, economic
development, and Chicago history
and politics.

holds a Ph.D. in political


science from the University of Ore¬
gon. He is a research associate at the
Center for Neighborhood Technology
in Chicago.

Jacket Resign by Julia Fauci


Remaking Chicago

“Challenges the
conventional wisdom
in urban political
economy.”
-Todd Swanstrom
author of
The Crisis of
Growth Politics

“A fresh and insight¬


ful view of the process
of downtown
redevelopment.”
—Paul Kleppner,
author of

You might also like