Journal of Rural and Community Development

The rural context of community

development in Canada1
Bill Reimer
Department of Sociology and Anthropology
Concordia University
Montréal QC Canada
[email protected]

Abstract
This paper examines contextual conditions that limit or enhance community
capacity processes. Four contextual conditions are considered in the analysis:
integration to the global economy, stability of the local economy, metro-adjacency,
and institutional capacity. Data from rural Canada are used to explore how these
contextual characteristics condition the relationship between the use of social
capital and four community outcomes: labour force participation, household
incomes, employment, and life expectancy. Results from the New Rural Economy
Project in Canada suggest that these contextual characteristics place important
conditions on the capacity processes considered. In some case, they accentuate the
strength of the relationship between social capital and the outcomes, in others they
reduce it, and in a few, they reverse the direction of the relationship between the
two. The paper concludes with some comments on the implications of the findings
for policy development and community development practice.

1.0 Introduction
Local community development initiatives provide considerable hope for places
under stress. Rather than passively suffer the consequences of external pressures,
community development approaches provide useful strategies and frameworks for
communities to take proactive measures to prepare for and build a better future.
They have shown how the identification of assets and liabilities, participation and
bottom-up capacity-building, democratic governance, and transparency can create
opportunities for groups, towns, and cities (Flora & Flora 2004; Kretzmann &
McKnight 1993; Green & Haines 2002).
But local development does not take place in a vacuum. Assets controlled from
outside the community can make a significant difference on the options available,
just as provincial or federal policies, natural events, and the historical legacy of the
region can modify the nature of their relations in ways that condition local
opportunities. Without identifying some of these contextual constraints and

1
The author thanks the Social Sciences and Humanities Research Council of Canada, The
Canadian Rural Revitalization Foundation, and my colleagues and partners on the New
Rural Economy Project for the support which has made this research possible. Mike Burns
and Moses Tiepoh have made major contributions to the paper through their work on the
NRE database and several of the indexes.

ISSN: 1712-8277 © Journal of Rural and Community Development

www.jrcd.ca
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 156

facilitators, community development that focuses only on the local assets and
liabilities is likely to result in frustration or failure. Once they have been identified,
on the other hand, the options and opportunities for local agency in the face of
external forces are likely to be more visible.
This paper provides a perspective and preliminary analysis for framing such local
approaches. Using research and insights from the New Rural Economy Project
(NRE) of the Canadian Rural Revitalization Foundation (CRRF), we outline some
of the key ways in which contextual effects may condition local development
processes.2 We then provide some research results that can be used to guide both
local decisions and general policies for rural revitalization.

Contextual Factors
Capacity is reflected in the ability of communities or groups to reorganize assets to
produce valued outputs. Figure 1 illustrates the key elements of this process as we
have conceived in within the NRE Project. This model treats assets (and liabilities)
as the basic endowments of a community or group. From a community
development perspective, these have been identified in terms of economic capital,
human skills and abilities, social capital, and natural resources but these are not
meant to be exhaustive of the types of assets and liabilities with which a group
may have at its disposal. Communities or groups can organize or reorganize these
assets and liabilities in various ways depending on their desired outcomes. In the
NRE, we have identified a number of outcomes of interest to researchers, policy-
makers, and rural people, but as with the assets, this list is not meant to be
exhaustive.

Figure 1: The NRE Capacity Model

The central part of the model represents the various processes through which
communities or group might reorganize their assets to produce outcomes. This
version of the model identifies four normative structures through which this is

2
Details on the NRE can be found via http://nre.concordia.ca.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 157

done: market-based, bureaucratic, communal, and associative. As discussed in

more detail when discussing the measurement of social capital below, they
represent some of the ways in which people co-ordinate their behaviour, justify
their activities, enforce standards and expectations, and produce outcomes (Reimer
2004; Tiepoh & Reimer 2004). These outcomes can, in turn, become new assets
and liabilities which may be used by communities or groups to produce new
outcomes in a cycle of capacity building or decline.
Although much of this capacity has been treated as locally based, it is important to
recognize that there are significant constraints on it due to conditions beyond local
control (Marsden 2004). These contextual conditions can impinge on or facilitate
the capacity process at many points as indicated in the Figure. In some cases this
occurs through the nature of the assets themselves (Stedman, Parkins, & Beckley
2004) while in others, it occurs through the institutions and entitlements
established for the management of those assets (Bird & Tassanyi 2001). The
existence of natural resources and the way they are exploited, for example, may be
determined by climate and natural endowment, institutional and political
regulations, or international markets. Local capacities must be imagined and
developed within these more general constraints.
External conditions may also affect local capacity through modification of the
action processes themselves (Salamon 2002; Williams, Sligo, & Wallace 2005).
National or provincial governments develop regulations that structure local
services, infrastructure, and jurisdictional boundaries, for example – and thereby
favor certain forms of organization over others. This may, in turn, modify the
assets-outcomes relations by facilitating or inhibiting certain types of capacity over
others. The recent round of amalgamations encouraged or enforced by Provincial
governments, for example, change the issues and agencies that local communities
might control, those that they don’t control, and the things they could control
(Vojnovic & Poel 2000).
In order to examine the relation between contextual effects and local processes, we
will focus on the relationship between social capital assets and selected community
outcomes. This choice is strategic since it builds on the strengths of the NRE
research and addresses issues that have been neglected in the literature on
community and economic development. Using the NRE capacity model, we will
first of all examine the general relationship between social capital assets and key
outcomes, then introduce contextual conditions in the analysis to see how the
general relationships might be altered. In this way, we not only contribute to the
general understanding about asset-outcome processes, but identify specific place-
related characteristics that can modify those processes.
The NRE project provides the basic data and research framework for our
discussion. It is a national, multi-disciplinary, collaborative project established in
1997 by researchers, policy-makers, and rural citizens. Among other things it was
designed to monitor four contextual characteristics that condition the options and
opportunities for local communities: the integration of local economies into the
global economy, the stability of the local economy, the proximity to major urban
centres, and the level of institutional capacity in the local region. These conditions
served as the basis for the sample frame from which the 32 sites in the NRE Rural
Observatory were drawn and they have been used as primary comparisons within
the profile, census, and survey data analysis of the project(Reimer 2002a). These
four conditions will be the primary focus for this paper.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 158

Global Economic Exposure and Integration

Canada has always been a nation with strong global connections. Its colonization
was largely driven by outside interests in our natural resources and our balance of
payments continue to be maintained by extensive commodity trade (Reimer 2005).
Rural areas have always been directly implicated in this process because of their
intimate connection to the natural resources we trade.
Recent concerns with freer trade and the globalization of markets, therefore, occur
in a background of a long history of global trade. What makes it different now is
the scale, competition, and ideology of the recent changes. The quantity of trade is
increasing, the range of actual and potential trading partners has grown, the
number of competing nations has increased through both technological change and
political agreements, and political discourse for more open and extensive trading
has become dominant.
The extent to which a rural community is exposed to these forces is likely to
condition its options in significant ways (Task Force on Persistent Rural Poverty
1993:314; Marsden 1998; Drummond & Marsden 1999). Our analysis, therefore,
will include this as a key dimension for comparison.
Global Economic Integration and Exposure indices were developed to measure
how much an area is integrated and exposed economically to the global or
international market (Makhija, Kim, & Williamson 1997; Krugman & Obstfeld
1991). Using Canadian census subdivisions (CSD) we identified their industrial
employment as a basis for estimating this exposure. An industry is considered
globally integrated if it engages in both exports and imports. An industry is
classified as globally exposed if it engages in exports only
Three types of indexes were created to measure global economic exposure and
integration for three key industry categories. These indexes were constructed as
follows:
Intra-Industry Trade (IIT): the ratio of net exports to total trade
Industry Exposure (EPTT): the ratio of exports to total trade
Industry Exposure (EPGDP): the weight of exports to total output or GDP
Since trade data was not available at the CSD level, we used Provincial values for
key industries, then weighted them by the distribution of employment in those
industries at the CSD level. These indexes were calculated for the period of 1993-
2002 for each of three key industries: (1) Agricultural/Fishing/Forestry/Hunting;
(2) Manufacturing; and (3) Utilities. Results from these indexes were then
averaged to give us an overall measure of global connectedness, for each year at
CSD level.3 For this project, results from 1996 and 2001 have been selected as
community database indicators. This index ranges from zero to one, with zero
indicating no global exposure and integration or connectedness and one indicating
“complete” global exposure and integration (Tiepoh & Burns 2004).

3
Tourism was not included in this analysis since it contains elements for global
connections (with international tourism and ownership) and local connections (with local,
regional, or national tourism and ownership).
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 159

Fluctuations in the local economy

Options and opportunities for places with very unstable economies will vary from
those in which the economic future is relatively predictable (Mankiw & Scarth
2001). Instability makes planning difficult and lowers the attractiveness of the site
for new industry and business. In addition, the stability of the economy is often
beyond the control of the local community – especially in rural areas. For this
reason, we included it as the second conditioning factor.
Economic stability was measured by examining fluctuations in inflation-
standardized Real Gross Domestic Product (GDP). Since these figures were not
available for CSDs, the Provincial values were weighted by local employment
levels. Employment trends were identified by using labour force survey estimates
to calculate GDP per capita and CSD industry shares in order to measure economic
stability at the regional level. The fluctuation index was based on the standard
deviation of weighted GDP figures for 11 industries over the 1987-97 period
(Mankiw & Scarth 2001; Sundrum 1990).
The results derived from calculating economic stability are expressed as an index
ranging from between zero to one. A value of zero indicates no economic
fluctuation (i.e. high economic stability); a maximum value of one indicates a high
degree of economic fluctuation (i.e. low economic stability) (Dressler & Burns
2004).

Adjacency to large metropolitan centres

Access to urban centres is a critical element for the economic and social condition
of communities and regions. Large urban centres provide a population base for
commerce and employment, a wide range of services and institutional resources,
and cultural aspects that are often glamorized in the popular media (Jacobs 1984;
Newby 1986; Sassen 2000). This geographically-based factor is usually beyond
local control.
Metro adjacency was measured using the Metropolitan Influence Zones (MIZ)
established by Statistics Canada (2006). These zones were established to reflect
the extent to which CSDs contained people who commuted to nearby Census
Metropolitan Areas (CMA) or Census Agglomerations (CA). Four zones were
identified: Strong MIZ (with commuting flows of 30% or more of the population
to any nearby CMA or CA urban core), Moderate MIZ (with commuting flows of
5% to 30% to any nearby CMA or CA urban core), Weak MIZ (with more than 0%
but less than 5% commuting flows to any nearby CMA or CA urban core), and No
MIZ (with less than 40 people in the labour force, or no people commuting to a
nearby CMA or CA urban core). Using these categories, we assigned those in the
Weak MIZ, No MIZ, and the Northern Territories to the ‘non-adjacent’ category
and all others to the metropolitan adjacency category.

The level of institutional capacity

Local communities also vary greatly in the number and size of major institutions
that are nearby. The existence of nearby schools, hospitals, government agencies,
and other major service institutions will considerably affect the opportunities for
employment and the attractiveness of the location for newcomers (Knack & Keefer
1997; Bollman 1999; Flora 1998). They will also have some important impacts on
the level of skills and abilities among the population(Granovetter 1985; Putnam
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 160

1993). In most cases, however, the existence of such institutions is beyond the
control of local people. For this reason, we include it as the fourth contextual
condition for our analysis.
There are few indicators in the public sphere that can be used to measure
institutional capacity. As a result, we were forced to use indirect means that focus
on individual characteristics and their employment in key institutions (Vansant
2000). This information was assumed to reflect the competence of participants
(demonstrated practical skills) and autonomy (legal and structural) of institutions
in the following activity areas:
Accessing and managing resources (financial, human, and technical,
including accessing and managing information);
Carrying out key functions (providing information, services and training;
contributing to social and economic progress);
Responding to exogenous stresses;
Introducing change when necessary in an effort to achieve the greatest
benefit possible from the first two capacities, and to enhance institutional
sustainability (through sound internal governance and inter-institutional
relations) (Hopkins 1996; Bhagavan & Virgin 2004; IMF 2002; Morgan &
Tascherwau 1996).
An index was constructed from the following items:
% of bilingual individuals in the CSD,
% of people in the CSD with a post-secondary education,
% of people in the Census Consolidated Subdivision (CCS)4 employed in
intellectual and managerial occupations,
% self-employed workers in the CSD5,
% of people in the CCS employed in education,
% of people in the CCS employed in government services, and
% of people in the CCS employed in health and social services.
The formula to measure local institutional capacity uses standardized scores for
each of the seven indicators listed above. The average was calculated to create a
single index. Results at the CSD level ranged from a low of –18% to a high of
18%. A positive percentage is an indication that a high level of institutional
capacity is present within the CSD (Briscoe & Burns 2004).
Table 1 identifies the number of rural CSDs that are high and low on each of these
indexes in 2001. The distinction was made by simply dividing the number of cases
in half for each of the dimension.

4
The CCS is a region larger than a CSD. It represents the labour force region associated
with a CSD. It was used in this indicator since many CSDs are unlikely to have major
institutions within their boundaries, but they could be found within the broader region
identified by the CCS.
5
This item was subtracted from the overall index since it reflects a competence that is less
likely to be directed to social institutions.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 161

Table 1: Rural CSDs by 4 Sample Frame Dimensions (2001)

Institutional Capacity
Economic Economic Metro
Connectedness Stability Adjacency Low High
Local Stable Not Adjacent 550 371
Local Stable Adjacent 187 154
Local Fluctuating Not Adjacent 196 166
Local Fluctuating Adjacent 229 115
Global Stable Not Adjacent 308 182
Global Stable Adjacent 110 102
Global Fluctuating Not Adjacent 333 420
Global Fluctuating Adjacent 308 378

A similar classification for 1991 data was used to construct the Rural Observatory
from which much of the analysis is derived. At that time an additional dimension
was added to reflect an overall outcome in the framework. This dimension divided
CSDs into leading and lagging categories depending on selected economic and
social indicators such as incomes, employment, and education levels. Case study
field sites were randomly selected from each of the 32 cells resulting from the
cross-classification of these 5 dimensions(Reimer 2002a).
Table 2 provides simple correlations between the four dimensions of the sampling
frame. It shows that these dimensions are not independent of one another, but most
of the correlations are small. This means that we are able to conduct an
examination of their independent effects with few methodological problems. This
is clearly less so for the sample of NRE field sites, because of its small size, but by
comparing from one data set to the other, we are able to minimize the limitations
of the small number of cases.

Table 2: Correlations between sample frame variables for rural CSDs, 2001
Fluctuating Metro Institutional Leading-lagging
Status (N) Adjacency (N) Capacity (N) status (N)
Global-Local .29** .03 ns .00 ns .25**
Status (3815) (3572) (3832) (3904)
(1=global)
Fluctuating .16** .07** -.02ns
Status (3476) (3774) (3815)
(1=fluctuating)
Metro -.02ns .11**
Adjacency (3533) (3538)
(1=adjacent)
Institutional .04**
Capacity (3832)
(1=high)
** p < .01; ns = not significant

Key Capacity Outcomes

Our analysis will focus on four key characteristics that are particularly important
for rural communities and sites: labour force participation, employment, income,
and life expectancy. These have been chosen since they are common indicators for
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 162

several outcomes in the NRE capacity model and they are relatively easily
accessible among national databases.
Three qualifications must also be made with respect to these characteristics,
however. First, they may be treated as indicators of assets as well as outcomes.
This is acknowledged in the NRE model as a feedback process and reflected in
much of the literature on community development and change (Wilkinson 1989;
Freudenberg 2004). For this reason, we will consider the current results as a first
step in a more complete analysis of CSD trajectories over time. Future work will
build on this first step to examine the longitudinal changes within rural places.
Second, our analysis is limited to synchronic techniques rather than the diachronic
ones implied by the model. Since the data has largely been collected in 2001, we
are limited to this indirect approach when analyzing capacity processes.
Finally, we do not imply that success in community development necessarily
means increases in all of the outcome indicators. Although they all have the
potential to significantly affect the well-being of rural places, there are many
instances where little or no increase in these variables may be preferable for
particular sites, by accident or design.

Labour Force Participation

The extent to which people participate in the labour force has long been a
preoccupation of governments and researchers. Labour force participation not only
reflects the extent to which people have access to incomes and resources through
employment, but it has been used to indicate the level of social inclusion and
relative contribution of the population to the economy as well. Low levels of
participation, whether through age, family structure, lifestyle, or alienation are
generally considered signs of a weakened economic and social system. In a small
town or rural context, however, they must be treated with caution if used in this
way, since many retirement or recreation-based communities may show relatively
low levels of participation without indicating local devitalization.
We use the Statistics Canada census variable of labour force participation as the
indicator for this characteristic. It identifies the proportion of people over 18 years
old who report that they are in the labour force or looking for work.

Employment Levels
Employment levels are generally used to indicate the extent to which the
population has access to the benefits of the dominant economy. They are also
indicators of the level of economic and social inclusion in a similar way to labour
force participation. Employment usually provides a basis for network
development, for example. As with labour force participation, however,
employment statistics must be treated with caution when generalizing to all rural
places, since low employment may not always reflect a weak economy in small
places. The Statistics Canada census variable for the proportion of people 18 years
of age and older who are in the labour force but unemployed is used as the
indicator for employment.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 163

Incomes
Incomes provide a key indicator of the financial resources and wealth available to
a local population. It is likely to be highly correlated to the labour force
participation and employment indicators above, but in addition, it includes income
from pensions, investments, and retirement sources that play a very important role
in many rural places. The access to adequate and stable incomes is likely to
considerably increase the development options for rural places.
The median household income is used in our analysis for two reasons. First, we are
assuming that the household is a primary economic unit within rural areas.
Incomes are typically drawn upon or contribute to all members of the household.
Second. It is the unit we have used in our NRE data collection. Using consistent
units of analysis for the field site and census-level data in this way will facilitate
the comparative analysis.

Health Status
The health condition of the rural population is a key outcome for both rural people
and policy-makers. Not only does it have direct relevance for the social condition
of the population, but it has indirect effects on the economy and political spheres
as well. Unfortunately it is very difficult to integrate health data into the CSD-level
information in our database, so we are left with few indicators to measure these
outcomes. At the moment, we only have information on life expectancy.
Life expectancy data is available only for health regions. These do not coincide
neatly with the boundaries of the sites we have identified, but they provide an
indication of the levels for the regions in which the sites are located. The values are
calculated as an average for the period between 1997 and 2001.

Capacity Processes – Social and Human Capital

Our analysis will be conducted in two stages. The first stage will be to examine the
relationship between capacity assets and processes and the outcomes we have
chosen above. The second stage will involve the examination of contextual effects
on the relationship between social capital and community outcomes. We will use
census and survey data to describe the general situation and trends, but will be
limited once again to the NRE field sites when considering their effects on the
capacity process.

Measuring social capital

The conceptualization of social capital has undergone considerable debate at both
national and international levels(PRI 2005; Grootaert & van Bastelaer 2001).
Rather than engage in those debates within this paper, we will adopt the
perspective and measures developed within the NRE Project and proceed with our
analysis on that basis. More detailed discussions of the rationales and justifications
for these measures can be found in several other reports and papers from our work
(Reimer 2002b; Reimer 2004; Beckley, Martz, Nadeau, & Wall 2000; Tiepoh &
Reimer 2004).
Within this framework, social capital is based on four fundamental types of
normative relations: market, bureaucratic, associative, and communal. They reflect
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 164

four ways in which norms guide the type of acceptable behaviours within social
relations, justify the distribution of resources to the actors, and assign sanctions
when those norms are violated. The four types operate to some extent in general,
but we usually find that one or the other tends to dominate in specific situations. In
some cases, they reinforce one another, but in others they may create tensions or
conflict that can undermine the smooth accomplishment of objectives. They may
produce outcomes that are positive or negative, particularly if we consider these
outcomes by level or time.
Available social capital can be measured by the institutions and organizations
within which the social relations are organized. A school, a baseball league, a food
bank, or a card club all represent social capital that may be used by people or
groups. For those outside the organization, however, the social capital they
represent may remain only potentially available. Similarly, for those who are
participants, the social capital of the school, food bank, or card club may remain
unused for achieving some objectives, even though it may be used for others. It
may not occur to them, for example, that their card club may be used as social
capital to advance community economic development objectives. Much of
community development practice is directed toward recognizing the potential
social capital that may be unused or unrecognized by community members then
mobilizing this social capital in new ways. Our analysis is designed to be sensitive
to this process by which the potential status of available social capital becomes
actualized.
We measure available social capital at the site or community level. The
institutions, businesses, organizations, and associations within the site are
identified and then classified with respect to one or more of the four dominant
norms they reflect. These institutions or groups are assumed to reflect the
organization of social relations according to the dominant normative structures
they reflect. Businesses and business-oriented associations, for example, are
considered to be organized primarily on market-based norms. Schools, hospitals,
and welfare offices are largely organized on bureaucratic-based norms. Voluntary
associations at local, regional, national, or international levels are considered to be
organized predominantly on associative-based norms while churches and
community family-oriented events are used as indicators of communal-based
norms.
On the other hand, we measure the use of social capital through the responses to
the NRE household survey conducted in 2001. A systematically selected sample of
households was surveyed in 20 of our research sites and one adult member
interviewed. The survey provided us with information regarding the activities of
household members with respect to employment, voluntary associations, social
support, local government, and the informal economy. From the responses, we
developed an index for the use of social capital that distinguishes the four types of
normative structures within which it was accessed(Reimer 2002b).
As we can see from table 3, there is only a moderate relationship between the
availability of social capital and its use (cf. highlighted diagonal cells). This means
we are forced to make a choice between measures of available or used social
capital when conducting our analysis. Since our primary concern in this paper is
with the implications of contextual conditions for local action, we will focus on the
used social capital for the time being.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 165

Table 3: Available Social Capital by Used Social Capital (NRE Household

Survey - 1995 cases)
Available Social Capital
Market- Bureaucratic- Associative- Communal- Total
Use of Social Capital based based based based
Market-based .37** -.20** -.20** -.35**
Bureaucratic-based .27** -.12**
Associative-based -.21** -.21** .42** -.12** .28**
Communal-based -.20** -.11** .40** .22**
Total -.35** -.17** .27** .21** .40**
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).

Table 4 provides basic correlations between the use of social capital and the
community outcomes we have selected. It includes separate indexes for the four
types of social capital, a general index for social capital (Total Social Capital), and
one representing the variance among the four types of social capital used by the
household. This last index reflects the extent to which the household makes use of
a wide variety of types of social capital (low variance) or depends primarily on one
type (high variance). It arises from the hypothesis that those sites which are able to
function well within more rather than fewer normative structures will be at an
advantage when it comes to resiliency and strategic action(Reimer 2002b). We
expect this would be reflected in better performance on these economic and health
outcomes.

Table 4: Correlations between Use of Social Capital and Outcomes in 2001

(NRE sites)
LF Unemployment Median HH Life
Participation rate (20) Income (18) Expectancy
Rate
Total Social Capital .48*
Market-Based SoCp .55* -.60**
Bureaucratic-based SoCp -.60** -.70**
Associative-based SoCp
Communal-based SoCp -.48* .77**
Variance among types -.64**
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).

As seen in Table 4, the overall measure for social capital shows a significant
relationship only to the unemployment rate. The positive value of the correlation
suggests that the use of social capital may act as compensation for unemployment
although we must be very cautious with such a conclusion since it is vulnerable to
the ecological fallacy.
A closer look at the various types of social capital reveals several important
relations with the outcome variables that were masked by the summated index.
Market-based social capital is positively related to the labour force participation
rate and negatively related to the unemployment rate in the site. Communal-based
social capital shows the opposite pattern, being negatively associated with labour
force participation and positively related to unemployment. This suggests that
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 166

market and communal-based social cohesion may serve to offset one another in
some way. Bureaucratic-based social capital is negatively related to both labour
force participation and household incomes. This may be a reflection of the
predominance of government institutions among those having a bureaucratic basis.
Retired persons, students, and welfare recipients are more likely to make use of
this type of social capital. They are also those who are likely to be outside the
labour force and have relatively low incomes. Once again, however, we must be
cautious with such conclusions because of the multiple levels represented by these
variables.
Life expectancy is only related to the variance among the four types of social
capital used. The negative sign suggests that life expectancy is highest within sites
that make use of many types of social capital. This is consistent with our
hypothesis. The processes involved are bound to be complex, but the consistency
of the result among various types of sites suggests that further analysis of these
processes would be valuable.
These results make the point that there are important and meaningful relationships
between social capital use within our field sites and some of the outcomes of likely
interest to community members and policy-makers. The processes behind these
relationships remain unclear at this point since the analysis is synchronic and has
been conducted at the site level while most of the processes are likely to be internal
to those sites. However, they serve as a good basis upon which we can consider the
primary focus of this paper: the role of contextual effects on the relationships we
have identified above.

Contextual Effects
An examination of simple correlations between the contextual variables and our
selected outcomes shows that none of them are statistically significant. One might
be tempted to conclude, therefore, that those contextual factors are irrelevant to
local development. However, the picture changes when we examine how
contextual variables affect the relationships between assets and outcomes within
our field sites.
To demonstrate these effects, the relationships between social capital and the
selected outcomes are computed separately for each condition of the contextual
variables. In Figure 2, for example, the positive relationship between the use of
market-based social capital and labour force participation is strongest among those
sites that are adjacent to major metropolitan centres. In non-adjacent sites the
relationship is reduced (it falls below levels of statistical significance).
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 167

Figure 2: Labour Force Participation by Use of Market-based Social Capital

and Metro-adjacency

90

80

70
LF Participation rate - 2001

60

50 Metro-adjacency
adjacent

40 not adjacent
.8 .9 1.0 1.1 1.2 1.3

Market-based Social Capital Used (ratio)

These results suggest that the context of the site is likely to affect the nature and
intensity of the relationship between the extent to which market-based social
capital is used in the site and the level of outcomes we have considered. The details
of the conditioning effects are likely to be valuable information for community
development strategies focused on social capital.
To explore these effects, the relationships between social capital and the selected
outcomes are computed separately for each condition of the contextual variables
(cf. Table 5). The most relevant conditions in the table are those where the
correlations exist for one of the categories of the contextual variable, but not for
the other. For example, we find that the correlation between the level of market-
based social capital and labour force participation is .87 among those sites with
relatively stable economies, but it is non-significant among those with fluctuating
economies. This suggests that increasing the level of market-based social capital
within the former type of sites is likely to have a greater impact on the labour force
participation than within the latter type. These conditions are illustrated in Figure
3.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 168

Table 5: Summary of significant contextual effects on the correlations between

the use of social capital and selected outcomes (N)
Type of Social Capital Used
Market- Bureaucratic- Associative- Communal- Variance
based based based based
LF Local
Partici- Global -.61* (14)
pation Stable .87* (6) .96* (6)
Rate Fluctuating -.62* (14)
Not Adj. -.70** (13)
Adjacent .79* (7)
Low Cap. -.78** (12) .60* (12) -.63* (12)
High Cap.
HH Local
Income Global -.65* (14) .53* (14)
Stable -.97** (5)
Fluctuating -.56* (13)
Not Adj. -.73* (11)
Adjacent
Low Cap. -.78** (10)
High Cap.
Un- Local
employ- Global -.55* (14) .82** (14)
ment Stable -.86* (6)
Fluct. .86** (14)
Not Adj. .82** (13)
Adjacent
Low Cap. -.75** (12) .83** (12)
High Cap.
Life Local
Expect- Global -.61*
ancy (11)
Stable
Fluctuating -.61*
(11)
Not Adj. -.71*
(10)
Adjacent
Low Cap. -.69* (9)
High Cap.
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 169

Figure 3: Labour Force Participation by Market-based Social Capital and

Economic Stability

90

80

70
LF Participation rate - 2001

60

50 Econ. Stability
fluctuating

40 stable
.8 .9 1.0 1.1 1.2 1.3

Market-based Social Capital Used (Ratio)

In some cases, the conditional effects are even more dramatic – producing opposite
results depending on the context. The use of associative-based social capital, for
example, appears to have quite different relationships with labour force
participation depending on whether the site economy has a relatively high or low
level of institutional capacity. This is illustrated by Figure 4. Although the
relationship among high capacity sites is not statistically significant, it suggests
that there may be important variations to consider when planning local strategies.
If the results in Figure 4 are generalizable, for example, it would suggest that
building associative-based social capital in low capacity sites will have positive
impacts on labour force participation rates, whereas in high capacity ones, its
effects will be negative.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 170

Figure 4: Labour force participation by Associative-based Social Capital and

Institutional Capacity

90

80

70
LF Participation rate - 2001

60

50 Inst. Capacity
high capacity

40 low capacity
.4 .6 .8 1.0 1.2 1.4 1.6 1.8

Associative-based Social Capital (ratio)

A closer look at Table 5 allows us to identify some of the most important

contextual variables to consider and their role for social capital. Sites with globally
integrated local economies seem most responsive to the use of various forms of
social capital, for example. In Table 5, we find that in globally-oriented economies,
market-based social capital is negatively related to unemployment, bureaucratic-
based social capital is negatively related to labour force participation and
household incomes, associative-based social capital is positively related to
household incomes, and communal-based social capital is positively related to
unemployment. In locally-based economies, none of these relationships appear to
be significant.6 If this pattern is robust, it suggests that globally-oriented
communities are more likely to benefit from attention to social capital in their
development strategies than locally-oriented ones.
The stability of the local economy shows similar conditional impacts on the
relationships between social capital and the outcomes considered. In this case,
however, the strongest impacts are not found so exclusively within one condition.
Relatively stable economies show positive relationships between market and
bureaucratic-based social capital and labour force participation and negative ones
between bureaucratic-based social capital and household incomes and between
market-based ones and unemployment. However, in fluctuating economies, the use
of communal-based social capital becomes more important. Within these types of
sites, communal-based social capital is negatively related to labour force
participation and positively related to unemployment. This reinforces the
conclusion that family and close friendship networks are more important where the
economic stresses are high, but it leaves open the question why this relationship

6
Since the number of cases for locally-based sites is rather small (6) we repeated the
analysis using non-parametric statistics (Spearman’s rho) and found the same pattern of
results.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 171

should be more important in sites with fluctuating economies over those where the
economies are stable.
Adjacency to metropolitan centres shows mixed results with respect to the use of
social capital and the outcomes considered. Non-adjacent sites are more likely to
show negative relationships between the use of bureaucratic-based social capital
and labour force participation and incomes and a positive relationship between
communal-based social capital and unemployment. On the surface, this suggests
that using bureaucratic and communal social capital in response to economic stress
is most likely to be found as a common strategy in non-adjacent sites. In adjacent
sites, on the other hand, the only significant relationship is between the use of
market-based social capital and labour force participation. What is remarkable
about these results is that the use of market-based social capital has little
relationship to such participation in non-adjacent sites. This challenges simplistic
claims that social capital enhances economic performance under all
circumstances(Putnam 1993).
We find that institutional capacity also shows conditional effects on the social
capital-economic outcome relationships. In this case, it is low capacity sites that
are most often the locus for such relationships. In low-capacity sites, the use of
bureaucratic and communal-based social capital is negatively related to labour
force participation, but the use of associative-based social capital is positively
related to the same outcome. Bureaucratic-based social capital is also negatively
related to household incomes. As with several of the other conditions,
unemployment is positively related to the use of both market-based and
communal-based social capital. In high capacity locations, however, none of these
relationships appear significant. This raises the question whether the use of social
capital is a secondary means of access to these economic benefits. Within sites
where there is institutional strength and high human capital, employment and
income are first of all sought through these avenues rather than through social
capital. Within sites with low capacity, social capital provides a more important
system of access to these outcomes.
Finally, we find that all four of the contextual variables are important for
examining life expectancy outcomes, but not in the ways above. They have no
impacts on the direct relationship between the use of each type of social capital and
health, but they become particularly important when we consider the range of
social capital types used. In global, fluctuating, not-adjacent, and low capacity
sites, for example, we find that the life expectancy is highest where there is a
relatively wide use of the various types of social capital. This is reflected in the
negative relationships found in all these cases. These results are consistent with our
hypothesis that advantages accrue to sites that are able to make use of a wide
variety of types of social capital, but it also raises a further question: Why should
this primarily be the case within global, fluctuating, not-adjacent, and low capacity
sites?

Conclusions
In general, these results provide strong evidence for the importance of contextual
conditions for local community development processes. They show that the
relationship between the use of various forms of social capital and selected
economic and health outcomes will vary depending on those contextual conditions.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 172

They support the NRE model that views local development as a mix of both
endogenous and exogenous factors – sometimes interacting in contradictory ways
– and they go on to highlight what some of those factors might be. Building
associative-based social capital may not always have positive impacts on
household incomes, for example, especially where the local economy is relatively
isolated from the global economy.
This analysis suggests that all four of our contextual conditions play an important
part in local processes. On their own, they appear unrelated to the outcomes
selected, but their importance emerges when we examine their conditioning
impacts on the relationship between the use of social capital and those outcomes.
The details of their roles differ significantly, however. The global connectedness
and institutional capacity variables appear to have their significant impacts within
one of their conditions (those sites that are globally connected and those that have
low institutional capacity). The other two (stability of the economy and
metropolitan adjacency) are related to the social capital-outcome relationship
under both of their conditions. In most cases they accentuate the relationships
found, but there is also evidence that some of those relationships may be reversed
depending on the context.
In all of these cases more research is required – especially research that permits in-
depth analysis of local dynamics within a framework where contextual differences
can be examined. This means we need databases that are multi-level, detailed, and
longitudinal. The results reported here are limited because they are synchronic –
thereby making it impossible to identify causal relations between the variables.
The theoretical model we use, however, proposes diachronic processes at both the
local and contextual level. If we are to explore the details and consistency of those
relationships, therefore, we need to develop the databases to make it possible.
The NRE project provides one model for doing this. From the beginning, we have
collected information about processes at a local level, but we have done so only
after ensuring that appropriate comparisons can be made. These results confirm the
wisdom of this strategy at the same time that they identify specific directions for
research, policy focus, and strategic planning.
These results have implications for researchers, policy-makers, and local citizens.
First, they should caution us if we take a strict focus on endogenous factors related
to local outcomes. Without taking into account the context, we are in danger of
misjudging the likely impacts of local changes. Building local capacity through
social capital may be a sound policy in general, but it needs to be modified
according to local conditions, the type of capacity considered, and the outcomes
desired. The variables considered in this paper are but a few of the possible ones to
be considered in the long term.
Second, we need to develop the research designs that are appropriate for the level
of complexity required. This necessarily means collaborative work since the detail
and number of cases required cannot be achieved by one or two researchers on
their own. But this approach requires more than inspiration – it needs the
institutional and financial capacity to put it into place. Once again, policy-makers
need to understand why this is so – in order to adapt such programs to either avoid
the negative effects or to provide compensation for those who may become
vulnerable to them.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 173

Rural citizens and leaders may use these results to better understand the constraints
and facilitators they identify and to prepare their strategic options for the future.
Those living in sites that are well connected to the global economy with plans to
increase this type of connection would be well advised to prepare for any negative
impacts likely to be felt on incomes, employment, or health. These data suggest,
for example, that all forms of social capital may be particularly important for
consideration in these plans.

References
Beckley, T., Martz, D., Nadeau, S., & Wall, E. (2000). Multiple outcomes,
multiple capacities: Delving deeper into the meaning of community
capacity. Sackville, NB.
Bhagavan, M. R., & Virgin, I. (2004). Generic Aspects of Institutional Capacity
Development in Developing Countries. Sweden: Stockholm
Environmental Institute.
Bird, R., M., & Tassanyi, A. (2001). Constraints on Provincial and Municipal
borrowing in Canada: Markets, Rules, and Norms. Canadian Public
Administration, 44(1), 84-109.
Bollman, R. D. (1999). Human Capital and Rural Development: What Are the
Linkages? In N. Walford, J. C. Everitt, & D. E. Napton (eds.) Reshaping
the Countryside: Perceptions and Processes of Rural Change. Wallingford,
UK: CAB International.
Briscoe, A., & Burns, M. (2004). Local Institutional Capacity. Seven Reports on
the Dientification of Rural Indicators for Rural Communities. Montreal:
The Canadian Rural Revitalisation Foundation. [http://nre.concordia.ca]
Dressler, J., & Burns, M. (2004). Economic Stability. Seven Reports on the
Identification of Rural Indicators for Rural Communities. Montreal: The
Canadian Rural Revitalisation Foundation. [http://nre.concordia.ca]
Drummond, I., & Marsden, T. (1999). The Condition of Sustainability. London:
Routledge.
Flora, C. B., & Flora, J. L. (2004). Rural Communities: Legacy and Change (2nd
ed.). Boulder, CO: Westview Press.
Flora, J. L. (1998). Social Capital and Communities of Place. Rural Sociology,
63(4), 481-506.
Freudenberg, N. (2004). Community Capacity for Environmental Health
Promotion: Determinants and Implications for Practice. Health Education
& Behavior, 31(4), 472-490.
Granovetter, M. (1985). Economic Action and Social Structure: The Problem of
Embeddedness. American Journal of Sociology, 91(3), 481-510.
Green, G. P., & Haines, A. (2002). Asset Building and Community Development.
Thousand Oaks, California: Sage.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 174

Grootaert, C., & van Bastelaer, T. (2001). Understanding and Measuring Social
Capital: A Synthesis of Findings and Recommendation From the Social
Capital Initiative. Washington, DC: The World Bank.
Hopkins, T. J. (1996). Capacity Assessment Guidelines The Programme Approach:
Assessment Levels and Methods. New York: UNDP, Management
Development and Governance Division.
[http://magnet.undp.org/cdrb/CAPLEV1.htm]
IMF. (2002). The Role of Capacity-Building in Poverty Reduction. Washington,
DC: International Monetary Fund.
[http://www.imf.org/external/np/exr/ib/2002/031402.htm]
Jacobs, J. (1984). Cities and the Wealth of Nations. New York: Vintage, Random
House.
Knack, S., & Keefer, P. (1997). Does Social Capital Have an Economic Payoff? A
Cross-Country Investigation. The Quarterly Journal of Economics, 112(4),
1251-1288.
Kretzmann, J. P., & McKnight, J. L. (1993). Building Communities from the
Inside Out: A Path Toward Finding and Mobilizing Communty Assets.
Chicago: ACTA Publications.
Krugman, P., & Obstfeld, M. (1991). International Economics: Theory and Policy,
2nd Edition. New York: HarperCollins Publishers Inc.
Makhija, M. V., Kim, K., & Williamson, S. (1997). Measuring Globalization of
Industries Using a National Industry Approach: Empirical Evidence
Across Five Countries and Over Time. Journal of International Business
Studies, May (Fourth Quarter).
Mankiw, G., & Scarth, W. (2001). Macroeconomics. New York: Worth Publishers.
Marsden, T. (1998). New Rural Territories: Regulating the Differentiated Rural
Spaces. Journal of Rural Studies, 14(1), 107-117.
Marsden, T. (2004). The Quest for Ecological Modernisation: Re-spacing rural
Development and Agri-Food Studies. Sociologia Ruralis, 44(2), 129-146.
Morgan, P., & Tascherwau, S. (1996). Capacity and Institutional Assessment:
Frameworks, Methods and Tools for Analysis. CIDA Policy Branch.
Newby, H. (1986). Locality and Rurality: the Restructuring of Social Relations.
Regional Studies, 20(3), 209-215.
PRI. (2005). Social Capital As a Public Policy Tool: Project Report. (Report No.
ISBN 0-662-69178-4). Ottawa: Policy Research Initiative.
[http://policyresearch.gc.ca/doclib/SC_Synthesis_E.pdf]
Putnam, R. (1993). The Prosperous Community: Social Capital and Public Life.
American Prospect, 13, 35-42.
Reimer, B. (2002a). A Sample Frame for Rural Canada: Design and Evaluation.
Regional Studies, 36(8), 845-859.
Reimer, B. (2002b). Understanding Social Capital: Its Nature and Manifestations
in Rural Canada. CSAA Meetings Canadian Sociology and Anthropology
Association.
Reimer, Journal of Rural and Community Development 1 (2006) 155-175 175

Reimer, B. (2004). Social Exclusion in a Comparative Context. Sociologia Ruralis,

44(1), 76-94.
Reimer, B. (2005). Rural and Urban: Differences and Common Ground. H. H.
Hiller (Ed.) Urban Canada: Sociological Perspectives (pp. 70-94). Don
Mills: Oxford University Press.
Salamon, L. M. (2002). The Tools of Government: A Guide to the New
Governance. New York: Oxford University Press.
Sassen, S. (2000). Cities in a World Economy. Thousand Oaks: Pine Forge Press.
Statistics Canada. (2006). Census Metropolitan Area and Census Agglomeration
Influenced Zones (MIZ) . Ottawa: Geography Division.
Stedman, R. C., Parkins, J. R., & Beckley, T. M. (2004). Resource dependancy and
community well-being in rural Canada. Rural Sociology, 69(2), 213-234.
Sundrum, R. M. (1990). Economic growth in theory and practice. Hampshire:
Macmillan Press.
Task Force on Persistent Rural Poverty. (1993). Persistent Poverty in Rural
America. Boulder: Westview Press.
Tiepoh, G. N., & Reimer, B. (2004). Social Capital, Information Flows, and
Income Creation in Rural Canada: A Cross-Community Analysis. Journal
of Socio-Economics, 33(4), 427-448.
Tiepoh, M. G. N., & Burns, M. (2004). Global Exposure and Integration. Seven
Reports on the Identification of Rural Indicators for Rural Communities.
Montreal: The Canadian Rural Revitalisation Foundation.
[http://nre.concordia.ca]
Vansant, J. (2000). A Composite Framework for Assessing the Capacity of
Development Organizations. USAID.
Vojnovic, I., & Poel, D. (2000). Provincial and municipal restructuring in Canada:
assessing expectations and outcomes. Journal of Regional Science,
XXIII(1 (special issue on provincial-municipal reforms)), 1-6.
Wilkinson, K. P. (1989). The Future for Community Development. J. A.
Christenson, & J. W. Robinson (eds.) Community Development in
Perspective (pp. 337-354). Iowa: Iowa University Press.
Williams, J., Sligo, F., & Wallace, C. (2005). Free internet as an agent of
community transformation. The Journal of Community Informatics, 2(1).