Levy Economics Institute of Bard College

Levy Economics
Institute

Strategic Analysis

of Bard College

March 2016

DESTABILIZING AN UNSTABLE
ECONOMY
. , ,
and

Introduction
Last year saw another year of continued growth for the US economy. GDP expanded and the unemployment rate decreased to 5 percent in the last quarter of 2015, its lowest level since the beginning
of 2008 and the start of the global financial crisis. The growth recovery and the decrease in unemployment led the Federal Reserve to change course and increase the federal funds rate in
December 2015. This was the first increase in the federal funds rate in almost 10 years, and it carried the implicit expectation that there would be further tightening in the near future.
However, the current US recovery, now well into its seventh year, is like no other. The Federal
Reserve press release announcing the rate hike was extremely cautious, and many economists
including ourselveshave warned about the possibility of an extended period of secular stagnation.1 Dissatisfaction with the performance of the US economy is dominating the 2016
presidential primaries: despite the relatively low 5 percent unemployment rate, the principal concern of voters and most of the candidates is economy/jobs (as it appears in the questionnaires
of those conducting the various polls).
It is not hard to understand this dissatisfaction. This has been by far the slowest recovery in
the postwar history of the United States. Compared to its precrisis peak in the fourth quarter of
2007, real GDP is only 11 percent higher (as of 2015Q4). Similarly, the number of employed civilian workers in December 2015 was only 3.3 million higherrepresenting an increase of 2.2 percentthan the corresponding employment level in November 2007, the peak of the previous
cycle. Finally, the civilian-employment-to-population ratio is now only 1 percent higher than its
trough after the crisis (2009Q2). This improvement took place between October 2013 and

The Levy Institutes Macro-Modeling Team consists of President Dimitri B. Papadimitriou and Research Scholars Michalis Nikiforos
and Gennaro Zezza. All questions and correspondence should be directed to [email protected]. Copyright 2016 Levy Economics Institute of
Bard College.

December 2014, and has since effectively stopped. The last

time the employment-population ratio was at the current

and the current account deficit. Net exports of petroleum

products in the period 201115 are the one serious exception

level was in April 1984.

In addition, the recent downturn in Brazil and Russia, the

to this. Had it not been for the improvement in the trade balance of petroleum products, the United States overall trade

economic slowdown in China and the crash of the Chinese

deficit would now be at its precrisis level of more than 6 per-

stock market (the Shanghai Stock Exchange Composite Index

is now 44 percent lower than it was last June), and, more generally, the fragile condition of the global economyespecially
the economies of US trading partnerspose another challenge for the US economy and threaten the already anemic
recovery.
The weak foreign demand for US exports is further
dampened by the appreciation of the dollar. In the course of
the last one and a half years, the broad trade-weighted nominal exchange rate of the dollar has appreciated more than 25
percent.
An important exception to the poor overall performance
of net exports during the current recovery is the net export
of petroleum products. The extraction of shale gas, together
with the drop in the price of oil, has led to a very significant
improvement in the trade of petroleum products. Shale gas
extraction has also contributed to aggregate demand through
investment. However, the benefits from the oil market downturn seem to have been exhausted. The price of oil is now
so low that new shale gas projects are not profitable, and there
is little room for improvement in the trade of petroleum
products.
Moreover, there are indications that the instability in the
financial markets can spread to the developed economies,
even the United States. The last couple of months have witnessed significant drops in the stock markets of Europe and
the New York Stock Exchange. Still, the S&P 500 Index is far
above the levels of early 2000 and 2007, and it is hard to see
how the fundamentals of the US economy justify that (in
the same way that they did not then).
However, we should not arrive at the hasty conclusion
that the fragility of the US economy emanates from some
exogenous shocks in foreign demand or the financial markets.
At its core, the US economy remains fragile because of three
deeply rooted structural characteristics. The first is the weak
performance of US net exports. Starting in the mid-1980s, but
especially since the 1990s, there has been a successful invasion
of American markets by foreign products, increasing imports

cent of GDP. This problem of unbalanced trade is now exacerbated by the slowdown in the emerging markets, stagnation
in the rest of the developed economies, and the appreciation
of the dollar. The existence of this structural external deficit
makes the achievement of a satisfactory growth rate for the
economy and full employment dependent on the accumulation of domestic deficits, public and/or private.
Second, over the last 25 years policymakers in Washington
have become increasingly fiscally conservative. The current
recovery is the only one in the postwar period during which
government expenditure has decreased in real terms. Fiscal
austerity, together with weak foreign demand, has put the
entire burden of supporting aggregate demand on the private
sector spending in excess of its income and borrowing. This
has led to a rapid increase in the private sector debt-toincome ratio in the United States.
This process is facilitated by asset inflation because rising
asset prices make the balance sheets of debtors (and creditors)
look better, enabling them to further increase borrowing and
pushing debt-to-income ratios higher. Moreover, nominal
increases in wealth also have a direct positive effect on consumption and aggregate demand. In that sense, the expansion
of the 1990s was supported by the (hyper)inflated stock market
of that period, and the expansion of the 2000s was supported
by the recovery of the stock market together with the real estate
market boom. Accordingly, the current recovery (weak as it is)
has been supported by an extraordinary increase in stock
prices. Therefore, a correction in the stock market will have a
seriously negative impact on growth and employment.
The third serious structural problem in the US economy
is the increase in income inequality over the last four decades,
which has continued uninterrupted after the crisis. Besides
the serious political ramifications it has, the increase in
inequality also has dire macroeconomic consequences. The
transfer of income shares from the middle class and lowerincome households toward households at the top of the income
distribution is a serious drag on demand, since the saving rate
of the latter is much higher than that of the former.

Strategic Analysis, March 2016

Moreover, the aforementioned increase in the debt-toincome ratio falls unevenly on households at the bottom of
the distribution. In a previous report (Papadimitriou et al.
2014), we showed that the debt-to-income ratio of the household sector as a whole increased from 0.6 in the mid-1980s to
1.1 on the eve of the crisis in 2007. This already striking
increaserelated to the developments in the foreign sector
and the fiscal stance of the governmentwas unequally
divided between the bottom and the top of the distribution.
In the top 10 percent of the distribution the ratio remained
virtually unchanged at a low level, fluctuating around 0.5,
while households in the bottom 90 percent saw their debt-toincome ratio increase from 0.7 to 1.6 in 2007. This uneven
distribution of debt has the dual effect of making the economy even more unstable while dampening aggregate demand
when overindebted households try to deleverage in periods
like the current recovery.
Thus, the fragile prospects for the US economy are not the
result of some exogenous shock but are, rather, based on inherent characteristics of the economy and need to be primarily
understood in terms of these three basic structural problems:
(1) weak foreign demand for US exports, (2) fiscal conservatism,
and (3) income inequality. It is these structural problems that

are now being compounded by the weak economic performance of US trading partners, the appreciation of the dollar, and
the possibility of a contraction in asset prices.
In the present report we discuss the state of the American
economy and its prospects for the near future. We show that,
given the current configuration of the US and global
economies, full employment in the United States will become
increasingly dependent on an implausible rise in private sector indebtedness, especially household indebtedness. Such a
process, even if it happens, cannot be sustained infinitely.
As always, we do not attempt to make short-run forecasts,
and our simulations of the possible path of the US economy
relate to the medium- and long-term future.

Figure 1 Index of Real GDP in US Recoveries,

1949Q42015Q4

Figure 2 Index of Employment-to-Population Ratio

Recoveries, 1949Q42015Q4

The Recovery So Far

GDP and Employment
The slow pace of the recovery can be understood with the help
of Figure 1, which depicts the path of real GDP from the
trough to the peak of each post-WWII economic recovery, at
quarterly frequency. Each line in the figure includes the
trough of each business cycle recessionnormalized to 100
and the peak of the subsequent recovery. The recession dates

160

112

150
108
Trough=100

Trough=100

140
130
120
110

104

100

100
90

10
15
20
25
30
Quarters since End of Recession

35

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: Bureau of Economic Analysis (BEA); National Bureau of Economic
Research (NBER); authors calculations

40

96

10
15
20
25
30
Quarters since End of Recession

35

40

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: Bureau of Labor Statistics (BLS); NBER; authors calculations

Levy Economics Institute of Bard College

Figure 4 Index of Real Personal Consumption Expenditures

in US Recoveries, 1949Q42015Q4

Figure 3 Index of Labor Productivity Recoveries,

1949Q42015Q3
135

160

130

150
140
Trough=100

120
115
110

120
110

105

100

100
10
20
25
15
30
Quarters since End of Recession

35

90

40

10
15
20
25
30
Quarters since End of Recession

Sources: BEA; NBER; authors calculations

we use are those published by the National Bureau of

Economic Research. The three colored lines correspond to the
three latest economic recoveries, including the current one.
The gray lines correspond to the previous postwar recoveries.2
It is important to keep in mind that by comparing the cycles
from trough to peak and not from peak to peak we present a
flattering picture of the current recovery, since the drop in
income and employment during the downturn was sharper
than in any other postwar cycle.
Two things stand out in Figure 1. First, we note that the
three most recent recoveries have been the shallowest in US
postwar history. Second, the current recovery is the weakest of
them all.
We notice the same picture in Figure 2, which examines
the recovery of the employment-to-population ratio over the
postwar business cycles. Again, we see that the three latest
recoveries have been the weakest in postwar history (with the
exception of the cycle in the 1960s that traces the recovery of
the 1990s). Most important, Figure 2 shows that in the latest
two cycles, 25 quarters into the recoverymore than six
yearsthe employment-to-population ratio had not recovered
to the level it was at in the trough of the cycle. As we mentioned
above, Figure 2 plays down the underperformance of the labor
market in the latest cycle, because the drop in the employment-

Figure 5 Top Income Shares, 19132014

35

30

5
1985

13

1976

40

1967

17

1958

45

1949

21

1940

50

1931

25

1922

55

1913

Percent

Sources: BLS; NBER; authors calculations

Strategic Analysis, March 2016

40

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2

35

Percent

2012

2003

95

130

1994

Trough=100

125

Top 10% Income Share (left scale)

Top 10% Income Share, Including Capital Gains (left scale)
Top 1% Income Share (right scale)
Top 1% Income Share, Including Capital Gains (right scale)
Source: Alvaredo et al. 2016

to-population ratio during the recession (from peak to trough)

was the steepest in the postwar period. Taken together, Figures
1 and 2 explain the reasons for the latent discontent with the
recovery despite the low level of unemployment.

Consumption, Inequality, and Debt

We can gain some further insights into the weak recovery if we
break down GDP into its major components. Figure 4 presents
the recoveries of real personal consumption expenditures. Not
surprisingly, since consumption is the largest part of GDP, the
pattern is similar to that in Figure 1. The recovery of consumption over the last three recoveries appears markedly weaker
compared to those in the past, and the current recovery is by
far the weakest in postwar US economic history.
The main reason for this behavior of consumption is the
high level of income inequality, with the high indebtedness of
the household sector and its effort to deleverage also playing
a significant role. As many researchers have convincingly documented, since the early 1980s there has been a very significant increase in income inequality in the United States.
According to the most recent data, this process, which was one

of the most important factors in the crisis of 2007, has continued unabated. Figure 5 shows the well-known data for the
income shares of the top 10 percent and top 1 percent of
households for the period 19132014 (Alvaredo et al. 2016).
The data show a clear decrease in the top income shares, and
thus in income inequality, in the late 1930s and early 1940s.
This situation continued until the late 1970s, when the top
income shares started increasing again. We observe that on the
eve of the 2007 crisis the income share of households at the top
of the distribution had reached pre-1929 levels. The important
difference between the post-1929 period, the so-called Great
Depression, and now is that in recent years income inequality
has maintained its upward, precrisis trend.
The increase in the income share of households at the top
of the distribution effectively meant that the average real
income of the remaining 90 percent stagnated. As Figure 6
shows, the rapid increase in the average real income of households at the top in the period after 1980 was accompanied by
stagnant average incomes for the remaining 90 percent. In
fact, the average real income of households at the bottom of
the distribution was lower in 2014 than in 1973.
Two more observations around Figure 6 are interesting.
As we could also infer from Figure 5, during the first three and
a half decades of the postwar period the growth of average
real income was identical for the two classes of households.

Figure 6 Index of Real Average Income, 19452014

(1945=100)

Figure 7 Household-Debt-to-Disposable-Income Ratio,

1960Q12015Q3

350

1.2

Moreover, as Figure 3 depicts, the recovery in labor productivity over the last six years has also been the slowest compared to all postwar business cycles. Indeed, compared to the
fourth quarter of 2010, labor productivity has increased by
only 2.6 percent. The sluggish growth in labor productivity
has allowed unemployment to shrink despite the slow recovery of output. However, it signifies that the new jobs that have
been created are largely low productivity and low paid.

1.1

300

1.0
250

0.9

200

0.8
0.7

150

0.6
100

0.5

50

2010Q1

2000Q1

1990Q1

1980Q1

1970Q1

2015

2005

1990

1960Q1

Top 10%
Bottom 90%

1975

1960

1945

0.4

Sources: Federal Reserve; BEA

Source: Alvaredo et al. 2016

Levy Economics Institute of Bard College

Figure 9 Index of Real Government Expenditure in US

Recoveries, 1949Q42015Q4

Figure 8 Index of Real Gross Private Investment in US

Recoveries, 1949Q42015Q4

200

240
220

180
Trough=100

Trough=100

200
180
160
140

160
140
120

120
100

100
80

10
15
20
25
30
Quarters since End of Recession

35

40

80

Earlier Recoveries
1991Q12001Q1

Earlier Recoveries
1991Q12001Q1

2001Q42007Q4
2009Q2

2001Q42007Q4
2009Q2

10
15
20
25
30
Quarters since End of Recession

35

40

Sources: BEA; NBER; authors calculations

Sources: BEA; NBER; authors calculations

Moreover, the difference between the period of stagnation of

the 1970s and the current one is striking. In the former case,
both classes of households shared the experience of stagnant
real incomes; in the latter, real incomes at the top bounced
back while incomes at the bottom kept falling.
The rise in income inequality is confirmed by other studies that approach the issue from a different perspective. For
example, the Levy Institute Measure of Economic Well-Being
(LIMEW)which takes into account all sources of disposable
income, noncash transfers, public consumption, imputed
income from wealth, and the value of household productionindicates that in accordance with the latest available
data, inequality of well-being also increased after the crisis of
20078, and reached historically high levels in 2013 (RiosAvila, Masterson, and Zacharias, forthcoming). Another
recent study (Berube and Holmes 2016) shows that inequality
at the city and metropolitan level in the United States was also
on the rise as of 2014.
From a macroeconomic standpoint, the increase in
inequality means a transfer of income from households with
high propensity to consume to households with lower propensity to consume, and as a result dampens consumption. From
this point of view, it is easy to understand the weak performance of consumption since 1990.

Given this stagnating average income level of the majority of households, the performance of consumption should
have been much worse. There were two factors that allowed
consumption to increase at the pace it did. The first was the
increase in the indebtedness of households. Figure 7 presents
the household sector debt-to-disposable-income ratio for the
period 19602015. It is no coincidence that the ratio was stable for the period before 1980, when inequality remained constant, and increased after 1980, when inequality started rising.
This increase in debt ratios was unevenly distributed
among the households at the top and the bottom of the
income distribution, with the households at the bottom
recording the higher increases in their debt-to-income ratios.
In effect, lower-income and middle-class households
increased their debt-to-income ratio in order to finance normal consumption expenditures in the face of stagnating
incomes.3 This increase in the debt-to-income ratio of the
household sector was one of the main reasons behind the crisis of 20078. As Figure 7 documents, this ratio reached 1.15
on the eve of the crisis, up from 0.55 two decades earlier.
In turn, the slow recovery of consumption in the post2009 period can be explained by the efforts of households to
reduce their indebtedness. Figure 7 shows the rapid decrease
in the debt-to-income ratio between 2008 and 2012, and the

Strategic Analysis, March 2016

stabilization of the ratio since. However, even today the ratio

remains at a high level by historical standards.
A related factor that allowed for and sustained the
increase in consumption was asset inflation. The period after
1980 and especially after 1990 was marked by rapidly increasing prices of stocks and real estate. Asset inflation has a positive impact on consumption through two different channels.
First, it hides the real vulnerability of highly indebted households (or firms) by inflating the asset side of their balance sheets.
As a result, the increase in the debt-to-income ratio of households did not seem that worrisome as long as the other side of
their balance sheets was growing at a similar pace. Moreover,
the increase in asset prices led to a large increase in the wealth
of the households at the top of the distribution, boosting their
consumption. Second, asset inflation made the balance sheets
of financial institutions look better than they actually were,
and sustained their willingness to increase lending.
In other words, in an economy where a rapid increase in
indebtedness is a precondition for growth, high asset inflation
becomes necessary to support the increase in indebtedness.
Thus, the increase in the debt-to-income ratio and asset price
inflation are two sides of the same coin, of the same underlying process.

Under the current trend (or even the current level) of

income inequality, a return to a normal rate of consumption
growth would require another round of increases in the debtto-income ratio of households, especially those at the bottom
of the distribution, and further inflation in asset markets to
support the increasing indebtedness.

Figure 10a Index of Real Federal Government Expenditure

in US Recoveries, 1949Q42015Q4

Figure 10b Index of Real Local and State Government

Expenditure in US Recoveries, 1949Q42015Q4

Investment
There are many economists who dismiss the significance of
the distribution of income and inequality. For example, the
American economist Robert Lucas famously wrote that, Of
the tendencies that are harmful to sound economics, the most
seductive, and in my opinion the most poisonous, is to focus
on questions of distribution (Lucas 2004).
The rationale behind this approach is that distribution is
determined by technology (the productivity of the factors of
production, etc.), and any attempt to change it creates distortions in the market that yield suboptimal economic results. If
distribution is left to be determined by market forces, profits
will increase and investment will boom, and, at the end of the
day, this improvement in economic activity will trickle down
to wage earners, rendering everyone better off. This rationale
has dominated economic and political debates in the United
States over the last 40 years, and has provided the intellectual

160

280

140
Trough=100

Trough=100

240
200
160

100

120
80

120

10
15
20
25
30
Quarters since End of Recession

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: BEA; NBER; authors calculations

35

40

80

10
15
20
25
30
Quarters since End of Recession

35

40

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: BEA; NBER; authors calculations

Levy Economics Institute of Bard College

Figure 11a Index of Real Government Public Investment in

US Recoveries, 1949Q42015Q4

Figure 11b Index of Real Government Expenditure on R&D

in US Recoveries, 1949Q42015Q4

300

250
200

200

Trough=100

Trough=100

250

150
100

100
50

50
0

150

10
15
20
25
30
Quarters since End of Recession

35

40

Earlier Recoveries
1991Q12001Q1

Earlier Recoveries
1991Q12001Q1

2001Q42007Q4
2009Q2

2001Q42007Q4
2009Q2

10
15
20
25
30
Quarters since End of Recession

35

40

Sources: BEA; NBER; authors calculations

Sources: BEA; NBER; authors calculations

justification against the need to change the patterns of income

distribution that developed over the same period.
However, this investment boom never took place. If anything, the opposite occurred, as can be seen in Figure 8, which
presents the postwar recoveries of investment. The break is
not as clear as in the case of consumption; still, the last three
cycles are on the low side of the postwar recoveries. The 2001
7 recovery of investmentfrom trough to peakwas by far
the slowest. The current recovery, six years after the trough,
has been the second slowest. The picture of the current recovery would be far worse if we compared the cycles from peak to
peak. It was not until the first quarter of 2015 that real investment reached its precrisis peak.

government expenditure is also very important. It can be seen,

in Figure 10a, that the fiscal stance of the federal government
in particular, though restrictive, has not been as restrictive as
it was in the 1990s. This is partly due to the American
Recovery and Reinvestment Act of 2009, which increased government expenditure in the first two years of the recovery.
A significant part of the fiscal consolidation comes from
local and state government (Figure 10b). Here, the difference
with the previous recoveries is again striking. The current recovery is the only one in the postwar history of the United State
with a reduction in local and state government expenditure.
Fiscal consolidation is far-reaching and affects almost
every category of government expenditure. One kind of expenditure that is particularly important for the long-run prospects
for US economic growth is public investment. Figure 11a shows
that government investment follows the same pattern as total
government expenditure, and that the current cycle has been
the most restrictive of the last seven decades.
Similarly, if we focus solely on government investment in
research and development (Figure 11b), the current cycle competes with the 19912001 recovery for last place. Recent studies
(e.g., Mazzucato 2013) have shown the complementarity
between private and public investment, especially public
investment in R&D. From this point of view, the current fiscal

Government Expenditure
Figure 9 shows that another major drag on aggregate demand
during the current recovery has been government expenditure. Today, real government expenditure is 8 percent lower
than it was when the recovery began in 2009Q2. This kind of
fiscal consolidation is unprecedented for the postwar period
and does not change even if we draw the same figure from the
peak of the previous cycle in 2007Q4.
It is important to note that this fiscal consolidation is not
confined to the federal government. The role of local and state

Strategic Analysis, March 2016

consolidation not only has a negative impact on aggregate

demand but also undermines the long-run prospects of the

in imports between 2011 and 2013. In the most recent period,

imports have started rising again, but they are still below the

US economy.

level of all previous economic recoveries.

The relatively good performance of imports is obviously
related to the relatively bad performance of GDP. Slow GDP
growth createdall other things equalslowly growing
demand for imports. However, this improvement is also
related to other factors, the most important being the development of new methods for extracting oil and gas (known as
hydraulic fracturing, or fracking), which led to a very significant drop in the import of petroleum products.
The performance of the foreign sector of the US economy can be better understood with reference to Figure 14,
which presents the overall trade balance in goods and services
and its decomposition into petroleum products, all other
goods, and services. The figure allows us to draw the following
conclusions:

Exports and Imports

The remaining components of GDP that need examination are
those associated with the foreign sector, exports, and imports.
Figure 12 presents the path of exports in the postwar recovery
cycles. As shown, exports performed well during the initial
phase of the current recovery, the result of dollar depreciation
and the relatively high growth rates of the US trading partners
at the time. However, as the recovery proceeded, the situation
reversed: the dollar stopped depreciating in 2011 and started
appreciating rapidly in 2014, while growth in many of the US
trading partners subsided. As a result, whereas the performance of exports was on the high side in the first phase of postwar recovery cycles, it is now on the low side. As the figure
shows, real exports have remained completely stagnant over
the last year. As we will discuss below, the factors that have led
to this stagnation are likely to intensify in the near future.
Finally, Figure 13 presents the postwar recoveries of
imports. Notice that since imports have a negative impact on

1.

The trade deficit of goods other than petroleum products

has been increasing during the current recovery at the
same pace as in the previous recovery cycle. It is now
close to its historical peak.

GDP, the higher the growth of imports, the lower the growth
of GDP. What is striking in this figure is the very low increase

2.

The major game changer has been the trade deficit in

petroleum products. In the early period of the recovery the

Figure 12 Index of Real Exports in US Recoveries,

1949Q42015Q4

Figure 13 Index of Real Imports in US Recoveries,

1949Q42015Q4

240

280
240
Trough=100

Trough=100

200

160

210
180
150

120
120
80

10
15
20
25
30
Quarters since End of Recession

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: BEA; NBER; authors calculations

35

40

80

10
15
20
25
30
Quarters since End of Recession

35

40

Earlier Recoveries
1991Q12001Q1
2001Q42007Q4
2009Q2
Sources: BEA; NBER; authors calculations

Levy Economics Institute of Bard College

trade deficit increased at a pace similar to previous cycles,

but in 2011 it reversed course. It has continued to decrease,
and now stands at around 0.5 percent, almost two percentage points lower than its postcrisis peak of 2.3 percent.
Another significant improvement has come from the
services sector, which recorded an increase in its surplus
of around 0.6 percent of GDP between 2008 and 2013.
This surplus has since remained steady.

3.

go much lower than it already is; and even if it did, the margins
for improvement would be small. In addition, at this low price
level, the exploitation of new oil fields becomes unprofitable, as
does the substitution of imported petroleum. It is indicative
that real imports of petroleum products stopped falling in
2014Q3 and have even risen slightly since then. This is also evident from investment in Mining exploration, shafts, and wells,
which in 2015Q4 was less than half its level just a year ago.
If the balance of petroleum products does stop improving, the overall trade balance will follow the path of the goods
except petroleum products. It is already evident from Figures
11, 12, and 13 that the appreciation of the dollar, together
with the weakening of growth of US trading partners relative
to the United States exerts significant pressure on that balance. This pressure is bound to continue if these factors persist. We will come back to this issue later.

As mentioned above, the improvement in the balance of

petroleum products is related to both the new extraction
methods for shale gas that decreased the importation of oil
and the collapse in oil prices over the last one and a half years:
it is now more than 70 percent lower than in June 2014. To
understand the importance of this improvement we can compare the current trade balance with a counterfactual where the
trade deficit of petroleum products would keep increasing
along its pre-2011 trend. In this counterfactual scenario, the
deficit in petroleum products would now be more than 3 percent of GDPa difference of 2.5 percentage points compared
to its actual current level. In this case, the overall trade deficit
would have exceeded its precrisis peak.
This improvement in the trade balance of petroleum
products is unlikely to continue, because the price of oil cannot

Asset Prices
As we mentioned earlier, the high levels of income inequality,
large external deficits, and fiscal conservatism of the last three
decades have made growth and employment in the United
States dependent on rising private indebtedness and asset
inflation that supports this rise in private sector debt. Asset

Figure 14 Trade Balance, 1985Q12015Q4

Figure 15 S&P 500 Index, 19702015

2500

1
2000

0
Percent

-1
1500

-2
-3

1000

-4
-5

500

Total
Goods except Petroleum and Related Products
Petroleum and Related Products
Services
Sources: BEA; authors calculations

10

Strategic Analysis, March 2016

Source: Federal Reserve Economic Data (FRED), St. Louis Fed

2014

2010

2006

2002

1998

1994

1990

1986

1982

1978

1974

1970

2015Q1

2011Q2

2007Q3

2003Q4

2001Q1

1996Q2

1992Q3

1988Q4

-7

1985Q1

-6

Figure 16 S&P/Case-Shiller Home Price Indices, 19752015

250

Figure 17a Shiller Cyclically Adjusted PriceEarnings Ratio

P/E10, 18812016
50

200
40
150
30
100
20
50

2009

2001

1985

1993

1977

1961

1969

1945

1953

1937

1929

1921

1913

1905

1897

1889

US National Home Price Index

20-City Composite Home Price Index

1881

2015

2011

2007

2003

1999

1995

1991

1987

1983

1979

1975

10

Source: www.econ/yale.edu/~shiller/data.htm
Source: FRED, St. Louis Fed

Figure 17b Market-Capitalization-to-GDP Ratio,

1971Q12015Q4 (1971Q1=100)
600
500
400
300
200

2015Q1

2011Q1

2007Q1

2003Q1

1999Q1

1995Q1

1991Q1

1987Q1

1983Q1

1979Q1

1975Q1

100

1971Q1

inflation, as long as it lasts, can have a positive direct effect on

aggregate demand through various channels. The expansion of
the 1990s was based on the stock market bubble of the time.
Figure 15 shows that there was a fivefold increase in the S&P
500 Index between 1990 and 2000. The expansion between
2001 and 2007 was based on the recovery in the equities market and the ever-inflating real estate bubble. Figure 16 presents
the S&P/Case-Shiller Home Price Indices, and shows that in
the six years between 2000 and 2006 these indices doubled.4
The problem with asset inflationfueled expansions is that
when the music stopswhen asset prices stop rising or, even
worse, falleconomic activity suffers. It is thus worth taking a
closer look at the main asset markets in the United States.
Figure 16 also shows that the real estate market has recovered from the crisis of 2007. Both indicesespecially the
National Home Price Indexare very close to their precrisis
peaks. Is this rebound justified by the fundamentals of the US
economy? If someone believes that there was a real estate bubble in 2006, and given the weak economic performance and low
inflation of the last six years, then her answer would be no.
What about the equities market? Figure 15 shows that the
S&P 500 Index has made a remarkable recovery over the last
six years. Between 2009 and its peak last year the index
increased by 270 percent, and it is now at historically high levels. Given the performance of the US economy over the same
period, it is hard to justify this increase.

Note: The index is calculated as the ratio of the end-of-period Wilshire 5000
index to nominal GDP.
Source: Federal Reserve

This becomes clear with two other indices, shown in

Figure 17, which normalize stock market prices to the earnings of firms and GDP. In the upper panel (Figure 17a) we
present the Shiller cyclically adjusted price-to-earnings ratio.
The index shows that, adjusted for earnings, the valuation of
the stock market is at precrisis levels, albeit lower than the levels it reached in the late 1990s. According to the other measure, which normalizes the market capitalization to GDP
(Figure 17b), we are now above the levels of the late 1990s.
These conclusions do not change even if we take into account

Levy Economics Institute of Bard College

11

The Foreign Sector

In our last policy report, issued in the spring of last year
(Papadimitriou et al. 2015), we stressed that one of the main
factors that exert negative pressure on the US economy is
weak foreign demand due to (1) the appreciation of the dollar
and (2) the weak demand for US exports due to the slowdown
in the economies of US trading partners.

To get an idea of the intensity of this pressure, we simulated a scenario where the growth rate of US trading partners
would be one percentage point lower than the projections in
the October 2014 World Economic Outlook (WEO) issued by
the International Monetary Fund (IMF 2014) and the exchange
rate would further appreciate 25 percent over the following
four years (slightly more than 6 percent each year). The result
of these simulations was an annual decrease in the growth rate
of slightly more than 1 percent compared to the baseline
scenario.
Figure 18a presents the difference between the IMFs
October 2014 projections for 2015 and the related estimates
from the January 2016 update to the WEO for the major
groups of countries. As one can see, world output growth was
0.8 percent lower than forecast one year ago. In general, the
IMFs projections proved overoptimistic, with the exception
of the eurozone (which was already in stagnation) and the
emerging and developing European economies.
Figure 18b depicts the same difference between the
October 2014 and more recent IMF projections for the major
US trading partners.5 The upward bias of the IMF projections
for last year is evident here, too. It is also significant that the
growth rate in Canada and Mexicothe United States top
two trading partnersslowed substantially. In total, the

Figure 18a Percent Difference between the October 2014

WEO Projections and January 2016 Update Estimates for the
Growth Rates of Selected Groups of Countries

Figure 18b Percent Difference between the October 2014

WEO Projections and January 2016 Update Estimates for the
Growth Rates of the Main US Trading Partners

1.0

0.5

-0.5

-1
Percent

-2

-1.5

-3

-2.0

-4

-2.5

-5

Sources: IMF; authors calculations

12

Strategic Analysis, March 2016

Italy

India

France

Taiwan

Belgium

Brazil

Netherlands

Foreign Demand

Sources: IMF; authors calculations

-6
South Korea

Emerging Emerging Emerging

Latin
Market and
and
and
America
Developing Developing Developing
and the
Economies
Asia
Europe
Caribbean

Germany

Eurozone

United Kingdom

Advanced
Economies

Japan

World

China

-3.0

Mexico

-1.0

Canada

Percent

the recent market correction (at the time of this writing, on

February 29, the S&P 500 Index was 5 percent below its peak
level two months before).
In conclusion, looking at the stock and real estate markets, we notice that they are at (or near) historically high levels. This recovery in asset prices has certainly helped the
economic recovery, which would have been much weaker had
it not been for asset inflation. However, current asset market
levels are hard to justify based on the fundamentals, and it
seems reasonable thatin the best-case scenariothe asset
price inflation of the last six years will slow down or stop. In
the worst case, the correction that began in January of this
year will continue. Given the reliance of the US economy on
asset prices, this will certainly have a serious negative impact
on macroeconomic performance.

growth rate of the export-weighted GDP of US trading partners was 0.9 percent lower than the IMF projections of
October 2014. At the same time, the broad trade-weighted US
dollar index appreciated by 9 percent between January and
December of 2015.
According to advance estimates from the Bureau of
Economic Analysis, the 2015 GDP growth rate (measured
from the fourth quarter of 2014 to the fourth quarter of 2015)
was 1.8 percent, down from 2.5 percent in the previous two
years and 1.1 percentage points lower than the rate projected
by the Congressional Budget Office last January (CBO 2015).
In light of our projections last year, this comes as no surprise.
Is the situation bound to improve in the near- or
medium-term future? The answer here is also, most probably,
no. The economic prospects for Canada, the largest importer
of US products, do not look encouraging. The drop in the
price of oil has put a huge strain on the Canadian economy;
in 2013, exports of energy products accounted for one-quarter of the countrys total exports. Moreover, the Canadian
economy is threatened by a high level of household debt,
which as of 2015Q3 had reached 165 percent of disposable
incomehigher than the precrisis-related ratio in the United
States. Finally, various measures show that the Canadian real
estate market is overheated.
Moving south of the US border, the situation in Mexico,
the second-largest importer of US goods, is not as dire,
although the performance of the growth rate in 2015 was also
one percentage point below the 3.5 percent forecast. This is
still the highest growth rate since 2012. However, the industrial production indexprobably the most important index
for the state of the Mexican economyshows signs of weakness. Mexicos economy is also vulnerable to the slowdown in
Canada, the emerging markets, and the United States.
The eurozone is the third-largest destination for US
exports. Figure 18a shows that, according to the IMF, the
growth rate for 2015 was slightly higher (0.15 percent) than
expectedbased, however, on already very low expectations
of only 1.35 percent. Besides these extremely low growth rates,
the dire state of the eurozone economies is exemplified by an
inflation rate of just 0.1 percent for 2015which turned negative (-0.2 percent) in February 2016and the European
Central Banks (ECB) negative base interest rate for overnight
deposits: the ECB charges banks 0.3 percent to hold their cash

overnight. Besides low growth, inflation and monetary policy

are also important for the US trade balance. Low inflation
abroad, all other things being equal, worsens the terms of
trade. Similarly, diverging monetary policy has contributed to
the recent appreciation of the dollar against the euro and
other currencies.
China accounts for almost 9 percent of US exportsa
significant level, but much lower than that of Canada, Mexico,
and the eurozone (the Federal Reserve export weights for
these three economies are 22.7 percent, 14.8 percent, and 17.5
percent, respectively). The Chinese economy is facing a transition from an export-led to a more domestic demandoriented growth model. The difficulties of this kind of transition
aside, China faces two other critical threats: first, a high level
of private debt; and second, the stock market plunge of the
last few months. Although Chinas share of US exports is not
as large as some people might assume, it remains very significant, and a slowdown in the Chinese economy will have a negative impact on overall US exports. A Chinese slowdown can
negatively impact US exports through a secondary channel as
well, by leading to a slowdown in other economies that trade
with the United States. Finally, the crisis in China will most
probably lead to a further depreciation of the yuan. Such
depreciation could have significant consequences for the US
external sector because, even though the export weight of the
Chinese economy is relatively low, the import weight is very
high, around 25 percent.
Japan completes the list of the top five US trading partners, with exports and imports weighing 4.7 percent and 7.1
percent, respectively. The Japanese economy continues to find
itself in a stagnant state; it did not grow at all in 2014 and
expanded by only 0.6 percent in 2015, while the Bank of Japan
(BoJ) recently joined the club of central banks that have
adopted a negative interest rate policy.
These five economiesCanada, Mexico, the eurozone,
China, and Japanaccount for around 70 percent of US
trade. Their weak economic performance is a general phenomenon and characterizes most global economies: Brazil is
heading for its worst recession in a long time, growth in the
UK and South Korea has slowed as well, and, finally, rounding
out the list of the United States main trading partners, Taiwan
barely grew in 2015.

Levy Economics Institute of Bard College

13

1.
2.

3.

Through a decrease in the foreign demand for US exports

because of lower GDP growth abroad.
Through the worsening of the US terms of trade, to the
extent that lower GDP is accompanied by (or leads to)
lower inflation abroad.
Through an appreciation of the dollar, due to either
diverging monetary policy (e.g., between the Fed, the
ECB, and the BoJ) or a reversal of the capital outflows of
previous years, which kept the dollar at low levels.

Furthermore, there may be feedback effects among these.

For example, there are many developing countries with a large
stock of debt denominated in US dollars. As the dollar appreciates, servicing this debt becomes more expensive, setting the
stage for a crisis in these countries that then leads to further
weakening of aggregate demand for US products.6

Baseline Scenario: The Unstable Economy

A more precise evaluation of the prospects for the US economy can be performed with the help of the Levy Institutes
macroeconomic model. As usual, we start with a baseline scenario and then build other scenarios upon it.
The baseline scenario builds on the projections of the
Congressional Budget Office (CBO) for the US economy and
examines both the prerequisites for and the implications of
the recent projections in The Budget and Economic Outlook:
20162026 (CBO 2016). More precisely, the CBO projects that
the growth rate of real GDP will be 2.7 percent in 2016 and 2.5
percent in 2017, dropping to 2 percent by 2020. These numbers are significantly below the CBOs January 2015 projections and even farther below those made in 2014 (CBO 2014a,
2014b, 2015), testifying to the significant downward pressures
that the US economy is subject to.
According to the CBO, the main contributor to growth
will be private consumption expenditure (1.8 percent and 1.9
percent of the total 2.7 percent and 2.5 percent, respectively, in
the next two years), followed by business investment (0.6 percent and 0.5 percent), residential investment (0.4 percent in
both years), and a small contribution by the government (0.2

percent). Net exports are thus projected to have a negative

contribution to growth (-0.3 percent and -0.5 percent).
At the same time, the CBO is projecting that the primary
government deficit as a percentage of GDP will remain stable
over the next five years. It also projects an increase in the overall deficit of roughly one percentage point over the same
period, as a result of the increase in government interest payments due to the increase in US Treasury debt yields. For our
simulations, we assume away this increase because (1) it is, in
our opinion, unlikely that yields will increase, and (2) to the
extent that this might happen, it is likely that there would be
further government spending cuts at the state and local level.
In other words, our baseline scenario examines the conditions necessary for the general government deficit projected
by the CBO to remain stable as a percentage of GDP, and for
the US economy to grow with the pace projected by the CBO.
For our simulations we make assumptions that are as
neutral as possible. We assume a mild increase in the price
level and a constant nominal exchange rate. The growth and
inflation rates of US trading partners follow the IMFs
October 2015 World Economic Outlook (IMF 2015) and its
recent January 2016 update (IMF 2016). Finally, we assume
that equity and real estate market prices will increase mildly
by 2 percent annuallyuntil 2020.

Figure 19 Baseline Scenario: Main Sector Balances, Actual

and Projected, 200520
15
10
Percent in GDP

This global slowdown can affect US trade in three ways:

5
0
-5
-10
-15

2005

2008

2011

Government Deficit
Private Sector Investment minus Saving
External Balance
Sources: BEA; authors calculations

14

Strategic Analysis, March 2016

2014

2017

2020

Figure 20a Private Gross-Debt-to-GDP Ratio, Actual and

Projected, 200020

Figure 20b Household-Debt-to-Disposable-Income Ratio,

Actual and Projected, 200020

1.80

1.15
1.10

1.70
1.05
1.00

1.60

0.95
1.50

0.90
0.85

1.40

Baseline
Scenario 1
Scenario 2
Scenario 3

2020

2016

2012

2008

2004

2020

2016

2012

2008

2004

2000

1.30

2000

0.80

Baseline
Scenario 1
Scenario 2
Scenario 3
Sources: BEA; Federal Reserve; authors calculations

Sources: BEA; Federal Reserve; authors calculations

The results of our baseline simulations are presented in

Figure 19. What we see is that the slowdown in the economies
of US trading partners, the appreciation of the dollar, and the
exhaustion of the economic benefits of the petroleum sector
lead to an increase in the current account deficit, which
reaches 6.3 percent in 2020half a percentage point above its
precrisis peak.
Given the stability of the government deficit, the increase
in the current account deficit is mirrored by an equivalent
decrease in the private sector balance. This balance becomes
negative in 2018, reaching -1.5 percent in 2020around its
level in 1998 and 2005. If this occurs, private sector balance
sheets will deteriorate rapidly. Figures 20a and 20b show that
the private sector gross-debt-to-GDP ratio as well as the
household-debt-to-disposable-income ratio will increase rapidly and converge toward pre-2007 levels. It is also important
to keep in mind that the current, historically high level of
income inequality implies that this increase in household
indebtedness will once more fall disproportionately on households at the bottom of the income distribution.
Is this likely to happen? Probably not. But it is exactly this
improbability that makes the baseline scenario interesting. In

other words, our baseline simulations show that, given the

current configuration of the US economy and the performance of its trading partners (as projected by the IMF), future
growth will, once again, have to be fueled by a rapid increase
in private sector indebtedness. Even if this happens, the experience of 2001 and 2008 shows how it will inevitably end.

Other Scenarios: Destabilizing an Unstable

Economy
Our baseline scenario shows that under its current structural
characteristics the US economy is unstable, and that raising
the private sector debt-to-income ratio to pre-2007 levels is a
necessary requirement for achieving the growth rates projected by the CBO in the period 201620.
As we explained in the previous sections, this unstable
configuration is further threatened by a possible weakening of
economic activity abroad, further appreciation of the US dollar, and a drop in asset market prices that could also trigger a
new round of private sector deleveraging. We evaluate these
possibilities by simulating three additional scenarios.

Levy Economics Institute of Bard College

15

Scenario 1 employs the same assumptions as the baseline,

with two important differences: (1) we assume that over the
course of 201620 the nominal exchange rate of the dollar will
increase by an additional 20 percent, and (2) that the growth
and inflation rates of US trading partners will be 1 percent
lower than the IMF projections. As we explained aboveand
in more detail in last years Strategic Analysis (Papadimitriou
et al. 2015) the projections of the IMF tend to be overoptimistic, and the global economic situation signals that a further slowdown is possible.
The further appreciation of the dollar and, most important, the slowdown in economic activity abroad lead to a
decrease in foreign demand for US exports. The impact of
these factors on the economic performance of the US economy can be seen in Figure 21. The growth rate in scenario 1 is
significantly below the baseline growth rate: 1.6 percent in
2017 (as opposed to 2.5 percent in the baseline), decreasing to
0.85 percent by 2020 (slightly more than 1 percent below the
baseline). These differences are similar to the ones we found
in last years estimations.
Moreover, in Figure 22 we see that the combination of
lower foreign demand and deterioration in the terms of trade

has a significant negative impact on the current account

deficit, which reaches 8.7 percent of GDP in 2020, as opposed
to 6.3 percent in the baseline. The slowdown will trigger automatic stabilizers, and thus the government deficit will
increase, reaching 6.7 percent by 2020. These changes in the
current account balance and the government deficit mean
that the deficit of the private sector will also be higher. Figure
22 shows that by 2020 the private sector deficit converges to 2
percent, higher than the 1.5 percent in the baseline. Finally, as
we can observe in Figure 20, the debt-to-income ratios are
higher in scenario 1 compared to the baseline since the private
sector is assumed to accumulate debt as in the baseline, but
the economic activity slows down.
In scenario 2 the nominal exchange rate, growth rate, and
inflation rate of US trading partners revert to their baseline
behavior. In this scenario it is assumed that the stock market
continues to fall throughout 2016 and then stabilizes for the
rest for the projection period. More precisely, the S&P 500
Index falls to around 1450 by the end of the year (comparable
to precrisis levels in 2000 and 2007).
In addition, scenario 2 assumes that the private sector
partly induced by the drop in the stock marketmoves at the

Figure 21 Real GDP Growth Rate, Actual and Projected,

201020

Figure 22 Scenario 1: US Main Sector Balances, Actual and

Projected, 200520
15

3.0

10

2.0
Percent of GDP

Annual Growth Rate (in percent)

2.5

1.5
1.0
0.5
0

0
-5
-10

-0.5
-15

-1.0

Baseline
Scenario 1
Scenario 2
Scenario 3
Sources: BEA; authors calculations

Strategic Analysis, March 2016

2020

2018

2016

2014

2012

2010

-1.5

16

2005

2008

2011

Government Deficit
Private Sector Investment minus Saving
External Balance
Sources: BEA; authors calculations

2014

2017

2020

Figure 24 Scenario 3: Main Sector Balances, Actual and

Projected, 200520

15

15

10

10
Percent of GDP

Percent of GDP

Figure 23 Scenario 2: Main Sector Balances, Actual and

Projected, 200520

5
0
-5
-10
-15

5
0
-5
-10

2005

2008

2011

2014

2017

2020

Government Deficit
Private Sector Investment minus Saving
External Balance

-15

2005

2008

2011

2014

2017

2020

Government Deficit
Private Sector Investment minus Saving
External Balance

Sources: BEA; authors calculations

Sources: BEA; authors calculations

end of 2016 toward a second round of deleveraging. As shown

in Figure 20, the pace of the deleveraging is relatively slow and
the debt-to-income ratios fall by 2020 to their early 2000s levels, which were already high by historical standards.7
The effect of the stock market slide andmore importantthe deleveraging of the private sector is shown in Figure
21. The growth rate falls in 2017 to below 0.4 percent and
remains at that level for the rest of the projection period.
Figure 23 shows that the lower growth rate leads to a better
current account balance compared to the baseline. On the
other hand, deleveraging increases the private sector balance,
which reaches 5.8 percent by the end of the projection period.
The fall in the growth rate leads to an increase in the government deficit, which reaches 7.8 percent in 2020, up from 5
percent in the baseline scenario.
Finally, in scenario 3 we assume that there is a combination of the negative factors in scenarios 1 and 2. The stock
market fall and private sector deleveraging are accompanied
by weaker growth abroad and an appreciation of the dollar
unfortunately, not a far-fetched scenario. Figure 21 shows that
this vicious alignment of adverse factors leads the growth rate
into negative territory, around -0.7 percent in 2017 and -0.9
percent by the end of the projection period. In terms of the
main sector balances, Figure 24 shows that the current
account deficit increases only slightly, since the negative

effects stemming from scenario 1 are counteracted by the

lower demand for imports due to the lower growth rates. The
private sector balance follows a trajectory similar to that in
scenario 2, reaching 5.3 percent in 2020. As one would expect,
the collapse in growth leads to an increase in the government
deficit, which reaches 9.8 percent in 2020.

Conclusion
The weakness in many economies around the world and the
turbulence in financial markets have induced many commentators to become cautious and warn about the possible risks
these developments might have for the United States and
global economies. We definitely agree with this position and
have warned about these possible destabilizing factors in our
previous reports.
However, as we explained above, it would be unwise to
conclude that an otherwise robust and stable US economy is
threatened solely by some exogenous shocks. The economys
instability is primarily structural, and is related to three main
problems: (1) high income inequality, (2) high external
deficits, and (3) the fiscal conservatism thatto paraphrase
Keyneshas conquered Washington as completely as the
Holy Inquisition conquered Spain.

Levy Economics Institute of Bard College

17

These fundamental structural characteristics make economic growth in the United States dependent on increasing
indebtedness and asset market inflation, which as the recent
experience has shown is a highly unstable process. It is this
unstable configuration that is now being further destabilized
by the weakening of foreign demand and the turbulence in
asset markets.
Therefore, achieving sustainable economic growth in the
United States requires, first and foremost, addressing these
fundamental issues: a decrease in income inequality, international cooperation to rebalance the global economy and
improve the US external position, and relaxation of the governments fiscal stance. The alternative is a future of secular
stagnation or debt-driven recoveries that will result in
increasingly severe financial and economic crises.

Notes
1. For our recent discussions of secular stagnation, see
Papadimitriou et al. (2014, 2015) and Nikiforos (2015).
Two of the most well-known proponents of secular stagnation theory are Larry Summers (2014) and Paul
Krugman (2014). A summary of the recent debates is
provided in Teulings and Baldwin (2014).
2. We have omitted the short cycle of the early 80s
(1980Q31981Q3).
3. Wolff (2012) provides a detailed analysis.
4. To be precise, the 20-City Index increased by 106 percent
and the National Index by 84 percent in the period
between 2000 and the first months of 2006.
5. The recent estimates can be found in the October and
January updates of the WEO (IMF 2015, 2016).
6. For a discussion of the links between US monetary policy,
the carry trade, and debt in emerging markets, see Bruno
and Shin (2015) and McCauley, McGuire, and Sushko
(2015).
7. For an examination of the debt levels at that time, see
Godley (1999).

18

Strategic Analysis, March 2016

References
Alvaredo, F., A. B. Atkinson, T. Piketty, and E. Saez. 2016. The
World Wealth and Income Database. Accessed January
25, 2016. www.wid.world/#Database.
Berube, A., and N. Holmes. 2016. City and Metropolitan
Inequality on the Rise, Driven by Declining Incomes.
Washington, D.C.: Brookings Institution.
www.brookings.edu/research/papers/2016/01/
14-income-inequality-cities-update-berube-holmes.
Bruno, V., and H. S. Shin. 2015. Global Dollar Credit and
Carry Trades: A Firm-Level Analysis. BIS Working
Papers No. 510. Basel: Bank for International
Settlements.
CBO (Congressional Budget Office). 2014a. The Budget and
Economic Outlook: 2014 to 2024. Washington, D.C.:
CBO. February 4.
. 2014b. An Update to The Budget and Economic
Outlook: 2014 to 2024. Washington, D.C.: CBO. August 27.
. 2015. The Budget and Economic Outlook: 2015 to
2025. Washington, D.C.: CBO. January 26.
. 2016. The Budget and Economic Outlook: 2016 to
2026. Washington, D.C.: CBO. January 25.
Godley, W. 1999. Seven Unsustainable Processes: Medium
Term Prospects and Policies for the United States and the
World. Strategic Analysis. Annandale-on-Hudson, N.Y.:
Levy Economics Institute of Bard College. Revised
October 2000.
IMF (International Monetary Fund). 2014. World Economic
Outlook: Legacies, Clouds, Uncertainties. Washington,
D.C.: IMF. October.
. 2015. World Economic Outlook: Adjusting to Lower
Commodity Prices. Washington, D.C.: IMF. October.
. 2016. World Economic Outlook Update: Subdued
Demand, Diminished Prospects. Washington, D.C.: IMF.
January.
Krugman, P. 2014. Four Observations on Secular
Stagnation. In C. Teulings and R. Baldwin, eds. Secular
Stagnation: Facts, Causes and Cures. London: CEPR Press.
Lucas, R. E. Jr. 2004. The Industrial Revolution: Past and
Future2003 Annual Report Essay. Federal Reserve Bank
of Minneapolis.
Mazzucato, M. (2013). The Entrepreneurial State: Debunking
Public vs. Private Sector Myths. London: Anthem Press.

McCauley, R. N., P. McGuire, and V. Sushko. 2015. Global

Dollar Credit: Links to US Monetary Policy and
Leverage. Economic Policy 30, no. 82 (April): 187229.
Nikiforos, M. 2015. A Nonbehavioral Theory of Saving.
Working Paper No. 844. Annandale-on-Hudson, N.Y.:
Levy Economics Institute of Bard College. July.
Papadimitriou, D. B., G. Hannsgen, M. Nikiforos, and G.
Zezza. 2015. Fiscal Austerity, Dollar Appreciation, and
Maldistribution Will Derail the US Economy. Strategic
Analysis. Annandale-on-Hudson, N.Y.: Levy Economics
Institute of Bard College. May.
Papadimitriou, D. B., M. Nikiforos, G. Zezza, and G.
Hannsgen. 2014. Is Rising Inequality a Hindrance to the
US Economic Recovery? Strategic Analysis. Annandaleon-Hudson, N.Y.: Levy Economics Institute of Bard
College. April.
Rios-Avila, F., T. Masterson, and A. Zacharias. Forthcoming.
Trends in American Living Standards and Inequality,
19592013. LIMEW Report. Annandale-on-Hudson,
N.Y.: Levy Economics Institute of Bard College.
Summers, L. H. 2014. U.S. Economic Prospects: Secular
Stagnation, Hysteresis, and the Zero Lower Bound.
Business Economics, 49, no. 2 (April): 6573.
Teulings, C., and R. Baldwin, eds. 2014. Secular Stagnation:
Facts, Causes, and Cures. London: CEPR Press.
Wolff, E. N. 2012. The Asset Price Meltdown and the Wealth
of the Middle Class. NBER Working Paper No. 18559.
Cambridge: National Bureau of Economic Research.

Levy Economics Institute of Bard College

19

Blithewood
PO Box 5000
Annandale-on-Hudson, NY 12504-5000
Address Service Requested

Nonprofit Organization
U.S. Postage Paid
Bard College