(OPE-L) Who pays for the "welfare" in the welfare state? A multicountry study. Anwar Shaikh.

From: Rakesh Bhandari (rakeshb@STANFORD.EDU)
Date: Sun Dec 14 2003 - 00:15:45 EST

Social Research, Summer 2003 v70 i2 p531(21)
Who pays for the "welfare" in the welfare state? A multicountry
study. Anwar Shaikh.

Full Text: COPYRIGHT 2003 New School for Social Research

I. Introduction

THERE was a time, not so long ago, when the welfare state was viewed
as a proud social accomplishment. But recently it has been under
considerable attack. At the heart of this attack has been the claim
that during its heyday, from the 1950s to the 1970s, the social
benefit expenditures of the welfare state led to subsequent economic
stagnation and persistent unemployment throughout the advanced world.
This paper brings to bear the empirical evidence from a multicountry
set of studies. I outline the issues involved, discuss the
methodology behind the empirical studies of six major Organization
for Economic Cooperation and Development (OECD) countries (Australia,
Canada, Germany, Sweden, the United Kingdom, and the United States),
and present the main findings. My central finding is that social
benefit expenditures were financed out of the taxes paid by
recipients of these very expenditures: in other words, by and large,
social welfare expenditures were self-financed, and could not have
been a source of fiscal deficits or a drag on growth.

1. The Rise and Fall of the Welfare State

The growth of welfare states is one of the characteristic features of
modern capitalist democracies. European welfare states began with
pension and social insurance programs in the late nineteenth and
early twentieth centuries and then grew into comprehensive systems of
social support between the 1930s and the 1950s. In the United States,
it took the Great Depression to spark similar initiatives in the form
of New Deal programs on social security, state-based unemployment
insurance, and limited federally subsidized public assistance (which
Americans call "welfare") for the elderly poor, dependent children,
and the blind.

After World War II, the role of the state expanded rapidly. From 1960
to 1988, in the OECD countries the average government share in gross
domestic product (GDP) rose by over one-half (from 27 percent to 42
percent), while the average government share in total employment rose
by about two-thirds (from 11 percent to 18 percent). Alongside this
came a shift in the types of government spending, away from
traditional expenditures on defense, public administration, and
general economic services and toward social welfare expenditures on
health, education, and transfer payments (social security and social
assistance payments, business subsidies, and interest on government
debt). By the 1980s, transfer payments had become the single largest
category of economic expenditure in most countries (OECD, 1985: 16).

But the rise in government expenditure was only one side of the
story. Taxes also rose sharply, and their composition shifted from
traditional sources, such as indirect business taxes, to social
security and personal income taxes (OECD, 1985: 16-17). Thus on the
whole, both government expenditures and the tax structure changed in
very similar ways.

The early part of the postwar period was the zenith of the welfare
state as the industrialized world grew at a rate of almost 5 percent
annually. However, by the mid-1970s the long underlying expansion had
peaked and by the late-1970s the average growth rate of the
industrialized world had fallen to half its previous level. By 1983,
the OECD countries as a whole were barely growing (OECD, 1991). In
this period of growth slowdown and eventual stagnation came rising
unemployment and poverty, which led to greater demands on social

In the United States, the postwar boom peaked in 1968-1969. The
economy moved into a phase of (initially inflationary) stagnation. A
major change took place in all major economic patterns at this point.
In the boom decades from 1947 to 1968, growth was strong,
unemployment averaged 4.8 percent, real wages grew almost 50 percent,
and the average annual federal budget deficit was a mere $1.7
billion. (1) In the subsequent two decades between 1969 and 1989,
unemployment rose to an average of 6.6 percent, real wages declined
by 14 percent, and the average budget deficit rose almost fiftyfold
to $82.4 billion (ERP, 1996). By 1980, eligibility for public
assistance had been restricted, and for those who did receive aid,
real benefits were 20 percent lower than what they had been in 1970.

 From 1980 to 1988, the Reagan administration carried out a
far-reaching and systematic policy of attacking workers and the poor.
It undermined unions and cut back on the level and duration of
unemployment benefits. Union membership declined rapidly during this
period, from about a quarter of the labor force to less than a sixth.
Real wages fell and worker concessions and givebacks became
commonplace. The number of people in low-wage jobs rose sharply: in
1970 only 20 percent of workers earned a real income of less than
$7,000 (in 1984 dollars); between 1979 and 1984 some 60 percent of
new jobs were in this category (Rosenberg, 1987). Military spending
increased dramatically even as social spending was slashed, so that
budget deficits continued to rise to new highs. Throughout all of
this, the rhetoric of the period was dominated by the notion of "tax
relief" for an overburdened population (a most familiar phrase once
again). Corporations did receive substantial tax relief, which added
to the benefits of a declining real wage. But as we shall see, for
workers the situation was different, since their tax rate continued
to rise (see figure 1 on page 1211).

The phase change outlined for the United States economy mirrored a
pattern evident across all advanced capitalist countries as the
decades of high growth gave way to decades of sluggish growth,
inflation, rising unemployment, and an attendant "fiscal crisis of
the welfare state" (Skocpol, 1987: 36). The European and United
States growth rates of real GDP are depicted in table 1, in which the
"oil shock" period from 1973 to 1974 is excluded to avoid biasing the
comparison across periods. The corresponding patterns for
unemployment rates are depicted in table 2.

Full Size PictureIn the light of these events, which were unfolding
across the advanced world, it is not surprising that there arose a
claim that the welfare state was to blame. Mainstream economists in
particular argued that its "social policy [led to] ... an
over-expansion of the government, which [was] ... a principal cause
of the economic slowdown and rising inflation of the 1970s (Buchanan
and Flowers, 1980, chap. 6, cited in Fazeli, 1996: 37). Thus the
welfare state was "a drag on economic activity and [had] ... reduced
economic performance" (Moudud and Zacharias, 2000: 7). Cutbacks in
social expenditures, particularly unemployment insurance and income
support for the poor and the elderly, were deemed necessary to
restore growth and reduce unemployment (Atkinson, 1999). And indeed,
such cutbacks began to spread across the developed world.

Full Size PictureMany different mechanisms have been proposed as the
sources of the putative negative effects of the welfare state. Within
orthodox economics, two are particularly important. The first of
these arises from the claim that the welfare state gives rise to
budget deficits in order to finance its social expenditures. These
deficits are said to stimulate consumption and reduce the household
savings rate, which in turn lowers the long-term growth rate of the
economy. A similar claim also arose on the left, in the form of the
"fiscal crisis of the state" identified by O'Connor (1973), and in
the claim by Bowles and Gintis (1982) that over the postwar period
the state induced a "substantial redistribution from capital to
labor," resulting in a "citizen wage" that grew so rapidly that by
the 1970s it played "a critical role in producing and prolonging" the
economic crisis of the 1970s and 1980s (Bowles and Gintis, 1982: 69,
84-85). In all of these arguments, the welfare state tends to reduce
and undermine growth over the long run, which at some point gives
rise to stagnation and unemployment (Fazeli, 1996, chap. 2; Moudud
and Zacharias, 2000: 8-14).

The second criticism is that the welfare state perpetuates the very
unemployment it creates. Orthodox economic theory says that
unemployment would be self-correcting as long as workers lowered
their real wages whenever there is any unemployment. But the various
social protection mechanisms of the welfare state are said to create
"labor market distortions" that interfere in this automatic
adjustment process. Unemployment insurance and income support reduces
the incentive of unemployed workers to accept lower wages and worse
working conditions, while payroll taxes and employee protections
reduce the wages that firms are willing to offer to workers. As
Krugman puts it, the relatively greater social protections in Europe
mean that "an unemployed European does not need to search for
employment with the desperation of his American counterpart"
(Krugman, 1994: 22). Thus the welfare state tends to prevent the
elimination of unemployment.

In the end we are told that the welfare state tends to undermine
growth and give rise to unemployment because it helps give rise to
budget deficits. At the same time it "distorts" labor markets by
making workers less desperate in the face of unemployment, which
tends to make unemployment persistent (Pear, 1995). In both cases,
the appropriate solution is to reduce the extent of the welfare state.

As far as the "labor distortions" argument is concerned, even Krugman
admits that this argument is unable to explain why the socially
conscious "European countries [were] able to achieve such low
unemployment rates before 1970" (23). Detailed micro- and
macroeconometric studies also indicate only a weak link between
benefits, labor market regulations, and unemployment (Bean, 1994:
594-595, 600-603). Most recently, a recent detailed cross-country
study finds no real empirical support for the view that the welfare
state's labor market institutions and policies played a key role in
the European unemployment crisis of the 1980s and 1990s (Baker, Glyn,
Howell, and Schmitt, 2002: 2-4, 54-57). Finally, it is a striking
fact that precisely during the period when the welfare state is
supposed to have most inhibited economic performance, from 1975 to
1993, per capita gross national product (GNP) in the OECD countries
increased by 314 percent, while per capita GNP in the United States
rose by only 234 percent. (2)

But might it not be still possible that in the heyday of the welfare
state, social spending, particularly social spending directed to
labor, was financed by government deficits, which in turn were large
enough to cause the subsequent slowdown of growth all over the
advanced world? We turn to this issue next.

2. Who Paid For Social Expenditures?

The notion that the welfare state was deficit financed relies on a
series of implicit claims. First, that the beneficiaries of the
social spending received more than they themselves paid in taxes.
Second, that this difference was largely financed by government
deficits, rather than transfers from other groups in society. And
third, that the resulting deficits were large enough to initiate a
slowdown in economic growth.

It is striking how little consistent empirical evidence has been
advanced for such claims. Individual country studies do not provide
much guidance for generalizations about the welfare state (Marmor,
1993). On the other hand, cross-national studies that analyze only
expenditures, (3) or only taxes, are equally inadequate because what
matters is the net balance between the two. For instance, if the
taxes paid by some group matched the social expenditures directed
toward it, then these social expenditures were self-financed, and
could not have been a source of fiscal deficits or a drag on growth.

The question thus becomes: who pays for the welfare spending of the
welfare state? In addressing this issue, we focus on the relations
between the state and labor (defined here as wage and salary earners,
excluding top management such as CEOs). This focus arises out of the
claim that it was the social support of labor in particular that
eventually undermined the welfare state. (4)

3. Our Principal Findings

Our findings are based on a series of cross-national studies of the
welfare state, conducted over the 1980s and 1990s. The framework
used, which is outlined in section two, was originally applied to the
United States (Shaikh and Tonak, 1987, 1994, 2000), and subsequently
to Australia, Canada, Germany, Sweden, and the United Kingdom, over
various intervals (Tonak, 1984; Bakker, 1986; McGill, 1989; Fazeli,
1992, 1996; Maniatis 1992).

The principal finding of these studies is that the taxes paid by wage
and salary earners closely parallel the social expenditures directed
toward them: for the estimated average of the advanced countries
between 1960 and 1987, the difference between the value of total
social benefits received and total taxes directly paid (the net
social wage) remains between 1 and 2 percent of GDP (3 to 5 percent
of total wages and salaries) in almost every year. Since this is
positive, it implies that wage and salary earners received more than
they paid. But this overall net subsidy is clearly small. And as we
shall see, while it is generally positive in the five OECD countries
studied, it is generally negative (that is, a net tax) in the United

The small size of this average net social wage ratio (net social wage
relative to GDP) does not support the claim that social benefit
expenditures hurt economic growth in the advanced world. Indeed, what
it instead shows is that the principle effect of net transfer flows
is to recirculate income among wage and salary earners as a whole.
But even here, the evidence indicates that any such intraclass
redistributive effect is quite limited. Detailed studies across
household income classes in various OECD countries seem to indicate
that redistributive effects concentrate on the lowest and highest
income ranges, so that the bulk of the income ranges are not much
affected by the net government intervention (OECD, 1985, chap. 7,
sec. B).

There are, of course, differences among countries, but even these are
not necessarily what one might have expected. For instance, in the
boom years the net social wage was negative in the United States,
which meant that wage and salary earner paid out more in taxes than
they received--they helped reduce any existing fiscal deficit. Over
the same interval in Sweden, the net social wage was roughly zero,
indicating that its generous social welfare expenditures were
actually self-financed. In neither the United States nor Sweden,
therefore, can welfare expenditures be indicted as the cause of
fiscal deficits or consequent economic stagnation. However, in
Germany the net social wage was generally positive in the boom years,
on the order of 4 percent of GDP. And even this modest proportion is
one of the highest in our sample. Here at least we can say that the
net social wage did have a substantial impact on government finances:
from 1950 to 1973, it accounted for roughly 42 percent of the
government deficit, which was itself about 7 percent of GDE Yet
Germany's growth rate was higher than that of either Sweden or the
United States in every subperiod (OECD, 1991). In fact, in every
period there is a positive correlation between the size of the net
social wage and economic growth: Germany has the highest growth rate,
Sweden is in the middle, and the United States has the lowest.

II Methodology of the Net Social Wage

Our concerns require us to identify the different parts of total
taxes that flow out of the aggregate wage bill, and the various
social benefits that flow back to the working population through the
medium of government expenditures; this will allow us to assess their
net balance (the net social wage). The category of "working
population" is not defined here in sociological or political terms,
but rather in terms of all those who earn wages and salaries, with
the exception of CEOs and other top management. This is in part
because wages play a central role in most economic processes, but
also because it allows us to get consistent results across countries.
We use national economic accounts and other sources to estimate and
track the net social wage in six advanced countries: the United
States, Canada, the United Kingdom, Germany, Sweden, and Australia.

The net social wage consists of the difference between social
expenditures directed toward the working population ("labor") and the
taxes directly levied on this same group. On the side of social
expenditures, we count all welfare expenditures (health, education,
welfare, housing, transportation, parks and recreation, transfer
payments to workers, etc.). We divide such expenditures into two
subsets. Expenditure Group I, which we assume is entirely received by
workers, and consists of items such as Labor Training and Services,
Housing and Community services, and Income Support, Social Security,
and Welfare (except the small items called Military Disability and
Military Retirement, which we treat as a cost of war). And
Expenditure Group II, which is directed toward workers and nonworkers
alike and comprises items such as Education, Health and Hospitals,
Recreational and Cultural Activities, Energy, Natural Resources,
Passenger Transportation, and Postal Services. The workers' share in
this is estimated by multiplying the group total by the labor share
in personal income (Shaikh and Tonak, 1987, 1994).

Full Size PictureOn the side of taxes levied directly on the working
population are income taxes, social security taxes, property and
other taxes. The primary monetary flow from which the taxes are
deducted is total Employee Compensation, which is the total cost of
workers to their employers. We consider this to be the gross wage of
workers, comprising wages and salaries as well as benefits such as
Employers Contributions for Social Insurance and Other Labor Income.
The total taxes to be considered then fall into two groups: Tax Group
I, which comes entirely out of gross wages, and consists of all
deductions for social security (that is, the sum of Employee and
Employer Contributions), and the more general Tax Group II, which
consists of Personal Income Taxes, Motor Vehicle Licenses, Personal
Property Taxes (primarily on homes), and Other Taxes and Non-taxes (a
very small category that includes passport fees and fines), and is
allocated between labor and nonlabor using the share of labor income
in personal income.

Comparing the expenditures on labor and the taxes paid by labor then
gives us the net social wage, defined as benefits received by workers
minus taxes paid (see Shaikh and Tonak, 1987, 1994, 2000, for further
details). The net social wage ratio then refers to its size relative
to GDP.

III. Empirical Results

1. The United States

Full Size PictureWe begin with our results for the United States,
with data taken from Shaikh and Tonak (2000). Figure 1 depicts the
benefit rate of labor (total social expenditures directed towards
labor relative to GDP) and the tax rate of labor (total labor taxes
relative to GDP). (5) It is clear that both rise sharply over the
postwar period. But what is particularly striking is that in the
United States, during the long boom from the end of World War II to
the late 1960s, the benefit rate is always below the tax rate. Thus
during the long boom the net social wage is negative that is, it is a
net tax on labor. Rather than dragging down the rest of the economy
in this interval, United States workers were subsidizing it.

It is only after the boom runs out in the early 1970s and the
unemployment rate rises sharply that the benefit rate overtakes the
tax rate and the net social wage become positive. This is simply
because increased numbers of unemployed and impoverished people
become eligible for payments, while at the same time their decreased
incomes reduce the taxes they pay. This same effect also raises the
government deficit. Thus a positive net social wage becomes
associated with a rise in the government deficit during the growth
slowdown, giving rise to the mistaken impression that the observed
correlation between the two was causal.

Full Size PictureBut the correlation did serve an important
ideological purpose, because it became fodder for the attack on the
welfare state. By the 1980s, beginning with Reagan and continuing
after his administration, this assault succeeded in dismantling the
safety net and sharply reducing the strength of workers'
organizations (Amott, 1987: 51). Even the sharp rise in the
unemployment rate at the beginning of the Reagan-Bush era (see table
2) barely changes the benefit rate because benefit and eligibility
cutbacks compensate for the greater numbers of unemployed. A second
surge of unemployment at the end of this era raises the benefit rate
once again because total benefits accelerate in the face of greater
unemployment while total employee compensation slows down for the
same reason and their ratio actually rise. Under the Clinton
administration it remains stable, but the tax rate rises, so that the
gap closes once more.

Figure 2 shows these same movements from the point of view of the
United States net social wage ratio (net social wage as a percentage
of GDP), which is simply the difference between the benefit and tax
rates depicted earlier. Over the entire postwar period, from 1952 to
1997, the average net social wage ratio in the United States is a
mere -0.33 percent. In effect, workers paid for their own social
benefits. (6)

2. Five OECD Countries

A similar analysis was undertaken for the other five countries in
this study (Canada, Germany, the United Kingdom, Australia, and
Sweden). Because of data limitations, starting dates for the series
range from the early 1950s for the first three countries, to the
early 1960s for the last two. Ending dates also vary, since the
country studies were undertaken by various people over different
interval in the 1980s and early 1990s. For this reason, it was only
possible to create comparable data for all countries for the period
from 1960 to 1987. (7) For our purposes this is sufficient to address
the question of whether or not, in Europe at least, the welfare state
might have caused the slowdown of growth that began in the 1970s.

Full Size PictureFigure 3 depicts the net social wage ratio in each
of the five OECD countries. Several concerns are immediately
apparent. First, unlike the United States, the other OECD countries
had generally positive net social wage ratios. Germany and the United
Kingdom had the highest ratios, though even they averaged only about
5 percent of GDP and about 8 percent of total wages. Second, Sweden,
the very paradigm of a welfare state, had an average net social wage
around zero over the boom period, and it climbed above zero only in
response to the stagnation in growth.

A third finding is that from 1960 to 1972, prior to the growth
slowdown in the mid-1970s, the net social wage ratio ranged between 0
and -2 percent in the United States (figure 2), and between 2 percent
and 4 percent in the other five countries combined (figure 4). (8)
Both these ratios rose substantially during the subsequent period of
stagnation, but this was an effect of the stagnation itself. Since
government deficits also rose for the same reasons, the rise in the
net social wage is correlated with a rise in government deficits.
This then gives rise to the mistaken belief that a rise in the social
wage was fueled by a rise in budget deficits, which then caused the
general decline in growth.

3. United States and Five OECD Countries Taken Together

Full Size PictureFigure 5 depicts the estimated net social wage ratio
of the advanced countries as a whole. To calculate this, the average
net social wage ratio of the five OECD countries (Australia, Canada,
Germany, Sweden, and the United Kingdom) was treated as a
representative sample of that in Europe. It was then averaged with
the United States ratio, using the respective shares of Europe's and
America's GDP in the sum of their GDPs, all expressed in United
States dollars (OECD, 1991). We see that for the entire group, the
net social wage ratio is roughly 1 percent for the boom period. It
really rises above that, to around 3 percent, in the period of
stagnation beginning in the early-1970s. Comparison of figure 3 and
figure 5 makes it clear that the average pattern is very similar to
that of Sweden. The Swedish welfare state, it seems, turns out to be
paradigmatic in more ways than one.

IV. Summary and Conclusions

After the great postwar boom in the advanced world gave way to
stagnation and inflation in the 1970s and 1980s, it became a
commonplace on both the left and the right to attribute the
disappearance of growth to the overexpansion of the state. It was
claimed that that the welfare state gave rise to budget deficits in
order to finance its social expenditures, and that these deficits in
turned reduced growth and spurred inflation. Most often, these claims
were supported by pointing to the fact that social expenditures had
risen as a share of GDP.

But a reference to the rising share of social expenditures is
inadequate, because the share of taxes in GDP also rose equally
rapidly. And when examined from this perspective, we find that the
benefit and tax rates closely parallel each other in all six advanced
countries studied here (Australia, Canada, Germany, Sweden, the
United Kingdom, and the United States). The difference between the
benefit and tax ratios, the net social wage ratio, is generally quite

In the United States from 1952 to 1997, the average net social wage
ratio in the United States is a mere -0.33 percent. Thus during the
entire postwar period, United States labor paid for its own social
benefits. Indeed, in the boom years of the postwar period, from 1950
to 1972, the United States net social wage was negative, which meant
that wage and salary earners paid out more in taxes than they
received. Rather than dragging down the rest of the economy in this
interval, United States workers actually subsidized it. It is only
after the unemployment rate rises when the boom runs out in the early
1970s that the net social wage ratio turns positive. But this is
because increased numbers of the unemployed and the poor led to
increased benefit payments, while at the same time their decreased
incomes reduced the taxes they pay. This same effect also raised the
government deficit. Thus a positive net social wage became associated
with a rise in the government deficit once the growth slowdown had
begun, giving rise to the mistaken impression that the observed
correlation between the two was causal.

In the other five OECD countries studied, over the long boom the
average net social wage was about 3.5 percent of GDP. Here too the
subsequent slowdown in growth raised the net social wage ratio to
about 5 percent, just as it raised the budget deficit.

Yet in all cases these correlations are a consequence of the
slowdowns in growth. On the deficit-as-drag argument, one would have
to at least demonstrate that countries with higher positive net
social wage ratios in the prior boom period have subsequently lower
growth rates, and that countries with negative net social wage ratios
have subsequently higher growth rates. But such a correlation does
not hold. We find that net social wage ratios vary considerably
across countries during the long boom, ranging from 5.5 percent in
the United Kingdom, 4 percent in Germany, about 2.5 percent in Canada
and Australia, a mere 0.4 percent in Sweden, to -1.2 percent in the
United States. Yet all of these countries--even the United States
with its negative net social wage--suffered a substantial slowdown in
growth over the next two decades. The putative causal link between
social expenditures and slowed growth simply does not hold up.


                  1961-1972    1975-1982    1983-1991

Europe             4.68%        2.23%        2.66%
United States      3.83%        1.97%        2.85%


                  1961-1972    1975-1982    1983-1991

Europe             2.59%        5.41%        9.51%
United States      4.94%        7.01%        6.74%


(1) Real wages are average annual earnings of nonagricultural workers
deflated by the consumer price index, both series taken from ERP
(1996), tables B-42 and B-56, respectively.

(2) The data are in United States dollars, Atlas methodology, from
the "World Tables" of the World Bank.

(3) For instance, Pampel and Williamson (1989) place a great deal of
emphasis on the demands placed on social expenditures by interest
groups such as the middle classes ("contradictory class locations"),
ascriptive groups (gender, race, ethnicity based), and the aged
(xiii-xiv), yet they seldom mention taxes at all. They thus end up
giving the false impression that social expenditures account for what
they call the "crisis" of the welfare state in the 1970s and 1980s

(4) This claim is explicit in Bowles and Gintis (1982). But this is
disputed by Shaikh and Tonak (1987, 1994, 2000) and Miller (1988,

(5) Since we are concerned with their impact on the economy as a
whole, we measure benefit and tax rates relative to GDP. But if we
are interested in their impact on the standard of living of workers,
then we should scale them relative to the gross income of labor--that
is, relative to employee compensation, as in Shaikh and Tonak (1987).

(6) In terms of gross labor income, the net social wage in the United
States from 1952 to 1997 averaged a mere -0.60 percent (Shaikh and
Tonak, 2000: 252). This is not only tiny, but also negative (i.e., a
net tax on labor).

(7) It was necessary to extrapolate benefit and tax ratios for
certain initial and final years (Australia, 1960, 1984-1987; Canada,
1987; Germany, 1987; Sweden, 1986-1987; United Kingdom, 1987).

(8) The combined ratio was calculated by converting all the net
social wages in each country in each year into United States dollars
through each year's exchange rate, summing them, and then dividing
them by the similarly converted total GDP of the five countries. This
is equivalent to taking a weighted average of the individual country
net social wage ratios, using the share of each country's converted
GDP in total converted GDP.


Amott, T. L. "Welfare Reform: A Workhouse without Walls." The
Imperiled Economy: Book II. New York: URPE, 1987.

Atkinson, A. B. The Economic Consequences of Rolling Back the Welfare
State. Cambridge: MIT Press, 1999.

Baker, Dean, and Andrew Glyn, David Howell, and John Schmitt. "Labor
Market Institutions and Unemployment: A Critical Assessment of the
Cross-Country Evidence." Center of Economic Policy Analysis, New
School University. Working Paper 2002-17. November 8, 2002.

Bakker, I. "The Reproduction of the Working Population in Canada."
Diss. The New School for Social Research, 1986.

Bean, C. R. "European Unemployment: A Survey." Journal of Economic
Literature 32 (June 1994): 573-619.

Bowles, S. and H. Gintis. "The Welfare State and Long-Term Economic
Growth: Marxian, Neoclassical, and Keynesian Approaches." American
Economic Review 72:2 (1982).

Buchanan, James, and Marilyn Flowers. The Public Finance. New York:
Irwin, 1980.

Economic Report of the President (ERP), 1996. Washington, D.C.: U.S.
Government Printing Office.

Fazeli, Rafat. "Transfers Between the State and the Working Class:
The Case of the U.K." Diss. The New School for Social Research, 1992.

--. The Economic Impact of the Welfare State and the Social Wage: The
British Experience. Aldershot, England: Avebury, 1996.

Krugman, P. "Europe Jobless, America Penniless? Forecast of Economic
Conditions. Foreign Policy 95 (Summer 1994): 19-34.

Maniatis, T. "State Revenues and Expenditures and Income
Redistribution in Sweden and Norway." Diss. The New School for Social
Research, 1992.

Marmor, T. R. "Understanding the Welfare State: Crisis, Critics, and
Countercritics." Critical Review 7:4 (1993): 461-478.

McGill, S. C. "The Social Wage in Australia: An Analysis by Social
Class." Diss. The New School for Social Research, 1989.

Miller, J. "A Negative Net Social Wage and the Reproduction Crisis of
the 1980s." Paper presented at the conference on International
Perspectives on Profitability and Accumulation. New York City, 1988.

--. "Social Wage or Social Profit? The Net Social Wage and the
Welfare State." Review of Radical Political Economics 21:3 (1989).

Moudud, Jamee K., and Ajit Zacharias. "Wither the Welfare State? The
Macroeconomics of Social Policy." Public Policy Brief no. 61. The
Levy Economics Institute of Bard College, 2000.

O'Connor, J. The Fiscal Crisis of the State. New York: St. Martin's, 1973.

Organization for Economic Cooperation and Development (OECD).

"The Role of the Public Sector: Causes and Consequences of the Growth
of Government." OECD Economic Studies 4 (1985).

--. National Accounts: Main Aggregates. Vol. 1: 1960-1989. 2 vols.
Paris: OECD, 1991.

Pampel, Fred C., and John B. Williamson. Age, Class, Politics, and
the Welfare State. Cambridge: Cambridge University Press, 1989.

Pear, R. "Welfare Debate Will Re-examine Core Assumptions." New York
Times, January 2, 1995: 1, 9.

Rosenberg, S. "Restructuring the Labor Force: The Role of Government
Policies." The Imperiled Economy, Book II. New York: URPE, 1987.

Shaikh, A., and E. A. Tonak. "The Welfare State and the Myth of the
Social Wage." The Emperiled Economy: Book 1. Eds. R. Cherry et al.
New York: URPE, 1987.

--. The Political Economy of National Accounts: An Alternate Approach
to the Measurement of the Wealth of Nations. New York: Cambridge
University Press, 1994.

--. "The Rise and Fall of the U.S. Welfare State." The Political
Economy Reader: Inquiries into Alternative Economic Theory and
Policy. Eds. R. Baiman et al. Armonk, N.Y.: ME. Sharpe, 2000: 247-265.

Skocpol, T. "America's Incomplete Welfare State: The Limits of New
Deal Reforms and the Origins of the Present Crisis" Stagnation and
Renewal in Social Policy. Eds. M. Rein, G. Esping-Andersen, and L.
Rainwater. New York: M. E. Sharpe, 1987.

Tonak, E. Ahmet. "A Conceptualization of State Revenues and
Expenditures: The U.S, 1952-1980." Diss. The New School for Social
Research, 1984.

Anwar Shaikh is a Professor of Economics at the Graduate Faculty, New
School University. He has been recently writing on the long-term
determinants of inflation in advanced economies. His recent
publications include "Is Personal Debt Sustainable?" (coauthored with
Papadimitriou et al., 2002) and published by the Levy Economics
Institute of Bard College.

This archive was generated by hypermail 2.1.5 : Mon Dec 15 2003 - 00:00:00 EST