Political Response to

Capitalist Crisis:
Neo-Marxist Theories of the State
and the Case of the New Deal
THEDA SKOCPOL

DESPITE all that has been observed since Marx’s time, as to the operations of
elites, bureaucracies, etc., Marxists generally seek to reduce political phenomena
to their &dquo;real&dquo; class significance, and often fail, in analysis, to allow sufficient
distance between the one and the other. But in fact those moments, in which
governing institutions appear as the direct, emphatic, and unmediated organs of
a &dquo;ruling class&dquo; are exceedingly rare, as well as transient. More often these institu-

tions operate with a good deal of autonomy, and sometimes with distinct interests
of their own, within a general context of class power which prescribes the limits
beyond which this autonomy cannot be safely stretched, and which, very gen-
erally, discloses the questions which arise for executive decision. Attempts to
short-circuit analysis end up by explaining nothing.
Thompsonl E. P.

This essay uses the history of New Deal politics during the De-
pression of the 1930s in the United States to assess the strengths and
limitations of several kinds of neo-Marxist theories of the capitalist
state.One purpose of the essay is to compare some alternative neo-
Marxist approaches, asking which raises the most fruitful questions
and offers the best explanations of New Deal politics. More basically,
the essay sketches some of the ways U.S. political institutions shaped

An earlier version of this article was presented in the session on &dquo;Political Systems&dquo; of the
seventy-fourth Annual Meeting of the American Sociological Association, Boston, August 1979.
In gaming historical background and formulating ideas for the paper, I especially benefited
from the research assistance of Kenneth Finegold and from individual discussions with him
and with Jonathan Zeitlin. Critical reactions and suggestions for revisions (not all of which I
was able or willing to accept) came from Fred Block, William Domhoff, William Hixson, Chris-

topher Jencks, John Mollenkopf, Charles Sabel, David Stark, and Erik Olin Wright.
1. E. P. Thompson, The Poverty of Theory and Other Essays (London: Merlin Press,
1978), p. 48.
 156

and limited the accomplishments of the New Deal and argues that
neo-Marxists of all varieties have so far given insufficient weight to
state and party organizations as independent determinants of political
conflicts and outcomes. The formulations here are meant to be sug-
gestive rather than conclusive. Theoretical debates on the state and
politics in capitalist societies are still wide open. An exploratory
essay that probes the interface between theories and a concrete his-
torical trajectory may help to push discussion away from abstract
conceptual disputes toward meeting the challenge of explaining actual
historical developments.

EXPLAINING THE NEW DEAL

Recent historiographical disputes have worried about the &dquo;con-

servative&dquo; versus &dquo;radical&dquo; nature of the New Deal, often asking, in
effect, whether or not the New Deal was intended to &dquo;save capitalism.&dquo;
In truth, U.S. capitalism was not fundamentally challenged-not by
the leaders of the New Deal and not by any powerful oppositional
political forces. While the question of why there was no such challenge
remains an important problem, we need to devote more attention to
understanding what actually did change politically and socially during
the 1930s and to understanding the limits placed on concretely present
tendencies within the New Deal.
The massive Depression of the 1930s not surprisingly stimulated
important political transformations in the United States. The Demo-
cratic party triumphed electorally and incorporated new popular
groups into its coalition. An unprecedented plethora of federal
agencies was established to implement new welfare and regulatory
policies. Labor militancy spread in the mass production industries.
The Congress of Industrial Organizations enrolled millions of workers
in industry-wide unions. And the federal government was transformed
from a mildly interventionist, business-dominated regime into an
active &dquo;broker state&dquo; that incorporated commercial farmers and
organized labor into processes of political bargaining at the national
level.
Still, certain changes that were conceivably possible failed to occur
in the New Deal. Although urban-liberal elements gained major new
ground within the Democratic party, they were not able to implement
a truly social-democratic program. And increased state intervention

in the economy, however significant as a break with the recent Ameri-

can past, nonetheless failed to achieve its overriding objective of full
economic recovery. Not until after the United States had geared up
 157

economically for World War II did extraordinary unemployment

disappear and national output fully revive
How can we account for New Deal transformations in the American
state and politics? And how can we explain why full economic recovery
was not induced despite the political changes? The orthodox &dquo;pluralist&dquo;

paradigm in political sociology is not well equipped to handle such

questions. Pluralism attempts to explain governmental decisions in
terms of the conflicting play of organized group interests in society
and as such it offers little that would help to explain major institutional
transformations in history. To be sure, some classic pluralist works,
such as David Truman’s The Governmental Process,33 include rich
descriptions of U.S. institutional patterns and allude to how they
encourage or block governmental access for different kinds of groups
and interests. Nevertheless, pluralists fail to offer (or seek) well-
developed explanations of how economic and political institutions
variously influence group formation and intergroup conflicts. Nor
do they feel the need to go beyond vague, evolutionist schemes that
posit institutional change in politics as an inevitable progression of
ever-increasing democracy, governmental effectiveness, and the special-
ization of political arrangements, all occurring in smooth, adaptive
responses to the &dquo;modernization&dquo; of the economy and society.4
Thus, to explain transformations such as those that occurred in the
U.S. during the 1930s, a pluralist would either have to refer to broad,
amorphous evolutionary trends at one extreme or to immediate
maneuverings of interest groups at the other. Explaining an increase
in state intervention, the specific forms this took, and especially the
limitations on the effectiveness of such intervention, would not be
a congenial undertaking.
More promising than pluralism for explaining the transformations

2. See Lester V. Chandler, America’s Greatest Depression 1929-1941 (New York: Harper
and Row, 1970).
3. David B. Truman, The Govemmental Process, 2nd ed. (New York- Knopf, 1971; origi-
nally 1951).
4. Even a very sophisticated developmentalist argument, such as Samuel Huntington,
"Political Modernization: America vs. Europe," chap. 2 in Political Order in Changing Societies
(New Haven: Yale University Press, 1968), still tends to explain macrohistorical change (or its
absence) in vague evolutionist terms, referring to the "needs" of society. Thus while Hunting-
ton brilliantly describes America’s enduringly peculiar "Tudor" polity, his explanation for why
America has no strong state is that it hasn’t "needed" one, either to maintain national unity
or to ensure economic growth. Aside from its crudeness, this explanation leaves one
wondering
about the Civil War, when unity broke down, and the Depression, when economic growth did.
Society’s "needs" must not be a sufficient explanation for political structures, their continuities
and transformations.
 158

of the New Deal era are neo-Marxist theories of &dquo;the capitalist state.&dquo;
They at least raise the right order of issues and establish some of the
analytical terms necessary for understanding such periods of institu-
tional change. At the very center of neo-Marxist analysis is the relation-
ship of political processes and state actions to the capitalist economy
and to basic class relations in capitalist society. According to neo-
Marxists, a period of economic crisis such as the Great Depression is
certain to spur socioeconomically rooted political conflicts and to
create pressures for unusual degrees and kinds of state action. More-
over, any neo-Marxist explanation of the ensuing political conflicts
and state actions would refer (in one way or another) to class actions,
class conflicts, class interests, and, above all, in advanced capitalist
societies, to the actions, conflicts, and interests of capitalists and the
industrial working class.
Beyond such fundamentals, however, there is not much agree-
ment among the various neo-Marxist approaches.5 To dicuss the
usefulness of these theories in any depth it is necessary to identify
sub-types and select examples appropriate to the particular purpose
at hand. My interest here is in the tools neo-Marxist theories may
have to offer for analyzing political conflicts and processes of state
intervention within the bounds of advanced (&dquo;monopoly&dquo;) capital-
ism. For this purpose, I shall explore particular examples of three
broad types of neo-Marxist theories: &dquo;instrumentalist,&dquo; &dquo;political-
functionalist,&dquo; and &dquo;class struggle.&dquo; For Marxist &dquo;instrumentalist&dquo;
theory, I shall especially emphasize works by U.S. historians of
&dquo;corporate liberalism.&dquo;6 For Marxist &dquo;political functionalism&dquo; (my
label), I shall use theoretical arguments from the early work of Nicos
Poulantzas.7 And for one kind of &dquo;class struggle&dquo; theory, I shall use
Fred Block’s essay &dquo;The Ruling Class Does Not Rule,&dquo; the arguments
of which were developed with reference to twentieth-century U.S.

5. Two good surveys of the literature are David A. Gold, Clarence Y. H. Lo, and Erik Olin
Wright, "Recent Developments in Marxist Theories of the Capitalist State," Monthly Review
27, no. 5 (October 1975): 29-43, and 27, no. 6 (November 1975): 36-51; and Bob Jessop,
"Recent Theories of the Capitalist State," Cambridge Journal of Economics 1, no. 4 (December
1977): 353-73.
6. I draw especially upon corporate-liberal arguments about the New Deal by Ronald
Radosh, "The Myth of the New Deal," in A New History of Leviathan: Essays on the Rise of the
Corporate State, ed. Radosh and N. Rothbard (New York. Dutton, 1972), pp. 146-87; and
G. William Domhoff, "How the Power Elite Shape Social Legislation," The Higher Circles: The
Governing Class in America (New York: Vintage Books, Random House, 1971), chap. 6.
7. Nicos Poulantzas, Political Power and Social Classes, trans. Timothy O’Hagen (London:
New Left Books, 1973); and idem "The Problem of the Capitalist State," Ideology in Social
Science, ed. Robin Blackburn (New York: Vintage Books, Random House, 1973), pp. 238-53.
After these works Poulantzas evoived away from political functionalism toward more of a
class-struggle approach. But I am only treating the earlier works here.
 159

history, including the New Deal.8 Each of the three theories to be

discussed posits a distinctive combination of class and state actions
through which a major capitalist economic crisis, such as the Great
Depression, should generate political transformations and government
interventions sufficient to ensure sociopolitical stability and renewed
capital accumulation. For each perspective, my concern will be to see
what the theory and the history of the New Deal have to say to one
another. Particular emphasis will be placed upon the National Industrial
Recovery Act (NIRA) of 1933-35 and the &dquo;Wagner National Labor
Relations Act, for these acts are especially relevant to any assessment of
the interrelations of the capitalist class, the industrial working class, and
the national government during the New Deal.
Passed in June 1933 at the end of the first &dquo;Hundred Days&dquo; of New
Deal legislation, the National Industrial Recovery Act was envisaged as
a joint business-government effort to promote national economic

recovery. To ensure the political support of organized labor, a provision

of the NIRA, section 7a, endorsed the right of industrial workers &dquo;to
organize and bargain collectively through representatives of their own
choosing.&dquo; However, most of the act addressed the declared needs of
businessmen. Industries were freed from antitrust restrictions and
prompted to draw up &dquo;codes of fair competition&dquo; to be approved and
enforced by a new National Recovery Administration (NRA). The
codes regulated hours of work and wage rates for labor, and they
raised prices and controlled levels of production in an effort to guar-
antee profits for capitalists. The idea was to bring about economic
recovery by stabilizing levels of production and employment and re-
storing &dquo;business confidence.&dquo;
The Wagner Act was the most innovative accomplishment of the
second, reformist phase of the New Deal. Enacted in July 1935, it
reiterated and put teeth into the promises earlier made in section 7a
of the NIRA. Industrial workers could join unions and bargain col-
lectively, free from harassment by their employers. A National Labor
Relations Board was established with full legal authority to determine
collective bargaining units, to hold elections to certify union representa-
tives on a majority basis, and to investigate allegations of unfair labor

8. Fred Block, "The Ruling Class Does Not Rule: Notes on the Marxist Theory of the
State," Socialist Revolution, no. 33 (May-June 1977), pp. 6-28. Class-struggle versions of
neo-Marxism are very diverse, and Block doesn’t represent all of them. He just represents
an example that I find particularly interesting because the state in Block’s theory is not col-

lapsed into class relations, as it appears to be, for example, in the more recent works of
Nicos Poulantzas and in some of the German neo-Marxist theories surveyed in John Hollo-
way and Sol Picciotto, eds., State and Capital. A Marxist Debate (London: Edward Arnold,
(1978).
160

practices by employers. Through the Wagner Act, in short, the U.S.

national government gave legal and administrative support to the
widespread establishment of industrial labor unions.

CORPORATE LIBERALISM AND THE NEW DEAL

In any discussion of neo-Marxist theories of the capitalist state,

Marxist instrumentalisms are the place to begin, for they have been
the take-off point for most recent debates. The somewhat dubious
honor of embodying the generic outlines of ths much-maligned ap-
proach has invariably been accorded to the pioneering book by Ralph
Miliband entitled The State in Capitalist Society. In truth this book
invokes arguments about politics and the state that have since been
crystallized into virtually every major neo-Marxist position on these
topics. Nevertheless, Miliband’s primary purpose in the book is to
debunk pluralism by showing that it systematically underestimates
the preponderate, self-interested political influence of members of the
capitalist class. As Miliband puts it: &dquo;What is wrong with pluralist-
democratic theory is not its insistence on the fact of competition
[that is, open political competition over state policies in capitalist
democracies] but its claim (very often its implicit assumption) that
the major organized ’interests’ in these societies, and notably capital
and labour, compete on more or less equal terms, and that none of
them is therefore able to achieve a decisive and permanent advantage
in the process of competition.&dquo;9 On the contrary, argues Miliband,
capitalists, particularly those who control major economic organiza-
tions, do enjoy decisive and stable political advantages because of their
privileged positions both &dquo;inside&dquo; and &dquo;outside&dquo; the state. Inside the
state, officials tend either to be from capitalist backgrounds or to enjoy
close career or personal ties to capitalists. Such officials, moreover,
almost invariably assume the inevitability and the legitimacy of a
capitalist economy. Furthermore, says Miliband, &dquo;business enjoys a
massive superiority outside the state system as well, in terms of the
immensely stronger pressures which, as compared with labour and
any other interest, it is able to exercise in pursuit of its purposes
For capitalists enjoy disproportionate access to organizational re-

sources, and they can credibly clothe their policy demands in &dquo;the
national interest&dquo; and back them up with potent threats of economic
or political disruption.
9. Ralph Miliband. The State in Capitalist Society (New York: Basic Books, 1969), p. 146.
10. Ibid.
 161

Miliband’s purpose is to sketch a broad frame of reference, not to

explain any particular kind of political outcome in capitalist societies.
However, since the emergence of a New Left historiography. in the
United States during the 1960s, there has been an especially &dquo;strong&dquo;
variant of instrumentalism specifically designed to explain why and
how state intervention has increased during the twentieth century in
U.S. capitalism. James Weinstein, William Domhoff, and Ronald Radosh
have been among the articulate proponents of this view,il and James
O’Connor relies in significant part upon it in his recent Fiscal Crisis of
the State.12 As Fred Block points out, &dquo; ... the heart of the theory is
the idea that enlightened capitalists recognize that crises of capitalism
can be resolved through an extension of the state’s role.&dquo;1~ Corporate

liberalism, Block notes, . &dquo;,

~

is a reinterpretation of the of American liberalism.... In [the liberal]

meaning
view, the expansion of the role of the state during the 20th century was a con-
sequence of popular victories that succeeded in making capitalism a more bene-
volent system. The new theory reversed the old view, arguing that liberalism was
the movement of enlightened capitalists to save the corporate order. In this view,
the expansion of the role of the state was designed by corporate leaders and their
allies to rationalize the economy and society. Rationalization encompasses all
measures that stabilize economic and social conditions so that profits can be made
on a predictable basis by the major corporations.~
Corporate liberalism seeks to explain particular episodes of capital-
ist political intervention, those that occur under crisis conditions and
involve the deliberate extension of state action by a class-conscious
vanguard. Under normal conditions, capitalists may influence the state
in disunified fashion in all of the ways Miliband outlines, often working
at cross-purposes through leadership posts in the state, personal ties
to particular policy makers, or interest-group pressures from without.
But, say the theorists of corporate liberalism, when crises of accumula-
tion occur or political challenges from below threaten, capitalists can
be expected to act as a class. As James O’Connor puts it, &dquo;by the turn
of the century, and especially during the New Deal, it was apparent
to vanguard corporate leaders that some form of rationalization of
the economy was necessary. And as the twentieth century wore on,

11. James Weinstein, The Corporate Ideal in the Liberal State: 1900-1918 (Boston: Beacon
Press, 1968). For relevant works by Domhoff and Radosh, see note 6. In a private communica-
tion, Bill Domhoff has convinced me that his views about the Wagner Act have developed away
from a strictly corporate-liberal position in publications subsequent to The Higher Circles. Thus
my arguments here apply only to the cited parts of that book.
12. James O’Connor, The Fiscal Crisis of the State (New York: St. Martin’s Press, 1973).
13. Fred Block, "Beyond Corporate Liberalism," Social Problems 24 (1976-77): 355.
14. Ibid., p. 352.
162

the owners of corporatecapital generated the financial ability, learned

the organizational skills, and developed the ideas necessary for their
self-regulation as a class.&dquo;1-5 Enlightened leadership for the capitalist
class is likely to come from those holding the most strategic and power-
ful economic positions. These vanguard leaders will realize the necessity
of stepped-up state intervention and will use their great resources
and prestige to persuade many capitalists to go along and to pressure
politicians to implement the needed programs. If necessary, minor
co-optive concessions will be offered to small businessmen and workers.
The resulting state intervention will, however, primarily work in the
interests of large-scale corporate capital.
For theorists of corporate liberalism, therefore, the New Deal is
envisaged of
as a set clever capitalist strategies to stabilize and revitalize
a U.S. economy dominated by large corporations. &dquo;The New Deal

reforms,&dquo; writes Ronald Radosh, &dquo;were not mere incremental gestures.

They were solidly based, carefully worked out pieces of legislation.&dquo; 16
Great stress is placed on the ultimate benefits that U.S. corporate
capitalism gained not only from government interventions to stabilize
particular industries, but also from such widely popular measures as
unemployment insurance, social security, and the legalization of
unions and collective bargaining. That such measures appeared to be
won by democratic pressure only represents an added advantage for

capitalists. For in this way, says Radosh, &dquo;the reforms were of such a
character that they would be able to create a long-lasting mythology
about the existence of a pluralistic American democracy....&dquo;
Business opposition to the New Deal is seen by theorists of corporate
liberalism as emanating from &dquo;small business types, with their own
conservative mentality, [who] responded to the epoch in terms of the
consciousness of a previous era.&dquo; 18 By contrast, far-sighted leaders of
big business strongly promote the necessary reforms. &dquo;The moderates
in the governing class had to put up a stubborn, prolonged fight until
the law would be able to reflect the realities of the new epoch of
corporation capitalism.&dquo;19
The corporate-liberal explanation of the New Deal is highly mis-
leading. It can only be substantiated through a purely illustrative and
selective citing of facts. When the theory is subjected to a rigorous,

15. O’Connor, Fiscal Crisis, p. 68.

16. Radosh, "Myth of the New Deal," in New History of Leviathan, ed. Radosh and Roth-
bard, p. 186.
17. Ibid.
18. Ibid., p. 187.
19. Ibid.
 163

skeptical examination, corporate liberalism fails to explain even those

aspects of the New Deal that seem most consonant with it.
There are facts that fit the corporate-liberal interpretation of the
New Deal; indeed, such facts are repeatedly recounted by proponents
of the theory. The self-stated aims of the leaders of the New Deal,
including Franklin Roosevelt himself, could easily be described as
&dquo;corporate-liberal,&dquo; in the sense that the top New Dealers were all
out to sustain American capitalism (and democracy) through reform,
not to propel the country toward socialism. More to the point, business
leaders and spokesmen were visibly involved in the New Deal. Their
presence was all-pervasive with respect to the early, comprehensive
measure for economic recovery, the National Industrial Recovery
Act of 1933-35,~ and under the. National Recovery Administration,
businessmen in each industry drafted and enforced the regulatory
codes. Later in the New Deal, the Social Security Act was endorsed
by a &dquo;Business Advisory Council&dquo; of prominent bankers and officers of
major corporations.21 Finally, corporate-liberal theorists correctly
point out that throughout the 1930s FDR never ceased wooing business
support and that there were always officials with business backgrounds
and ties holding high-level positions in the federal administration.
But facts such as these represent only the loosest conceivable evi-
dence for validating the hypothesis that capitalist plans and influence
caused the New Deal. (Indeed, by analogous criteria, one could &dquo;prove&dquo;
that Marxist theory was sponsored by capitalists because of Marx’s
close personal and financial ties to Engels, the son of a capitalist manu-
facturer !) If we hold corporate liberalism to more rigorous standards of
validation, then the key questions become: Was there at work during
the 1930s a self-conscious, disciplined capitalist class, or vanguard of
major capitalists, that put forward functional strategies for recovery
and stabilization and had the political power to implement them
successsfully? Were most corporate leaders (especially of big, strategic
businesses) prepared to make concessions to labor? Did business op-
position to the New Deal come primarily from small business? These
questions go to the heart of the corporate-liberal claims; if they cannot
be answered affirmatively, then the theory does not adequately explain
the New Deal.
No part of the New Deal better appears to fit the corporate-liberal

20. See Ellis W. Hawley, The New Deal and the Problem of Monopoly (Princeton, N. J.:
Princeton University Press, 1966), chaps. 1-3; and Arthur M. Schlesinger, Jr., The Coming of-
the New Deal (Boston: Houghton Mifflin, 1958), chaps. 6-8.
21. Radosh, "Myth of New Deal," in New history of Leviathan, ed. Radosh and Rothbard,
pp. 157-58.
 164

model than the National Industrial Recovery Act and the National
Recovery Administration (NRA) established under it. Capitalists, above
all, those who ran the major corporations in each industry, were by all
historical accounts able to get exactly what they wanted out of the
NRA, that is, fixed prices and stabilized production. Nevertheless,
despite the ubiquitous influence of big businessmen in the formulation
and implementation of the NIRA, this part of the New Deal does not
measure up as a class-conscious strategy for U.S. corporate capitalism.
The policies pursued were inadequate to the needs of the economy and
to the interests of big business in general. For the most important fact
of all about the NIRA, though, curiously, one that is never discussed
by corporate-liberal theorists, is that it failed to bring economic re-
covery through business-government cooperation. Yet if we probe a
bit into the actual state of consciousness and discipline among U.S.
capitalists at the time of the NIRA, this failure becomes more under-
standable. We can see that, ironically, the failure can be partially
attributed to the strong and misdirected political influence of (by
corporate-liberal criteria) insuffzciently class conscious capitalists.
U S. capitalists were ill prepared to act together as a class in the
early 1930s.22 During the height of the Progressive Era (and the heyday
of the National Civic Federation), there had been a measure of class
unity and discipline, at least among large-scale corporate capitalists.
In large part this was because the House of Morgan, autocratically
directed by J. P. Morgan himself, enjoyed strong influence on boards
of directors of the major corporations in many key industries. By
the 1930s, the Morgan hegemony was no longer so absolute. World
War I and the patterns of economic growth during the 1920s loosened
the dependence of industrial firms upon outside financing. Sheer
competitive disunity increased in many industries. Within others
there was more unity, as ties among firms were strengthened during the
mobilization for World War I and through the trade-association move-
ment sponsored by the Republican administrations during the 1920s.
Thus by the end of the decade the highest level of consciousness and
discipline that some (by no means all) capitalists enjoyed was focused
within single industries, especially those like textiles, where active and
reasonably effective trade associations had evolved
22. This paragraph draws especially upon David Vogel, "Why Businessmen Distrust Their
State: The Political Consciousness of American Corporate Executives," British Journal of
Political Science 8, no. 1 (January 1978): 70-72; and Gabriel Kolko, Main Currents in Ameri-
can History (New York. Harper and Row, 1976), pp. 100-117.
23. See Louis Galambos, Competition and Cooperation: The Emergence of a National
Trade Association (Baltimore: Johns Hopkins University Press, 1966); and Robert F. Himmel-
 165

Not surprisingly, therefore, when the Depression struck and busi-

ness spokesmen began to come forward with plans for government
programs to help business, even the most comprehensive visions of
&dquo;business planning&dquo; outlined by Gerard Swope, president of General
Electric and Henry Harriman, president of the U.S. Chamber of Com-
merce, called primarily for coordination within industries.24 The idea
was to give government backing to the efforts of trade associations

(or other industry-wide bodies) to regulate competition in the interests

of all (and especially the more established) businesses. But these plans
had very little to say about how problems of interindustry coordina-
tion were to be resolved in the interest of the capitalist class, and the
accumulation process, as a whole. At most, Swope and Harriman en-
visaged mutual consultation by representatives chosen from each
industry, but they did not say how particular industries (or dominant
enterprises within industries) could be persuaded to accept plans not
favorable to their short-term interests.
In truth, the supposedly class-conscious capitalists of the early
1930s were mesmerized by false analogies based upon their experience
of business-dominated government intervention during World War
1.25 The War Industries Board (WIB) of 1918 had been run by busi-
nessmen-administrators in the interests of the dominant firms in each
industry.~ As William Leuchtenberg puts its, &dquo;perhaps the outstanding
characteristic of the war [World War I] organization of industry was
that it showed how to achieve massive government intervention without
making any permanent alteration in the power of corporations.&dquo;27
U.S. capitalists applauded this experience of &dquo;government intervention.&dquo;
Not only was their autonomy respected, but the happy result was
economic prosperity and plentiful profits. In the face of another
national crisis after 1929, why not revive the methods that had worked
so well in World War I? For the capitalists there was the added ad-

berg, The Origihs of the National Recovery Administration (New York: Fordham University
Press, 1976).
24. Gerard Swope, The Swope Plan (New York: Business Bourse, 1931); Otis L. Graham,
Jr., Toward a Planned Society (New York. Oxford University Press, 1976), pp. 24-25; and
Kolko, Main Currents, pp. 116-20.
25. William E. Leuchtenburg, "The New Deal and the Analogue of War," in Change and
Continuity in Twentieth-Century America, ed. John Braeman, Robert Bremner, and E. Walters
(Columbus: Ohio State University Press, 1964). This article discusses the appeal of "analogies
of war" to virtually all of the sets of actors in the early New Deal.
26. Robert D. Cuff, The War Industries Board : Business-Government Relations during
World War I (Baltimore: Johns Hopkins University Press, 1973).
27. Leuchtenburg, "Analogue of War," in Change and Continuity, ed. Braeman, Bremner,
and Walters, p. 129.
 166

vantage that appealing to the &dquo;analogue of war&dquo; could provide an

occasion for nationalistic propaganda: it could pull Americans of
all classes together &dquo;to do battle against the economic crisis,&dquo; a crisis
that otherwise might have been blamed on the capitalists themselves.
Still, there was an enormous difficulty in modeling the NRA on
the WIB. Government mobilization of industrial resources for war,
which inherently involves increasing production through massive
federal spending, is not at all equivalent to using government authority
to raise prices and stabilize profits, production levels, wages, and
employment. In the WIB, businessmen-administrators were asked
to allocate plenty and to control rapid expansion in an orderly way,
with government purchases offered as inducements. But in the NRA,
the job was to discipline businessmen, and labor, in a situation of
scarcity and with few positive sanctions. Such policies were certain
to exacerbate and politicize economic conflicts. The WIB was therefore
a very poor model for capitalists to draw upon in their plans, and

demands, for government intervention in the Depression. Moreover,

it was unrealistic to expect that the efforts of individual industries
to enhance their own profits by increasing their prices would lead
to a rise in total real output and employment. Yet business spokesmen
convinced themselves and, initially, many politicians that recovery
would come in this way. Their &dquo;trade association consciousness&dquo; and
their infatuation with the inappropriate WIB model left U.S. capitalists
ill prepared to advocate any other more realistic plan for government
intervention in the 1930s. The result was prolonged economic crisis
and continuing political uncertainties for capitalists.
If the failure of the NRA raises questions about corporate liberalism
by demonstrating the inability of U.S. capitalists to pursue a class-
conscious strategy for economic recovery, New Deal labor politics
even more directly contradict the corporate-liberal model of
political
change. The history of these policies shows that major industrial
capitalists were not prepared to grant concessions to labor; instead
undesired policies were forced upon them through the workings of a
national political process that they could not fully control.
Even during the honeymoon between business and government
at the start of the New Deal, major capitalists and their spokesmen
were very reluctant to make concessions to labor,
especially not any
that would facilitate independent labor unions. As Arthur Schlesinger
points out, in the original passage of the NIRA through Congress, the
prolabor section 7a was constantly on the verge of being defined out
of existence: &dquo;The trade association group, evidently feeling that
organization was a privilege to be accorded only to employers, accepted
 167

the idea of 7a with reluctance; even the more liberal among them,
like Harriman and Swope, had made no provision for organized labor
in their own plans.... Only the vigilance of Jerome Frank, Leon
Keyserling and Senator Wagner and the fear [in Congress] of provoking
labor opposition kept it in.&dquo;2$
Such good friends as labor had in the early New Deal were not from
the ranks of major capitalists, but from within the government.~ Labor
Secretary Frances Perkins was a rallying point for politicians and
professionals who wanted to promote welfare measures and national
regulation of working conditions, wages, and hours. And Senator
Robert F. Wagner of New York spearheaded efforts to guarantee
labor’s right to collective bargaining with employers through independ-
ent union organizations. Wagner was certainly not anticapitalist, but
his ideas about reform, recovery, and the rights of labor went well
beyond the notions of even the most far-sighted U.S. capitalists. In
August 1933, Roosevelt made Wagner the chairman of a National
Labor Board (NLB) that was supposed to promote labor’s rights under
section 7a. The board tried to persuade employers to bargain with
unions that were to be certified, after board-supervised elections, as
representing a majority of workers. But major employers adamantly
refused, and the NLB could not persuade either Roosevelt or the
NRA to enforce its decisions. As Senator Wagner saw that friendly
persuasion would not work, he and his staff began planning and lobby-
ing within Congress and the Roosevelt administration for strong legisla-
tion to enforce union recognition. These efforts eventually culminated
in the (Wagner) National Labor Relations Act of July 1935.
By the second half of 1934, business opposition to the New Deal
was spreading and becoming more vocal and organized .30
To be sure,
smaller businessmen were among the earliest opponents, as they reacted
against the competitive advantages secured for large corporations
under the NRA .31 But businessmen across the board were souring

28. Schlesinger, Coming of New Deal, p. 99.

29. Good background on the labor policies of the New Deal is especially to be found in
Irving Bernstein, The New Deal Collective Bargaining Policy (Berkeley and Los Angeles: Uni-
versity of California Press, 1950); and Murray Edelman, "New Deal Sensitivity to Labor Inter-
ests," in Labor and the New Deal, ed. Milton Derber and Edwin Young (New York: DaCapo
Press, 1972), pp. 157-92.
30. Ellis Hawley, "The New Deal and Business," in The New Deal, ed. Braeman. Bremner.
and Brody, vol. 1, The National Level, pp. 64-66; and Schlesinger, Coming of New Deal,
chap. 30; William H. Wilson, "How the Chamber of Commerce Viewed the NRA: A Reexamina-
tion," Mid-America 44 (1962): 95-108; and Kim McQuaid, "The Frustration of Corporate
Revival in the Early New Deal," The Historian 41 (August 1979): 682-704.
31. Small business opposition, pressed especially through Congress, is a major theme in
Hawley, New Deal and Problem of Monopoly, chap. 4.
168

on &dquo;bureaucracy&dquo; and were growing increasingly apprehensive about

government regulation as political demands were voiced on behalf
of farmers, consumers, and industrial labor. While a few major capital-
ists continued to speak out for the Roosevelt administration as individ-
uals and through the Business Advisory Council, a much larger number
of major capitalists were becoming increasingly hysterical in their
opposition to the New Deal. The American Liberty League, launched
in 1934, drew its most important support from major financial and
industrial interests clustered around DuPont Chemical and General
Motors-hardly &dquo;small business&dquo;!32 What is more, the National Asso-
ciation of Manufacturers (NAM), which prior to the 1930s was pre-
dominately a spokesman for small and medium businesses, was
transformed during the early 1930s into an anti-New Deal vehicle
dominated by big businesses.~ This is important to note because
supporters of corporate liberalism repeatedly dismiss NAM opposition
to New Deal policies as representing only the stubbornness of small
businessmen.
Many U.S. businessmen ended up opposing the Social Security
Act, and virtually all large-scale industrial employers unequivocally
opposed the Wagner Act.34 To say this is not to deny that, eventually,
these measures were accepted by most corporate capitalists in the U.S.
Nor is it to deny that these measures ended up, after the 1930s, creat-
ing conditions favorable to smoother capitalist economic growth and
more stable industrial labor relations. But U.S. capitalists did not plan
or promote these measures: they could not foresee their ultimately
favorable effects, and, in the depressed economic situation and un-
certain political climate of the mid-1930s, they feared the immediate
ill effects of increased government power within the economy. In the
area of labor relations, in particular, U.S. capitalists were comfortably

accustomed to running their corporations with a free hand; independ-

32. George Wolfskill, The Revolt of the Conservatives: A History of the American Liberty
League (Boston: Houghton Mifflin, 1962); and Frederick Rudolph, "The American Liberty
League, 1954-1940," American Historical Review 56, no. 1 (October 1950): 19-33. Rudolph
writes (p. 22): "The membership of its national advisory council was drawn largely from the
successful business interests of the industrial states of the North and East ..."
33. Philip H. Burch, Jr., "The NAM as an Interest Group," Politics and Society 4, no. 1
(Fall 1973): 100-105.
34. On social security see Edwin E. Witte, The Development of the Social Security Act
(Madison: University of Wisconson press, 1962), pp. 88-90; and Schlesinger, Coming of New
Deal. p. 311. On the Wagner Act see Bernstein, New Deal Collective Bargaining Policy, passim.
The resistance of industrial employers is documented in Daniel A. Swanson, "Flexible Individ-
ualism: The American Big Business Response to Labor, 1935-1945," (Undergraduate thesis,
Harvard College, Social Studies, 1974). See also Vogel, "Why Businessmen Distrust Their
State," pp. 63-65.
 169

ent labor unions, they correctly understood, would only circumscribe

one of the areas of managerial autonomy that they had enjoyed in
the past. That the labor unions would be established through increased
federal regulatory powers only made the prospect worse.
To be sure, U.S. capitalists enjoyed great political influence during
the 1930s, as they did in previous decades, and have ever since. But
corporate liberalism greatly overestimates what U.S. capitalists were
able and willing to do during the 1930s to engineer effective, congenial
political responses to a capitalist economic crisis. As Ellis Hawley
writes:
Since they [capitalists] seemed have benefited most from the innovations of the
to
period, the temptation was to conclude that they must have planned it that
strong
way and used the New Dealers either as their tools or as camouflage for their
operations. In reality, so the evidence at hand indicates, they had neither the
power, the unity, nor the vision to do this. They could, to be sure, push an initial
program upon the new administration, limit the efforts at structural reform, and
secure desired stabilization measures for certain types of industries. But they could
not make the initial program work or retain the initiative; and instead of seeing that
their long-range interests lay with the pattern taking shape after 1934 and moving
quickly to adopt it, most of them spent the next six years fighting a bitter and
expensive delaying actions. 35
Major New Deal measures were passed and implemented over the
opposition of capitalists. Not only did capitalists fail to control the
political process during the mid-1930s, they even lost their ability to
veto major legislative enactments that touched directly upon their
accustomed prerogatives. Corporate-liberal theory cannot explain
why or how this could happen, just as it cannot account for the failures
of the business-sponsored NIPA.
Quite evidently, U.S. politics in the 1930s was more complex than
the corporate-liberal perspective maintains. Let us therefore proceed
to the &dquo;political functionalism&dquo; of Nicos Poulantzas.

POLITICAL FUNCTIONALISM AND THE NEW DEAL

Nicos Poulantzas is well known for some basic disagreements with

instrumentalist approaches. For Poulantzas, &dquo;the direct participation
of members of the capitalist class in the State apparatus and in the
government, even where it exists, is not the important side of the
matter. The relation between the bourgeois class and the State is an
objective relation. This means that if the function of the State in a
35. Hawley, "New Deal and Business," in The New Deal, ed. Braeman, Bremner, and Broay,
voL 1, The National Level, pp. 75-76.
 170

determinate social formation [that is, in a given society] and the

interests of the dominant class in this formation coincide, it is by
reason of the system itself....&dquo;36 In Poulantzas’s view capitalists do
not need to staff the state apparatus directly; nor must they put delib-
erate political pressure on government officials. Even without such
active interventions, capitalists will still benefit from the state’s activi-
ties. For the state, by definition, is &dquo;the factor of cohesion of a social
formation and the factor of reproduction of the conditions of pro-
duction of a system.&dquo;-37 State interventions will, in other words, neces-
sarily function to preserve order in capitalist society and to sustain and
enhance the conditions for capitalist economic activity.
According to Poulantzas, the state and politics work in opposite
ways for the dominant capitalist class and for the working class (and
other noncapitalist classes). 38 Because working-class unity is a threat
to capitalism, Poulantzas posits that the state functions most funda-
mentally to &dquo;disunite&dquo; the workers. It does this, in part, by transform-
ing them into privatized individual citizens, competitive in their eco-
nomic relations and members of a classless &dquo;nation&dquo; in political terms.
At the same time, the political system (in a democratic capitalist
state) allows workers to vote and form interest groups and political
parties through which they may be able to achieve limited concessions
through nonrevolutionary political struggles. Thus the capitalist state
controls workers (indeed all nondominant classes) by promoting their
&dquo;individualization&dquo; and by making necessary co-optive concessions
in ways that tend to divide noncapitalists into competing sub-groups.
Poulantzas’ state functions the opposite way for the capitalist
class. In sharp contrast to Miliband’s instrumentalism and to corpor-
ate liberalism, Poulantzas holds that neither the political interventions
of self-interested capitalists nor the policies formulated by an enlight-
ened corporate vanguard will ensure that the political system functions
in the interests of capitalists. Instead this happens only because of
the interventions of a &dquo;relatively autonomous&dquo; state not directly
controlled by capitalists. This state organizes the unity of the capitalist
class itself, because it is &dquo;capable of transcending the parochial,
individualized interests of specific capitalists and capitalist class frac-

36. Nicos Poulantzas, "The Problem of the Capitalist State," in Ideology in Social Science,
ed. Robin Blackburn (New York: Vintage Books, Random House, 1973), p. 245.
37. Ibid., p. 246.
38. Nicos Poulantzas, Political Power and Social Classes, trans. Timothy O’Hagen (London:
New Left Books, 1973). See also Simon Clarke, "Marxism, Sociology and Poulantzas’s Theory
of the State," Capital and Class, no. 2 (Summer 1977), pp. 1-31.
 171

tions.&dquo;&dquo;’ Simultaneously, the relatively autonomous state also enforces

whatever concessions the current state of the political class struggle
makes necessary if the dominated classes are to be kept in line.
Poulantzas is not greatly interested in explaining exactly how
the capitalist state goes about performing its inherent functions. Histor-
ical contingencies such as divisions within the capitalist class and the
vagaries of the political class struggle apparently determine, in Poul-
antzas’s view, specifically how the state is structured and how it func-
tions. But, short of revolution, functional outcomes are certain to
occur: the bottom line for Poulantzas always seems to be the stability
of the capitalist system, the reproduction of capitalist production
relations, and the continuation of capitalist class domination. Poul-
antzas’s capitalist state is basically a vehicle of system maintenance.
Corporate-liberal instrumentalists, as we have seen, attempt to
explain New Deal measures as the strategies of class-conscious capital-
ists. More appropriately, Poulantzian theory would stress the political
provenance of policies, even of a measure such as the NIRA, which
closely conformed to the declared preferences of capitalists. From a
Poulantzian vantage point, New Deal economic policies were not simply
a response to the demands of capitalists; rather they addressed the

interests of competing groups both within the ranks of the capitalist

class and between capitalists and noncapitalists. This process of placat-
ing competing interests through active state intervention was mediated
by the Democratic party after its massive electoral victories in 1932.
If business strategies and political influence had been all that was
necessary to produce an NIRA-type program, then it should have
been enacted by the Republicans in late 1931 or during 1932, when
Swope and Harriman were first urging their plans. Instead, the NIRA
came only after the Democrats and Roosevelt came to power, and even

then, the NIRA was not formulated until the end of the &dquo;Hundred
Days&dquo; of early New Deal legislation. The NIRA itself was presented
as a consensual, national effort to help businessmen, workers, and

consumers, all together. And it represented the culmination of a sweep-

ing legislative program in which the needs of farmers, bankers, home
owners, the unemployed, and local governments had been addressed.40

39. Gold, Lo, and Wright, "Recent Developments," p. 38.

40. Actually, those whose needs were addressed earliest and most thoroughly in the New
Deal were commercial farmers. This was not incidental in political terms, for Roosevelt had
gained the presidential nomination in 1932 only through support from the South and West.
Also, given the disproportionate weight of rural areas in the U.S. national political system of
the 1930s, Roosevelt was bound to be especially desirous of attracting voters and placating
Congressmen from these areas. It is noteworthy that neo-Marxist theories of politics in ad-
 172

Just as Poulantzas’s theory would suggest, this over-all process of inter-

est aggregation and consensus building, all within the bounds of a
taken-for-granted effort to save the existing capitalist economy, no
doubt could be undertaken only by the Democratic party led by
Franklin Roosevelt. Compared to the Republicans led by Herbert
Hoover, the FDR-led Democrats in 1933 were more popularly rooted
and sufficiently &dquo;relatively autonomous&dquo; from pure business domina-
tion to enable them to take strong state initiatives in the economic
crisis. These initiatives, in turn, promised benefits for practically every-
one and rebuilt national morale.
Within the political context thus created, it was possible to formu-
late and enact the NIRA. In the drafting process, various proposals for
the government action to promote recovery were melded together:41
business schemes for government-backed industrial cartels were the
primary basis for Title I of the act, yet there was also a Title II estab-
lishing the Public Works Administration, which incorporated planss
calling for major government spending to stimulate industry. Moreover,
section 7a of Title I made promises to labor, and there were also
rhetorical concessions to consumers and small businessmen.
The origins, the political context, and the legislative content of
the NIRA thus fit Poulantzian political functionalism very well: the
state undertook to organize business to promote recovery, and it did
so with all of the symbolic trappings and concessions to popular groups

that were necessary to present the entire effort as a unified, national

battle against the Depression. Indeed, if the early New Deal and the
NIRA had only quickly achieved their declared purposes of socio-
political stability and full economic recovery, then Poulantzian theory
would appear to offer a perfect explanation for the New Deal. But in
actuality, the NIRA blatantly failed to bring economic recovery, and
it deepened conflicts within the capitalist class, between capitalists
and the state, and between capitalists and labor. Poulantzian theory
predicts functional outcomes of state policies and interventions. It
offers little direct theoretical guidance for explaining why and how
failures of state policies could occur, especially not failures threatening
to capitalists. It offers little guidance for dissecting the concrete course

vanced capitalism tend to ignore farmers, concentrating instead only on industrial workers and
industrial and financial capitalists. But a complete analysis of class and politics in the New
Deal would definitely have to look at the relationships of farmers and agricultural
laborers to industrial capitalists and workers.
41. The best discussion is Hawley, New Deal and Problem of Monopoly, chaps. 1-2. See
also Schlesinger, Coming of New Deal, chap. 6.
 173

of political and social struggles over time, especially not struggles that
lead toward deepening political contradictions as opposed to functional
resolutions of crises through stabilizing political actions.42
In the context of his polemic against instrumentalism, Nicos Poul-
antzas has insisted that it does not matter whether capitalists staff
or pressure the state. Thus Poulantzas declared in a critique of

Ralph Miliband, &dquo;If Miliband had first established that the State is
precisely, the factor of cohesion of a social formation and the factor
of reproduction of the conditions of production of a system that itself
determines the domination of one class over the others, he would
have seen clearly that the participation, whether direct or indirect,
of this class in government in no way changes things. ,,43
Obviously,
this is an extreme formulation. Poulantzas is saying that, no matter
what, the state functions automatically to stabilize and reproduce
the capitalist system. Yet Poulantzas has also made statements about
relations between capitalists and the state apparatus that, if posed
as hypotheses about conditions that could vary historically and cross-

nationally, would help us explain the failures of the NIRA. The &dquo;cap-
italist State,&dquo; Poulantzas has suggested, &dquo;best serves the interests of
the capitalist class only when the members of this class do not partic-
ipate directly in the State apparatus, that is to say when the ruling
class is not the politically governing class. ,,44
We have already seen that the NIRA as a piece of legislation was
drafted and enacted in a way that conforms to the Poulantzian notion
of a &dquo;relatively autonomous&dquo; political process producing state policies
that both support capitalism and are democratically legitimate. This
was possible in 1933 because the electoral system, the Democratic

party, the Roosevelt administration, and the Congress operated to

produce the &dquo;Hundred Days&dquo; and the NIRA legislation. But, interest-
ingly enough, the &dquo;relative autonomy&dquo; of politics was much less evi-
dent in the implementation of the NIRA than it was in its enactment.
A silent assumption of Poulantzas’s functionalist theory is that there
will always be a centralized, bureaucratic administrative apparatus to
manage economic interventions on behalf of the capitalist class as a
whole 45 But for understandable historical reasons, the U.S. federal

42. Analogous criticisms (focused on Poulantzas’s own concrete analyses of fascism) are
offered by a historian in Jane Caplan, "Theories of Fascism: Nicos Poulantzas as Historian,"
History Workshop 3 (Spring 1977): 83-100. As Caplan says (p. 98), "we must ... not turn
history into a prolonged tautology." Poulantzas’s functionalism constantly tends to do this.
43. Poulantzas, "Problem of Capitalist State," in Ideology, ed. Blackburn, p. 246.
44. Ibid., p. 246.
45. This is obviously a very "French" assumption!
174

government in the early 1930s lacked any such administrative capacity.

The failures of the NIRA to promote economic recovery and to stabi-
lize relationships among businessmen can be attributed in significant
part to the absence of effective capacities for autonomous economic
intervention on the part of the U.S. federal administration.
The national government with which the U.S. entered the Great
Depression was basically formed during the Progressive Era partly in
reaction against, and partly upon the foundations of, the uniquely
&dquo;stateless&dquo; governmental system that had held sway in America during
the nineteenth century This nineteenth-century system has aptly
been called a &dquo;state of courts and parties,&dquo; because its basic govern-
mental functions were divided between, on one hand, a potent judiciary
branch and, on the other, a network of government offices staffed by
locally rooted political parties according to their electoral fortunes and
patronage requirements. In the expanding, decentralized capitalist
economy of nineteenth-century America, this governmental system
functioned remarkably well. The courts regulated and defended prop-
erty rights, and the party-dominated electoral-administrative system
freely handed out economic benefits and loosely knit together a diverse
society. With the advent of corporate concentration and the emergence
of a truly national economy and society, the government of courts and
parties began to face national-administrative and policy-making tasks
for which it was poorly suited. But the old system managed to remain
intact, and block governmental &dquo;modernization,&dquo; as long as its
mass-mobilizing political parties were relatively balanced in their
political competition (at the national level and in many states outside
the South). Only after the massive electoral realignment of 1896
decreased party competition in many formerly competitive states
and created a national imbalance strongly in the Republicans’s favor,
was the way opened for the building of new national administrative

systems.
Such administrative expansion came slowly and in fragmented ways
during the Progressive Era. Unlike Continental European nations with
bureaucratic states inherited from preindustrial, monarchical times,
the U.S. national government, starting late in the game and from a low
level, developed autonomous administrative capacities only imperfectly
and under central executive coordination and control. Presidents (along
with groups of professionals) took the lead in promoting federal admin-

46. This and the following two paragraphs build on Stephen Ice Skowronek’s excellent
"Building a New American State: The Expansion of National Administrative Capacities, 1877-

1920" (Ph.D. diss., Department of Political Science, Cornell University, January 1979), to be
published soon by Cambridge University Press.
 175

istrative expansion and bureaucratizing reforms. But Congress resisted

efforts at administrative expansion and, at each step, contested the
executive branch for control of newly created federal agencies. For, by
the early twentieth century, the U.S. had, if anything, even more of a
&dquo;Tudor polity,,47 -a polity of divided sovereignty among the legis-
lative, executive, and judicial branches and among federal, state, and
local governments-than it had during the nineteenth century. In the
earlier government of courts and parties, political party discipline had
provided a kind of coordination in government. But once parties were
weakened and once administrative realms began to be set up beyond
the direct patronage controls of the parties, institutional struggles
between the president and Congress were unleashed, above all, over
how much administrative expansion should occur and under whose
control.
In this context, no centrally coordinated, executive-dominated
national bureaucratic state could emerge, not even during World War I.
Administrative expansion during that crisis was ad hoc and staffed by
officials predominately recruited from business.&dquo; Moreover, it was
rolled back by Congress right after the end of the war. What remained
were a few (increasingly uncontrollable) independent regulatory agen-

cies and some restricted realms of federal administration with over-

lapping, cross-cutting lines of control and access to the executive and to
Congress. During the 1920s, the Republicans governed within this
system, making minimal efforts at institutional innovation and under-
taking few federal interventions in the affairs of states, localities, or the
private economy. When the Depression hit, therefore, the U.S. had
(for a major industrial nation) a bureaucratically weak national govern-
ment, and one in which existing administrative capacities were poorly
coordinated.
This historical background on the U.S. state can help us understand
what happened in the implementation of the NIRA during 1933-35
in two main ways. First, it becomes easy to see why the National
Recovery Administration, set up under Title I of the NIRA to regulate
the industrial economy, had to be created from scratch and through
the emergency recruitment of administrators from business back-
grounds. There was no pre-existing federal bureaucracy with the man-
power and expertise needed to supervise a sudden, massive effort to
draw up hundreds of codes to regulate wages, working hours, prices,

47. This phrase comes from Huntington, "Political Modernization: America vs. Europe,"
in Political Order in Changing Societies, chap. 2.
48. Cuff, War Industries Board.
 176

and production practices in every U.S. industry from steel and auto-
mobiles to textiles and consumer services.
The head of the NRA appointed by Roosevelt was General Hugh
Johnson, a man with business experience and connections, who had
served in the WIB during World War I. Johnson moved quickly to set
in motion the process of approving and enforcing codes of fair competi-
tion for the various industries. Not taking time to define regulatory
standards, Johnson appointed many &dquo;deputy administrators.&dquo; He
followed the WIB precedent of recruiting officials from business, very
often drawing his deputies from the same industries with which they
were then supposed to negotiate over the codes. Moreover, once the
codes were approved by the NRA, their enforcement was typically
delegated to code authorities dominated by representatives selected by
trade associations or other major interests within each industry.49
Hastily assembled in these ways from extragovernmental sources of
manpower, expertise, and organization, the NRA ended by being, as a
contemporary observer noted, little more than &dquo;a bargain between
business leaders on the one hand and businessmen in the guise of
government officials on the other.&dquo;50 Obviously there was no &dquo;auto-
nomy of the state&dquo; in relation to capitalists within the Recovery
Administration. In consequence, economically powerful corporations
and established trade associations were able briefly to gain legal backing
for their own short-term interests, disregarding the legislated provisions
for labor to have its own union organizations and representatives on
code authorities and overriding the interests of smaller businesses or be-
leaguered competitors.51 Yet these short-term advantages came at a
price. Government regulation without state autonomy soon left busi-
nessmen quarreling among themselves, with the winners unable to en-
force their will except through cumbersome legal procedures, and with
the losers able to bring counterieverage on the NRA through Congress
and courts. (Indeed, the NIRA was eventually declared unconstitutional
in response to a suit brought by a small poultry-processing company!)
And the NRA codes, once captured by big business, simply functioned
to freeze production, guarantee monopoly prices to dominant firms,
and undermine general economic expansion. Arguably, a more auto-

49. Good accounts of NRA administration include Hawley, New Deal and Problem of
Monopoly, chap. 3; Schlesinger, Coming of New Deal, chap. 7; and Leverett S. Lyon et al.,
The National Recovery Administration: An Analysis and Appraisal, 2 vols. (Washington D.C.:
The Brookings Institution, 1935), chap. 4-9.
50. Quoted in Hawley, New Deal and Problem of Monopoly, pp 56-57.
51. On the economic and political consequences of the NRA, see ibid., pp. 66-71 and
chaps. 4-6; Chandler, America’s Greatest Depression, pp. 229-39; and Schlesinger, Coming of
New Deal. chaps. 8-10.
 177

nomous form of state regulation could have kept prices down and
facilitated expanded production. In any event, when businessmen
themselves became the state, government intervention could do little
more than reinforce and freeze the economic status quo, while simul-

taneously politicizing conflicts among businessmen.

The limited capacities of the U.S. federal government can be used
to explain the failure of the NIRA in a second way. Title II of the
act called for massive federal expenditures on public-works projects,
something that supporters felt would provide employment and help
stimulate industry through construction contracts and purchases of
materials. General Hugh Johnson fondly hoped to head both the NRA
and the Public Works Administration (PWA), and he envisaged quickly
spending the $3.3 billion PWA appropriation to help expand the
economy even as the industrial regulatory codes were put into effect.
Historians often imply that Roosevelt made a mistake in putting the
gung-ho General Johnson in charge of the NRA, while handing the
PWA to &dquo;Honest Harold&dquo; Ickes, who proceeded to spend his agency’s
appropriation very slowly and only on projects of unquestionable
soundness.52 Apparently, historians dream (a bit anachronistically)
of a quick, Keynesian fix to the Depression-if only the PWA had had
the right man as director. But this fails to take sufficient account of the
given administrative and political realities of the time. The administra-
tive means to implement a huge, speedy public-works program were
simply not available, as Herbert Stein points out in his remarks on the
proposals for $1 billion- to $8.5 billion-dollar programs that were
offered from 1929 on:
As proposals of amounts of money be spent for federal construction in a short
to
period, perhaps a year or two, these could not be taken seriously. The
suggestions
federal government could simply not raise its construction expenditures quickly
by one or two billion dollars a year, for instance, and have any structures to show
for it. Federal construction expenditures were only $210 million in 1930-a small
base on which to erect a program of several billion dollars. The larger proposals
were intended to finance expansion of state and local public works expenditures,
in addition to federal.... But even combined federal, state, and local construction
expenditures in 1930 were less than $3 billion.’3
Arguably, the Roosevelt administration should have thrown adminis-
trative regularity to the winds and, in the interests of promoting eco-
nomic recovery, simply handed public-works funds to businesses or

52. Schlesinger, Coming of New Deal, pp. 103-9, is the locus classicus of the Johnson-
versus-Ickes comparison.
53. Herbert Stein, The Fiscal Revolution in America (Chicago: University of Chicago Press,
1969), pp. 23-24.
178

to local governments. This might or might not have rapidly expanded

productive economic investments. But politically it would have been
disastrous. Virtually everyone inside and outside government at the
time believed in &dquo;balanced budgets,&dquo; and especially in a time of na-
tional economic crisis, federal expenditures were subject to close critical
scrutiny for signs of waste or corruption. In choosing the cautious
Harold Ickes to head the PWA, Roosevelt acted in the knowledge that
conspicuously wasteful or foolhardy public works expenditures could
put the entire program in danger in Congress and perhaps raise doubts
about other New Deal legislation as well, including relief expenditures.
The NIRA, in short, failed to regularize relationships among busi-
nessmen and failed to promote the economic expansion needed by the

capitalist class as a whole in significant part because there was little

autonomous administrative capacity in the U.S. national government
of the early 1930s. Insofar as Poulantzian theory tends to assume
that all capitalist states will automatically have this capacity, or will
rapidly generate it if it is needed, the theory becomes misleading.
Governmental capacities vary with the political histories of various
countries; in turn, these governmental capacities affect what can be
done for capitalist economies and for capitalists both in &dquo;nor~nal&dquo;
times and in crisis situations. Functional, adaptive state interventions
do not always occur, and a large part of the explanation for whether
they do or not, and for the exact forms of state interventions, lies in
the prior histories of the state structures themselves.
If Poulantzas’s functionalist theory overestimates the automatic
ability of capitalist states to unify, organize, and serve the class interests
of capitalists, it also underestimates the extent to which political
struggles and state actions in capitalist democracies can actually stimu-
late or accelerate challenges to capitalist prerogatives from below, rather
than merely averung challenges from below through minimal, non-
threatening concessions. For labor policies under the NIRA ended up
promoting the emergence, through bitter conflicts, of independent
industrial labor unions. These unions not only encroached upon the
formerly near-absolute control of capitalist managers over the work-
place, they also organizationally unified industrial workers to a greater
degree than ever before in U.S. history.
Of course, the INTRA as originally passed in 1933 was intended to
pacify labor, not’ to encourage industrial conflicts. But the actual
effect of the act was to intensify conflicts over its labor provisions
among workers, businessmen, and politicians.54 In part, the declared
54. See Chandler, America’s Greatest Depression, pp. 231-32; and, in general, the references
cited in note 49.
 179

goals of the NIRA were simply not implemented in the business-

dominated NRA. Labor representatives appeared on less than 10
percent of the industrial code authorities, and the probusiness NRA
administrators refused to disallow the company unions that managers
organized to circumvent section 7a’s declaration of labor’s right to
organize. In many industries, workers went on strike in efforts to
secure their rights under the NIRA. Indeed, the impact of the NIRA

upon labor was not simply the denial of promises in the actual ad-
ministrative practices of the NRA. The mere passage of the act raised
hopes among labor-union organizers and industrial workers, who
redoubled their efforts in the field. Given employer resistance, the
predictable result was an accelerating strike wave. Thus the NIRA
and the NRA, by their mere existence, tended to encourage and polit-
icize industrial labor disputes-hardly a &dquo;functional&dquo; outcome for
capitalists.
Finally, and perhaps most important of all, the NIRA set in motion
an effort to put the full legal backing of the state behind independent
labor unions. During the life of the NRA Senator Wagner failed to get
cooperation from businessmen, or backing from Hugh Johnson, for his
plan to have the National Labor Board supervise elections and certify
labor unions on a majority basis as representatives for workers in
negotiations with employers. But Wagner’s failure simply spurred
him to redouble his efforts to legislate this solution over business
opposition. And given the independent political leverage available to
members of the Congress, Wagner was able to pursue a policy-making
strategy at odds with the official attitude of the Roosevelt administra-
tion.
For the sake of organized labor, it was a good thing that Wagner,
was able to push ahead of official New Deal policies. If Roosevelt as

president or as head of the Democratic party had been able to control

all policy initiatives, the stance of the U.S. state toward labor in the
New Deal might have better conformed to Poulantzas’s theoretical
expectations. For FDR and his labor secretary, Frances Perkins, favored
only &dquo;paternalistic&dquo; concessions to labor, such as legislative measures
to regulate wages, hours, and working conditions. They were not
particularly friendly to organized labor; nor would they sponsor gov-
ernment measures to increase its poweL55 Roosevelt invariably wanted
to &dquo;balance&dquo; existing pressures from capital and labor, even though

capital’s economic and organizational power was vastly preponderant.

During 1934, Roosevelt intervened in the NRA’s consideration of
55. Edelman, "Sensitivity to Labor Interests," in Labor and the New Deal, ed. Derber and
Young, pp. 161-64, 178-82.
180

disputes in the auto industry, essentially backing up management’s

position in favor of &dquo;proportional representation&dquo; for company versus
independent unions, rather than supporting the policy of Wagner’s
National Labor Board in favor of unified representation through
majority election.56 Not until Wagner had carried his proposals to
the verge of sure Congressional victory did FDR endorse strong pro-
labor measures.
Robert Wagner politician bred in the New York Tammany
was a
Machine during period
a when the machine had begun to sponsor some
measure of social reform.5~ During his political career, Wagner devel-

oped ties to the American Federation of Labor and to liberal policy

associations. And as a senator during the 1920s, he was one of the
first congressmen to build an independent staff and to employ and
consult professional experts in drafting legislation. Wagner established
himself even before the Depression as a major formulator of national
economic policies, and he was unusually effective in piloting bills
through the Senate.
Thus Wagner had the policy-formulating capacity, the political
connections, and the legislative skills to promote what eventually be-
came the Wagner National Labor Relations Act. His frustrations with
the NRA sharpened Wagner’s perception of the need for strong, one-
sided legislation in favor of unions. And the looseness of presidential
control over Congress and of discipline on policy matters within the
Democratic party allowed him the political space to move aggressively
ahead of Roosevelt. From a capitalist point of view, this meant that the
U.S. political system not only inadvertently stimulated industrial
conflict, as an unintended effect of the NIRA, but also generated an
autonomous political effort, spearheaded by Wagner, to array state
power against capitalist prerogatives and preferences. This kind of
development does not seem to be envisaged by Poulantzas’s theory,
which predicts that the capitalist state will invariably act to make the
working classes less, not more, threatening to capitalists. Poulantzas’s
perspective probably underestimates the potential democratic respon-
siveness of elected politicians in all capitalist democracies. It certainly
underestimates for the U.S. political system of the 1930s both the
responsiveness of liberals such as Wagner and the autonomous room for
maneuvering that members of Congress could enjoy within the U.S.

56. Bernstein, Turbulent Years, pp. 172-85.

57. This paragraph (and other discussion of Wagner) draws on J. Joseph Huthmacher,
Senator Robert F. Wagner and the Rise of Urban Liberalism (New York: Atheneum, 1971);
and Leon H. Keyserling, "The Wagner Act: Its Origin and Current Significance," The George
Washington Law Review 29, no. 2 (December 1960): 199-233.
 181

&dquo;Tudor polity.&dquo; In state&dquo; did not act in a unified way

effect, &dquo;the
toward labor, and some elements within it were prepared to promote
an entirely new system of industrial labor relations for the USA.
This system would ultimately-a decade later-prove acceptable
to U.S. capitalists. But in the meantime, as the accompanying figure
suggests, New Deal labor legislation (especially section 7a and the
Wagner Act) tended to stimulate rather than dampen labor disputes.
From 1934 through 1939 labor disputes were primarily about union
recognition. Leftist scholars sometimes interpret union recognition
as a conservative goal, because they treat struggles for socialism or

total workers’ control of industry as implicit alternatives, or because

they anticipate the stabilizing functions that unions eventually came
to play in the U.S. political economy. But in the context of the U.S.
in the 1930s, the growth of (noncompany) unions threatened capitalist
prerogatives in the workplace, and the disproportionate expansion of
industrial as opposed to craft unions united broader sectors of in-
dustrial labor than at any previous period in U.S. history. Capitalists
(with very few exceptions) regarded all of this as threatening and
believed that the federal government was encouraging labor offensives
whose results were not fully predictable or necessarily controllable. For
the decade of the Depression, the capitalists were correct in their
analysis: &dquo;functional&dquo; outcomes were by no means certain.
How shall we draw the balance on Poulantzas’s political functional-
ism as an approach to explaining the New Deal? A very welcome
aspect of this variant of neo-Marxism is its stress upon what politics
and the state do ’’relatively atuonomously&dquo; for the economy, for
the capitalist class, and for (or to!) the noncapitalists, especially work-
ers. But, unfortunately, what the theory offers with one hand it im-

mediately takes back with the other, for it wants us to believe that the
state and politics always do just what needs to be done to stabilize
capitalist society and keep the economy going. If this were really true,
then state structures, state interventions, and political conflicts would
not really be worth studying in any detail, and politics as such would
have no explanatory importance. All that we would need to know
about the New Deal, or about the NIRA, would be that these were part
of a flow of history that eventually worked out splendidly for U.S.
capitalists. But, of course, this is not all we need to know, or all we
need to explain. To penetrate more thoroughly into the political dy-
namics of the 1930s, we need a perspective that pays more attention to
class and political conflict and assigns greater importance to the autono-
mous initiatives of politicians. Fred Block’s &dquo;The Ruling Class Does
Not Rule&dquo; meets these criteria, and we turn now to an exploration
of the strengths and limitations of this third neo-Marxist approach.
 182

CLASS STRUGGLE, STATE MANAGERS, AND THE NEW DEAL

Like Poulantzas, Block launches his attempt to build a theory of

the capitalist state with a critique of instrumentalism. Block accepts
the reality of capitalist influence in the political process but argues
that Marxists must &dquo;reject the idea of a class-conscious ruling class&dquo;
and posit instead &dquo;a division of labor between those who accumulate
capital and those who manage the state apparatus.&dquo;’8
Within this
division of labor, capitalists are conscious of the specific, short-term
economic interests of their firms or sectors, but &dquo;in general, they are
not conscious of what is necessary to reproduce the social order in

changing circumstances.&dquo;59 Capitalists do not directly control the

state, however, for the state is under the direction of the &dquo;state man-
agers,&dquo; defined as &dquo;the leading figures of both the legislative and
executive branches- [including] the highest-ranking civil servants, as
well as appointed and elected politicians.&dquo;60
Once we accept the idea
of a real division of control between the economy and the state, Block
asserts, &dquo;the central theoretical task is to explain how it is that despite
this division of labor, the state tends to serve the interests of the
capitalist class.&dquo;61 It is not sufficient simply to posit the &dquo;relative
autonomy of the state.&dquo; Instead, an adequate structural theory must
spell out causal mechanisms in two distinct areas. &dquo;It must elaborate
the structural constraints that operate to reduce the likelihood that the
state managers will act against the~ general interests of capitalists.&dquo;62
And the theory must also try to explain why, on occasion, state mana-
gers actually extend state power, even in the face of capitalist resist-
ance, in order to rationalize or reform capitalism. Rationalization and
capitalist reform, Block tells us, refer &dquo;primarily to the use of the
state in new ways to overcome economic contradictions and to facili-
tate the [nonrenressivel integration of the working class.&dquo;
According to Block, it is fairly easy to explain why state mana-
gers would normally be very reluctant to act against capitalist interests,
even without assuming that class-conscious capitalists run the state. The
state in a capitalist society does not directly control economic produc-
tion, and yet the state managers depend for their power and security
in office upon a healthy economy. The state needs to tax and borrow,
and politicians have to face re-election by people who are likely to hold
58. Block, "The Ruling Class Does Not Rule," p. 10.
59. Ibid.
60. Ibid., p. 8 (fn.).
61. Ibid., p. 10.
62. Ibid., p. 14.
63. Ibid., p. 7 (fn.).
 183

them responsible if the economy falters. Thus state managers willingly

do what they can to facilitate capital accumulation. Given that most
economic investment decisions are controlled directly by private
capitalists, state managers are especially sensitive to the overall state
of &dquo;business confidence,&dquo; that is, &dquo;the capitalist’s evaluation of the
general political/economic climate.&dquo; &dquo;Is the society stable; is the
working class under control; are taxes likely to rise; do government
agencies interfere with business freedom; will the economy grow? These
kinds of considerations are critical to the investment decisions of each
firm. The sum of all of these evaluations across a national economy can
be termed the level of business confidence.&dquo;&dquo; Normally, state man-
agers will not want to do anything that might hurt business confidence,
and so instead of pursuing social reforms or attempting to expand the
state’s role in the economy, they will confine themselves to formulating
policies that are generally supportive of capital accumulation and not
seriously objectionable to any major sector of the capitalist class.
But &dquo;if the state is unwilling to risk a decline in business con-
fidence, how is it then that the state’s role has expanded inexorably
throughout the twentieth century?,,6-5 Block is unwilling to accept the
answer offered by corporate-liberal instrumentalists. On the contrary,

Block holds that capitalists are almost by definition too short-sighted

initially to accept, let alone to promote, major reforms or extensions
of state power. Such changes come primarily in opposition to capitalist
preferences. And they mostly come when, and because, state managers
are strongly prodded to institute reforms by the working class. Class

struggle, says Block, pushes forward the development of capitalism.

It does this economically by raising wages and thus prompting capital-
ists to substitute machinery for workers. It does it politically by pres-
suring state managers to institute economic regulations and social
reforms.
According to Block, the biggest spurts forward in state activity
come during major crises such as wars or depression. During wars

capitalists cannot easily undercut state managers, and during depres-

sions the decline of business confidence is not such a potent threat.
Moreover, especially during economic crises, class struggle and pres-
sures from below are likely to intensify. Thus state managers may find

it expedient to grant concessions to the working class. Yet they will

do so only in forms that simultaneously increase the power of the
state itself. What is more, over the longer run, especially as economic
 184

recovery resumes or a wartime emergency ends, the state managers will

do the best they can to shape (or restructure) the concessions won by
the working class in order to make them function smoothly in support
of capital accumulation and existing class relations. Thus it can come to
pass that reforms and extensions of state power originally won through
&dquo;pressure from below&dquo; can end up being &dquo;functional for&dquo; capitalism
and accepted by many of the very capitalists who at first strongly
resisted the changes.
Block’s theory of the capitalist state, especially his explanation of
major thrusts of capitalist rationalization, is elegant and powerful.
In a number of ways, Block’s ideas add to our understanding of why
the New Deal occurred as it did. Where imprecisions or ambiguities do
remain in Block’s approach as applied to the New Deal, we can readily
pinpoint some promising lines for future theories about political
reforms and rationalization within capitalism.
Thinking first about the early New Deal-the &dquo;Hundred Days&dquo; of
legislation culminating in the NIRA-the Block perspective shares
some inadequacies with Poulantzian political functionalism, yet also

improves upon it in significant ways. To take the inadequacies first:

Block does not seem to do any better than Poulantzas in explaining
why the NIRA failed to achieve its objectives. Like Poulantzas, Block
pitches his theorizing at a high level of abstraction, talking about
capitalism in general, and does not investigate existing state structures
as constraints upon what state managers can do when they attempt to
facilitate capital accumulation (whether through reform or not).
Compared to Poulantzas, the strength of Block’s approach as (it
might be) applied to the early New Deal lies in Biock’s greater attention
to specific causal mechanisms. Instead of merely positing that the
state &dquo;must&dquo; intervene to save capitalism, Block suggests that U.S.
governments during the Depression were spurred to do all they could to
facilitate capitalist economic recovery because public revenues and
politicians’ electoral fortunes depended on such efforts. It also makes
sense in terms of Block’s theory that both the Hoover administration

and, at first, the Roosevelt administration tried to promote recovery

in close cooperation with capitalists. For Hoover and Roosevelt wanted
to revive business confidence and thus resuscitate private investments.66
What is perhaps less clear from Block’s perspective is why only
FDR and the Democrats (and not Hoover and the Republicans) were
willing to promote recovery specifically through increased state inter-
vention in the economy. We can, however, derive a kind of explanation

66. Stein, Fiscal Revolution, chaps. 2-5, passim.

 185

from Block’s theory-one that resembles Poulantzas’s tendency to root

the relative autonomy of the state from capitalists in the pressures
generated by &dquo;political class struggle&dquo; in favor of concessions for non-
capitalists. We can argue that the Democratic victory over the Repub-
licans in the elections of 1930 and 1932 was an expression, albeit
highly politically mediated, of class-based pressures from below on the
U.S. federal government. In response to this pressure, the argument
would go, the Democrats were urged, and enabled, to use state power
for reformist and regulatory efforts well beyond what the business-
dominated Republicans had been willing to undertake, even after the
crash of 1929. Since at first hopes were high for quick economic
recovery through revived business confidence, Roosevelt made every
effort to fashion his first &dquo;Hundred Days&dquo; of legislation (including the
bold National Industrial Recovery Act) not only to meet the reformist
demands of farmers, workers, and the unemployed, but also to meet
the preferences of many specific groups of capitalists and to enhance
the general confidence of businessmen as rapidly as possible. In short,
a judicious use of Block’s theoretical perspective can go a long way

toward explaining the special blend of popular responsiveness and

willingness to cooperate with capitalists that characterized the fledgling
Roosevelt administration in 1933-34.
Yet, of course, where Block’s theory really comes into its own is
in the analysis of the major social welfare and labor reforms of the
New Deal era. Using Block’s theory, there is no need to attribute
measures such as the Social Security Act and the Wagner Act either
to the far-sighted planning of the capitalist class or to the automatic
intervention of a capitalist state smoothly functioning to preserve
order and promote economic recovery. Instead, according to Block,
these measures were made possible by a conjunction of working-class
pressures with the willingness of state managers to increase their own
institutional power at a time when capitalist veto power was unusually
weak. The reforms provided benefits to many members of the working
class, strengthened the state in relation to the working class, and
increased the state’s capacity to intervene in the capitalist economy.
The eventual result was that working-class struggle ended up con-
tributing to the further development of American capitalism.
As a general &dquo;explanation sketch&dquo; of the causes of the major social
reforms of the New Deal and their eventual consequences, this is im-
peccable. Still it must be emphasized that at the very points where
Block’s class struggle theory of capitalist rationalization becomes most
analytically relevant, it also becomes quite vague. What exactly is
meant by &dquo;class struggle&dquo; or &dquo;working-class pressure&dquo; for reforms?
 186

How do varying forms and amounts of working-class pressure affect

political and economic outcomes? Equally important, what are the
likely interrelationships between working-class pressures and the
activities of state managers? Do the latter only respond, or are they
likely, under certain kinds of circumstances, to stimulate pressure
from noncapitalists as well?
Merely raising questions such as these suggests that to go from the
general &dquo;structural&dquo; dynamics outlined by Block to actual causal
explanations, one would need to specify cross-nationally and histori-
cally variable patterns in such things as: the occupational and com-
munity situations of the industrial working class; degrees and forms
of union organization; the actual and potential connections of working-
class voters and organizations to political parties; and the effects of
party systems and state structures on the likelihood that politicians’
responses to working-class unrest or demands will be reformist rather
than repressive.
In the case of the New Deal labor reforms, a number of conditions
influenced the nature and pattern of change. The U.S. industrial work-
ing class entered the Depression without strong unions and without a
labor-based political party. There was, however, a large semiskilled
manufacturing labor force consisting of relatively settled immigrants
and their children-a force ripe for mobilization into industrial
unions, The Democratic party, meanwhile, was &dquo;available&dquo; to come
to power in the Depression. And insofar as its leadership took a reform-
ist course, it was splendidly positioned to attract industrial working-
class votes into a national political party that, nevertheless, could and
would remain procapitalist and in many ways quite conservative.
Working-class pressure leading to reforms can entail very different
scenarios. It can mean that strong labor unions and a labor or social-
democratic or Communist political party impose a more or less anti-
capitalist reform program on (and through) the government. Or it can
mean that spontaneous working-class &dquo;disruption,&dquo; especially strikes,
forces specific concessions out of a reluctant government. 58
Neither of

67. On the industrial working class see David Brody, "The Emergence of Mass-Production
Unionism," pp. 223-62 in Change and Continuity in Twentieth Century America. ed. John
Braeman, Robert Bremner, and E. Walters (Columbus Ohio State University Press, 1964);
Irving Bernstein, The Lean Years (Baltimore: Penguin Books, 1966); Bernstein, Turbulent
Years ; Derber and Young. eds., Labor and the New Deal; and Ruth L. Horowitz, Political
Ideologies of Organized Labor: The New Deal Era (New Brunswick, N. J.: Transaction Books,
1978). I draw generally on these works in the discussions of labor that follow.
68. Such a portrayal of the process of change is offered by Frances Fox Piven and Richard
Cloward in Poor People’s Movements: Why They Succeed., How They Fail (New York: Pan-
theon Books, 1977), chap. 3. An excellent critique of Piven and Cloward’s analyses of events in
 187

these scenarios really fits what happened in the United States during
the 1930s. Obviously the industrial working class was organizationally
too weak for the first scenario to happen. And we have already seen
enough historical evidence on what happened with labor reforms in the
New Deal to see that the &dquo;disruption&dquo; scenario is also inaccurate.
Section 7a of the NIRA was originally achieved not by working-class
disruptions, but by political lobbying by the American Federation of
Labor and by the legislative efforts of Senator Robert Wagner. Once
enacted, even though not successfully implemented, section 7a as a
specifically political measure had a strong positive impact on labor-
union growth in two main ways. First, it encouraged union organizers
and emboldened rank-and-file workers with the belief that the national
government would now support their efforts to unionize and bargain
for economic gains. Second, it started Senator Wagner and others on
the road toward the formulation of the Wagner Labor Relations Act,
which ended up enforceably legalizing labor unions.
While labor’s experiences under the NRA also fueled a growing
strike wave in 1933-34, it cannot be plausibly argued that these strikes
directly produced the Wagner Act of 1935. By 1934, rising numbers of
labor disputes, and the inability of the National Labor Board (set up
under the NIRA) to resolve them, did cause Roosevelt and Congress
to become increasingly concerned with legislating new means for

managing such disputes But, as the figure on page 188 shows, the
immediate fruit of this concern, passed just as strikes came to a peak
in mid-1934, was not the Wagner Act. Rather it was Public Resolution
number 44, a measure cautiously designed to offend as little as possible
major industrialists and conservative politicians. 70 A National Labor
Relations Board was set up under the resolution, but it enjoyed no
greater power to make employers accept independent unions than had
the National Labor Board. Nevertheless, for whatever reasons, strikes
fell off significantly (see the figure on page 188) from the time of the
resolution until after the enactment of the Wagner bill a year later, in
July 1935.
If working-class pressure helped to produce the Wagner Act, it was
pressure registered not only through strikes but also through the major
Democratic electoral victories in Congress during the fall elections of

the 1920s is offered in Timothy George Massad, "Disruption, Organization and Reform: A
Critique of Piven and Cloward" (Undergraduate thesis, Harvard College, Social Studies, March
1978).
69. Bernstein, New Deal Collective Bargaining Policy, pp. 71-72, 77.
70. On the equanimity of a U.S. Steel vice president about Public Resolution no. 44, see
the quotation in ibid., p. 81.
 189

1934. This election strengthened liberals in the Congress and virtually

eliminated right-wing Republicans It came, moreover, just as business
opposition to the New Deal was becoming bitter and vociferous. Half a
year later, in June 1935, the Supreme Court declared Title I of the
NIRA-the early New Deal’s major piece of recovery legislation-un-
constitutional. It was at this extraordinary political conjuncture, one
marked by rhetorical class conflict as well as by FDR’s last-minute
conversion to the Wagner Act (both because of its Congressional
support and because he saw it as a partial substitute for the NIRA),
that the various politicians and labor-board administrators who had
all along pushed for pro-union legislation were finally able to carry
the day and put through Congress the single most important reform
of the New Deal era.
Block’s theory fits these historical developments, to be sure, but
with the important provisos that independent initiatives by liberal
politicians within the Democratic party were decisive in the entire
sequence from section 7a of the NIRA to the Wagner Act and that
the significant working-class pressures were registered electorally as
well as expressed in industrial unrest. The U.S. industrial working
class of the 1930s was not strong enough either to force concessions
through economic disruption alone or to impose a comprehensive
recovery program through the national political process. It depended
greatly on friendly initiatives and support from within the federal
government and the Democratic party. It got these not simply through
disruption, but because it represented an attractive potential for elec-
toral mobilization and an interest group that, among others, FDR
wanted to conciliate if possible. It also helped labor enormously that
business panicked about the New Deal after 1933 and, by going into
political opposition, gave liberals like Wagner even more space for
legislative maneuver.
Turning from the causes to the consequences of the labor reforms
of the New Deal, we arrive at other important queries that can be
raised about Block’s general explanation sketch. Do reforms enacted
partially in response to demands from below invariably end up &dquo;ration-
alizing&dquo; capitalism by allowing state power to be used &dquo;in new ways
to overcome economic contradictions and to facilitate the integration
of the working class?&dquo; What determines and, possibly, limits the capac-
ities of state managers to turn reforms and the concomitant enhance-

71. It is highly unusual for a party in power to gain in "off-year" Congressional elections,
yet the Democrats did this in 1934. See John M. Allswang, The New Deal and American Politics
(New York: Wiley, 1978), p. 30; and James T. Patterson, Congressional Conservatism and the
New Deal (Lexington: University of Kentucky Press, 1967), pp. 32-33.
190

ment of state power into effective supports for the capitalist economy
as awhole?
In his theoretical statements on these issues, Block is not only
imprecise; he actually points in contradictory directions. At various
places in his article, he asserts that state managers almost by definition
have the capacity to work for the general good of the capitalist class
and economy:
Unlike the individual capitalist, the state managers do not have to operate on the
basis of a narrow profit-maximizing rationality. They are capable of intervening
in the economy on the basis of a more general rationality. In short, their structural
position gives the state managers both the interest and the capacity to aid the
investment accumulation process
The more power the state possesses to intervene in the capitalist economy, the
greater the likelihood that effective actions can be taken to facilitate investment. 73

But, at other points, Block admits that the pattern in which working-class
pressures that ultimately enhances the state’s capacity to manage
capital accumulation &dquo;is not a smoothly working functional process
He rather weakly alludes to &dquo;time lags&dquo; and &dquo;friction&dquo; in the over-all
process, yet also offers some stronger antifunctionalist formulations:
There might, in fact, be continuing tensions in a government program between its
integrative interest and its role in the accumulation process. [S] ome concessions
to working-class pressure might have no potential benefits for accumulation and
might simply place strains on the private economy,.75
The increased capacity of state managers to intervene in the economy during
these periods [of crisis and reform] does not automatically rationalize capitalism.
State managers can make all kinds of mistakes, including excessive concessions
to the working class. State managers have no special knowledge of what is necessary
to make capitalism more rational: they grope toward effective action as best they
can within existing political constraints and with available economic theories

With this last statement, Block has come 180 degrees from the notion
that the state managers are &dquo;capable of intervening in the economy on
the basis of a more general rationality&dquo; because &dquo;their structural
position&dquo; gives them &dquo;both the interest and the capacity to aid the
investment accumulation process.&dquo;~~ Obviously, Block can’t have it
both ways. Either state managers enjoy the automatically given capac-

72. Block, "Ruling Class Does Not Rule," p. 20.

73. Ibid., p. 26.
74. Ibid., p. 23.
75. Ibid.
76. Ibid., pp. 25-26.
77. Ibid., p. 20.
 191

ity to rationalize capitalism or their capacities are determined by

&dquo;existing political constraints.&dquo; Given that Block footnotes the final
quote reproduced above with a reference to the New Deal, it is
apparent that he believes &dquo;existing political constraints&dquo; were im-
portant then
Indeed, during the New Deal, U.S. state managers could not auto-

matically use state power to ensure full economic recovery (and there-
by ensure social and political stability). Their efforts were always
channeled, shaped, and limited by existing state and party structures
not conducive to fully effective state interventions. We have already
seen the limitations that an absence of strong, autonomous adminis-
trative institutions placed upon the effectiveness of the National
Industrial Recovery Act of the early New Deal. Furthermore, even
after pressures from below and state initatives had combined to put
through the major social and labor reforms of the post-1934 New
Deal, &dquo;existing political constraints&dquo; still stymied the efforts of the
state managers to turn such measures and the political and state power
generated by them into effective means for promoting capitalist eco-
nomic recovery.
Block’s theory of capitalist rationalization through class struggle
and state-sponsored reforms silently assumes that social democracy is
the highest evolution of capitalism and that any major crisis of ad-
vanced capitalism resulting in important reforms will ipso facto entail
a breakthrough toward what C. A. R. Crosland once called a &dquo;full

employment welfare state.&dquo;79 Perhaps the pattern of state intervention

implied here could aptly be called &dquo;social-democratic Keynesian,&dquo;
because state interventions, regulatory and fiscal together, would pro-
mote private accumulation of capital and, at the same time, further the
social welfare, the full employment, and the political-economic leverage
(that is, through strong unions) of the working class. Now, the interest-
ing thing about the reformist phase of the New Deal (1935-38) is that,
although tendencies existed that might conceivably have added up to
a social-democratic Keynesian breakthrough, the existing governmental

and party structures so patterned political conflicts and so limited the

possibilities for political transformations as to prevent any such break-
through from actually occurring. Instead, the direct consequence of
78. Ibid., p. 26.
79. From C. A. R. Crosland, The Future of Socialism (London: Jonathan Cape, 1956),
p. 61. I have taken the expression via Andrew Martin, "The Politics of Economic Policy in the
United States: A Tentative View from a Comparative Perspective," Sage Professional Paper.
Comparative Politics Series (Beverly Hills: Sage Publications, 1973). My thinking has been
greatly influenced by Martin’s paper.
192

the labor reforms, and of the enhanced labor and liberal power that
accompanied them, was not economic recovery through social-demo-
cratic Keynesianism but increased social and political tension, leading
by 1938-39 to an insecure impasse for the domestic New Deal.
What were the tendencies present in the reformist New Deal that
pointed toward a social-democratic Keynesian breakthrough? For one
thing, as Block’s theory would suggest, the social and labor reforms of
the New Deal generated new nexes of political power and interest,
binding liberal-Democratic politicians into symbiotic relationships with
the welfare and relief agencies of the federal government and binding
both urban-liberal politicians and trade-union leaders into a symbiotic
relationship with another new organ of state power, the National
Labor Relations Board (NLRB). NLRB administrators achieved their
entire raison d’être through the spread of labor unions; they had a
natural institutional bias (as well as a legal mandate) in favor of pro-
tecting all legitimate unions and union drives Labor unions, in
turn, certainly benefited from the protections thus afforded them.
Union membership had declined to below 3 million in 1933 but started
rising thereafter and shot up, especially from 1935, to over 8 million
in 1939.81 Both labor unions and the NLRB depended upon liberal-
Democratic support in Congress and within presidential administra-
tions. And, of course, liberal-Democratic politicians benefited from
monetary contributions and from the electoral mobilization of workers’
votes. This mobilization was organized, from 1936 on, through political
committees established by union leaders grateful for benefits from the
Roosevelt administration and hopeful that further gains could be made
for workers (including those unemployed and on relief) through the
strengthening of liberal forces in the Democratic party.82 Indeed, it
did seem that liberal Democrats, workers, unemployed people, people
on welfare, and federal administrators could all work together fruitfully
after 1935-36. Voices like Robert Wagner’s called for economic recov-
ery through bolstering the spending power of the working class. Urban
liberals in general advocated increased welfare and public works ex-
penditures and proposed substantial new efforts in public housing
Another major tendency that might have helped produce a break-
through to social-democratic Keynesianism came from the president
80. Edelman, "New Deal Sensitivity to Labor Interests," in Labor and the New Deal,
ed. Derber and Young, pp. 170-72.
81. Derber and Young, eds., Labor and the New Deal, pp. 3, 134.
82. On unions’ political activities, see esp. J. David Greenstone, Labor in American Politics
(New York: Vintage Books, 1970), chap. 2.
83. Huthmacher, Wagner and Urban Liberalism, chaps. 12-13.
 193

himself. After the 1936 election, Roosevelt introduced an executive

reorganization plan drafted by a committee of academic experts on
public administration. As drafted, this plan was well designed to over-
come many built-in obstacles to any strategy (whether social-Demo-
cratic or not) of coordinated state intervention in the society and
economy.84 The reorganization plan called for the consolidation of
federal administrative organs (including formerly independent regula-
tory bodies) into functionally rationalized, centrally controlled cabinet
departments. It would have enhanced presidential powers of planning
and policy coordination by: increasing executive versus Congressional
control over the expenditure of budgeted federal funds; establishing a
National Resources Planning Board for comprehensive, long-range
planning and coordination of federal programs; and greatly expanding
the White House staff directly responsible to the president. All of these
measures, taken together, would have gone a long way toward reversing
the historically inherited fragmentation of the U.S. federal administra-
tion. Since the start of the New Deal, the U.S. federal bureaucracy had,
of course, expanded significantly. But the expansion had been piece-
meal and, given the competing lines of authority between Congress and
the president and the paucity of institutional means of executive
coordination available to Roosevelt, the expanded federal government
was becoming increasingly unwieldy.
Given that FDR and the Democrats were re-elected by huge popular
margins in 1936, the time looked ripe to enhance the Roosevelt adminis-
tration’s ability to institutionalize and expand its reforms. And, to
engage in a bit of counterfactual historical speculation, imagine that
such a system not only could have been achieved but also that it could,
in turn, have been harnessed to a disciplined and thoroughly liberal
Democratic party, one devoted to the pursuit of full economic recovery
through social spending. Then the institutional and political conditions
for a true social-democratic Keynesian breakthrough in the U.S. would
have been established. But, of course, things did not turn out this way,
largely because of the nature of the Democratic party and the structure
of the U.S. national government. The Democratic party absorbed,
contained, and rendered partially contradictory the political gains of
organized labor and liberals in the 1930s, and the national government
blocked F1~R’s comprehensive scheme for administrative reorganiza-
tion.

84. Richard Polenberg, Reorganizing Roosevelt’s Government: The Controversy over

Executive Reorganization 1936-1939 (Cambridge: Harvard University Press, 1966). Polenberg
writes (p. 26): "A powerful president equipped with the personnel, planning, and fiscal control
necessary to implement his social program—this was the Committee’s aim."
194

To begin with the failure of comprehensive reorganization, Roose-

velt’s plan met defeat in Congress above all because it threatened to
disrupt well-established patterns of institutional power.~ Senators and
representatives were jealous of Congress’s traditional power to &dquo;pre-
audit&dquo; the expenditures of federal agencies, for through such power
individuals and committees in Congress could influence programs
affecting their state or local constituencies. In addition, both members
of Congress and administrators of federal agencies were very nervous
about administrative reorganizations that would disrupt existing sym-
biotic relationships among Congress, bureaucrats, and organized interest
groups in the society at large. Predictably, the most intense opposition
came from conservatives, for these people worried about the kinds of

programs Roosevelt (and the New Dealers in general) might generate

through a reorganized federal executive. But not only conservatives
opposed the plan; some of Roosevelt’s own Cabinet members, acting
like any traditional U.S. cabinet heads jealous of their bureaucratic
domains, opposed reorganization or failed to support it. Interestingly
enough, the &dquo;urban liberal&dquo; par excellence, Senator Robert Wagner,
also opposed the reorganization plan. He was worried about Congres-
sional prerogatives and the administrative independence of the National
Labor Relations Board.~
The Democratic party also functioned as an obstacle to a social-
democratic Keynesian breakthrough in the 1930s. The most basic
fact about the New Deal Democratic party was simply that it worked
pretty much as all major U.S. political parties have done since about
the 1830s.8~ U.S. parties do not formulate explicit, coherent national
programs that their members in Congress are then disciplined to sup-
port. Rather U.S. parties &dquo;aggregate&dquo; blocs of voters through very
diverse appeals in different parts of the country. Their overriding
function is to compete in S)g$e-1I1embO-COD15fibLienOy, majority-take-all
elections in order to place politicians in office from the local to the
state to the national level. Once in office, parties exhibit some disci-
pline (especially on procedural matters), but members of Congress are
85. Ibid., pts. 2, 3. Eventually, parts of the original executive plan were passed by Con-
gress, but these parts simply strengthened the president’s staff and modified the cabinet depart-
ments somewhat. The changes accomplished were piecemeal-far from the original plan for

comprehensive reorganization.
86. Ibid., pp. 139-40. Polenberg cites Keyserling, "Wagner Act," pp. 203, 207-8, 210-12,
on Wagner’s views on the NLRB’s administrative location. See also Huthmacher, Wagner and
Urban Liberalism, pp. 243-45.
87. Theodore J. Lowi, "Party, Policy, and Constitution in America," in The American
Party Systems, ed. William Nisbet Chambers and Walter Dean Burnham, 2nd ed. (New York:
Oxford University Press, 1975), pp. 238-76.
 195

remarkably free as individuals (or ad hoc coalitions) to pursue whatever

legislation or whatever administrative measures they believe will appeal
to local constituents or to organized groups of financial contributors
and voters. Members of Congress may, of course, be unusually attentive
during periods of crisis to directives from the president and from House
and Senate leaders; this was the case during the &dquo;Hundred Days&dquo; of
1933. But such extraordinary coordination rarely lasts for long, and there
is still unlikely to be a disciplined pursuit of an over-all party program,
if only because U S. presidents, senators, and representatives are elected
in diverse ways and by differently structured constituencies.
These institutional givens meant that as liberals, especially from
northern urban areas, made gains within the national Democratic party,
they could hope to gain some legislative and administrative leverage for
their various constituents. They could not expect, however, to formu-
late comprehensive national strategies for a Democratic president and
Congress to enact. The existing Congressional-executive system would
frustrate any such efforts. Even more important, the Democratic
party could maintain a semblance of national unity only by avoiding
too many head-on clashes between its entrenched, overwhelmingly
conservative southern wing and its expanding liberal urban-northern
wing. There were also nonsouthem rural interests in the party, and
their representatives were willing to support government programs for
farmers but were usually not sympathetic to the needs of urban areas or
labor.
Franklin Delano Roosevelt, an immensely popular President,
gained the Democratic nomination in 1932 through southern and
western support. By 1936 he also enjoyed strong support in urban
areas and from organized labor.~ But it became increasingly difficult

for Roosevelt to manage all of these diverse interests under the Demo-
cratic umbrella. For a brief time in 1938, he attempted to make the
party more consistently liberal by &dquo;purging&dquo; various very conservative
Democrats who had opposed New Deal reforms.89 But machines or
special agglomerations of organized interests at local and state levels
control nominations within U.S. parties, which are &dquo;national&dquo; only in
label and for the purpose of electing presidents. So FDR’s purge failed,
and liberalism-even as an attitude, let alone as a comprehensive pro-

88. Allswang, New Deal and American Politics, chaps. 2-4, passim.
89. On the attempted purge and the reasons for its failure, see ibid., pp. 121-26; and
Patterson, Congressional Conservatism, chap. 8. On the paradoxical explanation for the one
apparent success that FDR had in purging a conservative, see Richard Polenberg, "Franklin
Roosevelt and the Purge of John O’Connor. The Impact of Urban Change on Political Parties,"
New York History 49, no. 3 Uuly 1968): 306-26.
196

gram of recovery and reform-could not take over the Democratic

Party. z

Moreover, given the way that the operations of the Democratic

party (and the Republican party) intersected with the operation of
Congress, liberals (especially urban liberals) were unable to translate
electoral support into truly proportional Congressional leverage. Iron-
ically, efforts by labor and liberals to elect Democrats could even
hurt politically, as Edelman explains:
Both the House and the Senate greatly underrepresent urban areas. The emphasis
on local needs therefore means that labor groups are underrepresented, for con-
stituency pressures are more often exerted disproportionately on behalf of farmers
and dominant local industries rather than on behalf of worker residential areas in
the cities. This bias is greatly emphasized because the constituencies that are
&dquo;safe&dquo; in the sense of predictably returning incumbents to office are also to be
found almost entirely in predominantly rural areas: the South, Vermont, Maine,
and some parts of the Middle West. Because an individual’s influence in Congress
depends so enormously on his seniority there, the congressmen from these areas
gain disproportionately in influence when their parties win control. Thus a hard-
fought Democratic victory in Pennsylvania, New York, Illinois, and California
sufficient to enable the Democrats to organize the House and Senate does not
benefit these states nearly as much as it benefits Alabama, Mississippi, Florida and
other states in which there is little hard campaigning. To this extent AFL and
CIO activity in congressional campaigns in the urban areas places some of the
position
anti-labor rural congressmen in a stronger

By the very nature of politics, conflicts intensify as a formerly weak

interest in a situation of diverging interests grows stronger. As labor and
liberal interests gained ground in U.S. politics during the 1930s, con-
servatives, whether businessmen or rural interests, were bound to op-
pose them, whatever the institutional arrangements had been. Yet
above and beyond this inevitable opposition, the structure and opera-
tions of Congress and the Democratic party facilitated the formation of
a &dquo;conservative coalition&dquo; to obstruct or modify many liberal New Deal

measures, even as these arrangements made it difficult for liberals to

become a national policy-making force.91 Nor, as we have seen, could
liberal programs supported by members of Congress always count on
solid support from Roosevelt or his administration. For Roosevelt and
his administrators had to worry about maintaining broad support in
Congress and about balancing pushes and pulls within the Democratic
party as a whole.

90. Edelman, "New Deal Sensitivity to Labor Interests," in Labor and the New Deal,
ed. Derber and Young, pp. 185-86.
91. Patterson, Congressional Conservatism, pp. 334-35.
 197

Given the structures of politics within which they had to operate,

liberals during the New Deal had to settle for doing what pays off in the
U.S. political system. They carved out domains of legislation and ad-
ministration favorable to well-organized constituents and then did their
best to defend these against bureaucratic encroachments or opposition
within Congress or from the administration.g2
In sum, &dquo;existing political constraints&dquo;-specifically, the U.S. gov-
ernment and political parties of the 1930s-limited the possibilities for

politicians to use the new political energies and the new domains of
state power generated by the reformist New Deal for the successful
resuscitation of capitalist accumulation through state spending for
(new or much expanded) domestic social programs. By the late 1930s,
full economic recovery still eluded the national economy under the
New Deal. In the absence of recovery, who could say where U.S. poli-
tics might go? Conservatives in Congress were gaining political strength
and were increasingly able and willing to block policy initiatives coming
from &dquo;urban liberals&dquo; inside and outside the government and to whittle
down welfare and public-works measures sponsored by the president.9~
Labor unions were still not accepted by many industrial employers, and
the NLRB and the Wagner Act were coming under increasingly vocifer-
ous political attacks
Yet, of course, the basic social and labor reforms of the liberal
New Deal did survive. And full recovery, indeed, spectacularly accel-
erated growth, was achieved by the U.S. economy. But the intrana-
tional structural forces specified in Block’s theory, that is, working-class
pressures plus the initiatives of state managers, were not alone respon-
sible for this outcome of the 1940s and after. Of the three neo-Marxist
theoretical perspectives examined in this paper as a whole, only Block’s
repeatedly alludes to ways in which transnational structures and con-
junctures affect the course of domestic politics in advanced capitalist
nations. Even Block, however, fails to accord such transnational factors
the systematic explanatory weight they deserve. It is not only the
interplay of capitalists’ economic decisions, working-class pressures, and
state managers’ initiatives that shapes political conflicts and trans-

92. On the resulting patterns of power see esp. Morton Grodzins, "American Political
Parties and the American System," Western Political Quarterly 13, no. 4 (December 1960):
974-98; and Grant McConnell, Private Power and American Democracy (New York: Vintage
Books, 1966).
93. Patterson, Congressiortal Conservatism, chaps. 6-10; and Richard Polenberg, "The De-
cline of the New Deal," pp. 246-66, in The New Deal, ed. Braeman, Bremner, and Brody,
vol. 1, The National Leuel.
94. Ibid., pp. 315-24; and Huthmacher, Wagner and Urban Liberalism, chap. 14.
 198

formations in advanced capitalism. International economic and politico-

military relations also matter.
The case of the New Deal certainly dramatizes the importance of
both international economics and international politics. Without the
virtual collapse of the international monetary system and the sharp
contraction of international trade in the early years of the Depression,
it is highly doubtful that &dquo;political space&dquo; could have opened up in the
United States for the pursuit of economic recovery and reforms
through greater state intervention. Even more important was the
impact at the end of the New Deal of the coming of World War II,
starting in Europe. Everyone knows that &dquo;military Keynesianism&dquo;-
growing federal expenditures on military preparedness and then war-
jolted the national economy out of the lingering Depression at the end
of the 1930s.95 What is perhaps less well known are the crucial ways in
which the turn toward military and foreign policy preoccupations and
then into war overcame many of the domestic social and political
impasses of the later New Deal.96 There were several key developments.
As the Roosevelt administration planned for (and later implemented)
wartime industrial mobilization, it sought a rapprochement with big
businessmen and tacitly agreed to cease pushing for new labor and
social reforms in return for business cooperation. The institutionaliza-
tion of labor unions was, at the same time, solidified during the war. In
return for a no-strike pledge from union leaders, the government
required employers to deal with unions and facilitated the enrollment
of union members in industries doing war production. Finally, the
issues of politics in Congress were partially transmuted from the do-
mestic and economic questions that exacerbated conflicts between
rural-conservatives and urban-liberals into more foreign-policy and
military related issues that averted (or delayed) some of the old dis-
putes over social reform. Reformers cotua no longer use economic crisis
as a rationale for social welfare measures. Instead, both liberals and
conservatives had to maneuver with the symbols of national unity
evoked to sustain the wartime efforts at home and abroad.
What all of this, taken together, amounted to was not at all a
defeat or a roll back for the reforms of the New Deal. Rather, most
of the reforms were consolidated and retained, but within a new na-
tional political context in which business and government were again

95. Robert L. Heilbroner, with Aaron Singer, The Economic Transformation of America
(New York: Harcourt, Brace Jovanovich, 1977), pp. 205-7.
96. David Brody, "The New Deal and World War II," pp. 267-309 in The New Deal, ed.
Braeman, Bremner and Brody, voL 1, The National Level. See also Richard Polenberg, War and
Society: The United States, 1941-1945 (Philadelphia: J. B. Lippincott, 1972).
 199

reconciled. Capitalists learned to live with and use many aspects of the
New Deal reforms, even as they retained leverage in the U.S. political
system to counter the power of organized labor and to limit future
advances by liberals and labor. The full economic recovery, the new
governmental tasks, and the new political climate brought by the
war, all were important in allowing this denouement. Without the
timely arrival of the war, the basic reforms of the New Deal might
not have survived-and they might not have proved so &dquo;functional&dquo;
for U.S. capitalism.

CONCLUSION

At the outset I declared three basic aims for this exploratory essay:
to show that some current lines of neo-Marxist analysis are more
promising than others for explaining the New Deal as an instance of
political conflict and transformation within advanced capitalism; to
argue that, for such explanatory purposes, no existing neo-Marxist
approach affords sufficient weight to state and party organizations as
independent determinants of political conflicts and outcomes; and to
sketch some of the ways in which U.S. political institutions shaped and
limited the accomplishments of the New Deal. In closing, I shall sum-
marize my conclusions on each of these major themes and point out
important areas for further investigation and critical discussion.
The evidence of the New Deal reveals the basic inadequacy of those
lines of neo-Marxist reasoning that treat political outcomes in advanced
capitalism as the enactments of a far-sighted capitalist ruling class or
as the automatically functional responses of the political system to the
needs of capitalism. Yet there are also more promising lines of analysis
within neo-Marxism. Those theorists who assign explanatory impor-
tance to class struggle recognize that conflicts between capitalists and
noncapitalists affect politics in capitalist societies. They also recognize
that politicians (especially in liberal democracies) must respond to the
demands of noncapitalists as well as to the demands of capitalists and
the conditions of the economy. Moreover, those theorists who refer
to the &dquo;relative autonomy of politics&dquo; or who assign &dquo;state managers&dquo;
an independent explanatory role are moving toward an approach that

can take seriously the state and parties as organizations of specifically

political domination, organizations with their own structures, their

own histories, and their own patterns of conflict and impact upon
class relations and economic development.
Nevertheless, so far, no self-declared neo-Marxist theory of the
capitalist state has arrived at the point of taking state structures and
 200

party organizations seriously enough. 97 Various ways of short-circuiting

political analyses have been too tempting. Political outcomes are
attributed to the abstract needs of the capitalist system, or to the will
of the dominant capitalist class, or to the naked political side-effects of
working-class struggles. It is often assumed that politics always works
optimally for capitalism and capitalists, leaving only the &dquo;how&dquo; to be
systematically explained. Even those neo-Marxists who have tried to
incorporate state structures into their modes of explanation tend to
do so only in functionalist or socioeconomically reductionist ways.
What is more, almost all neo-Marxists theorize about &dquo;the capitalist
state&dquo; in general, thus attempting to explain patterns of state inter-
vention and political conflict in analytic terms directly derived from a
model about the capitalist mode of production as such.
But capitalism in general has no politics, only (extremely flexible)
outer limits for the kinds of supports for property ownership and
controls of the labor force that it can tolerate. States and political
parties within capitalism have cross-nationally and historically varying
structures. These structures powerfully shape and limit state inter-
ventions in the economy, and they determine the ways in which class
interests and conflicts get organized into (or out of) politics in a given
time and place. More than this, state structures and party organizations
have (to a very significant degree) independent histories. They are
shaped and reshaped not simply in response to socioeconomic changes
or dominant-class interests, nor as a direct side-effect of class struggles.
Rather they are shaped and reshaped through the struggles of politi-
cians among themselves, struggles that sometimes prompt politicians to
mobilize social support or to act upon the society or economy in
pursuit of political advantages in relation to other politicians. In short,
states and parties have their own structures and histories, which in turn
have their own impact upon society.
The New Deal is a difficult case, almost ideal for developing a
macro-analytic perspective that points to all of the necessary factors
to be considered in explaining transformations of the state and politics
within advanced capitalism. Real political changes occurred in the

97. There are, however, some very promising pieces of comparative-historical analysis,
including Martin, "Politics of Economic Policy in the United States"; Ira Katznelson, "Con-
siderations on Social Democracy in the United States," Comparative Politics 11, no. 1 (October
1978): 77-99; Peter J. Katzenstein, ed., Between Power and Plenty (Madison: University of
Wisconsin Press, 1978), esp. conclusion; and, perhaps the nicest example of all of an approach
that combines class and political-structural analysis, Susan S. Fainstein and Norman I. Fain-
stein, "National Policy and Urban Development," Social Problems 26, no. 2 (December
1978): 125-46.
 201

New Deal: the inclusion of some popular interests in a more interven-

tionist &dquo;broker state,&dquo; the incorporation of the industrial working
class into the Democratic party, and the expansion of urban-liberal
influence within the party. But the political changes that occurred did
not please capitalists or ensure capitalist economic recovery or con-
solidate the basic reforms that were enacted-not until World War II
intervened to transform the over-all political and economic context.
By using various neo-Marxist explanatory approaches as foils for
probing the politics of the New Deal, I have argued that many of the
limitations on effective state intervention and on liberal reforms in the
New Deal can be traced to the existing U.S. national administrative
arrangements, governmental institutions, and political parties. These
same structures, moreover, shaped the piecemeal reforms and the

partially successful efforts to proffer relief and to promote recovery

in response to the Depression. Capitalists, industrial workers, and farm-
ers certainly helped to shape and limit the New Deal, as did the con-
tours of the massive economic crisis itself. But economic and class
effects were all mediated through the distinctive structures of U.S.
national politics. The immediate changes were not fully intended by
anyone, were not consistently in conformance with the interests of any
class, and were not smoothly functional for the system as a whole.
But the changes did make sense as the product of intensified political
struggles and undertakings within given, historically evolved structures
of political representation and domination. Such structures are the
keys to any satisfactory explanation of the New Deal-and to other
episodes, past, present, and future, of political response to economic
crisis within capitalism.