Paul Pierson

The Scope and Nature of Business Power Employers and the American Welfare State,

1900-1935

ZeS-Arbeitspapier Nr.14/95

Zentrum für Sozialpolitik Universität Bremen

Postfach 330440 0-28334 Bremen

Prof. Paul Pierson, Russell Sage Foundation, New York, was a visiting research fellow at the

Centre Far Social Policy Research in July/August 1995.

Contents

I. The Analysis of Business Power .... ..... ..... ... ... ........ .......... ................... ................ ..... 8

11. Business Power and the Formation of the American

Welfare State .... ........ ... .... ... ..... ...... ... .. .... ... ..... ..... ................. .................................. 18

111. Conclusion: Politics, Markets and Social Policy .................................... ....... .... 45

Abstract

The argument of this essay is that Ilew illstitutiollalism offers the analytical tools to free the

discussion of business power from the intellectual cul-de-sac reached by the late I 970s. In the

first section theoretical arguments are developed about the study of political power and about

the nature of business power in advanced industrial societies. In the second section these

arguments are applied to a topic of intense scholarly controversy: the sources of the American

welfare state's peculiar early development. The key is to understand the interactions among

three factors: the intensely fragmented character of American institutions, the instrumental and

structural power of private corporations, and the transformations of state-market relations

induced by the shock-wave of the Great Depression.

5

The analysis of business power in advanced industrial democracies has reached an impasse. Within

political science, and especially in the fields of American and comparative politics, there was a

vigoraus discussion of the sources and scope of business influence during the I 960s and 1970s. With

lhe growing interest in "the state" and institutions, now commonly known as the "new

institutionalism" , this earlier discussion was allowed lO lapse. 1 Ironically, this shift in intellectual

orientations occurred at precisely the time when observers ranging from international relations

scholars to journalists c1aimed that the power of business was rapidly increasing . There is a clear need

to revisit the business power debate and restore the study of political influence -- the question of who

governs -- to a central place in political sc ience.

The argument of this essay is that rather than constituting a theoretical alternative to a focus on

business influence, the new institutionalism offers the analytical tools to free the discussion of

business power fram the intellectual cul-de-sac reached by the late I 970s . In short, this essay seeks to

contribute to the development of an institutional political economy.2 Rather than simply substituting a

"some of both" argument for the traditional "either/or" formulation, however, I argue for a set of

spec ific propositions about how political in stitutions mediate the extent of business influence over

publ ic policy.

The essay proceeds in two stages. The first section develops theoretical arguments about the study of

political power and about the nature of business power in advanced industrial democracies. After

briefly reexamining earlier debates over business influence, I show how insights drawn from

institutionalist analysis provide the basis for so me significant advances. In particular, institutional

analysis can help to c1arify some of the indirect as weil as direct paths of business influence, and to

suggest why the scope of this influence might shift over time.

The second section of the essay applies these arguments to a topic of intense scholarly contraversy:

the sources of the American welfare state's peculiar earl y development. The nature and extent of

business influence on this pracess has been the subject of considerable dispute. While so me analysts

have treated business as the central actor, a number of recent institutionalist studies have sharply

criticized this claim. Both positions, [ argue, have some merit. "New institutionalists" have

2

There has been only one major study in American poli tics in the past decade that takes business inf]uence

as its central subject: David Vogel's Fluctuating Fortunes: The Political Power of Business in America (New York : Basic Books), 1989. This is an effort that is already underway in discussions of Western European politics, though that literature has paid Iittle attention to my central cancern: the institutional determinants of business influence. See for example Peter Hall . Governing the Economy (Oxford: Polity), 1986; and David Soskice, "The Institutional lnfrastructure for International Competiti veness: A Comparative Analysis of the UK and Germany," in A.B. Atkinson and R. Brunetta, eds., The Economics of the New Europe (London: Macmillan), 1991 , pp. 45-66.

6

persuasively argued that business lobbying alone cannot explain many aspects of "American

exceptionalism," and that the peculiarities of political institutions played an important role in shaping

the American welfare state. Yet "new institutionalist" accounts are themselves unsatisfying. Absent

any orientation toward political economy the analysis of social policy seems strangely disembodied.

Political activity in early twentieth-century America was embedded in a vast market economy. The

resources of private actors at least rivaled those available to public authorities.3 Major social policy

initiatives required substantial interventions in the economy and had considerable implications for

both the financial position and authority of powerful private actors. Despite the elegance of many new

institutionalist accounts, there remains a fundamental implausibility to analyses that deal with such

prominent economic issues while exhibiting such limited attention to economic interests.4

How are we to explain the origins of America's exceptional welfare state? The key, I argue, is to

understand the interactions among three factors : the intensely fragmented character of American

political institutions, the instrumental and structural power of private corporations, and rhe

transfonnation of state-marker relations induced by the shock-wave of the Great Depression. The

particular design of American political institutions has had profound effects on the exercise of

business influence. Prior ro the New Deal, rhe United States was an extremely decentralized polity.

This institutional environment was an ideal setting for the exercise of what has been called the

structural power of business. In an institution al environment where relatively small political

jurisdictions shared authority over an increasingly integrated economic market, concems about capital

mobility effectively limited state interventions to those that business either favored or regarded as

essentially harmless.

During the New Deal, however, the site of social policy activity moved to the national level.

Combined with the fact that business confidence had already plummeted, this institutional shift

dampened fears of capital mobility and hence diminished the structural power of employers. Business

influence then largely depended upon the application of instrumental power in a volatile political

climate. The unusual mobilization of popular pressures during the New Deal meant that employers

faced considerable political competition. This popular mobilization, along with the highly-fragmented

3

4

"Large private companies [in the early twentieth century] employed far more people and spent much more Illoney than the largest and wealthiest state government. One railroad with offices in Boston employed 18,000 people rand] had gross receipIs of about $40 million a year ... At the time, the State of Massachusetts employed only 6,000 people land] spent about $7 million .. ."' Edward Berkowitz and Kim McQuaid, Creating the Welfare State: The Political Economy of Twentieth-Century Reform (New York: Praeger), 1980, p. 25. An otherwise thoughtful essay by Ann Shola Orloff, a prominent institutionalist, is instructive. In a detailed review of early American social poliey, the economic consequences of various proposals receive no discussion. Business groups are simply lumped together in the general cat.egory of "elites", and opposition to major extensions of the social wage is attributed exclusively to fear of patronage. Ann Shola Orloff, "Gender in Early U.S. Social Policy," Journal of Policy History, Vol. 3, No. 3, 1991. pp. 249-281.

7

nature of national institutions, limited the business community's overall ability to shape legislation. At

the same time, this setting strengthened particular factions of business with leverage over institutional

veto points.

Thus both the scope and nature of business influence was radically different in the two periods. As Ira

Katznelson has observed, "the politics of social welfare in the United States is hardly fixed: it

expands, contracts, and shifts its axes of conflict over time."5 Indeed, the current analysis suggests

that a framework focusing on economic actors and political institutions need not be static. What

stands out in the early development of the American welfare state is not overwhelming business

strength or weakness, but the marked variation in business influence over time.6 Drawing again on a

theme emphasized by new institutionalists, but linking it to a focus on the role of economic structures,

I stress the importance of treating the Depression itself as a "critical juncture" in the development of

American social policy.7

Discussions of American social policy history often have had an almost teleological bent, seeing the

1930s as the inevitable moment when the United States finally began to develop a modern welfare

state. The common description of the United States as a "Iaggard" implies that the U.S. was only

"Iate" in embarking on roughly the same course travelled by other industrial countries. But the

Depression was not simply the moment when the American welfare state partly "caught up" with

European practices; it was an upheaval that fundamentally altered the interplay of politics and

markets. Major changes in the functioning of the private market economy and the operation of key

political institutions opened the door to dramatic policy reforms. This economic cataclysm greatly

diminished the scope of business power. Government officials, backed by an unusually-mobilized

electorate, were able to set social policy development on a new path. Absent the Depression, there is

little reason to think that American social policy would have evolved on anything like the course it

followed from the New Deal onwards.

The Social Security Act was a watershed in the development of the American state; six decades later,

it remains a defining feature of the American political economy. Placing the Act's enactment in a

broader historical context reveals that both before and after 1929 institutional features mediated the

expression of business power in crucial ways, and hence helped 10 shape the peculiar structure of the

American welfare state. How that mediation occurred, however, changed drastically between the two

5

6

7

lra Katznelson, "Rethinking the Silences of Social and Economic Policy," Political Science Qual1erly, Val. 101 , No. 2, 1986, p. 308. This analysis thus shares the thrust of Vogel's argument in Fluctuating Fortunes, although I place more emphasis on the fole of institutions in general (and federalism in particular) in mediating business power. On critical junctures see for example Stephen A. Krasner, "Sovereignty: An lnstitutional Perspective," in James A. Caporaso, ed., The Elusive State: International and Comparative Perspectives (Newbury Park, CA: Sage Publications, 1989), pp. 69-96.

8

periods. While business interests were far from irrelevant in the 1930s, the introduction of Social

Security depended on extraordinary circumstances that placed business actors on the defensive. The

institutional and economic conditions that had ensured a privileged position for business interests

ceased to operate in the mid-1930s. Depression-induced changes in the functioning of institutions

opened a window for reform.

I. The Analysis of Business Power

One of the striking features of the debate over the role of business in the development of social policy

is the absence of serious discussion of how political influence is exerted and wh at kinds of evidence

are relevant to judging the extent of a group's power. Analysts have tended to limit their theoretical

discussions to an outline of alternative independent variables (including business influence) that might

determine policy outcomes, before proceeding directly to a discussion of the historical record. Yet

influence is an extremely tricky concep!. In fact, much of the debate over the American welfare state

has misfired because it has failed to pay sufficient attention to methodological issues concerning how

influence might be identified and measured. I thus begin with a few general comments about the

nature of politica1 influence, be fore reviewing past discussions of business power.

Studying Political Influence. The enormous literature on the concept of power suggests the need to

reflect brieflyon what it is that we are trying to identify. Political influence means something far

broader than the triumph in a "head-to-head" conflict over a particular amendment or piece of

legislation. The principal justification for broadening our view is the concept of anticipated

reactions.8 Political actors are in a position to calculate (with at least some degree of accuracy) the

reactions of other actors. Given this capacity, a group's actions often will not reflect its true

preferences, but rather its strategie cakulations of what is the best that can be accomplished given

existing circumstances. Measuring influence requires that the analyst look beyond an actor's publicly­

expressed policy preferences.

Consider Figure I, which depicts the policy preferences of organized labor and business over a range

of policy options stretching from LO to BO. Business actors favor (to an increasing degree) those

8 Among the rnany literatures that one might consult on this issue, a particularly useful one is the recent debate Qver Congressional/agency relations. This discussion focuses specifically on the complex interactions that result when each side acts with same knowledge of the political resources cf other actors, and in anticipation of the Iikely reactions to a particular strategy. See for example Mathew D. McCubbins, RügeT Noll and Barry Weingast, "Structure and Process, Policy and Politics: Administrative Arrangements and the Politieal Control of Ageneies," Viriginia Law Review, Vol. 75, 1989, pp. 431-82; and Terry M. Moe, liAs Assessment of the Positive Theory of Congressional Dominance," Legislative Studies Quarterly, Vol. 12, November. 1987, pp. 475-520.

9

policies grouped toward the right end of the spectrum, while labor favors those at the left end . At any

point in time, however, only a narrow subset of these policy alternatives is likely to be on a

government's decision agenda. At one moment, the political conflict may be between options LI and

L2, while at another time the choice is between options Bland B2. Although the final choice between

these two "viable" alternatives is the most visible part of the policy-making process, it is often not the

crucial one. Rather, the most significant aspect of influence involves moving the decision-making

agenda toward an actor's preferred end of the spectrum. A particular group may get its preFerred

outcome at this final stage, but only because it has recognized its weak position and made strategic

concessions that allow it to hold off a still-worse alternative. Thus, focusing only on the actual

conflict between contending alternatives may give a very misleading impression of the influence of

relative actors. For example, if Congress chose Blover B2, we would be mi staken to conclude that

labor actors have more political influence than business actors do, even though labor obtained its

preferred outcome within this particular subset. The crucial issue is to determine why the "viable"

options were those at the right end of the spectrum, rather than why one of the two options relatively

favorable to business interests was finally selected.

Figure J: Ranked Policy Preferences of Lobor alld Business

Labor __ ~ ________ ~ ____ ~ ____________ ~~ ______ ~ __________ ~__ Business Prefers Lo LI L 2 8 2 BI Bo Prefers

As I will explore in Part 11 of thi s essay, the debate on the role of business in the development of the

welfare state has often failed to embrace this fundamental point. Much of the discussion has focused

on political fights over particular pieces of state and national legislation. Who wrote the legislation,

and who provided the expertise needed to formulate particular proposals? Who backed wh ich bill ?

These are significant questions, but they necessarily draw attention to the particular choices made

within a narrow range of options, rather than to the matter of why only that particular sub set was

considered feasible. Not surprisingly, this focus also encourages confusion about what the hi storical

record shows. In the fight between relatively narrow alternatives, there will be some victories for each

side, and even the lines di viding the "sides" are likely to become ambiguous as actors take a variety of

stances for strategic reasons . Thus, both advocates and opponents of the "business dominance" view

will find evidence to support their claims.

10

Recognizing the significance of anticipated consequences makes the study of influence more difficult.

The exertion of power often means convincing opponents that they cannot hope to accomplish their

true goals, and must instead aim for something far less ambitious. We wish to determine whether a

particular policy stance reflects a genuine preference or reluctant acquiescence in light of a weak

political position9 Analysts must pay crucial attention to the agenda-setting and alternative

specification processes through wh ich the effective "decision" comes to involve a choice between two

rather narrowly-differentiated options. I 0 We wish to know why at a particular point in time the set of

"feasible" initiatives comes to rest at a particular pI ace, and why that place may later change. When

applied to business, a crucial implication of this change in focus is a reconsideration of the important

indirect sources of business influence on policy outcomes.

The "Business Power Debate" Revisited. Two decades ago, the significance of business influence was

perhaps the central issue in political science. From the late I 960s to the early I 980s, both pluralists

and Marxists focused on the nature and extent of business power in advanced industrial democracies.

In each of these largely self-contained debates, one group emphasized the "instrumental" aspects of

business power. I I For instrumentalists, the power of business sterns from its ability to staff

governments with business supporters and to exert direct influence on government decision-makers

through campaign contributions and lobbying efforts.

Instrumentalist views of business power were in turn subjected to a number of telling criticisms.

Instrumentalists greatly exaggerated the extent to wh ich business controlled access to high

governmental positions; they ignored the significant financial and (especially) electoral resources of

competing interests; finally, they glossed over the deep c1eavages that divided the business

community on many issues. By the mid-1970s, the case far some sort of cohesive, unchallenged

"power elite" was widely dismissed. There was, however, little consideration of the obvious next

question: the conditions under which the instrumental power of employers is likely to expand or

contrac!. 12 Having couched the debate in such sweeping terms -- either business was omnipotent or

it was just another societal interest -- the discussion failed to move forward.

9 One of the reasons this essay focuses on the pre-World War n development of social policy is the exiSlence of a substantial and thoroughly-investigated archival record, which rnakes it possible LO trace the strategie behavior of important actors Qver time. In many eases, one can document private acknowledgements that public positions represent a strategie accommodation ta political realities.

10 John Kingdon, Agendas, Alternatives, and Public Policies (Boston: Little, Brown), 2d ed. , 1995. 11 Within Marxist debates on the "capitalist state" this view was advanced in Ralph Miliband, The State and

Capitalist Society (New York: Basic Books), 1969. In the pluralist/elitist debate, "elite theorists" presented a somewhat similar argument. See e.g., C. Wright Mills, The Power Elite (New York: Oxford University Press), 1956; G. William Domhoff, Who Rules America? (Englewood Cliffs , NJ.: Prentice Hall), 1967. Miliband dedicated his book to Mills.

12 Again, the notable exception is Vogel, Fluctuating Fortunes.

11

As the analysis of instrumental power ground to a halt, a different (although not incompatible)

conception of "structural" business power emerged -- again almost simultaneously within both these

academic discussions. Among Marxists, the "structural" position was initially (if hazily) advanced by

Nicos Poulantzas, and greatly refined in aseries of essays by Fred Block. 13 In the pluralistJelitist

debate, disillusioned pluralists elaborated a very similar set of arguments. In Charles Lindblom's well­

known formulation, business occupied a "privileged position" that invalidated many of the key

assumptions of pluralist theory . 14

Both Block and Lindbiom begin with the observation that in a market economy, crucial authority is

vested in private firms. These organizations playa critical role in modern societies: they control the

production of wealth, and the wages they pay provide the principal economic resources for most

citizens. Given the centrality of business organizations, Lindbiom argues, their managers parallel

public officials in their capacities to make authoritative decisions governing the lives of millions of

citizens. In addition - and this is crucial to the argument about structural power - firms face a very

c1ear incentive structure: they must strive to maximize profits. This conclusion does not rest on

assumptions about individual greed, but on a recognition that market systems are powerful

mechanisms for inducing decision-makers to adopt profit-maximizing behavior. Market competition

is a Darwinian struggle in wh ich profitability is the measure of fitness. Firms that fail to be profitable

will, in time, go to the wall.

In market systems, firms decide where, when, and how much to invest based on their calculations of

profitability. Absent expectations of profit, investments will not occur. Without investment, firms can

neither produce goods nor create jobs. It is these consequences of investment decisions that give firms

structural power. The decisions of individual firms, when aggregated, have a profound impact on the

state of the economy as a whole and, in turn , on the quality of life of those who reside in that

economy.

To the extent that individual citizens can bring their concerns to bear on politicians, fear of economic

deterioration resulting from declining investment is likely to generate severe pressures on public

policymakers. The prospect of a public backlash gives policymakers a strong incentive to maintain the

profitability of private investment. This power is "slructural" because the pressure to protect business

13 Nicos Poulantzas, Political Power and Social C1asses (London: Verso), 1973. Fred Block, Revising State Theory: Essays on Politics and Postindustrialism (Philadelphia: Temple University Press), 1987. The key essays in Block's volume were first published in the mid-1970s.

14 The first statement of this position was in the revised introduction to the 1953 work of two leading pluralists, Robert nahl and CharIes Lindbiom, Politics, Economics and Welfare, published in 1976. See al so CharIes E. Lindbiom, Politics and Markets: The World's Political-Economic Systems (New York: Basic Books), 1977; CharIes E. Lindbiom, "The Market as Prison," Journal of Politics, vol. 44, no. 2. 1982, pp. 324-36; Robert A. Dahl, Dilemmas of Pluralist Democracy: Autonomy vs. Control (New Haven: Yale Uni versity Press), 1982.

12

interests is generated automatieally and apolitieally. It results from private, individual investment

deeisions taken in thousands of enterprises, rather than from any organized effort to influenee

polieymakers . The market, LindbIom eoneludes, is a prison, in whieh polieymakers are quiekly and

effeetively punished for aetions that undermine the prospeets for profitability in the private sector.

There are a number of problems and ambiguities in thi s argument about the nature of business

influenee. In partieular, the assertion of business' "privileged position" has appeared to be ill-suited

for eomparative investigations of poliey development. LindbIom presented hi s argument as a general

claim about the nature of private power in market eeonomies, but as erities have pointed out, market

systems are eompatible with widely divergent relations between business and the state. 15 Patterns of

government intervention vary greatly aeross countries and over time within partieular countries.

Business does not always get what it wants; governments c\early have the capacity under so me

eireumstanees to adopt polieies over the objeetions of some, even most businesses. If the market is a

prison, Lindblom's erities say, the exereise yard seems to be large enough to allow room for

tremendous poliey diversity.

The seeming ineapaeity of struetural arguments to deal with these issues has led seholars interested in

explaining poliey variation to either fall back on instrumentalist eoneeptions of business power or

move away from an emphasis on business altogether. This response is unfortunate, beeause the

argument about struetural power, when properly understood and linked to an analysis of other

politieal and eeonomie variables, is entirely eompatible with an emphasis on poliey variation aeross

issues, countries, and over time. Several propositions follow logieally from the struetural power

argument, all of whieh suggest that business influenee will depend upon partieular features of the

politieal and eeonomie eontext. The following four propositions ean be used to integrate arguments

about struetural power into an examination of poliey variation:

I. The struclural power of business is a variable, not a cOllstant. If influenee depends on fear of disinvestment, then it will vary depending on how eredible polieymakers believe that threat to

be. One major aspeet of this eredibility is the ease of moving investment to another politieal

jurisdietion. Fear of disinvestment will also depend on the extent to whieh polieymakers believe

that they will be punished for poor eeonomie performance. Hypothetieally , in a eontext where

eitizens were unable to infliet such punishment, the eonstraint on polieymakers would be greatly redueed.

2. This struclural power is a signalling device; by itself it does not dietate poliey ehoiees .

Governments must deeide what polieies will sustain investment. This requires eomplex ealculations that are a produet of external pressures (ineluding eultural ones) and internal deeision-making systems. The prospeet or aetuality of disinvestment ean seI the agenda for

15 See especially David Vogel, "Political Science and the Study of Corporate Power: A Dissent from the New Conventional Wisdom," British Journal of Political Science, Vol. 17, 1987, pp. 385-408 and Vogel, Fluctuating Fortunes.

13

governments and help to define (or rule out) alternatives, but this "signal" cannot tell

governments what to do. The extent to wh ich business influences specific policy choices will be a function of instrumental rather than structural power. I 6

3. Not all soeial polieies activate the signalling device . The key is whether policies have (or are

expected to have) a negative impact on profits. Wh ether costs are likely to be born by business,

the impact on labor market conditions, and the presence or absence of possible offsetting benefits for employers are all factors that will determine wh ether policy initiatives discourage

investment.

4. The structural power argument does not rest on an assumption that business interests Qre monolithic. Because the circumstances of firms are quite different, social policy initiatives are

Iikely to pose stronger threats to some firms, sectors or regions than to others. lndeed, social

policies often benefit some businesses at the expense of others. In such circumstances, the investment signal is likely to be mixed.

In short, the argument about the structural influence of business does not imply a simple prohibition

of government policy initiatives or even of initiatives that run counter to business interests. It is fully

compatible with an emphasis on policy variation, both between count ries and over time. To help

account for variation, though, arguments about the structural properties of market systems need to be

Iinked to analysis of the interaction of those features with other political and economic factors,

especially the role of political institutions .

Nor, as these propositions make c1ear, is an analysis of structural power an alternative to a focus on

instrumental power. Both pathways to influence are relevant in advanced industrial democracies -­

although one or the other is often more significant in a particular time or place. If structural power is

sufficiently great, there may be little need to resort to instrumental efforts to modify or block public

policies. I 7 Where structural power fails, instrumental influence becomes more important. The impact

of both factors will also vary in different stages of the policy-making process. Since structural power

acts as a signalling device, its importance is feit primarily in setting the public agenda and in ruling

out options that are particularly objectionable to potentially mobile firms. Instrumental power is often

more relevant in determining the specific design of legislative proposals.

16 On these distinct stages of policymaking and their different political dynamics see Kingdon, Agendas, Alternative and Public Policies. On the idea that policy-makers must "construct" responses to a perceived 1055 of business confidence see David Plotke, "The Political Mobilization of Business," in Mark Petracca, ed., The Politics of Interests (Boulder, CO: Westview Press), 1992, pp. 175-98.

17 For this reason, I am skeptical of Jeff Frieden's claim that it is firms with the least mobile assets (in his terms, high assel specificity) which will have the greatest polilical influence. Jeffrey Frieden , Debt, Development, and Democracy (princeton: Princeton University Press), 1991. As Hirschman noted, those lacking the "exit" option may find it more necessary to employ "voice", but greater political activity should not be conflated with greater intluence. Again, the anticipated reactions of potentially-mobile finns will tend to make policymakers attentive to their concerns. Albert Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Finns, Organizalions, and States (Cambridge: Harvard University Press), 1970.

14

The Institutiollal Mediation of Business Power. The thrust of these propositions about business

influence is that market pressures do not create a complete barrier to policy activity; instead,

economic and political conditions as weil as the characteristics of particular policy proposals are

relevant. The next stage in the argument is to link these propositions, along with a complementary

analysis of instrumental power, to the characteristics of political institutions. Many of the questions in

the original debate over business power were posed at a very high level of abstraction. What,

participants asked, was the relationship between capitalism and democracy? The institutionalist

riposte -- to this as to many questions -- would be "it depends." Capitalist democracies take widely

divergent forms, and the specifics of institutional design will have a significant impact on what

business actors seek, how they pursue their goals, and the likelihood that their concerns will be

addressed.

Three important hypotheses follow from the preceding analysis:

(I) The struetural power of business will inerease in deeentralizedfederal systems. Capital mobility is

a key - and highly variable - element of business' structural position. Because mobility is much greater

between states than between countries, state policymakers are especially eager to create a "healthy

business c\imate." The role of capital mobility is thus likely to be greater in federal systems, and

especially pronounced in decentralized federal systems where national authorities make only limited

efforts to "Ievel the playing fjeld" among competing states. 18 Fiscal arrangements in federal systems

can minimize financial disparities among the states, increasing their capacities to pursue autonomous

policies and limiting their vulnerability to the movement of economic actors across state lines. These

equalizing arrangements have always been weak in the United States. As we shall see, this was

especially true prior to the New Deal, making the threat of capital mobility unusually vivid for state

governments. Political institutions mediate the exercise of business influence. In federal systems,

prospects for social policy are thus likely to be influenced by whieh level of government - state or

national - is in a position to pursue reform. 19

(2) Fragmented institutions make it harder for business to playa eonstitutive rather than blocking

role in poliey formation. Business is not always opposed to social policy initiatives; on occasion,

business interests may actively promote government intervention 20 At other times, business actors

may seek to steer policymakers toward the more palatable of two unappealing options. Yet as I have

18

19

Paul Pierson, "Fragmented Welfare States: Federal Institutions and the Development of Social Policy," Governance, Vol. 8, No. 4, October, 1995 (forthcoming). Für an excellent example of the use of this kind of argument to explain cross-national poliey variation. eolleen A. Dunlavy, "Mirror Images: Political Structure and Early Railroad Poliey in the United States and Prussia," Studies in American Political Development, Vol. 5, No. I, Spring, 1991, pp. 1-35, esp. pp. 28-34.

20 Cathie Martin, "Basic Instincts? The Formation of Firm Preferences for National Health Insurance," American Political Science Review, December, 1995 (forthcoming).

15

stressed, the struetural power of business funetions only as a "signalling deviee", prompting

polieymakers to promote a favorable investment climate. While clearly quite important, thi s is a

limited form of influenee. It struetures governmental agendas and excludes possible alternatives, but

does not dietate poliey choices. Whether business is able to eompel speeifie positive actions or to

shape the details of poliey design depends on instrumental power.

The fragmented Ameriean politieal structure makes it hard for any group, including business aetors, to

sueeessfully pursue a eohesive policy agenda21 In addition to federalism, the separation of powers

and deeentralization within the branehes (espeeially in Congress) ereate multiple opportunities for

aecess -- as weil as multiple veto points . These structures have eneouraged Ameriean business to

organtze In a relatively deeentralized fashion, obstrueting the development of European-style

eorporatist peak assoeiations22 The tremendous hurdles to legislation also eneourage lowest­

eommon-denominator polieies and inerease the influenee of those seeking to block reform. Under

normal eonditions, national poliey-making institutions in the United States advantage those playing

"defense" over their opponents. Groups will rarely be eohesive and powerful enough to eontrol the

development, enaetment, and implementation of important pieces of legislation .

(3) Fragmel/ted institutions may eI/haI/ce the struc/ural and instrumental illfluellce of particttlar

sectors of business. Business interests are sometimes divided over soeial poliey . In a fragmented

institutional setting, however, the magnitude and direetion of business power will not neeessaril y be

simply the sum total of gainers and losers within the business eommunity. Some employers may have

influence over a particular actor who has access to a veto point - for example, if these firms are

regionally concentrated. If so, they may be in a position to block reforms that more prominent

business groups (and other political actors) support23 Again, there is an important interaction

between institutional structures and the power of economic actors.

This section has suggested a framework for the study of business power. In contrast to Vogel, I have

argued that core claims about the structural power of private corporations are compatible with an

2 1

22

On this point see David Vogel, "Why Businessmen Distrust 'Their' State: The Political Consciousness of American Corporate Exeeutives," British Journal of Political Seience, Vol. 8, 1978, pp. 45-78; Terry Moe, "The PoJitics of Structural Choice: Toward a Theory of Public Bureaucracy," in Oliver Williamson, ed., Organization Theory frorn Chester Barnard to the Present and Beyond (New York: Oxford University Press), 1990, pp. 116-53. Fiorina? Robert H. Salisbury, "Why No Corporatism in America?" in Philippe C. Schmitter and Gerhard Lehrnbruch, eds., Trends toward Corporatist Interrnediation (Beverly Hills: Sage), 1979, pp. 213-30; and Graham K Wilson , "Why is There 110 Corporalism in the United States?" in Gerhard Lehmbruch and Philippe C. Schmitter, eds., Patterns of Corporatist Policy-making (Beverly Hills: Sage), 1982, pp. 2 19-36.

23 The influence of geographically concentrated interests may be both structural and instrumental. In the relevant case for this essay - the South - both kinds of influence were at work. Southern politicians were subjected to both direct busi ness lobbying and to the more indirect pressures stemming from their efforts to lure (and retain) potentially-mobile capital.

16

emphasis on policy variation. These arguments are a supplement, not an alternative, to arguments

about the instrumental power of firms. Linking these arguments about the distinct paths of business

influence to an analysis of institutional structures makes it possible to generate propositions about

variations in the scope and nature of business influence. The next question is wh ether this framework

is useful for explaining the evolution of American social policy. In the next seetion, I suggest that it

iso When joined to an understanding of the Great Depression's transformative role, the shifting pattern

of marketlstate interactions can account for the most distinctive aspects of American welfare state

formation.

11. Business Power and the Formation ofthe American Welfare State

The central historiographical controversy over the American welfare state's formation has concerned

the importance of business power. A number of scholars have argued that business actors largely

designed policy in this period, or at least that public policies were intended to serve business

interests24 Those emphasizing institutional factors, most notably Theda Skocpol, have discounted the

role of employers in policy development, stressing instead such factors as the orientations of

patronage-based political parties, efforts of state-building elites inside and out of government, the

impact of policy legacies from earlier periods, the institutional power of southern politicians, and

(most recently) the role of broad-based women's groups25

Both views are overdrawn. More precisely, each is relatively convincing for only part of the period in

question. The history of the American welfare state before World War 11 must be divided into two

eras, punctuated by an economic cataclysm. In the decades before the Depression, social policy in the

United States was remarkably limited, especially at the federal level. Few ob servers could have

looked at the politicallandscape of the late I 920s and drawn the conclusion that the United States was

evolving towards a system of social supports even faintly comparable to what had already emerged in

24

25

See especialty Edward Berkowitz and Kim McQuaid, Creating the Welfare State: The Political Economy ofTwentieth-Century Reform (New York: Praeger), 1980; Colin Gordon, New Deals: Business, Labor, and Politics in America, 1920-1935 (Cambridge: Cambridge University Press) 1994; Frances Fox Piven and Richard A. Cloward, Regulating the Poor: The Public Functions of Sodal Welfare (New York: Vintage Books), 1971; and Jilt Quadagno, The Transformation of Old Age Security: Class and Polities in the American Welfare State (Chieago: University of Chieago Press), 1988. See Theda Skoepol and John Ikenberry, "The Political Formation of the Ameriean Welfare State in Historieal and Comparative Perspeetive," Comparative Soeial Research, Vol. 6, 1983, pp. 87-147; Ann Shola Orloff and Theda Skocpol , "Why Not Equal Proteetion? Explaining the Polities of Publie Soeial Spending in Britain, 1900-1911 , and the United States, I 880s-I920s," Ameriean Soeiologieal Review, Vol. 49, Deeember, 1984, pp. 726-50; Margaret Weir, Ann Shola Orloff and Theda Skoepol , eds. , The Politics of Soeial Poliey in the United States (Princeton: Princeton University Press), 1988; and Theda Skocpol , Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (Cambridge: Belknap Press of Harvard), 1992.

17

many less industrialized European countries. The Depression, however, sent a shock-wave through

the American political economy, shifting the balance among political actors. In this greatly changed

environment, previously dominant business groups were forced to make their peace with a vastly

expanded federal social role, and the basic structure of the American welfare state was put in place

between 1933 and 1939.

lt is crucial to understand that the Depression fundamentally changed the operation of American

political institutions, and, in doing so, altered both the extent and character of business influence.

Accounts that minimize the role of business prior to 1929 are untenable. The radical decentralization

of policy initiative prior to the New Deal maximized the value of a major dimension of business

power: the threat of capital mobility . In this context, policy very c10sely conformed to the preferences

of most business actors. The Depression, along with a massive electoral realignment, undermined the

institutional foundations of business hegemony. Policy initiative shifted to the federal level. Here,

business had to rely on its capacity to deploy instrumental power in a highly-fragmented policy­

making process at a time of high popular mobilization. While hardly without influence, employers

c1early played a far more modest role du ring this crucial period than they had prior to the New Deal.

All Unfeuered Common Market: American Social Policy Before the New Deal. Pre-New Deal social

policy was state and local policy. The federal government showed little enthusiasm for social reform.

The major parties were loose coalitions of regional and sectoral interests, dedicated to the promotion

of patronage rather than coherent programs of policy reform. Reformist forces were weak in the

Democratic party, and the Republicans, who were in a preponderant position after 1896, were c1osely­

aligned to major business interests. With the important exception of civil war pensions (to be

discussed below), the GOP was hostile toward any federal social role. The Supreme Court shared this

antipathy, and sharply circumscribed federal interference with the "freedom" of workers and

employers to settle contracts together, regardless of how unequal the terms might be.

This party orientation, the blocking role of the Supreme Court, and anational institutional structure

wh ich required broad consensus produced a pattern of minimal ist policymaking, strongly oriented

towards distributive rather than redistributive initiatives26 This distributive focus included strong

efforts to foster the development of the American "common market." The federal government doled

out land, invested in communication and transportation networks, and tried to reduce other barriers to

interstate commerce. lt sought to facilitate national economic integration by encouraging the

unrestricted mobility of goods, labor and investment. Yet the federal government generally refused to

26 This distinction is from Theodore Lowi's c1assic artic1e. "American Business, Public Policy, Case-Studies, and Political Theory," World Politics, Vol. 16, July, 1964, pp. 677-715. On the distributive emphasis of federal policy in this period see Richard L. MeCormiek, "The Party Period and Publie Poliey: An Exploratory Hypothesis," Journal of Ameriean History, Vol. 66, 1979, pp. 279-98; and Skoepol , Proteeting Soldiers and Mothers, eh. 2.

18

become involved in soeial policy . Table I details the most striking aspects of "American

exceptionalism": the absence of so me national social policies that were common elsewhere and the

very late arrival of others. While social insurance initiatives received Iimited attention at the national

level during and after the Progressive era, none had any realistic chance of passage.27

Table 1: Time of Adoption of Five Major Welfare Programs

(US Rank among 15 Developed Nations')

Initial Adoption Binding o r Extensive Binding and Extensive Program

Old Age, Disability and Survivors

Sickness and Maternity

Workers' Compensation

Unemployment Compensation

Farnily Allowances

12th

15th

15th

11th

none

11th 11th

none none

13th 10th

11th 9th

none none

*The fifteen countries are Australia, Austria, Belgium, Canada, Denmark, France, Germany, Italy, the Netherlands, New Zealand, Norway, Sweden. Switzerland. the United Kingdom, and the United States. The United States is the only country in the group without binding and extensive health eare benefits and family allowances.

Saurce: Alexander Hicks. Joya Misra and Tang Nah Ng, "The Programmatic Emergence of the Social Security State," American Sociological Review, Vol. 60, June, 1995, pp. 329-49, p. 337.

There was one enormous exception to this federal policy vacuum: the extensive system of ci vii war

pensions which f10urished in the United States in the late 19th and early 20th century28 At its peak,

the pension system accounted for 37% of federal expenditures, providing pensions -- which compared

favorably to the benefit levels of early 20th century European pension systems -- to one million union

veterans and their survivors. Skocpol has argued that Civil War pensions reflect the patronage

27

28

The federal government did become involved at the margins of social policy during thi s period, but not on any scale. Excluding aid to veterans, federal welfare expenditures - $.25 per capita in 1929 - were roughly 6% of total national expendilures. Berkowitz and McQuaid, Creating the Welfare State. p. 66. The few federal social policies that were able to pass, such as the Sheppard-Towner program for infant and maternal health, were tiny -- clüseT to demonstration projects than entitlements. And Sheppard-Towner failed 10 achieve pennanenl authorization. On Sheppard-Towner see Skocpol, Protecting Mothers and Soldiers. eh. 9. The nexi two paragraphs draw on Richard FrankIin Bensei. Sectionalism and Political Development: 1880-1980 (Madison: University of Wisconsin Press), 1984, Ch. 3; and Skocpol, Protecting Soldiers and Mothers, Ch. 3.

19

orientation of late 19th century American politics. Because northern, American-born males were the

recipients, benefits could be effectively targeted on precisely the regions and groups most crucial to

the Republican party's patronage/electoral machine.

Everything we know about national politics in this period, however, suggests that widespread

business support was aprerequisite for the remarkably extensive civil war pension system. The lack

of a business outcry against these massive income transfers is in stunning contrast to its position on

far-Iess intervention ist social policy issues. The reason is not hard to identify 29 The pensions were

financed by revenues from tariffs that were of vital interest to Northern industry, but which had the

awkward consequence of filling the government's coffers to overflowing. It is no accident that

pension spending took off around the same time that Congress faced growing demands to cut tariffs in

the light of the government's remarkably healthy finances 30 The pension system not only disposed of

the politically inconvenient surplus, but did so in a way that strengthened the business-dominated

Republican coalition.

Despite reformers' aspirations, the Veteran's pension system was not transformed into a universal or

means-tested system. While Skoepol emphasizes elite concerns about corruption as a result of their

experience with the program for veterans, it is also true that the "tariff engine" was being phased out

during this period. Customs duties provided 37% of government revenues in 1902, but only 32% in

1913 and just 7% by 192231 With this dec\ine there was no basis for business support of federal

soeial initiatives. As the cohort of veterans dwindled, the federal government returned to its practice

of non-intervention in the realm of social policy. Far from becoming more centralized and extensive

in the early 1900s -- the pattern in almost every other industrial society -- American social policy

moved in the other direction.

Federal inactivity meant that efforts to pass social legislation focused on the states. In state capitols,

initiatives resembling those common in Europe were on the public agenda in the early 19OOs. Yet

wh at is striking again is the feeble progress of even state-Ievel social legislation. While the scope of

state initiatives before 1930 was far greater than national activity, it remained extremely limited. This

was true despite the fact that the balance of political forces in many states seemed more promising

than it was at the national level, and in many respects comparable to the conditions that proved

conducive to reform abroad. The more industrialized states boasted levels of economic development

29 See Bensei, Sectionalism and American Political Developrnent, eh. 3. 30 Skocpol emphasizes that the Arrears Act of 1879, which greatly expanded the program, occurred at a lime

"when there was praclically no surplus" (p. 113). This was true only because the federal government was using its large surplus to rapidly retiTe the debt left fram the Civil War -- a use of revenues that could not be sustained for 10ng.

31 Figures on tariff revenues frorn United States Census, Historical Statistics. p. 1 t 22.

20

and unionization rates equal to or greater than those of many European nations32 Unlike the national

leadership of the AFL, state level union federations were often eager to promote sociallegi slation 33

A significant barrier was that political authority within most states was as fragmented as it was in

Washington . The need to pull together "super-majorities" sufficient to overcome the many possible

obstacles to legislation hindered government action. One of these obstacles, the courts, sometimes

could not be overcome by any action short of constitutional reform. There are a number of examples

during this period of legislative victories undone by court verdicts34

Yet there is substantial evidence that the principal source of state inactivity was fear that social reform

would have an adverse impact on business . Individual states were embedded within an increasingly

integrated national economy, making concerns about lost competitiveness highly salient. The

possibility that social legislation would discourage new private investment, encourage firms to

relocate, or undermine the ability of local firms to compete with out-of-state rivals was a major

constraint on state-Ievel policy initiatives. In the context of decentralized federalism, business

possessed significant structural power.

Recent studies have produced a wealth of anecdotal evidence to suggest the weight of this factor in

state-Ievel discussions of social legislation . As David Brian Robertson notes in an important study ,

"[sltate policymakers' fear of placing their state at a competitive disadvantage permeates social policy

discourse during this period .',35 Government-established commissions pointed to the adverse

economic consequences as an explanation for deferring social reform. Political interests that might

32

33

34

See for eX31l1ple the interesting comparison between Massachusetts and Britain in Orloff and Skocpol , "Why Not Equal Proteetion?" See Christopher Anglim and Brian Gratton, "Organi zed Labor and Old Age Pensions," International Journal 01" Aging and Human Development, Val. 25, No. 2, 1987, pp. 91-107; Skocpol, Protecting Soldiers and Mothers, Ch. 4. Thi s blocking role can be exaggerated. As Melvin Urofsky notes in a review of state court responses to protective legislaLion " ... statc courts moved consistently toward approval of a wide range of reform legislation. In atlempLing Lo enact their program, Progressives, atLhough occasionally delayed in the courts, were not blocked there." Melvin I. Urofsky, "State Courts and Protective Legislation during the Progressive Era: AReevaluation," Journal of Americ.n History, Vol. 72, No. I, June. 1985, p. 64. While a complete account of policy outcomes in this period would certainly give significant weight to the role of the courts, my goal is simply to demonstrate that business interests were extremely powerful , and Lo consider the reasons why that power diminished considerably after 1929.

35 David Brian Robertson, "The Bias of American Federalism: The Limits of Welfare-State Development in the Progressive Era," Journal of Policy History , Val. I, No. 3, 1989, p. 274. See also Colin Gordon, "New Deal, Old Deck: Business and the Origins of Social Security, 1920- 1935," Politics and Society, Vol. 19., No. 2, 1991 , pp. 165-207; and Gordon, New Deals. Robertson's and Gordon's important work provide copious documentation of these pressures on state governments. I have relied considerably on their analyses in my treatment of pre-New Deal social policy, though as discussed below I disagree strongly with Gordon's analysis of the New Deal period. On the impact of feder.lism before the New Deal see also Edward D. Berkowitz and Kim McQuaid, Creating the Welfare State: and William Graebner, "Federali sm in the Progressive Era: A Structural Interpretation of Refonn," The Journal of American History, Val. 64. no. 2, September, 1977, pp. 33 1-57.

21

otherwise have been neutral or even favorable to such initi atives were often pushed into opposition

for the same reason.

The evidence to support this explanation of state inactivity in social policy is not only anecdotal. A

number of aspects of the hi storical record provide significant support for my claim that fear of

business disapproval was a principal obstacle to reform36 The most important piece of ev idence is

the distinctive pattern of success and failure of social reformers. Social legislation was not entirely

absent in the pre-New Deal period. Those reforms that were successful, however, were ones that were

either in the interest of local businesses or promised only marginal disruptions of market conditions.

Especially in industrial states, initiatives that did not have the acquiescence or active support of

employers were invariably either blocked or severely watered-down to mitigate concerns about

interstate competition.

Workmen's compensation provides the clearest example.37 Distressed by the unpredictability and

expense of injury-related litigation, most firms were eager for a legislative approach. Business groups

moved strongly in favor of reform after the turn of the century. As James Weinstein notes, "ninety­

fi ve percent of the 25,000 employers to whom the N.A.M. sent questionnaires in 19 10 favored

automatic compensation for industrial accidents; few had the resources to institute such programs

privately. The Association therefore endorsed the idea of workmen's compensation legi slation.',38

The National Civic Federation, in which prominent businessmen played a guiding role, became a

forceful advocate of compensation laws. Beginning in 1910, legislation to establi sh regulatory

mechanisms for handling compensation of workplace accidents spread rapidly. Although business

interests sometimes failed to obtain the precise legislation that they preferred , none of the systems

adopted would have posed a threat to the profitability of local businesses even if other states had

failed to acl. As Roy Lubove summarizes the historical record, " ... employers often assumed the

initiative in sponsoring workmen's compensation and the program ultimately was shaped by their

objectives rather than the needs of the injured.,,39 Other successful social initiatives also show

the importance of business support or acquiescence. Programs that offered proteclion to the

"deserving" poor (i.e., those who were not expected to participate in waged labor), such as aid for the

blind or disabled, met with some success, but only if public outlays were expected to be very modest.

Old age pensions made just slight progress. The aged held a marginal status in labor markets during

36 It is hard to imagine how one could prove the case definitively. The technique employed hefe is 10 provide evidence for several distinct propositions, each of which is consistent with the argument advanced earlier about the structural power of business. See Gary King, Robert Keohane and Sidney Verba, Designing Sociallnquiry: Scientific [nference in Qualitative Research (Princeton : Princeton University Press), 1994.

37 See Roy Lubove, The Struggle for Social Security, 1900- 1 935 (Pittsburgh: University of Pittsburgh Press), 1968, pp. 45-65 ; Berkowitz and McQuaid, Creating the Welfare State, pp. 33-4 I.

38 James Weinstein , "Big Business and the Origins of Workme n's Compensation," Labor History, Vol. 8, 1967, pp. 161-2.

39 Lubove, Struggle for Social Security, pp. 61-62.

22

the period, but pensions threatened to be costly. Thus while pension initiatives fared beller than

proposed social insurance schemes for workers, they nonetheless remained extremely limited in

scope. As Lubove concludes after summarizing the experience of a range of states, "the main reason

for the failure [of pension reform advocates] was the effective opposition inspired by business

organizations.',40 By 1929, only six states had adopted pension laws. Even these laws, it must be

stressed, were so watered-down as to be largely symbolic. They caUed only for smaU, means-tested

benefits for the destitute, and were passed only after reformers agreed to make them "county

optional." Because counties faced even greater constraints on redistributive spending than states, this

arrangement assured that expenditures would be very low in the few cases where counties agreed to

provide pensions . In the entire country, there were roughly 1,000 recipients of old age assistance in

1929; total spending was about $200,00041

As Skocpol has persuasively argued, the most widespread activity on social legislation before 1933

involved "maternaiist" policies. A large-scale mobilization of women's groups and other social

reformers, sometimes acting in concert with organized labor, succeeded in generating a wave of

initiatives during the Progress ive Era. Legislative action included maximum hours regulation for

women workers, mother's pensions, and (to a lesser degree) minimum wages for women workers.

Nothing about the scope and content of this effort, however, chaUenges the claim that business

influence played a critical role in regulating legislative activity. The legislation that passed was of

little concern to most businesses. Employers undoubtedly would not have pursued maternaIist policies

on their own, but they had little objection to many of these initiatives. On the other hand, where such

legislation posed areal threat to a large number of employers that legislation invariably failed to pass .

Mothers ' pensions, which spread rapidly, were unproblematic for employers. Because the mothers of

young children were not generaUy expected to participate in the paid labor market, these proposals

had a limited impact on business. Indeed, by aUowing children to stay in their mothers' homes rather

than state-financed orphanages, these policies might even result in lower tax levies. Reliance on

county-level finance guaranteed that spending on such programs would be very modest. Jt is simply

not accurate, as Orioff asserts, that the reason these programs offered only a pillance in benefits was

"the lack of administrative capacities of American state governments ..... 42 As she acknowledges,

reformers often sought more generous laws . The problem was not that state administrative capacities

were weak, but that reformers could not pass legi slation without first accepting a financial structure

that guaranteed such programs would be extremely modest. As Mark Leff concludes, "unlike much

40 Ibid, p. 140. 41 Louis Leotta, "Abraham Epstein and the Movement for Old Age Security," Labor History, Vol. 16, No. 3,

Summer, 1975, pp. 362-63. 42 Orloff, "Gender in Early U.S. Social Policy," p. 267.

24

this period. What the influence of business does explain is the narrow limits of such policies. None of

the maternaIist policies that passed posed much of a threat to business interests.

Further evidence of business power can be seen in the realm of social insurance legislation (covering

risks associated with illness, unemployment, or old age). These policies, which would have required

high taxes or offered significant protection to people who were expected to be in the labor market,

met stiff resistance. Although at various times there was considerable popular agitation at the state

level for social insurance, no proposals were able to make headway before the Depression47

Business opposition to these reforms was widespread, and fear of handicapping local businesses was

frequently advanced as ajustification for rejecting legislation .

In the face of widespread business hostility no state passed old age insurance, health insurance, or

unemployment insurance before 1929. The battles over unempIoyment insurance were typical.

Despite sporadic attention, the only state were unemployment compensation legislation made much

progress in the 1920s was mostIy-rural Wisconsin . While popular pressures mounted during the

Depression, state-Ievel initiatives remained constrained prior to federal intervention in 1935. Even

with growing public support for government action, only Wisconsin (in July or 1934) succeed in

implementing unemployment compensation before the passage of the Social Security Act was

imminent48 Most business representatives opposed the Wisconsin bill , but the final legislation had

been carefully tailored to meet their most serious objections. It fell far short of the aspirations of most

soeial insurance advocates, and was rar weaker than the unemployment insurance legislation that had

previously spread across Europe. As Daniel Nelson observes, after its passage employers "soon

became enthusiaslic backers ... As neighboring slales seriously considered more coslly plans,

Wisconsin employers rejoiced thai they had gotten off so easily.'049

Thus while Skocpol and others have rightly noted a significant record of state-Ievel social policy

aClivity prior to 1929, the pattern of that activity is revealing. Some very modest policies of labor

market regulation were possible; so were absolutely minimal provisions for the "deserving poor" (that

is, those outside the labor market), if these provisions were combined with strict proteclions against

fiscal laxity. Policies that involved significant redistributive spending, however, or that might have

provided real protection for jobless workers, were beyond the pale. In short, policy outcomes during

47 Lubove, Struggle for Social Seeurity, Chs. 4, 6-7; Gordon. New Deals. Ch. 7. 48 Daniel Nelson. Unemployment Insuranee: The Ameriean Experienee 1915-1935 (Madison: University 01'

Wisconsin Press); Edwin Amenta el al , "The Political Origins of Unemployment Insurance in Five Ameriean States," Studies in Ameriean Politieal Development, pp. 137-82. Amenta et al (p. 143n) add that New York passed a bill in April, 1935, "c1early ahead" of federal legislation , but the legislative progress of the Social Security Act was far advanced by then and there was widespread anticipation of federal action.

49 Nelson, Unemployment Insuranee, p. 128.

25

the pre-New Deal period eonform to a simple rule of thumb: reforms that threatened to raise eosts

signifieantly for a large number of firms engaged in interstate eompetition were very unlikely to pass .

Nor is the pattern of soeial legislation prior to the Depression the only indieation of employers'

struetural power. The seope of state aetivity outside the sphere of social poliey eonstitutes a seeond

souree of evidenee. The broader poliey reeord suggests that the limited seope of soeial polieies did not

refleet an antipathy for government action per se. Thus, the emphasis that some authors have plaeed

on fear of patronage as a eonstraint on publie spending in this period is unpersuasive.50 States eagerly

pursued -- and powerful interests supported -- aetivities that provided opportunities for patronage, as

long as these polieies also seemed likely to promote eeonomie development. As the Ameriean market

beeame more integrated in the early twentieth eentury, the states greatly inereased spending on

polieies that enhaneed the opportunities for profitable investment, such as the expansion of

transportation networks and edueational systems (see Table 2). Just as a foeus on eapital mobility

would suggest, the retieenee of state governments was limited to redistribwive pOlieies51 State and

loeal spending on edueation and highways expanded almost ten-fold in the first quarter of the 20th

eentury. By 1927, these developmental programs aeeounted for 52.1 % of state and loeal spending, up

from 41.5% in 1902. During the same period, the share of state and local expenditure devoted to

publie welfare measures aetually dropped from 3.3% to 1.9%. Thus, the broader reeord of state

aetivity, like the pattern of sueeess and failure within the field of soeial poliey itself, suggests that

only when it ca me to cerlain kinds of aelivity were state governments inciined to sir on their hands.

Tabte 2: State and Loeal Spending Oll Distributive (md Redistributive Programs (millians of dollars)

1902 1927

~ 12=1. cOmbined ~ 1.Q=1. cOmbined

Education 62 238 300 510 2017 2527

Highways 4 171 175 514 1295 1809

Public Welfare 10 27 37 46 111 157

Total Spending 186 959 1145 1976 6359 8335

Souree: U.S. Department of Commerce. Bureau of the Census, Historical Statistics of the United States: Colonial Times to 1970, 93rd Congress, I st Session, House Doeument No. 93-78 (Part 2), pp. 1131, 1134.

50 See for example Orloff, "Gen der in Early U.S. Soeial Poliey," who argues (p. 265) that "Iarge-seale soeial spending ... was polilically doomed by elite reactions against patronage."

51 This argument about the "developmenlal " rather than "redistributive" orientalion of state and IDeal spending is developed in Paul E. Peterson, City Limits (Chicago: University of Chicago Press), 1981 ; and Paul E. Peterson , The Price ofFederalism (Washington, D.C.: Srookings), 1995.

26

A third source of evidence for the structural power thesis is the extent to which exposure to interstate

competition helps to account for both the geographic distribution of reform and differences in

employer responses. Where the local economy remained isolated from the national economy, reform

was easier to pursue . Thus, for example, states in the then-relatively remote regional economy of the

Pacific coast were able to pursue some initiatives with little complaint from business that provoked

vehement business opposition in the industrial heartlands of the midwest and northeast52 lsolated

Montana, with no industry to speak of, was the only state where a majority of counties actually

approved "county-optional" pensions before 192953 As al ready noted, minimum wage laws for

women had alm ost no success in regions that had an industrial base . Within industrial states, social

legislation was more likely to be acceptable if its major impact was not on economic sectors that were

exposed to strong competition from out-of-state rivals . Exemptions were frequently offered to those

sectors that were most vulnerable.

A final strand of evidence is the clear indication that states regarded social regulation as a collective

action problem, and the persistence of state efforts to confront that problem through co-ordinated

action. If state governments worried that loeal business would lose out to less regulated competitors in

other states, a logical response was to try to limit the opportunities for free-riding by forming so me

sort of collective agreement. The obstacles to such action were staggering. In the absence of federal

intervention, coordination required legislation in many different states, each with a distinctive

economic profile, partisan complexion and legislative calendar54 The conflicting interests of the

states made collective action even more difficult. Especially in the South, states that hoped to attract

investment were extremely unlikely to accept such agreements.

Ultimately, these formidable barriers doomed collective efforts. What is remarkable is that despite the

huge problems involved, attempts to achieve co-ordination were widespread 55 Progressives

organized much of their activity around the promotion of "model statutes", which were seen as a

potential halfway house between the ineffectiveness of single-state initiatives and the implausibility of

federal action. Efforts to negotiate inter-state compacts covering various aspects of soeial regulation

were common. The investment of significant resources in such a hopeless political strategy was an act

52 Berkowitz and McQuaid, Creating the Welfare State, pp. 30-33; Graebner, "Federalism in the Progressive Era," pp. 34 t -2.

53 Leotta, "Movement for Old Age Security." p. 363. 54 The fact thm 22 legislatures in this period met for 60 days cr less in each two year session was a signiticant

drag on interstate cooperation. See James T. Patterson, The New Deal and the States: FederalisI11 in Transition (Westport, Ct.: Greenwood), 1969, p. 55.

55 Graebner, Federalism and the Progressive Er." pp. 341-55; Gordon, "New Deal , Old Deck," pp. 182-83; Berkowitz and McQuaid, Creating the Welfare State, pp. ; Jane Clark, "[nterstate Compacts and Social Legislation," Political Science Quarterly, Vol. 50, 1935, pp. 502-24; Jane Clark, "[nterstate Compacts and Social Legislation 11: lnterstate Compacts after Negotialion ," Political Science Quarterly, Vol. 51, 1936, pp. 36-60.

27

of desperation -- an act of policymakers eager for reform but trapped In an extremely hoslile

institutional environment.

Theda Skocpol has recently stressed that neither business power nor federalism conslituted an

"absolute bar to strong legislation" at the state level 56 This conclusion is accurate, but the argument

advanced here about the slructural power of business in a federal system does not suggest otherwise.

What the argument suggests -- and what the evidence supports -- is that policy initiatives under such

conditions generally will occur only where the threal to business profitability is small or non-existent,

or in jurisdictions where business is not yet weil established. More economically intrusive initiatives

either will not be seriously attempted (given anticipated reactions), or will be accompanied by

vigorous (if futile) efforts to forge a collective response to the imprisoning effects of interstate

competition. The historical record offers strong evidence for all these propositions. The

decenlralizalion of policymaking institutions in the United States prior to the New Deal greatly

enhanced the political leverage of business and contributed heavily to the feebleness of social

regulation before 1929.

The Impact 0/ the Great Depression on MarketlState Relations. The Great Depression shook political

systems throughout the industrialized world, but the collapse in the United States was particularly

profound. Unemployment, already high in the late I 920s, exploded after 1929. By 1933, fully one­

third of American workers were jobless. Farm income fell from $7 billion in 1929 to $2 billion in

1932. Cases of starvalion were regularly reported in the press. Urban unrest grew, and many

localilies, slrelched 10 the limit and unable to raise tax es, faced insolvency57

This economic catastrophe transformed American politics. The Depression constitutes such a massive

fact of American polilical development that it is somehow easy to pass over the matter quickly.

Analysts of social policy often seem to view the Depression as a moment which simply facilitated the

American welfare state's efforts to "catch up" with its European counterparts. To counter this view, it

is worth emphasizing the conclusion of the previous section. By 1929, the United States was the

world's leading economic power. Yet in sharp contrast to all other nations near the same stage of

industrialization, the United States on the eve of the Depression showed no signs of building a

nalional welfare state. lndeed, the evidence of a public social role even at the state level was at best

extremely patchy58 American policymakers were locked in an institutional framework that

56 Skocpol, Prolecting Soldiers and Mothers, p. 51. 57 See lrving Bernstein, The Lean Years: A Hislory 01' Ihe American Worker, 1920-1933 (Baltimore: Penguin

Books), 1970; Piven and Cloward, Regulating the Poor, Ch. 2; Frances Fox Piven and Richard A. Cloward, Poor People's Movemenls: Why They Succeed, How They Fail (New York: Random House), 1977, Ch. 2.

58 Thus I disagree wilh Skocpol's conlention thai Ihe United States "ca me close to forging a maternaIist welfare state" after the turn of the century. Skocpol. Protecting Soldiers and Mothers. p. 2. While her analysis of political mobilization around a "rnaternalist" program offers a persuasive account of the

28

foreclosed national action and - by shifting responsibility to the states - maximized business'

privileged position . Analysts need to explain not just the American welfare state's "Iateness" but the

fact that it arrived at all. In the late I 920s, the United States showed no signs of moving toward even a

weak national welfare state. Instead, it was becoming a radically different kind of political economy,

in which a common economic market co-existed with only the most feeble and decentralized

structures of social regulation . 59

The shoek-wave of the Depression , however, led to substantial shifts in the balance of political power

among the major parties, in stitutional actors, and economic groups. The interaction between economic

interests and institutions was quite different in this new context than it had been before the New Deal,

and broke (at least partially) the logjam in American social policy . In the rest of this section, I outline

the shifts in power relations that accompanied the Depression . The next section will chart the impact

of these shifts on soeial policy development.

The Depression profoundly affected the power of different actors. The most dramatic and easil y

documented shift was in the balance between the two major parties. AFter the critical election of 1896

the Republican party had established a clear ascendancy in national politics, wh ich continued through

the I 920s. Closely allied with business interests and deeply hostile to social legislation, Republican

predominance was a major barrier to domestic reform. The e lections of the 1930s, however, totally

reversed the parti san balance. Like ruling parties elsewhere, Republicans paid a high price when the

economy collapsed, and Hoover's ineffectual governance magnified the electoral penalty. Table 3

shows the results. The election of 1932 gave Roosevelt a huge majority . The Democrats' majorities in

Congress -- 3 I I -116 in the House and 60-35 in the Senate -- were the largest the party had ever

received and the largest for any party since 19 I 060 The 1934 and 1936 elections cemented

Roosevelt's remarkable victory. 1934 was only the second mid-term election in American history

when the majority party picked up seats in both the House and Senate. 1936 completed this

extraordinary electoral sweep. FDR buried Alf Landon , losing only Maine and Vermont. The line-up

in Congress was stunning: in the Senate, Democrats outnumbered Republicans 76-16, and werejoined

by four left-Ieaning Senators from other parties; in the House, 33 I Demoerats and 13 Farmer­

Laborites and Progressives faced only 89 Republicans. The Republican Party had lost its institutional

leverage in Washingtoll.

legislation that did pass before 1932, her position exaggerates the scope and significance of these interventions. What most distinguished pre-New Deal social policy was its eXlremely minimal nature.

59 The closest parallel might be the contemporary European Community, though in Europe the decentralized sodal policy structures (Le., pre-existing national welfare states) are obviously far stronger. See Stephan Leibfried and Paul Pierson, eds. European Social Policy: Between Fragmentation and Integratjon (Washington, D.C.: Brookings), 1995.

60 Same progressive Republicans were also willing to support New Deal legislation. James T. Patterson, Congressional Conservatislll and the New Deal: The Growlh of the Conservative Coalition in Congress, 1933-1939 (Westport, Cl.: Greenwood), pp. 17, 33.

29

Table 3: Electoral Realignment, 1929-1937

~ Senate

Yl1= Congress lliill! ~ Qtb.tl lliill! B!m Other

1929-1931 71st 167 267 1 39 56 1 1931-1933 72nd 220 214 1 47 48 1 1933-1934 73rd 310 117 5 60 35 1 1935-1936 7 4th 319 103 10 69 25 2 1937-1938 75th 331 89 13 76 16 4

Source: V.S. Department of Comrnerce, Bureau of the Census. Historical Statistics of the United States: Colonial Times to 1970, 93rd Congress, Ist Session, Document No. 93-78 (Part 2), p. 1083.

Among the ehief benefieiaries of this realignment were southem politieians . While the North and

West had long experieneed intense partisan eompetition, the South had been solidly Democratie for

deeades . Combined with the elosed nature of the southem one-party system, this tradition meant that

southern representatives had gained suffieient seniority to move rapidly into positions of power in the

newly Demoeratic Congress. Throughout the New Deal, this strong institutional position was to give

southern politicians influence greatly disproportionate to the South's population or economic strength .

The main institutional beneficiary of the Depression was the Presidency. The electoral result of 1932

gave FDR a strong hand. Although Congress gradually reasserted itself, Roosevelt was able to parlay

the need for forceful action and his own immense popularity into a dramatic expansion of Presidential

power.

If some actors found their relative power enhanced by the economic crisis, others were suddenly

vulnerable . A crucial group among the laller were state governments, which had been the centers of

domestic policy-making before the New Deal. States were severely constrained in the harsh economic

climate of the 1930s. State and local tax revenues fell from $6.8 billion in 1930 to $5.7 billion in

1932. Reliant on regressive tax levies that dwindled with the sharp fall in production, many would

have been hard-pressed to maintain previous levels of activity. lnstead, they were asked to respond 10

unprecedented demands for assistance. A number of southem states teetered on the edge of

bankruptcy, and most states faced rising interest eharges from fearful lenders61 In this environment,

even state governments inclined to vigorously protect their poliey prerogatives were forced to turn to

the federal govemment for leadership.

Most important, the Depression and ensuing political events dramatically altered the position of

business. The earlier discussion of the roots of business power suggests that these transformations

would reduce the leverage of employers in a number of ways. First, the indirect, structural power of

61 Patterson , The New Deal and the States, pp. 31 -2.

30

business declined. Because private investment had already plummeted, the usual impact of an implied

or actual "investment strike" was diminished. These circumstances gave policymakers an unusual

amount of room to maneuver. Even more important was the changing role of political institutions . As

the center of po licymaking shifted from state capitols to Washington, policymakers were freed from

the shackles of interstate competition. The capacity to formulate national policies gave federal

official s an opportunity to cope with the problem of capital mobility, and sharply reduced business'

structural power.

The direct, instrumental power of business also declined after 1932. The electoral realignment of the

early 1930s greatly weakened the business community's most important political allies in the national

Republican party . Although Democrats in Congress were hardly immune to business pressure, they

were far less malleable than the previous Republican majori ty62 Their susceptibility to business

influence was further reduced by another development : the widespread popular mobilization on social

issues. The ri se of unions, growing unrest among the unemployed, and the emergence of the

Townsend movement for old age pensions gave political force to calls for social reform. The

appearance of political figures like Huey Long, who stood ready to capitalize on these mobilizations,

made politicians much more sensitive to alternative political demands than had been the case a few

years earlier. Thus, while employers retained substantial instrumental influence in national politics,

their agendas now faced much greater political competition.

In short, the fundamental change in economic conditions had major repercussions for the balance of

political power. The shifts, it must be emphasized, were relative ones. Employers, state governments

and even, on occasion, Republicans, continued to playapart in the construction of social policy . But

the influence of all these actors declined, and this decline created openings for national policy

initiatives that had been impossible be fore 1929.

II is difficult to overstate the importance of the Great Depression in the development of American

social policy . The Depression altered the structure of the American economy, the functioning of

political institut ions, and the interaction between the two. In a matter of a few years, an environment

that had been relentlessly hostile to social initiatives was transformed into one that permitted a flurry

of policy activity.

Social Policy Durillg The New Deal: Scope and Limits 0/ Business Influellce. How much of a role,

then, did business play in the development of social policy du ring this crucial period? This quest ion

has been the subject of considerable dispute, with some analysts maintaining that business played a

62 It is revealing that Gordon, who sees business as a powerful actor in shaping New Deal sodal policy, finds it necessary to mention the 1932 election only twice in passing, and then only to nole that Roosevelt was elected and that there was "a large turnover in congressional ranks." Gordon, New Deals, pp. 163-4. 168.

31

central part in shaping the Soeial Security Act63 It is elear, however, that most employers strongly

opposed the Social Security Act. In a soc iety where business had played such a pervasive role in

political development, some businessmen inev itably remained involved, directly or indirectly, in

policy formulation. A handful , at most, were genuinenly enthusiastic. Even among the small faction

of employers who supported the SSA, however, were many who did so only because they feared that

the alternative would be worse. In other words, support for the SSA often represented a strategic

response to a loss of political power. Once one attends to the significance of anticipated reactions, it

becomes elearer that only a sharp decline in business power made the Social Security Act possible.

In the years before 1935, employers were almost universally opposed to any proposals for social

insurance, particularly at the federallevel. Although so me among the ti ny proportion of firms that had

their own welfare capitalist arrangements backed efforts to level the playing field, even these

employers rejected policies that would restrict their own ability to run their private programs as they

chose. Berkowitiz and McQuaid, who generally play up the role of business, acknowledge that "".

Dennison, Swope, Folsom, Edward A. Filene, and others ". all retained elear preferences for

corporate, as opposed to govemmental , activity in the welfare field . All shared the assumption that

government -- as opposed to private industry -- operated without the necessity of profit-maximization

and would thus oversupply welfare services. ,,64 Marion Folsom of Eastman Kodak, a leading welfare

capitalist, argued in the early 1920s that "it would be almost impossible to prevent government from

making those changes [from self-financing to an outright dole] so far as unemployment in surance type

programs were concerned.,,65 Gerard Swope of General Electric, while far more open to government

intervention than most of his competitors, sought only something Iike a "nightwatchman" federal role

in social policy, with govemment providing loose oversight over alrade association-dominated

system of social regulation. As late as 1931, Swope was testifying be fore Congress to this effecl: "I

think that industry ought to take care of its own [social welfare] difficulties and problems. You see,

the moment government begins to help there is no economic restraint. You can vote money.,,66 Of

course, even the modest government role in soeial policy which Swope envisioned was anathema to

63 See, among many, Gordon, "New Deal , Old Deck"; Berkowitz and MeQuaid, Creating the Welfare State; and Jill Quadagno, "W elfare Capitalism and the Social Securily Act of 1935," American Sociological Review, Val. 45 , Oetober 1984, pp. 632-47; J. Craig Jenkins and Barbara G. Brems, "Social Protest, Hegemonie Competition, and Social Reform: A Politieal Struggle Interpretation of the Origins of the Ameriean Welfare State," Ameriean Soeiologieal Review, Val. 54, 1989, pp. 891-909.

64 Ed Berkowitz and Kim McQuaid, "Businessman and Bureauerat: The Evolution of the American Social WelareSystem, 1900-1940," Journal ofEeonomie History, Val. 38, No. I, March, 1978, p. 128.

65 Quoted in Nelson , Unemployment lnsurance, pp. 45-46. 66 U.S. Congress, Senate Seleet Committee on Unemployment lnsurance, Hearings on S.R. 483 , 72nd

Congress, Ist Session, 1932, pp. 29-30. On the Swope plan see Kim MeQuaid, "Corparate Liberalism in the Ameriean Busi ness Community, 1920-1940," Business History Review, Val. 52, No. 3, Autumn, 1978, pp. 353-4.

32

most business actors. While NAM and the Chamber of Commerce favored so me of Swope's price­

fixing ideas, they had no interest in soc iallegislation.

In short, very few in the corporate sector advocated anything remotely like the sweeping expansion of

federal social po licy authority contained in the SSA, including compulsory old age and

unemploy ment insurance, a maj or new federal role in social assistance, and very substantial payroll

taxes. There is no real dispute that by late 1934, despite widespread popular agi tat ion for major social

legislation, the overwhelming majority of business actors opposed not only the New Deal in general

but the Social Security Act in part icular67 According to Folsom, a participant in the CES advisory

council who had little reason to understate business support, "only about five per cent of employers"

were for "anything along the lines" of Social Security . Far from having chosen a representative group

of employers, Roosevelt was "Iucky to get five to go on a committee" advising the CES6 8 Even these

wanted the absolute minimum federal role. The National Association of Manufacturers expressed

opposition from the beginning. The Chamber of Commerce initially pursued dialogue and negotialion,

but repudiated the cautious stance of its President Henry Harriman in May 1935 and passed aseries of

resolutions expressing strong oppositi on to the New Deal, including the SSA69

In gaugi ng employers' attitudes toward reform, it is crucial to remember that the business community

had strong reasons to understate its degree of opposition to what was c1earl y a popular program. A

typical gambit of business representatives was to express vague support for the concept of soc ial

security before launehing into virulent critiques of the "details", urging delay or the break-up of the

omnibus legislat ion into discrete pieces. As Edwin Witte, the chief architect of the SSA observed,

these strategie stances were clearly intended to scuttle .11 or most of the leg islation 70 They refl ected

a recognition on the part of prominent business actors that out-and-out public opposition was

imprudent.

Those who depict business as a powerful force in the formulalion of the Act emphasize that the

ad mini stration drew support from within the business community, and point to the role of busi ness

leaders on the Advisory Council to the Committee on Economic Security. Yet the fact that the

administration could find a handful of supporters from the enormous ranks of corporate execulives is

67 For fu rther ev idence see Theda Skocpol and Edwi n Al1lenta, "Did Capitalists Shape Social Security?" American Sociological Review, Vol. 50, August, 1985, pp. 572-75; and Edwin Amenta and Sunita Parikh, "Capitalists Did Not Want the Social Security Ac!; A Critique of the Capitalist Dominance Thesis," American Sociological Review, Vol. 56,1991, pp. 124-29.

68 Marion B. Folsom, Columbia Oral History Project memoir, Social Security volume (typescript), pp. 9-1 1, Columbi a Univers ity Libraries, New York.

69 On this c\ash see Roben M. Collins, The Business Response to Keynes (New York; Columbia University Press), 198 1, pp. 36-40. This dynamic was repeated almost 60 years later, as the Chamber's leadership initial wi llingness to consider Clinton's national health care proposals was repudiated by an oUlraged membership.

70 Schlabach, Cautious Reformer, p. 145; WiUe, Development ofthe Social Security Act, pp. 78-9.

33

hardly impressive71 Even if it could be shown that these actors were crucial (wh ich, as I will argue,

it cannot), it would still be true that the Social Security Act passed despite the fact that it could claim

backing from only a tiny minority among the ranks of American business.

It is, however, important to examine the scope and significance of this modest circle of "supporters."

Two points are crucia!. The first is that even the administration's thin support appears in many cases to

have reflected a strategic accommodation to federal legislation rather than any genuine enthusiasm. It

reflected, in other words, the weak position of business in the debate over domestic reform. As

already mentioned, no one in the business community was advocating such sweeping reforms before

popular pressures put them on the political agenda. Against a background of tremendous social

ferment, "sympathetic" employers stressed time and again the dangers of opposition to social reform.

Folsom, criticizing the Chamber of Commerce's decision to oppose the Social Security Act, argued

that "employers must realize that the country is facing an old age and an unemployment problem ...

and that legislation to meet these problems is inevitable.,,72 As Witte observed of Harriman, one of

the most prominent business supporters, his "general attitude was that so me legislation on social

security was inevitable and that business should not put itself in the position of attempting to block

this legislation, but should concentrate its efforts upon getting it into an acceptable form." 73 The

private comments of Robert Lund, the Chairman of N.A.M., to a group of prominent employers in

October 1934 sum up the predicament of employers:

71

I don't think any of us have been aggressive enough to think we could go down there [to the

White House] and demand this or demand the other thing. If we could demand it -- and get it -­

that would be the quickest way to straighten out this situation. What we are doing insteead is

attempting to adjust ourselves and modify or change or meet half-way, the situation. I have in

mind, particularly, this question of unemployment insurance ... The attitude of [the Department

of Commerce's Business Advisory Council] is not to do the thing which I think almost

unanimously they think should be done -- that is, to oppose unemployment insurance as a thing

that is impractical, that it will not accomplish the result; that it is merely a superficial attack and not the sound solution of unemployment. ... Their disposition is to support something which they themselves don't believe in. Of course, I wouldn't want to have what I am saying repeated, but

some of us in that group are very strongly of the opinion that instead of attaeking it in that way,

that is, meeting half-way what we know is fallacious, and trying to make it as bearable as

possible, the strong position for us to take is for us to stand for what we know is sound in the

way of soeial insurance. 74

Four of the five executives on the Advisory Council were frorn finns that were among the liny minority of American corporations al ready operating private unemployment plans. The fiflh , Walter Teagle. was a well-known welfare capilalist.

72 Quoted in George Wolfskill and John A. Hudson. All but the People: Franklin D. Roosevelt and His Critics (Landon: MacMillan), 1969, p. 159. They seem to eile: "Business and Finance," Time magazine .. May 11 , 1936, pp. 73-74.

73 Witte, Development of the Social Seeurity Aet, p. 89n. 74 Statement of R. Lund, National Industrial Conferenee Board Meeting, Oetober 25, 1934, pp. 54-55 . Hagley

Museum Library .

34

Lund's comments reflect a position of weakness, not power. The administration shrewdly exploited

the conservatives' sense of vulnerability wherever possible. In a key congressional showdown over

Old Age Insurance, Witte's successful attempt to sway wavering Senators avoided any stress on the

virtues of the administration's plans. Instead, he emphasized that given political realities (he

alternative would be worse than the administration's proposal 75

In addition to recognizing the degree of accommodation to political realities among Roosevelt's

business "supporters", one should not exaggerate the impact of these employers on the content of the

bill. As I have emphasized, it is amistake in any event to draw too many conclusions about relative

influence from conflicts over details of Iegislation . Nonetheless, since those who point to business

influence over the SSA have stressed the impact of employers on legislative design, it is worth

examining their efforts. Two main paths of business influence on the specifics of the SSA have been

proposed. The first was the role of Industrial Relations Councilors (an organization with links to

Rockefelier interests) in providing much of the research for the unemployment insurance provisions

of the SSA 76 The second was the presence of a number of employers on the Advisory Council of the

CES.

None of these paths of influence was of great importance. The key features of the unemployment

insurance proposal eventualiy enacted had emerged weil before the hiring of Industrial Relations

Councilors. The "tax-offset" sc he me that was the plan's cornerstone was developed by Wisconsin

economist Paul Raushenbush and his wife Elizabeth Brandeis in Wisconsin during late 1933, and had

been included in the Wagner-Lewis unemployment insurance bill in 1934.77 The detailed accounts of

major participants (Witte and Altmeyer) show little indication of IRC influence. Indeed, it is difficult

to identify a single significant provision of the Act that was suggested by IRe.

The Advisory Council role was also modest. The group was not organized until November 1934, weil

after most of the work on the SSA had been completed. Its timing and make-up strongly suggest that

it was constituted largely for the purpose of public relations. More important, there is no sign that any

of its actions had an impact on the content of the SSA7 8

75 United States Senate, Commiuee on Finance, Sen are Hearings on the Economic Security ACI, Hearings on S. 1130, 74th Congress, I st Session, January 21-February 12, 1935; Wille describes this as "the only possible argument" 10 win over conservative vOles. Development of the Social Security Act, p. 10311.

76 See especially G. William Domhoff, "Corporate-Liberal Theory and the Social Security Act: A Chapter in the Sociology ofKnowledge," Politics and Society, Vol. 15, no. 3 (1986-87), pp. 113-16.

77 Schlabach, Cautious Reformer, pp. 93-4. Justice Louis Brandeis, Elizabeth Brandeis' father, had brought to her attention the factthatthe Supreme Court had upheld the constitutionality of a federal tax-offset system in the 1927 caseofFlorida v. Mellon.

78 Witle, Development of the Social Security Act is instructive on both the composition and role of the Advisory Committee.

35

Indeed, the evidence from the few clear instances where the policy preferences of these and other

business actors were articulated is telling. Factions of employers made three visible efforts to

infIuence specific features of the Social Security Act: they pushed for greater federal control over

unemployment insurance, advocated employee contributions to unemployment insurance, and favored

the "Clark Amendment," which would have exempted firms with existing pension systems from the

aAl program. Each 01 these proposals was deleated. Bryce Stewart, the head of IRC, lobbied for a

more centralized, "subsidy" plan for unemployment insurance and received backing from all the

employers on the Advisory Council, as weil as majority support on the Council as a whole79 The

CES rejected this proposal, expressing doubts about the plan's constitutionality and anticipating

heated Congressional opposition. The President backed the CES, and the subsidy plan received no

attention in Congress, where discussion centered on even more decentralized options. Employers on

the Advisory Council also pushed for employee contributions to unemployment compensation. Again,

the CES refused to go along.

A broader push for changes came from outside the small circIe of business leaders who worked with

the bill's drafters. During congressional debate, insurance lobbyists fought for the Clark Amendment,

which would have exempted firms with pension systems from Old Age Insurance. Many employers

backed the amendOlent, and it was adopted in the Senate. Had this revision remained in the

legislation, it would have dramatically altered the character of aAl. The option of "contracting-out"

might have led to a two-tiered system of insurance, and saddled the public sector with the worst

risks80 Facing strong opposition from both FDR and the House participants in the conference

committee on the SSA, however, the Senate eventually capitulated.

There is no convincing evidence that the power of business actors was important in producing this

path-breaking piece of national soeial legislation. The political coalition that initiated these reforms

79 The reason for this stance is unclear. Skocpol and others have implied that the employers on the advisory council simply wanted more centralizatiol1 and uniformity Lo avoid "balkanizalion." Skocpol and Ikenberry, "Political Formation of the American Welfare State," p. 128. Before, during, and after the 19305, however, employers have almosl never taken that stance. Rather, these large employers wanted even more freedom LO experiment with alternative plans that they could operate independently or by seetOT,

and feared that distinctive state laws would make that difficult. See Testimony of Marian Folsom . Senate Hearings on S. 1130 (Washington, D.C.: U.S. Government Printing Office), 1935, p. 582. Thanks to Peter Swenson fOT this point. Most businesses outside the "welfare capitalist" circles of the BAe probably preferred the more decentralized approach eventually adopted. Domhoff thus regards this as only a minor dispute between "two slightly different business plans." Dom hoff, "Corporate-Liberal Theory and the Social Security Act," p. 3 16. But of course, the crucial point is thai the businesses preferring less centralization were only comparing IwO bad options: their true preference was for no unemployment insurance plan at all.

80 Indeed, given the difficult time that OAl eventually had supplanting the means-tested OAA program. it is more likely that the ultimate result would have been a system divided between a public means-tested program and a private system of earnings-related pensions. On the dynamic between OAA and OAI see Jerry R. Cates, Insuring Inequality: Administrative Leadership in Social Security, 1935-54 (Ann Arbor: University of Michigan Press), 1983.

36

was not a truly "cross-class coalition" in which major segments of business were enthusiastic

advocates of government intervention. Different factions of business had distinct interests, and so me

were much more hostile to federal action than others. Yet the fact remains that the overwhelming

majority of business actors opposed the Social Security Act or anything like it. Among the tiny

minority who were supportive, many were motivated primarily by fear of the likely alternatives.

There is, furt her, absolutely no sign that the even smaller minority who generally favored such a

major expansion of federal activity played anything like a decisive political role.

This does not mean that employers were powerless. The strongest case for business influence at this

juncture concerns two less direct effects. First, policymakers' concerns about business confidence

continued to set so me broad restrietions on the range of policy initiatives considered. There is

evidence that this background, structural influence of employers remained a factor in the formation of

the SSA. Witte, for ex am pie, acknowledges the concerns of Treasury officials that proposals for

soeial insurance were "alarming" to business, and he adds that "many people close to the committee

[on economic security 1 were ... anxious to allay these fears." He argues that this anxiety influenced

the timing of the administration's proposals, as weil as the decision to phase in payroll tax es gradually

over three years81 Neither of these decisions involved crucial features of the legislation . One could

speculate that such concerns about business confidence also placed broader constraints on wh at the

administration regarded as feasible, but lhere are no indications that this was an important factor in

shaping social security.

A second source of employer influence on the SSA is both more important and more difficult to

measure. This influence was expressed through the critical interventions of southern Democrats in

Congress82 One of the distinctive consequences of federalism is the way in wh ich it activates lhe

territorial component of economic interests83 In the United Stales, this territorial division made the

peculiar Democratic alliance particularly vulnerable on issues related to restructuring lhe political

economy. To summarize crudely, the northern wing of the party, drawing on urban working-class

support, was inclined towards policies of social demoeratization, including generous and centralized

soeial programs. This programmatic agenda was alarming to the southern wing of the Democratic

coalition. The Soulh during this period constituted a truly distinctive political economy. Networks of

political and economic domination, backed up by the volatile element of racial hatred, were tightly

81 Witte, Development of the Social Security Act, p. 6611. 82 The following draws particularly on Bensei, Sectionalism and American Political Development, chs. 4 and

5; 1ill Quadagno, "From Old-Age Assistanee 10 Supplemental Seeurity Ineome: The Politieal Eeonomy of Relief in the South. 1935-1972," in Weir, Orloff, and Skoepol, Polities of Social Poliey, pp. 235-63 ; and Kenneth Finegold, "Agrieulture and the Polities of V.S. Social Provision: Social Insuranee and Food Stamps," in Weir. Orloff. and Skoepol, Politics of Social Poliey. pp. 199-234.

83 See Paul Pierson, "Fragmented Welfare States: The Development of Social Policy in Federal Systems," Governance, Vol. 8, No. 4 , Oetober, 1995.

37

linked and mutually reinforeing. Sorting out the role of raee and c1ass in this system is diffieult, but

both led the southern politieians who represented the ruling white oligarehy to guard loeal

prerogatives fiercely . This was particularly true when it eame to policies governing the labor market.

Ruling interests in the South had a strong desire to avoid any national policies that might weaken

blaek shareeroppers' dependence on low-wage work, or threaten the low-wage, non-union

environment that made the South attractive to potentially mobile capita!. Southern employers thus

strongly advocated that the design of the Social Security Act be as deeentralized as possible.

Again, the ability of partieular economie aetors to exert influence depended heavily on institutional

variables. The partieular structure of American federal institutions both activated specific policy

preferences (by making the advantages of decentralized policies to southern economic interests

evident) and strengthened their advocates (by placing southern representatives in a strong politieal

position). America's peeuliar system of seniority-based Congressional eommittees gave southern

representatives tremendous leverage. Consequently, the administration's initial proposals in the Social

Security Aet bowed to many of the South's concerns. For example, participants' accounts suggest that

the decision to adopt a deeentralized system of unemployment insurance was taken partly in response

to concerns about constitutionality and partly in anticipation of congressional preferences.84 Once the

legislation reached Congress, southern representatives suceessfully pushed for further modifieations

to limit the impaet of the bill on wage levels in the region. Indeed, almost all the changes introduced

by Congress were designed to weaken federal authori ty85 States were given increased flexibility to

design their own unemployment compensation systems. Agrieultural and domestie workers were

dropped from the soeial insurance part of the system (wh ich meantthatthree-fifths of blacks would be

excluded). Provisions governing the means-tested Old Age Assistance program were modified to

enhanee local administrative control and to exclude any national standards on minimum benefits.

Thus, southern employers were part of a coalition that achieved significant changes in the Social

Security Act.

These programmatic choices have had enduring repercussions.86 The hopes of reformers that the SSA

would spur further efforts to develop a national system with stranger citizenship rights (including

national health insuranee) were not fulfilled. The New Deal proved to be an extraordinary but fleeting

moment of opportunity. After the Depression the United Stales fell baek into normal polities, and

acute institutional fragmentation prevented dramatic reform. The effort to nationalize social policy,

and henee to minimize the structural leverage of business over soeial policy, was only partially

84 See in particular WiUe. Developmenl of the Social Security ACL 85 Witte, Developmenl of the Social Security Act, p. 62; Quadagno, "Political Ecollomy of Relief," pp. 237-

47. 86See especially Theda Skocpol, "The Limits of the New Deal System and the ROOlS of Contemporary Welfare

Dilemmas," in Weir, Orloff and Skocpol , Politics of Social Policy, pp. 293-311.

38

successful. The continued decentralization of those policies particularly consequential for labor

market conditions helped to sustain a regionally bifurcated political economy. The growing

competitive pressure of a low-wage, non-union Southern alternative severely constrained the

prospects for "social democratic" reforms in the North after the New Deal87

That the transformation of American social policy during the New Deal was incomplete, however,

should not distract attention from the magnitude of the changes introduced. The Social Security Act

created a national system of old age insurance, as weil as essentially mandatory state programs of old

age assistance, unemployment insurance, and aid to dependent children. Equally important, the SSA's

provision for payroll taxes and federal matching grants for state social programs laid the fiscal

foundation for the modern welfare state. These reforms, I have argued, were possible only because of

the great change in the scope and nature of business influence during the I 930s. With the shift of

political action to Washington, the structural power of business declined dramatically . The rise of

social movements meant increasing competition for political power. Business interests remained a

force in politics, especially when they had access to strategically-Iocated veto points. But employers

ceased to occupy the privileged position they had held in struggles over social legislation prior to

1929.

111. ConcIusion: Politics, Markets and Sodal Poliey

Competing versions of "new institutionalism" have swept political science. It is widely accepted that

the "rules of the game" in political life have profound effects on the formation of group identities and

organization, the distribution of political resources, and the prospects for alternative political

strategies. As is always the case, however, bringing important phenomena "in" to social analysis tends

to push other relevant features "out." While discussions of political and policy development have

already proceeded beyond simple dichotomies between "state-centered" and "society-centered"

explanations, efforts to integrate the role of social and economic actors with an understanding of

institutional constraints remain halting.

The key to moving forward is to be clear about the pathways through which business actors influence

public policy . There are two distinct dimensions of business influence: instrumental power, exerted

87 Later developments confirm that it was not simply the desire to sustain a racial oligarchy that lay behind the South's distinctive posture. The fate of further efforts (in the early and late 1970s and again in the mid-1980s) to "nationalize" welfare were largely unsuccessful, primarily because of continued southern opposition. The desire 10 retain a low-wage, non-unionized economy attractive to mobile capital has been an enduring southern strategy. See Paul Pierson, "The Creeping Nationalization of Income Transfers in the United States," in Leibfried and Pierson, eds., European Social Policy.

39

through lobbying, and structural power, exerted through the political repercussions of private

investors' control over mobile capita\. The leverage of business over policy development depends on

both these elements. Each is subject to variation across countries, over time, and from issue to issue.

Institutional factors playa crucial role in determining these variations, thus mediating the power of

employers in political Iife.

The interaction between changing market dynamics and the fragmented political institutions of the

United States can clarify the origins of America's exceptional welfare state. There is no doubt that the

unusual institutional structures that govemed decision-making profoundly influenced the formation of

American social policy. This was true both during the early period of limited governmental action and

then, following the onset of the Depression, in the shaping of the Social Security Ac!. However, this

observation need not be placed in opposition to a view that stresses how economic forces contributed

to the shaping of social reform.

Many of the key puzzles of social policy development can be addressed once one complements an

examination of the instrumental power of business with a focus on the political significance of capital

mobility in different institutional settings. The responsibilities allocated to business in market

economies create a climate in which private investment decisions have profound political

consequences. Because employers control investment, they influence policymaking even without

consciously attempting to do so. Political institutions, however, mediate this influence. In a federal

system, market dynamics are translated into intense interstate rivalries. Before the Depression this

rivalry inhibited state level social policy formation . Because the Depression opened the door for

federal action, it sharply reduced the structural power of business. In this changed environment, a

massive expansion of the federal role proved possible. During and after the New Deal, institutional

arrangements, including federalism, continued to mediate the efforts of economic actors to exercise

political power. Perhaps most important, southern politicians eager to attract business and turn

interstate rivalries to their advantage were able to block the establishment of a fully national welfare

state. But the larger point of this essay is that the changing role of institutions altered both the scope

and character of business power.

This effort to wed new institutionalist and political economy approaches to social policy development

is not simply a call for compromise. It is an attempt to link together the insights of both perspectives,

showing how each can be strengthened by the other's arguments. Hopefully, it offers persuasive

evidence that a framework incorporating a significant role for economic actors and processes as weil

as institutional structures can address fundamental questions in a compelling way. The "rules of the

game" clearly matter a great deal, but it is equally true that the development of economic policies

cannot be understood without close attention to the nature of modern market economies and the

interests, strategies. and influence of economic actors.