Theda Skocpol, “A Society without a ‘State’? Political Organization, Social Conflict, and Welfare Provision in the United States”
“A Society without a ‘State’? Political Organization, Social Conflict, and Welfare Provision in the United States”
Journal of Public Policy, Vol. 7, No. 4 (Oct. – Dec., 1987), pp. 349-371
This article was originally presented at the Institute on ‘Foreign Perspectives on the U.S. Constitution,’ sponsored by the American Council of Learned Societies at the Wingspread Conference Center, Racine, Wisconsin, on 29 September 1987. It draws upon some material presented in the introduction to The Politics of Social Policy in the United States, edited by Margaret Weir, Ann Shola Orloff, and Theda Skocpol (Princeton, NJ: Princeton University Press, 1988).
‘State and society’ are terms of reference bound to seem out of place in a discussion of the Constitution and governance of the United States of America. As an insightful observer once put it (Pollard 1925, 31), ‘Americans may be defined as that part of the English-speaking world which instinctively revolted against the doctrine of the sovereignty of the State and has … striven to maintain that attitude from the time of the Pilgrim Fathers to the present day.’ Citizens of the United States view themselves as fortunate not to be subject to any overbearing ‘State.’ And foreign observers rightly have trouble identifying elements of concentrated sovereignty in the American political system – except, perhaps, when the USA acts aggressively on the world stage.
This article will touch upon historical reasons why Americans lack a sense of the state. Primarily, however, I shall argue that we can learn a surprising amount about American society and politics by treating state-society relationships in more analytical terms. It is an ethnocentric illusion to imagine that the United States has been a dynamic society and capitalist economy ‘unencumbered’ by any state. Instead, the specific organizational forms that state activities have taken in America have profoundly affected the social cleavages that have gained political expression, and helped to determine the sorts of public policies that US governments have – and have not – pursued from the nineteenth century to the present day. Drawing upon my own current research, I can illustrate this argument by exploring why US patterns of public social provision differ from those associated with European welfare states. If the US constitution is construed broadly to mean not just a document drawn up in 1787-8 but an entire configuration of governance with associated cultural meanings, then this constitution has much to tell us about why American social policies have come to be as they are.
The American Version of Public Social Provision
Modern welfare states, as they eventually came to be called, had their start between the 188os and the 1920s in pension and social insurance programs established for industrial workers and needy citizens. Later, from the 1930s through the 1950s, such programmatic beginnings were elaborated into comprehensive systems of income support and social insurance encompassing entire national populations. In the aftermath of World War II, Great Britain rationalized a whole array of social services and social insurances around an explicit vision of ‘the welfare state,’ which would universally ensure a ‘national minimum’ of protection for all citizens against old age, disability and ill health, unemployment, and other causes of insufficient income. During the same period, other nations – especially the Scandinavian democracies – established ‘full employment welfare states’ by deliberately coordinating social policies, first with Keynesian strategies of macroeconomic management, and then with targeted interventions in labor markets.
American ‘social policy’ prior to the 1930s included state and local support for mass public education, along with generous federal benefits for elderly Civil War veterans and their dependants. In contrast to European tempos of welfare state development, neither old-age pensions nor social insurance made much headway in the United States before the Great Depression. Then, in what has been called a ‘big bang’ of national legislation (Leman 1977), the Social Security Act of 1935 established nation-spanning social insurance and public assistance programs, creating a basic framework for US public social provision that has remained in place down to the present. Health insurance was omitted from the Social Security Act and later schemes for universal national health benefits also failed. But three major kinds of social provision were included in the 1935 legislation: federally required, state-run unemployment insurance; federally subsidized public assistance; and national contributory old-age insurance.
Unemployment insurance was instituted in 1935 as a federal-state system. All states were induced to establish programs, but each individual state was left free to decide terms of eligibility and benefits for unemployed workers, as well as the taxes to be collected from employers or workers or both. Unemployment benefits and taxation became quite uneven across the states, and it remained difficult to pool risks of economic downturns on a national basis or to coordinate unemployment benefits with Keynesian demand management. Despite efforts in the 1930s and 1940s to nationalize unemployment insurance and join its operations to various measures of public economic planning, no such explicit joining of ‘social’ and ‘economic’ policy developed in the postwar United States.
Public assistance under Social Security was a set of programs already existing in certain states by the early 1930s, for which the federal government would henceforth share costs. Assistance for the elderly poor and for dependent children were the most important programs. Free to decide whether they would even have particular programs, the states were also accorded great discretion to decide matters of eligibility and benefits and, in practice, methods of administration. Over time, as old-age insurance expanded to cover virtually all retired employees in the United States, federal old-age assistance became proportionately less important than it was originally. Meanwhile, by the 196os, the Aid to Dependent Children program (now ‘Aid to Families with Dependent Children,’ providing benefits to caretakers as well as the children themselves) expanded enormously with a predominately female adult clientele.
Labelled ‘welfare,’ AFDC has very uneven standards of eligibility, coverage, and benefits across the states, generally providing the least to the poorest people in the poorest states, and leaving many impoverished men and husband-wife families without any coverage at all. Since 1935, the one program originally established on a purely national basis, contributory old-age insurance, has usurped the favorable label ‘social security’ that once connoted the whole, and has become the centerpiece of US public social provision. Payroll taxes are collected from workers and their employers across the country. Ultimately, retired workers collect benefits roughly gauged to their employment incomes, with some redistribution toward the low-wage contributors to the system.
After 1935, additional programs were added under this contributory insurance rubric: for surviving dependants in 1939; for disabled workers in 1956; and for retirees in need of medical care in 1965. Equally important, ‘social security’ grew in coverage and benefits, as more and more employees and categories of employees were incorporated during the 1950s, and then benefit levels were repeatedly raised by Congress. By the 1970s, the United States, uneven and often inadequate in the help provided to unemployed and dependent people, had nevertheless become reasonably generous in the benefits offered to retired people of the working and middle classes.
Properly understood, public social provision in the United States also includes many federal measures undertaken to further the wellbeing of broad sectors of the populace, including middle-strata employees and small property-owners. Some of these measures, such as agricultural price supports to shore up the incomes of commercial farmers, were launched in the New Deal. Others – such as educational aid to military veterans and grants to localities to encourage hospital construction – were established or significantly expanded in the aftermath of World War II, when higher peacetime revenues became permanently available through the extended federal income tax. Without carrying the explicit label of ‘welfare’ or ‘social policy,’ these policies often rely upon indirect means of governmental action, such as ‘tax expenditures’ (that is, incentives in the form of forgiven tax payments) rather than direct public expenditures. In other cases, federal subsidies flow to third-party providers, such as private businesses or associations or local governments, as a way to encourage the provision of goods or services to favored groups or to members of the citizenry at large. US housing policies especially exemplify the use of indirect instrumentalities. Both federal mortgage insurance and the deductibility of mortgage interest payments from federal taxes have spurred widespread private home ownership, and even federal programs to house the poor have relied on regulatory breaks and subsidies to private real estate developers.
‘Welfare’ became an explicit area of US political controversy and policy innovation only during the 1960s, when the ‘War on Poverty’ and the effort to create a ‘Great Society’ were declared. For the first time since 1935, major new programs of needs-tested public assistance were established in the form of in-kind aid through Food Stamps and Medicaid. In 1972, moreover, old-age and other assistance programs (originally established as federal programs under Social Security) were nationalized, ensuring more standardized benefits. Still, the much larger AFDC program remained federally decentralized – and standards for other benefits such as medical care are often tied to this uneven standardbearer of the American ‘welfare’ system. In turn, welfare remains, as always, both institutionally and symbolically separate from national economic management, on the one hand, and from non-means-tested programs benefiting regularly employed citizens, on the other.
In sum, a number of features of US social provision require explanation. Core welfare-state policies such as old-age pensions and social insurance were much slower to emerge in the United States than in Europe; and some – such as national health insurance – have never been instituted at all. We need to know why America’s initiation of such modern social policies was (by international standards) so belated. We also need to make sense of the two major periods of explicit social policy innovation at the national level in the twentieth-century United States: the New Deal of the 1930s, which witnessed the passage of the Social Security Act, and the Great Society and Nixon era reforms of the 1960s and early 1970s, which included new programs for the poor.
Aside from matters of historical phasing, the programmatic structure and modes of implementation of US social provision also require explanation. The public assistance and social insurance programs established under the Social Security Act have remained disjoint and nationally uneven, and these programs operate in conjunction with a variety of indirect federal incentives and subsidies that provide scattered social benefits outside of any comprehensive vision of the goals or effects of an American welfare state. Unlike the situation in Britain since the Beveridge reforms instituted after World War II, national standards have not been established in the United States for public benefits (not even the direct ones). And ‘social security’ has remained firmly bifurcated, both institutionally and symbolically, from ‘welfare.’ Equally, in contrast to
‘full-employment welfare states’ in Scandinavia, there has been little coordination of US social policies with nationally rationalized public interventions in the macroeconomy or labor markets. In the United States, various social benefits remain operationally, fiscally, and symbolically separated from one another; and they are all kept quite apart from other things the national government may be doing in relation to the economy and society. We need, in short, to better understand why the twentieth-century United States developed not a Western European-style welfare state but a much more disjoint system of public social provision.
Existing Explanations and their Shortcomings
Among many of those seeking to understand the development of social policies in the United States, several approaches currently hold sway. Each offers important insights but falls short of offering a satisfying explanation, in ways that can be briefly indicated.
One school of thought can be dubbed the ‘logic of industrialism’ approach (e.g., Cutright 1965; Wilensky and Lebeaux 1965; and Wilensky 1975: ch. 2), because it posits that all nation-states respond to the growth of cities and industries by creating public measures to help citizens cope with attendant social and economic dislocations. Once families are off the land and dependent on wages and salaries, the argument goes, they cannot easily cope with disabling accidents at work, or with major episodes of illness, unemployment, or with dependent elderly relatives unable to earn their keep. Social demand for public help grows, and all modern nations must create policies to address these basic issues of social security without forcing respectable citizens to accept aid under the demeaning and disenfranchising rules of traditional poor laws.
Plausible as this sounds, recent cross-national studies on the origins of modern social insurance policies have demonstrated that urbanization and industrialization (whether considered separately or in combination) cannot explain the relative timing of national social insurance legislation from the late nineteenth century to the present (cf. Collier and Messick 1975; Flora and Alber 1981). The United States in particular does not fit well into the logic of industrialism schema. Not incidentally, proponents of this perspective have tended to include data for ‘the US case’ only when doing cross-national analyses for the period after 1935. Before the 1930s the United States is an awkward outlier; for this country was one of the world’s industrial leaders, yet ‘lagged’ far behind other nations (even much less urban and industrial ones) when it came to instituting public pensions, social insurance, and other modern welfare measures.
Another school of thought – let’s call it the ‘national values’ approach – accepts many underlying dynamics posited by the logic of industrialism argument, but introduces a major modification to explain why some nations, such as Bismarck’s Imperial Germany in the i88os, initiated modern social policies at relatively early stages of urbanization and industrialization, while others, most notably the United States, delayed behind the pace of policy innovation that would be expected from the tempos of urbanization and industrialization alone. The answer, say proponents of this approach (e.g., Rimlinger 1971; Kaim-Caudle 1973; King, 1973; Gr0nbjerg, Street, and Suttles 1978), lies in the values and ideologies to which each nation’s people adhered as urbanization and industrialization gathered force. Cultural conditions could either facilitate or delay action by a nation-state to promote social security; and cultural factors also influenced the shape and goals of new policies when they emerged. Thus Gaston Rimlinger, one of the ablest proponents of the national values approach, argues (1971, 91) that early German social insurance policies were facilitated by the weakness of liberalism and the strength of ‘the patriarchal social ideal’ and ‘the Christian social ethic’ in nineteenth-century Germany. In the United States, however, laissez-faire liberal values were extremely strong and a ‘commitment to individual achievement and self-help’ led to a ‘tenacious’ ‘resistance to social protection…’ (Rimlinger 1971, 62) . The unusual strength and persistence of American liberal values explain why social insurance measures failed to be instituted before the 1930s, for ‘it was only when the Great Depression revealed in a shocking manner the utter defenselessness of the citizen in the American industrial state that the still potent individualistic tradition could be overcome’ (Rimlinger 1971, 193).
Arguments like Rimlinger’s offer a credible general gloss on what happened. Yet general deductions from national values simply cannot give us the answer to many crucial questions about the timing and programmatic structure of American social provision, as a couple of historical examples can readily dramatize. Laissez-faire liberal values were in many respects more hegemonic and popular in nineteenthcentury Britain than they were in the nineteenth-century United States, yet in the years before World War I Britain enacted a full range of social protective measures, including workers’ compensation (1906), old-age pensions (1908), and unemployment and health insurance (1911). These innovations came under the auspices of the British Liberal Party, and they were intellectually and politicallyjustified by appeals to ‘new liberal’ values of the sort that were also making progress among educated Americans around the turn of the century. Under modern urbanindustrial conditions, the ‘new liberals’ argued, positive governmental means must be used to support individual security; and this must be accomplished without undermining individuals’ dignity or making them dependent on the state. If British Liberals could use such ideas to justify both state-funded pensions and contributory social insurance this way in the second decade of the twentieth century, why couldn’t American progressives do the same? In both Britain and the United States sufficient cultural transformation within liberalism had occurred to legitimate fledgling welfare states without resort either to conservative-paternalist or socialistjustifications (see Orloff and Skocpol 1984 for further elaboration of this argument).
Then, too, when American New Dealers of the 1930s at last successfully instituted nationwide social protections justified in ‘new liberal’ terms, why did they end up with the specific array of policies embodied in the Social Security Act? Why was health insurance left aside, despite the availability of liberal rationales for it just as good as those put forward for unemployment and old-age insurance? And why did the public assistance programs subsidized under Social Security actually cement the dependence of many individuals on the arbitrary discretion of state and local authorities, rather than furthering individual dignity and the pre dictable delivery of citizen benefits as a matter of ‘rights’? A final query is perhaps the most telling: Given the clear value-priority that Americans have always placed on individuals getting ahead through work, why did the New Deal as a whole fail to achieve proposed measures to guarantee jobs for everyone willing to work? Arguably, the social security measures that were achieved were less in accord with longstanding American values than governmental commitments to full employment would have been.
Arguments stressing the impact of either industrialism or national values on social policy development tend to downplay political struggles and debates. During the last decade, however, many historians and social scientists have analyzed the political contributions of capitalists and industrial workers in shaping patterns of social policy in the capitalist democracies. As part of this trend, two sorts of class politics perspectives have been applied to American social politics; one highlights the initiatives of ‘welfare capitalists’ and the other stresses ‘political class struggles’ between workers and capitalists.
Proponents of the welfare capitalism approach (e.g., Domhoff 1970; Radosh 1972; Berkowitz and McQuaid 1980; Ferguson 1984; Quadagno 1984) take for granted that corporate capitalists have dominated the US political process in the twentieth century, and they look (in various ways) for economically grounded splits between conservative and progressive capitalists as the way to explain social policy innovations. Early in this century, the argument goes, certain American businesses preceded the public sector in evolving principles of modern organizational management, including policies for stabilizing and planning employment and protecting the social welfare of loyal employees. Prominent ‘welfare capitalists’ then pressed their ideas upon policy intellectuals and public officials, so that public social insurance measures in key states and at the federal level were supposedly designed to meet the needs of progressively managed business corporations.
This perspective has served as a good lens through which to view the complementarities that have often developed between public social policies – once enacted – and the labor-management practices of American corporations. For example, many American corporations accommodated nicely to Social Security’s contributory old-age insurance program, meshing it with their own retirement benefits systems, especially after World War II. But business groups originally opposed the passage of the Social Security Act, as well as the passage of most other federal and state-level social and regulatory measures favorable to workers – and this creates problems for the welfare capitalism thesis (see the full discussion and references in Skocpol and Amenta 1985). Its proponents have failed to demonstrate convincingly that genuine groups or categories of US capitalists – as opposed to a handful of maverick individuals not represen tative of any class or industrial sector – actually supported mandatory public pensions or social insurance. However adaptable American capitalists have proven to be after the fact, the historical evidence is overwhelming that they have regularly fiercely opposed the establishment of public social policies in the first place. Political processes other than the initiatives of capitalists have nearly always been the causes of US social policy innovations.
The second class politics perspective takes for granted that capitalists everywhere tend to oppose the emergence and expansion of the welfare state. This ‘Social Democratic’ or ‘political class struggle’ approach has predominated in recent cross-national research on the development of social policies in Europe and the United States (cf. Castles 1978, 1982; Bjorn 1979; Stephens 1979; Korpi 1983; Shalev 1983; Myles 1984; Esping-Andersen 1985). To explain why American public social provision commenced later and has not become as generous as European public social provision, this approach underlines the relative weakness of US industrial unions and points to the complete absence of any laborbased political party in US democracy. Given these weaknesses of working class organization, US capitalists have been unusually able to use direct and indirect pressures to prevent governments at all levels from undertaking social-welfare efforts that would reshape labor markets or interfere with the prerogatives or profits of private business. Only occasionally – most notably during the New Deal, and afterwards through the liberal wing of the Democratic Party – have American workers or unions been able to muster sufficient strength to faciliate some innovations or expansions of public social provision.
Certainly this emphasis on political class struggle between workers and capitalists helps to explain why the United States has not developed a comprehensive full-employment welfare state. Just as the political class struggle approach would have it, US unions have frequently supported extensions of public social provision, while business groups have opposed them. Thus, policies more similar to those of Sweden probably would have come about in the United States had American workers been as highly unionized or had the modern Democratic Party been a truly socialdemocratic party based in the organized working class.
Nevertheless, if our intention is not merely to contrast the United States to Europe but also to explain the specific patterns of modern American social policy, then the Social Democratic perspective on political class struggles is insufficient in several ways. Strict attention to political conflicts of interest between capitalists and industrial workers deflects our attention from other socioeconomic forces that have intersected with the US federal state and with decentralized American political parties to shape social policymaking. Until very recently, agricultural interests in the South and West were crucial arbiters of congressional policymaking.
And struggles over social welfare or labor market interventions have often involved regional, ethnic, and racial divisions. We need a mode of analysis that will help us understand why these conflicts have been equally or more telling than industrial class conflicts in the shaping of social provision in the United States. Political class struggle theories have been argued with certain state and party structures in mind, namely centralized and bureaucratized states with parliamentary parties dedicated to pursuing policy programs in the name of entire classes or other broad, nationspanning collectivities. For much of Europe, the existence of such features of political organization has given substance to the presumption that the industrial working class may translate its interests into social policies, whenever ‘its’ party holds the reins of national power over a sustained period. But, of course, the United States has never had a centralized bureaucratic state or programmatic parliamentary parties. Thus the American case highlights the importance of bringing much more explicitly into our explanations of social policymaking the historical formation of each national state – as well as the effects of that state’s institutional structure on the goals, capacities, and alliances of politically active groups.
American State Formation and the Limits of Social Provision
‘State formation’ includes constitution-making, involvements in wars, electoral democratization, and bureaucratization – large-scale historical processes, in short, whose forms and timing have varied significantly across capitalist-industrializing countries. In sharp contrast to many European nations, the United States did not have a premodern polity characterized by monarchical absolutism, a locally entrenched standing army and bureaucracy, or recurrent mobilization for land warfare against equal competitors. Instead, the American colonies forged a federalist constitutional republic and (after some years of continued sparring with Britain) the fledgling nation found itself relatively geopolitically sheltered and facing toward a huge continent available for conquest from militarily unformidable opponents. Wars have never had the same centralizing effects for the US state as they have had for many European states, in part because America’s greatest war was about itself, and also because mobilization for both the World Wars of the twentieth century relied heavily on the organizational capacities of large business corporations and trade associations (Cuff 1973; Vatter 1985). Only after World War II, when the United States took on global imperial functions, did a federal
‘military-industrial complex’ emerge, nourished by the first persistence into peacetime of substantial direct federal taxation.
The American Revolution was a revolt not only against the British Empire but also against any European-style notion of concentrated political sovereignty – whether focused in a supreme parliament, as in Britain after the English Revolution, or in an official bureaucracy built up under absolute monarchy, as in much of Continental Europe. After years of political skirmishes between colonists and royal governors, a confederation of thirteen colonies separated Americans from Britain; then the founding fathers sought to cement a precarious national unity by designing a new federal government. Under the Constitution adopted in I 788, the powers of the states and the central government were carefully divided and balanced against one another in a ‘compound’ arrangement (Scheiber 1978) that left many ambiguities for the future, while the new rules for the federal government spread cross-cutting responsibilities among Congress, the President, and a system of courts. In Samuel P. Huntington’s words (1968, 11o), ‘America perpetuated a fusion of functions and a division of power, while Europe developed a differentiation of functions and a centralization of power.’
Skeptical of the desirability of any one center of authority acting in the name of the national good, the sponsors of the Constitution aimed for a governmental system that would protect against executive arbitrariness and check selfish interests in the state legislatures, yet at the same time maintain informal hegemony for gentlemen holding civic ideals of disinterested republican virtue (Wood 1987). Antifederalists were even more distrustful of centralized authority and almost defeated the proposed Constitution. Within a decade after ratification, however, early American political factions converged on a shared language of ‘constitutionalism’ to articulate sharply conflicting ideals about the scope of governmental authority; each contending faction argued that it was correctly implementing the Constitution’s provisions.
Henceforth, Americans looked to ‘the Constitution’ and ‘the rule of law,’ as the loci of fundamental sovereignty. Especially in pre-Civil War America, these functioned as a ‘roof without walls’ in the apt words of John Murrin (1987), as ‘a substitute for any deeper kind of national identity,’ because ‘people knew that without the Constitution there would be no America.’ While neverending rounds of legislation in Congress and the states expressed shifting sets of special interests, the sovereign ideals of constitutionalism and the rule of law could reign impersonally above an economically expansionist and socially diverse country. Only during the Civil War did a Republican-run crusade to save a Northern-dominated nation temporarily transfer the locus of sovereignty to an activist federal government. But even the Civil War did not generate an autonomous federal bureaucracy; and the forces of localism, divisions of powers, and distrust of government activism never disappeared even in the North. The American ‘Tudor polity’ (Huntington I968) re-emerged full force after the Southern states rejoined the union in the 1870s.
Europeans had concentrated sovereignties and a sense of ‘stateness’ while Americans had neither. Stephen Skowronek has placed the totality of early American political arrangements in a framework that helps to highlight their distinctive features. Skowronek points out (I982, 19, 24) that America certainly did have a state, both in the sense of ‘an organization of coercive power’ and in the sense of ‘stable, valued, and recurring modes of behavior within and among institutions’:
The early American state maintained an integrated legal order on a continental scale; it fought wars, expropriated Indians, secured new territories, carried on relations with other states, and aided economic development. Despite the absence of a sense of the state, the state was essential to social order and social development in nineteenth-century America (Skowronek 1982, 19).
To be sure, this early American state was not a set of locality-penetrating bureaucracies headed by a monarch or a parliament. Rather, in Skowronek’s telling phrase it was a ‘state of courts and parties.’ Operating across state and federal levels, courts and parties were the key organizations – and judges and party politicians were the crucial ‘officials in action’ – that made up the American state in the nineteenth century (Skowronek I 982, 24, borrowing from Commons 1968 ). ‘Party procedures lent operational coherence to the disjointed institutions of the governmental apparatus, [and] court proceedings determined the meaning and the effect of the law itself’ (Skowronek 1982, 27).
Courts were not very prominent in the original debates over Constitutional design, yet as the nineteenth-century progressed they carved out a more authoritative role than the Founders had envisaged or than British courts enjoyed. ‘There is hardly a political question in the United States,’ observed Alexis de Tocqueville (1969 , 270), ‘which does not sooner or later turn into ajudicial one.’ To be sure, early Americanjudges and lawyers needed to adjust English common law precedents to US circumstances, and they had to fend off various movements to codify the laws and reduce judicial discretion. Yet these elites and the courts through which they operated also enjoyed important advantages. They could take advantage of their countrymen’s regard for the Constitution and legal procedures as common points of reference in a polity wracked with jurisdictional disputes, where fundamental issues regularly required adjudication. And there was no national civil bureaucracy that could compete with the courts by promoting ‘the national interest’ in a more substantive fashion.
Along with courts, political parties and vocationally specialized partisan politicians became the pivots of the nineteenth-century American polity. Ironically, this happened even though the Constitution made no mention of them, given that the Founders disapproved ‘the baneful effects of the spirit of party’ (George Washington, as quoted in Wallace 1968, 473). Foreign observers of the actual workings of American government noticed the increasing centrality and distinctiveness of US parties.
James Bryce (1895, 5) observed in the 188os that in ‘America the great moving forces are the parties. The government counts for less than in Europe, the parties count for more….’ ‘A description of them is therefore a necessary complement to an account of the Constitution and government’ since ‘their ingenuity, stimulated by incessant rivalry, has turned many provisions of the Constitution to unforseen uses…’ (Bryce 1895, 3).
‘The party organizations in fact form a second body of political machinery existing side by side with that of the legally constituted government . . .,’ such that ‘the whole machinery, both of national and of state governments, is worked by the political parties’ (Bryce 1893, 6). American parties, Bryce noted, ‘have been organized far more elaborately than anywhere else in the world, and have passed more completely under the control of a professional class’ (Bryce 1893, 6).
The regular American parties of the nineteenth century managed the complex, neverending processes of nominations and elections for local, state, and national offices. Party conventions became the typical means for nominating candidates, and the nineteenth century’s frequent elections required that party supporters be kept in a high state of enthusiasm and readiness through canvasses and rallies. Crucially, from the Jackson era through the end of the century, parties also controlled the staffing and functioning of public administration in the United States (Shefter 1978).
Administrative staffing through patronage was complementary to the intensified electoral activities of the new political parties. The opportunity to control the allocation of public offices inspired party cadres and allowed national and state party brokers to offer local loyalists influence over appointments allocated from their levels of government. In turn, public office-holders were highly motivated to contribute portions of their salaries and their time to foster the popularity of their party. For only if their party won the next election would their jobs be safe. Otherwise, the opposite party and all of its appointees would claim the spoils of office.
Once in place by the 1840s, the parties and their managers proved remarkably resilient, dominating US politics and knitting together the branches and levels of the ‘Tudor polity,’ throughout the nineteenth century (Keller 1977; McCormick 1986). The local roots of the parties sunk deep into particular neighborhoods; yet party efforts simultaneously spanned localities within states and, to a remarkable degree, reached across the nation as a whole. Certainly the party organizations were not top-down hierarchies; rather they were ramified networks fuelled by complex and shifting exchanges of favors for organizational loyalty. As such, however, they successfully linked local to state politicians and kept state politicians in touch with one another and with whatever national office-holders their party might have.
Not until the twentieth century – decades after electoral democratization and well after capitalist industrialization had created private corporate giants operating on a national scale – did the US federal, state, and local governments make much headway in the bureaucratization and professionalization of their administrative functions (Shefter 1978; Skowronek 1982). With the greatest changes coming first at municipal and state levels, bureaucratic-professional transformations happened piecemeal through reform movements spearheaded by the new middle classes. As the various levels of government were thus partially reorganized, the fragmentation of political sovereignty built into US federalism and into the divisions of decision-making authority among executives, legislatures, and courts, was reproduced in new ways throughout the twentieth century. American political parties have remained decentralized in their basic operations, and in many localities and states the major parties uneasily combine patronage-oriented and interest-group oriented modes of operation. Within the federal government, Congress, with its strong roots in state and local political establishments, has remained pivotal in national domestic policymaking – even during periods of strong executive initiative such as the New Deal and the World Wars and the Cold War (Grodzins 1960; Patterson 1967; Huntington 1973; Fiorina 1977; Amenta and Skocpol 1988).
The patterns of US state formation just summarized have conditioned the rhythms and patterns of social policymaking from the nineteenth century to the present. America’s nineteenth-century patronage democracy had a strong proclivity for widely distributing benefits (McCormick 1979).This system fuelled the late nineteenth-century expansion of de facto disability and old-age benefits for those who could credibly claim to have served the Union forces during the Civil War. By the early twentieth century, some 28 per cent of all American men over 65 years old, and over a third of elderly men in the North, were receiving regular pensions from the federal government- making the United States, in a sense, a precocious welfare state for a privileged generation (Skocpol et. al. 1987)!
Yet once American government finally began to bureaucratize and professionalize, the surviving structures of patronage democracy and elite perceptions of ‘corruption’ in the Civil War pension system discouraged US progressive liberals from imitating the pension and social insurance innovations of their English contemporaries (Orloff and Skocpol 1984). America’s governmental ‘responses to industrialism’ during the Progressive Era were primarily local and state regulatory laws that often established agencies to take over functions from the courts.
During the New Deal and in its aftermath, the United States finally launched a kind of modern welfare state, including public assistance and social insurance measures. Nevertheless, the Social Security Act was rooted in prior state-level laws or legislative proposals under active debate in the I930s; and congressional mediation of contradictory regional interests ensured that national standards could not be established in most programs. Subsequently, American national mobilization for World War II – a mobilization less total and centrally coordinated by the state than the British mobilization for the same war – did not overcome congressional and local resistance against initiatives that might have pushed the United States toward a nationalized full-employment welfare state. Instead, this pivotal war enhanced federal fiscal capacities and created new possibilities for congressionally mediated subsidies and tax expenditures, but did not permanently enhance public instrumentalities for labor market intervention or executive capacities for coordinating social spending with macroeconomic management (Amenta and Skocpol 1988).
Basic structural features of the US state have thus powerfully set overall institutional limits for social provision in the United States. Yet fundamental patterns of state formation are only the starting point for analysis. In addition, political struggles and their policy outcomes have been conditioned by the institutional leverage various social groups have gained, or failed to gain, within the US state and political party systems. By analyzing ways in which America’s distinctive state structure has influenced possibilities for collective action and for political alliances among social groups, we can improve upon arguments that simply highlight struggles of capitalists versus industrial workers.
American Political Institutions and Social Groups
America’s precociously democratized federal polity has made it difficult for either capitalists or industrial workers to operate as a unified political force in pursuit of class projects on a national scale. Ira Katznelson (1981, 1985) and Martin Shefter (1986) have spelled out the situation for workers in a series of important publications. Because in the United States white manhood suffrage and competing patronage parties were in place at the very start of capitalist industrialization, American workers learned to separate their political participation as citizens living in ethnically defined localities from their workplace struggles for better wages and employment conditions. No encompassing ‘working class politics’ emerged; and American trade unions developed no stable ties to a labor-based political party during the period around the turn of the century when European social democratic movements were forged.
Nationally, American workers were left without the organizational capacities to push for a social democratic program, including generous and comprehensive social policies. In localities where they did have considerable political clout, American workers tended to gain advantages on ethnic rather than class lines. Only during and after the New Deal was this situation modified, as alliances developed in many places between urban-liberal Democrats and industrial unions. Yet the Democrats and the unions never went beyond flexible and ad hoc partnerships. Particular Democratic politicians put together unique constellations of supporters, sometimes including certain unions and sometimes not, while unions retained the option of supporting friendly Republicans as well as Democrats.
Although the US polity has thus hardly served as a tool of broadly defined working class interests, American capitalists nevertheless – in the apt phrase of David Vogel (1978) – ‘distrust their state.’ In part, this is their understandable response to a long-democratized polity prone periodically to throw up moralistic reform campaigns challenging the prerogatives of large-scale property holders. In addition, it reflects the frustrations that American capitalists recurrently experience in their dealings with this or that part of the decentralized and fragmented US state structure. For not only does US federal democracy impede unified working class politics; it also gives full play to divisions within business along industrial and geographical lines.
Conflicts within the ranks of American business are readily politicized, and US corporate interests have always found it difficult to provide unified support for national initiatives that might benefit the economy as a whole on terms favorable either to most sectors of business or to economically dominant sectors. Within the ranks of business as well as beyond, the losers can always ‘go to court’ – or back to the legislatures, or to a new bureaucratic agency – for another round of battle in the interminable struggles that never seem to settle American policy questions, especially where government intervention in the economy is at issue, as it usually is. Understandably, therefore, for American capitalists the US state has seemed neither coherent nor reliable. National episodes of democratic reform leading to strengthened state interventions have been resisted with the aid of any coalitional allies at hand – even though most businesses have proven able to accommodate, reshape, and even benefit from the policy results after the fact.
Within a state structure so discouraging of unified, persistent class politics as US federal democracy has always been, it is perhaps not surprising that social policy breakthroughs have clustered in widely separated ‘big bangs’ of reform: during the Progressive Era of about 1906 to 1920; during the New Deal of the mid- 1930s; and between the mid 1960s and the mid-1970s, during and right after the ‘Great Society’ period. Each cluster of new policies was heterogeneous, a product of disparate plans previously pursued by networks of policy intellectuals (perhaps buttressed by voluntary associations or particular governmental bureaux) rather than by any political party or nationwide economic group. What is more, each cluster of policy innovations was put through by coalitions of groups in touch with sets of elected legislators, coalitions that crystallized during nationally perceived ‘crises’ widely understood to call for positive governmental solutions. But in each episode the coalitions favoring new social policies were temporary, fragile, incapable of any permanent institutionalization – and very soon undone by conservative backlashes that drew on localist plus business and other resistance to enhanced state power in the United States.
In the histories of many modern welfare states, agricultural classes have proven to be important political allies of urban-industrial groups that would either promote or resist enhanced state interventions. In the United States agricultural interests have been even more pivotal than elsewhere, and not only because there have always been large numbers of family farmers in American society. The American federal state with its decentralized and nonprogrammatic political parties has provided enhanced leverage to interests that could associate across many local political districts. Such widespread ‘federated’ interests – including organizations of farmers from the Grange to the American Farm Bureau Federation, along with locally rooted businessmen linked to the Chamber of Commerce, and certain professional associations including the National Education Association and the American Medical Association – have been ideal coalition partners for any more nationally focused forces that might want to promote, or obstruct, or rework social policies, especially as proposals have had to make their way through the House of Representatives in Congress.
Occasionally, such widespread federations have spurred the passage of national social policies; the impact of the Townsend Movement on the old-age provisions of the Social Security Act would be a case in point.
More often, widespread federations – especially those involving commercial farmers and small businessmen prominent in many communities – have obstructed or gutted proposed national social policies. Thus, for example, during the early New Deal of I933 to I935, federal agricultural policies had the not fully intended effect of strengthening interest-group association among commercial farmers across the disparate crop-areas of the South and Midwest (Finegold and Skocpol 1984). In turn, this meant that the American Farm Bureau Federation was better able to ally with business organizations, including the Chamber of Commerce, to pressure congressional representatives against one liberal New Deal social welfare proposal after another, from 1936 onwards.
Pinpointing institutional leverage through Congress also helps us to make sense of the special role of ‘the South’ in modern American social policymaking – a role that certainly rivals that of either capitalists or the industrial working class. To be sure, the South’s role cannot be understood without underlining the class structure of Southern cotton agriculture as a landlord-dominated sharecropper system from the late nineteenth century through the 1930s (Alston and Ferrie 1985). Nor could we possibly ignore the explicit racism that ensured minority white dominance over black majorities in all sectors of economic and social life. Yet the South was militarily defeated in the Civil War, and by the 1930s this region was not very weighty in the national economy as a whole; nor were its social mores typical of the nation. Thus socioeconomic factors and generalized references to racism will not alone tell us why Southern politicians had so much leverage during and after the New Deal that they could take a leading role in congressional alliances opposed to national welfare standards and any strong federal presence in economic planning.
The influence of Southern agricultural interests in the New Deal depended on the insertion of their class power as landlords and their social power as white racial oligarchs into federal political arrangements that from the 189os to the 1960s allowed an undemocratized single-party South to coexist with competitve two-party democracy in the rest of the national polity (Key 1949). Above all, Southern leverage was registered through a congressionally centered legislative process in Washington that allowed key committee chairmen from electorally ‘safe’ districts to arbitrate precise legislative details and outcomes. From the New Deal onward the ‘national’ Democratic Party used congressional committees to broker the internal divisions between its Southern and urban-liberal Northern wings (Bensel 1984, ch. 7). This prevented the often contradictory orientations of the two wings from tearing the national party apart, but at the price of allowing the enactment of only those social policies that did not bring the national state into direct confrontation with the South’s nondemocratic politics and racially embedded systems of repressive labor control.
In short, taken together, the US state structure as it had been formed by the 1930s and 1940s, along with the operations of the New Deal party system, magnified the capacities of Southern economic and social elites to affect national policies – at the same time that the capacities of other interests, including those sections of organized industrial labor allied with urban Democrats in the North were simultaneously enhanced by the same US state structure and party patterns. Many features of the New Deal Social Security system – and indeed of the entire disjoint configuration of social and economic policies with which the United States emerged from the political watersheds of the New Deal and World War II – can be understood by pinpointing the social interests and the political alliances that were able to gain or retain enhanced leverage through the longstanding federal and congressional institutions of the US state. The New Deal certainly brought social policy innovators to the fore through the newly active federal executive. It also energized urban-liberal forces and created new possibilities for political alliances through the electorally strengthened and partially realigned Democratic Party. Nevertheless, in the end, America’s federal state and regionally uneven democracy placed severe limits on the political alliances and policies that could prevail as the original foundations were laid for nationwide public social provision in the United States.
Finally, to understand major developments in social policies since the New Deal, it is crucial to remember that the United States was – paradoxically – both the ‘first’ and the ‘last’ to democratize its electorate among the longstanding capitalist democracies. It was the first for white males, who were irreversibly enfranchised by the I 830s; and it became the last for all citizens, because except briefly during Reconstruction and its immediate aftermath most blacks in the United States could not vote until after the migrations from the South after the 1930s and the Civil Rights upheavals of the 196os. For all of the twentieth century until the 196os, the United States was a regionally bifurcated federal polity: a mass two-party democracy in the East, North, and West, coexisting within the same national state with a single-party racial oligarchy in the South. Only since the 196os, through major transformations that are far from completed, have American blacks been mobilized into US national democracy, and has two-party electoral competition made headway in the deep South.
The Civil Rights revolution of the 196os began the process of mobilizing blacks into the Southern electorate, and on new terms into the national electorate and the Democratic Party. These are processes whose effects have been tumultuous – both on agendas of debate over social policy, and on political alliances concerned with policy alternatives from the Great Society to the present. Yet the incorporation of blacks into the national polity has not been happening in a social policy vacuum; it is taking place in the context of the configuration of social policies inherited from the New Deal. Within this configuration of policies, ‘social security’ for the stably employed majority of citizens had become by the 196os institutionally and symbolically bifurcated from ‘welfare’ for the barely deserving poor (Skocpol 1988). And for socioeconomic and political reasons alike, working-age blacks were disproportionately clients of the vulnerable ‘welfare’ components of US social provision.
During the social policy reforms of the 1960s and early 1970s, welfare clients temporarily benefited from the widespread recognition that the New Deal system of social policies had not adequately addressed issues of poverty or responded to the needs of blacks, who could now vote in greater numbers. Liberal Democrats tried to use welfare extensions and new ‘anti-poverty’ programs to incorporate blacks into their – otherwise undisturbed – national political coalition. But many social policy reforms of the 196os and 1970s soon backfired to disturb rather than reinforce Democratic coalitions. National politics underwent a sea change, and since the 1970s conservative forces hostile to enhanced public social provision have found renewed sources of strength within and beyond the Deocratic Party. This has left impoverished people, including many blacks, increasingly isolated in national politics. And it has left ‘welfare’ programs more vulnerable than ever to attacks by those who question the US federal government’s role in providing for support for vulnerable citizens.
A Continuing Ambivalence about the Role of the State
Down to the present, Americans remain ambivalent about concentrated political authority, as they were two centuries ago when the Constitution was framed. They are quick to see the ills that government can inflict, and slow to perceive the good things that a responsible national state can do for all citizens. But it would be wrong to suggest that Americans have ever been a people ‘without a state’ either in fact or in fancy. As long as we are prepared to specify the peculiar sets of institutions that have added up to America’s distinctive versions of the modern state in different historical periods, we can identify the way in which political officials have shaped policies and the ways in which state structures have patterned the conflicts and alliances of major social groups.
The nineteenth-century American ‘state of courts and parties’ presided over the widespread distribution of economic and social benefits, including the Civil War pensions that served as de facto disability and old-age pensions for many around the turn of the century. In the twentieth century, America’s unevenly bureaucratized and democratized federal state discouraged class politics and placed severe limits on comprehensive provision for the poor and unemployed. Yet this twentieth-century American state has, nevertheless, allowed the elaboration of many public benefits for the broad working and middle strata – benefits that these groups have eagerly accepted and politically supported, even when it means paying visible federal taxes as in the case of social security’s contributory retirement insurance.
Sadly, continuing American ambivalence about the role of the state in social provision focuses most acutely on public programs for the poor and for blacks. This should not be surprising, given what we have learned about historical processes of American state formation and the effects of political institutions on the political leverage of different social groups. Comparative-historical studies of modern welfare states teach us that vulnerable groups do best when bureaucrats and national political parties have worked together to build universal systems of public social provision, stretching from the upper middle classes to the poor. As we have seen, US state structures have rarely allowed such coalitions to shape social policies.
Instead, from the nineteenth century to the present, American political arrangements have recurrently facilitated political efforts to provide generous social policies for those in the American majority who can help themselves – and strong doses of ‘rugged individualism’ for the minorities who cannot! I doubt that this was quite what the Founding Fathers had in mind when they wrote a Constitution meant to guard against the tyranny of majorities, as well as the abuses of sovereigns. But, then, the Founders were men imbued with a sense of history, and they must have understood that American politics would not always work out as they hoped and planned. In the final analysis, moreover, the Founding Fathers were so obsessed with frustrating the tyranny of concentrated sovereignty that they – comfortable men of the propertied classes that they were – would probably have been willing to sacrifice the gains that the less privileged could make from a European-style state. To the present day, most Americans probably still agree. And that is why grievous problems of racism and poverty will likely remain festering dilemmas of American social provision for many years to come.