State Governmental Capacity and Welfare Reform - CiteSeerX [PDF]

State Governmental Capacity and Welfare Reform

Lawrence M. Mead Department of Politics New York University 715 Broadway, #446 New York, NY 10003-6806 Phone: 212-998-8540 Fax: 212-995-4184 E-mail: [email protected]

Prepared for delivery at the 2002 Annual Meeting of the American Political Science Association, August 29-September 1, 2002. Copyright by the American Political Science Association.

2

State Governmental Capacity and Welfare Reform

I investigate the link between the general capacity of state governments and their ability to reform welfare. I define what governmental quality means and measure it is using Elazar’s political cultures. I define what successful welfare reform means, drawing on past research and experience.

My criteria stress

process, the avoidance of political and administrative problems. I then test the link between the Elazar cultures and the performance criteria using recent case studies of state implementation of Temporary Assistance for Needy Families. As expected, Elazar’s “moralistic” states perform best. Results hinge, however, on how good government and successful reform are defined. [99 words]

3 Introduction Welfare reform has been a leading issue in American politics for four decades. “Welfare” here means the nation’s controversial aid program for needy single-parent families. First enacted in 1935 and once called Aid to Families with Dependent Children (AFDC), it is run by states under loose federal rules using both federal and state monies. Controversy about welfare arose because of rising welfare rolls. AFDC grew rapidly in the 1960s and early 1970s to support over 11 million people. After changing little for 15 years, the rolls again jumped in the early 1990s to exceed 14 million people. Liberal proposals to raise benefits and coverage were made in the 1960s and 1970s, but most were defeated. Instead, in the 1980s and 1990s Congress adopted conservative proposals to limit eligibility and require adult recipients to work. The most radical reform occurred in 1996, when AFDC was recast and renamed Temporary Assistance for Needy Families (TANF) by the Personal Responsibility and Work Opportunity Reconciliation Act. PRWORA changed the program from a federal entitlement to a block grant to states, limited families to five years on the rolls, and stiffened work requirements, among other changes. States had always controlled benefit levels. TANF gave them, as well, substantial control over eligibility and other details. Since 1994, the welfare rolls have fallen sharply to 5.4 million people, due partly to PRWORA, but also to good economic conditions and rising subsidies for low wages and child care. Past Research This paper addresses how the general capacity of state governments affects their ability to reform welfare. Little past research bears on this issue. The political science literature on welfare is mostly about the politics of reform at the national level (Moynihan 1973; Weaver 2000). The policy literature on welfare is mostly about the

4 social and economic condition of the recipients, including the effects of reforms (Cottingham and Ellwood 1989). Much less is known about how local institutions run or change the program. State politics scholars investigate how various features of states influence their policies. But in the welfare area, the main focus has been on what makes benefits generous. Typically, states that are rich, urban, liberal, and with competitive parties offer higher welfare benefits than those with the opposite characteristics (Cnudde and McCone 1969; Dawson and Robinson 1963; Hanson 1983; Plotnick and Winters 1985). But showing which states are generous does not establish which can perform the more complex task of reforming welfare. In recent years, that has meant chiefly, not changing benefits, but adding work requirements to them. States must support needy families while also inducting the adults into work programs and providing child care and other support services for them. There are studies of the implementation of welfare work programs (Chadwin, Mitchell, and Nightingale 1981; Mead 1985, 1997), but they cover at best a few states and say little about politics. Studies of state reform that do address politics tend to describe state actions without assessing them (Hagen and Lurie 1994; Norris and Thompson 1995). Conversely, evaluations of welfare work programs assess their impacts on clients but usually say little about either politics or administration (Gueron and Pauly 1991). In this study, I link the general capacity of state government to the ability to reform welfare. I first define what state governmental capacity means and measure it using Elazar’s theory of political cultures. I also define what successful welfare reform means, drawing on past research and experience. My definition is oriented to process, to avoiding political and administrative problems, rather than policy effects. I then test the

5 link between the Elazar cultures and these criteria by using recent case studies of statelevel reform under TANF. I hypothesize that the states that Elazar called “moralistic” will perform best, and the results confirm that. The Meaning of State Capacity What state capacities should we expect to influence the ability to reform welfare? Strong government could mean having institutions that in some general sense look capable. One study from 1971 rated state legislatures in terms of whether they were well-staffed, clearly accountable, well-informed, independent of the executive or lobby groups, and representative. (There is no later study). States often thought to be wellgoverned—such as California, Minnesota, and Wisconsin—ranked high. But no tie was made to concrete policy outcomes (Citizen’s Conference on State Legislatures 1971). A later study linked states with strong legislatures to more generous welfare policies (Grumm 1970, 451-5). But again, making welfare generous is not the same as reforming it. Administrative quality is tougher to define and measure. Studies suggest tentatively that quality runs higher in richer and more urban states (Barrileaux, Feiock, and Crew 1992; Sigelman 1976). Two recent studies rate the states in terms of public management capabilities (Barrett and Greene 1999, 2001). Another study assesses states by the broader criteria of “accountability and information management,” “executive centralization,” “staffing and spending,” and “representation,” with each factor assembled out of multiple measures (Bowman and Kearney 1988). These assessments, however, say little about how well a government uses its assets to run actual programs such as welfare.

6 Other research speaks of the capacity of states to innovate, a talent certainly tested by welfare reform (Gray 1973, Savage, 1978; Walker 1969). But innovation is only one trait of capable government. Political and administrative mastery would appear more central. What makes government capable is not that it necessarily does new things but that it does well what government is supposed to do, which is to govern. This means first to make policy and second to implement it (I assume that policy is in part made during the implementation process). Good policy is not necessarily novel; rather, it is rational or prudent in light of a state’s circumstances; it is also politic, that is broadly acceptable. To implement policy well means to carry out the agreed such measures effectively and efficiently. The best way to characterize capable government in this sense in not in terms of specific institutions but rather by using the broader concept of culture. I use that term here to mean the basic attitudes that shape how public institutions actually operate, as against their forms. Culture teaches leaders and citizens the way things are in politics, but also how they ought to be. What is seen as a proper or improper purpose for government, and how should one conduct public affairs? Political culture clearly differs among the states (Eckstein 1988; Patterson 1968). The best-known theory of state political culture is the scheme developed by Daniel Elazar. He divided the states into three types. In “moralistic” states, positions were justified by appeals to the “public interest,” rather than narrower interests, and public administration was strong. In “individualistic” states, in contrast, government tended to serve more specific groups. Parties were strong, each standing for coalitions of interests. Bureaucracy was well-developed but weaker than in the moralistic culture. Finally, in the “traditionalistic” culture, chiefly in the South, government was limited to preserving traditional values (originally the racial caste system). Parties counted for less than in the

7 other cultures and the bureaucracy was underdeveloped and distrusted (Elazar 1984, chap. 5). Elazar based his scheme on his own observation of the politics and the religious and ethnic background of the states, but later scholars using more systematic methods have produced parallel results (Johnson 1976; Lieske 1993; Morgan and Watson 1991). Elazar found all three cultures to be present throughout the United States, yet each state was dominated by one of them. Table 1 lists the states Elazar placed under each heading, plus sub-groupings that he identified within each culture. The scheme is shown as of 1984, the last Elazar published. This is just before serious welfare reform begins in the late 1980s. [Table 1 here.] The Elazar cultures are distinct from other political differences among the states, for example in political ideology or partisanship (Erikson, Wright, and McIver 1993, 150-76). At the same time, the cultural types are linked to many other features of state politics and government. Moralistic states tend to show high political participation, competitive parties, strong merit systems, and liberal and innovative programming. Traditionalistic states show less of all these things. Individualistic states tend to fall in the middle in these respects while showing strong parties and more centralized administration than the moralistic group (Fitzpatrick and Hero 1988; Johnson 1976, 499507; Morgan and Watson 1991, 36-47; Sharkansky 1969). The Elazar scheme expresses well what most people mean by capable government, or lack thereof. Indeed, states commonly referred to as “good government” fall within the moralistic category. Elazar believed that each culture contributed something valuable to American politics. Traditionalism celebrated timeless values, individualism the values of compromise and freedom, moralism the quest for the good society (Elazar 1984, 141-

8 2). But the moralistic style is clearly the most high-minded and the most oriented to achieving specific objectives. Moralistic politics stresses problem-solving, not partisan rivalry, while bureaucracy is well-funded. In the individualistic style, political agreement is more limited, more based on compromise, although bureaucracy is well-developed. In the traditionalistic style, both social policymaking and administration are less developed. One criticism of Elazar is that his categories appear to capture the beliefs of political elites and activists better than those of the voters (Kincaid 1980, 6-11, 13-14; Welch and Peters 1980). But that is appropriate for the case of welfare, where the views of leaders seem to matter a good deal more than those of voters (Uslaner and Weber 1975; Tweedie 1994). Another criticism is that moralism and traditionalism may capture little more than the differences between Northern and Southern states. Some also suggest that the moralistic states differ from others chiefly in showing less ethnic and racial pluralism (Hero and Tolbert 1996). While this is true, it is the specifically political attitudes defined by Elazar that most directly influence politics and policy. The Welfare Test The idea that political culture affects governmental performance is not new. Robert Putnam showed that regional governments in Italy were markedly more effective and responsive in the Northern, more civic regions of the country than they were in the less civic South (Putnam, Leonardi, and Nanetti 1993). An analysis paralleling Putnam’s showed that civic culture was similarly related to government performance among the American states. Here civic culture meant qualities of civic engagement, beliefs in political quality, trust in others, and cooperative social structures, and was clearly related to moralism in Elazar’s sense. Government performance included qualities of policy liberalism, innovation, and administrative effectiveness (Rice and Sumberg 1997).

9 This analysis, however, is highly aggregated. What it implies concretely is unclear. How the three cultures govern may be better shown by how they handle a specific issue, such as welfare. At the outset, the hypothesis that the moralistic style might promote successful welfare reform is plausible. Reform in the currently meaning chiefly involves grafting complex work programs onto welfare. The problem-solving bent of the moralistic states, and their strong administration, should favor them over the other cultures in that endeavor. In comparison, the individualistic culture should display a more interest-centered and conflictive political style and a less enterprising bureaucracy, in welfare as in other matters. Among the traditionalistic states, we should find less interest in—and capacity for—reforming welfare. Such patterns appeared during welfare reform prior to PRWORA. Successful mandatory work programs in welfare first developed in the 1980s, when the federal government allowed some states to experiment with more demanding work requirements than normally allowed. Experimental evaluations of some of these programs, principally by the Manpower Demonstration Research Corporation (MDRC), showed that they had positive effects on the employment and earnings of their clients. These findings justified the expansion of mandatory work programs under the Family Support Act of 1988 (Wiseman 1991). The positive evaluation of later work programs, in turn, promoted their further expansion under PRWORA. That reform, however, was also driven by Republicans’ desires to curb welfare, promote marriage, and devolve control to the states (Weaver 2000). Moralistic states led in the development of these programs. The most noted early efforts appeared in San Diego and Riverside Counties, in California. Other exceptional local programs appeared in Colorado, Iowa, Oregon, and Utah (Pavetti, Holcomb, and Duke 1995; Scrivener et al. 1998). The leaders in efforts to rebuild welfare around work

10 on a statewide basis were Michigan and Wisconsin. The Wisconsin reform, which combines severe work tests for welfare adults with generous child care and other services is conspicuously radical (Mead 2000). All these states fall within Elazar’s moralistic group. Conversely, states that encountered problems during the early phases of reform tended to come from the other cultures. In Chicago, an attempt to enforce job search on AFDC adults failed to achieve any clear impact on their employment or earnings, probably because of weak administration (Friedlander et al. 1987). In Florida, a work program of the early 1990s recorded only small impacts, probably due to inadequate child care funding (Kemple, Friedlander, and Fellerath 1995). In Massachusetts, the much-touted Employment and Training Choices (E.T.) program failed to have enduring influence because its evaluation was not experimental and the state failed to rebuild the bureaucracy around work (Hill and Main 1998; Nightingale et al. 1991). All these states are individualistic or traditionalistic. My question is whether these patterns held during the implementation of TANF in the mid 1990s. This era of reform is better test of the role of government capacity than earlier phases. The demands on state governments were larger. For them, to experiment with welfare was no longer a choice. Washington demanded tougher work tests and other changes. It also offered states new latitude to set other welfare policies for themselves. States thus had to make and implement more welfare policy than ever before. They did so, furthermore, in remarkably similar settings. The earlier developments occurred in different states under varying conditions over more than decade. All the states, however, implemented TANF in the mid to late 1990s. Even Wisconsin, which started to reform earlier than most states, carried out its most radical changes in this

11 period. Economic conditions were also unusually favorable almost everywhere. And, although TANF allowed states some latitude in work policy, virtually all states chose the goal of moving adult recipients rapidly into employment. Because of this unusual parallelism in timing, conditions, and goals, more of the differences in outcomes among states should reflect differences in governmental capacity than would normally be the case. The Meaning of Successful Implementation What concretely did successful welfare reform mean? In this analysis I define it in process terms, as the avoidance of harmful political and administrative problems. (Meanings more oriented to policy content are discussed further below). My criteria draw upon research prior to TANF (especially Norris and Thompson 1995) and my own experience researching and consulting about welfare reform in several states (MD, MA, MO, NJ, NY, PA, RI, SC, WV, and WI). The result was these six standards. Political performance: Policymaking. Did the state government frame its own policy and legislate sensibly to achieve it, in light of experience? Or was it pushed into reform by Washington, or did it make mistakes that could have been foreseen? Policymaking means taking a real decision of one’s own about AFDC, not simply passing enabling legislation required by TANF. Consensus. This means not just enacting reform but doing so with enough agreement so that the key changes will not likely be reversed at the next election. That implies avoiding divisive opposition either in the legislature or among outside lobby groups.

12 Resources. Was the state willing and able to fund the programs implied by its goals, such as additional bureaucracy, child care, and other support services? The focus here is on basic adequacy rather than the detailed capacities of the bureaucracy. Administrative performance: Commitment. Were senior administrators implementing reform committed to it? Or were they pushed into reform by politicians? Other commitment issues include morale among lower-level staff, the degree of cooperation between the state and local government, and union cooperation with reorganization in the bureaucracy. Coordination. Were the units or agencies charged with implementing welfare reform able to work together smoothly? Or were there serious divisions, conflicts, or communication problems among them? Such problems could occur either within the welfare department or between it and other agencies? Capability. Did the welfare bureaucracy have the expertise and facilities needed to implement reform? Such as the capacity for case management and adequate reporting systems? The focus here is on the skills and equipment of staff, which tend to be longterm problems, as against funding specifically for reform or bureaucratic divisions. The three political and the three administrative criteria run roughly parallel. Policymaking and commitment express the political direction of reform, the goals that policymakers or administrators choose. Consensus and coordination express the degree of unity that officials succeed in building around their decisions, either in the political arena or during the administrative process. Resources and capability express at two levels whether a program “has the horses” to get the job done. The Sample We now need evidence to test the connection between the Elazar cultures and these performance criteria. As mentioned, much less research exists about the local

13 conduct of welfare reform than about its national politics and economic effects. Fortunately, two national projects have tracked the implementation of TANF in some depth in about two dozen states. The Assessing the New Federalism study at the Urban Institute, in Washington, DC, covers thirteen states. The State Capacity Study at the Rockefeller Institute of Government, at the State University of New York, covers twentyone states. The two together cover 23 states. I have added a twenty-fourth—Oregon— using other sources. As Table 1 shows, the sample includes several states from each of Elazar’s three cultures, and at least one from each of his subcultures. This evidence has obvious limitation for rating state performance. The studies are not all by the same authors. The evidence is second-hand. The authors did not explicitly assess the states by my criteria, or indeed any criteria. Rather, they give accounts of the reform process in the states, and I infer performance from these. (The state scores I assign using the studies are entirely my own and should not be attributed to these projects or their sponsors). .Substate details are based on only some localities, which may not be representative. The appendix lists the specific documents I reviewed on each state. The number of sources on each state varies from one to six. Due to the uncertainties I attempt only to judge whether a state met a criterion or not. I make no judgment about degrees. I assume that success is more likely to go unmentioned than problems. So I assume that a state did meet a criterion unless the documents indicate otherwise. On the other hand, the fact that I have multiple documents on most states gives reason for confidence. So does the fact that the two national projects overlapped substantially. Eleven states were covered by both, including most of the larger states and all of the moralistic states, other than Oregon. The accounts from different documents and studies generally agreed closely. In no case did one source claim institutional

14 success while another spoke of failure. At most, problems mentioned in one document go unmentioned in another. In these cases, I judged a state to have had the indicated problem. While a specific rating might be questioned, this would be unlikely to alter the overall results, which are clearcut. Results My hypothesis was that the moralistic states would prove best at satisfying the performance criteria, followed by the individualistic states, with the traditionalistic states the least capable. Figure 1 largely confirms this. It shows where states from the three cultures fell in overall performance, with one point given for each of the six criteria satisfied. Generally, the moralistic states score high, the traditionalistic states low, and the individualistic states in the middle. However, there are exceptions and considerable variation within the groups, for reasons explained below. [Figure 1 here.] Figure 2 shows that the political and administration dimensions of performance, as I have defined them, were strongly linked to each other. States that were high or low on one tend to be high or low on the other. But again, the association is far from straightline. [Figure 2 here.] Table 2 gives more detail about how the sample states scored on the six performance criteria. States generally performed better politically than they did administratively. [Table 2 here.]

15 Political performance Policymaking. Most states satisfied this criterion. Most took a significant decision about reform, and this decision was sensible in light of state goals and experience. Welfare work programs prior to TANF had stressed training recipients for well-paying jobs. Most states shifted toward “work first,” or placing recipients in available jobs, even if lowpaid, as TANF also required. The distinction of the moralistic states was mostly that they had most often run experiments in welfare reform and thus had the most systematic evidence to go on. Ohio and Colorado delegated decisions heavily to their counties, but within a statewide framework. A few states did not seriously make reform policy. New York was so deeply divided that it took no serious decisions about AFDC, but rather delegated how tough to be about work to the counties. Alabama and Missouri were pushed into reform by federal action and appeared to have little welfare policy of their own. In several other Southern states (FL, GA, NC), policymaking appeared to be casual and personalized, with the governor or legislators offering reform plans with, apparently, little inquiry or evidence behind them (a regional style noted by Key 1949). Texas policymaking was incoherent; the state claimed to pursue work first but based its policy on an experimental program focused far more on education and training.. Consensus. On this standard, states had more trouble. States in the leading two groups were able form a majority behind reform that stretched behind the legislature. They came to terms with advocates who questioned reform. In the moralistic states, consensus was feasible because critics accepted that work requirements were inevitable; they questioned the details rather than the principles of reform. This reflected the earnestness of the moralistic style, where leaders feel they should focus on solving a recognized problem

16 rather than defend narrower interests. In New Jersey and Rhode Island, divisions ran deeper, but consensus was achieved using elaborate consultative processes. The other states had more difficulty. In Arizona, a proposal to privatize part of reform sparked acute controversy. In Alabama and Ohio, efforts to consult with community groups became too contentious to support change. With its legislature unable to act, Alabama implemented reform chiefly through administrative means. Ohio delegated the key issues to the counties. In New York, which did the same, and in California, legislative debate on reform was intensely partisan. In New York, and also in Georgia, Massachusetts, and Mississippi, advocates or civil rights groups rejected reform decisions as racist or illegitimate—the sort of hard-core opposition not seen in the moralistic states. Resources. Most states devoted sufficient funds to afford the programs and services that their decisions required. A key issue here was whether they would or could fund sufficient child care to prevent long waiting lists for mothers seeking subsidized care, either to leave welfare or among the larger low-income population. Some states (CO, FL, TX) failed do this due to fixed limits on state spending or revenues that were entrenched in state law or constitutions. These curbs evinced anti-government attitudes that overshadowed reform. In other cases (AL, MS, NC, WV), the documents suggest, states were simply unwilling or unable to “pay the piper” for political reasons. Administrative performance Commitment. In the top two groups of states, the welfare bureaucracy was fully engaged in reform. Administrators accepted the goals as their own, at least at a senior level, and significant resistance lower down was absent. Indeed, in some cases, as in Utah and Wisconsin, local officials had worked with state leaders to create reform.

17 In Florida and Georgia, however, officialdom was dragged into reform from above and showed little commitment to it. In Arizona and California, the agency or major localities had been heavily committed to a skills-oriented approach to welfare and resisted the shift toward work first. In Texas, welfare reform was a lower priority to administrators than rebuilding non-welfare employment programs and other initiatives. In Colorado and New Jersey, local agencies had a history of defiance toward the state government, and this prevented them from fully endorsing reforms decided in the capital. Coordination. Bureaucratic divisions emerged as the most serious institutional problem faced by reform, at least in this sample. One important division, within the welfare agency, was between the income maintenance clerks that determined eligibility for benefits and the case workers who supervised clients in their search for work, either directly or through referrals to contractors. The other important divide was between welfare as a whole and non-welfare employment agencies, to which welfare referred clients. The latter included the Employment Service (ES), a federally-funded job placement agency, and training programs under the federal Job Training Partnership Act (JTPA). Wisconsin’s solution to harmonizing the elements was the most radical. Inside welfare, the eligibility and case worker roles were merged in a single staff person. Externally, in most localities welfare kept control of the welfare work mission, using the outside employment agencies, if at all, only as contractors. New York and Tennessee also kept welfare in control of local implementation. In Oregon, Washington, and Minnesota, all the agencies worked collaboratively to implement reform, apparently with few differences. In Utah, welfare and employment programs had already been merged in a single agency.

18 All the other states encountered significant problems. In some instances (AZ, CA, GA, KS, WV), there were troublesome divisions between the income maintenance and case worker or employment side of welfare itself. In California, for example, welfare work programs had been run almost apart from the eligibility operation, and this obstructed the implementation of work first policies under TANF. More often, however, the problem was the interface between welfare as a whole and the non-welfare work agencies. After national welfare work programs were first enacted in 1967, the ES had run them in collaboration with welfare, in practice dominating them in most localities. But because the ES’s routines stressed serving job seekers who came to it voluntarily, it generally performed poorly with welfare clients. These jobseekers came to it on a mandatory basis, as a condition of receiving aid. To succeed with them, the agency had to enforce work but also support employment with special services. The ES often found both these roles uncongenial.. Welfare could better run work programs without the ES, many concluded. In the 1980s, Washington mandated that policy, and the ES was demoted to the role of contractor to welfare. With TANF, however, Congress lifted those restraints. Meanwhile, the Workforce Investment Act (WIA) of 1998 required that states merge the administration the ES, JTPA, and other non-welfare work programs. Local boards were supposed to create unified offices, or “one stop shops,” that would serve the clienteles of all the programs. Many states now thought that the welfare work mission should be entrusted to this new structure, in part to avoid the cost of a separate welfare work operation. The change ignored the earlier experience. In the majority of states in this sample (AL, CA, CO, FL, GA, KS, MA, MI, MS, MO, NJ, NC, OH, RI, TX), the attempt to place welfare work under the WIA structure created serious confusion. The problems included lack of clear procedures to refer

19 clients to WIA, to serve them there, or to report results back to welfare. The attempt to use WIA was the one blemish on the performance of Kansas and Michigan, two of the moralistic states. In Florida and Texas, the state’s determination to turn welfare work over to WIA seems to have paralyzed welfare reform, at least in the short run. Capability. In the top group of states, administrators appeared to have the talents and facilities they needed to cope with reform. Welfare staffs took on the employment role, often unifying it with eligibility determination, and they had adequate information systems to run the new work programs. But in the bottom groups, the bureaucracy was often overwhelmed by the demands of reform. The states that had difficulty unifying case management included Massachusetts, Rhode Island, Tennessee, Washington, and West Virginia. Those where management information systems were inadequate included Florida, Georgia, and Tennessee. One cause of these struggles in the South is a longstanding inability to draw many capable people into public service. This reflects the low prestige of public service in the traditionalistic culture. In Mississippi, the problems led to a failed attempt to turn over welfare to the private sector. Discussion These results support the idea that capable government did promote successful welfare reform—at least as I have defined those terms—during the implementation of TANF. And yet capability was not sufficient. California and Colorado are two moralistic states that scored below average. Despite manifest administrative talent, they were pulled down by their internal divisions, at both the policymaking and administrative levels. Conversely, Tennessee, although a traditionalistic state, scored high.. Culture seems to matter less for the performance of American states than it did in Putnam’s Italy.

20 That is because the variations in civic capacity around the United States are smaller than among the Italian regions (Rice and Sumberg 1997, 109-10). All states have some potential to be governed well. In realizing that potential, much depends on the decisions of leaders. Some might say that these analyses are definitional. Elazar framed his cultures largely in terms of the political practices of states. Since welfare reform is one of those practices, it is hardly surprising that culture “predicts” it. However, Elazar based his scheme on state behavior in general and over long periods of time, whereas welfare reform is a recent and specific issue. If culture predicts reform success in welfare reform, that at least suggests that political traits have enduring and pervasive influence. One might also object that the assessment is too focused on welfare and too shortterm. If one examined some other policy area, or judged over a longer period, the verdict might be different. If the criterion were success in general employment policy, for example, states that chose to combine welfare reform with WIA would get credit for that, rather than being penalized. And over the longer term, some of the problems that afflicted welfare reform in the 1990s might get solved. The lower-performing states might close the gap on the leaders. This assessment is oriented to process, to how the states went about reform. That does not establish that reform had good effects on the recipients or the society. Did reform raise work levels among the poor or reduce poverty? Many factors besides welfare reform affect this. No state has evaluated TANF. However, the reform appears to have promoted higher work levels among single mothers while reducing welfare receipt and poverty (Blank and Haskins 2001). Whether the moralistic states excelled in these terms is unclear.

21 The Urban Institute did survey social conditions in its thirteen study states. In 1999, the five states with the highest work levels among low-income single parents were, in order, Wisconsin, Minnesota, Michigan, Florida, and Colorado. The five states with the lowest child poverty rates in 1998 were Wisconsin, Minnesota, Massachusetts, Michigan, and Colorado (Urban Institute 2000). All these states except Florida and Massachusetts are moralistic. That does not prove that good government produced the good results—something else might have—but it is at least consistent with that theory. Most important, my results depend critically on how government capacity and welfare reform are defined. Perhaps the moralistic states excel only because today’s principal meaning of “welfare reform” demands the creation and management of complex welfare work programs. That is a task where the moralistic states excel. On other matters, the meaning of successful implementation is different, as is the type of government likely to promote it. States can, for example, vary benefit levels. Liberal reformers favor high benefits, conservatives low ones. In research not reported here, I found the first policy to be associated with moralistic but also individualistic states, the second with traditionalistic states. Similarly, states can choose how tough to be in sanctioning clients (that is, reducing their benefits) for noncooperation with work requirements; individualistic states typically are lenient, traditionalistic states severe. Typically, the individualistic style is to subsidize single-parent families without asking much of them in return. The traditionalistic style is to deny aid for fear of promoting dependency. In contrast to either, the moralistic states try to be both generous and demanding. They typically pay high benefits, but they also demand that recipients work, and also that absent fathers pay child support. This effort to enforce good behavior through government reflects the origins of the moralistic style in the pro-government but intolerant politics of Puritan New England (Elazar 1984, 127; Fischer 1989, 13-205).

22 If one focuses on a range of welfare policies, then, the meanings of successful implementation and good government are less clear. Good government in the moralistic sense, however, still has a certain primacy. For the combination of generosity with demands seems to be the welfare policy that most ordinary Americans prefer. In surveys, large majorities of respondents say that they want welfare to help needy families while assuring that the adult recipients work (Gilens 1999, chaps. 2, 8). Perhaps for this reason, most states are moving in this direction. Under TANF, most have stiffened demands and incentives promoting work, but few have changed benefit levels other than to strengthen work incentives (Gais and Weaver 2002). This combination of aid with demands is what requires the development of elaborate welfare work programs. This is the meaning of welfare reform where the moralistic states have clear advantages. To pursue reform in this sense will probably require the other states to become more moralistic over time.

23 Appendix: Sources for Welfare Case Studies [To save space, this appendix might be made available to readers on request.] After each state, I indicate whether it was included in the Urban Institute or Rockefeller Institute of Government samples and then the specific sources used to estimate its performance. Sources other than UI or RIG are used for Oregon. Both UI and other sources are used for Massachusetts. Many of the RIG reports appeared in two edited volumes: Sarah F. Liebschutz, ed., Managing Welfare Reform in Five States: The Challenge of Devolution (Albany: Rockefeller Institute Press, 2000). Carol S. Weissert, ed., Learning from Leaders: Welfare Reform Politics and Policy in Five Midwestern States (Albany, NY: Rockefeller Institute Press, 2000). Alabama Urban Institute state. Sandra J. Clark, Sharon K. Long, Krista Olson, and Caroline Ratcliffe, Income Support and Social Services for Low-Income People in Alabama (Washington, DC: Urban Institute, January 1998). Pamela A. Holcomb, Kathryn Schlichter, Stefanie R. Schmidt, Gina Adams, and Jacob Leos-Urbel, Recent Changes in Alabama Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, October 2001). Arizona Rockefeller Institute state. John Stuart Hall and Gerald J. Kubiak, “Arizona’s Welfare Reform Experience” (Albany: State University of New York, Rockefeller Institute of Government, November 2, 1999). John Hall, “Second Report Form: Arizona” (Albany: Rockefeller Institute of Government, 2000). California Urban Institute state, Rockefeller Institute state. Rob Geen, Wendy Zimmermann, Toby Douglas, Sheila Zedlewski, and Shelley Waters, Income Support and Social Services for Low-Income People in California (Washington, DC: Urban Institute, June 1998).

24 Deborah Montgomery, Laura Kaye, Rob Geen, and Karin Martinson, Recent Changes in California Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, May 2002). Cristy Jensen, “First Report Form: State of California” (Albany: Rockefeller Institute of Government, n.d.). Colorado Urban Institute state, Rockefeller Institute state. Nancy Pindus, Randy Capps, Amy-Ellen Duke, and Karin Malm, Income Support and Social Services for Low-Income People in Colorado (Washington, DC: Urban Institute, September 1998). Jeffrey Capizzano, Robin Koralek, Christopher Botsko, and Roseana Bess, Recent Changes in Colorado Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, October 2001). Malcolm L. Goggin, “Report: Colorado Welfare Reform” (Albany: Rockefeller Institute of Government, July 2000). Florida Urban Institute state, Rockefeller Institute state. Pamela A. Holcolm, Kimura Flores, Marta Pernas, Carla Herbig, Karen C. Tumlin, and Christopher A. Botsko, Income Support and Social Services for Low-Income People in Florida (Washington, DC: Urban Institute, February 1999). Christopher Botsko, Kathleen Snyder, and Jacob Leos-Urbel, Recent Changes in Florida Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, July 2001). Robert E. Crew, Jr., and Belinda Creel Davis, “Florida Welfare Reform: Cash Assistance as the Least Desirable Resource for Poor Families,” in Managing Welfare Reform in Five States, ed. Liebschutz, chap. 2. Robert Crew, “Second Report Form: Florida” (Albany: Rockefeller Institute of Government, 2000). Georgia Rockefeller Institute state. Michael Rich, “First Report Form: Georgia” (Albany: Rockefeller Institute of Government, 1997). Michael Rich, “Second Report Form: Georgia” (Albany: Rockefeller Institute of Government, 2000).

25 Kansas Rockefeller Institute state. Jocelyn M. Johnston and Kara Lindaman, “Kansas Carves Out a Middle Ground,” in Learning from Leaders, ed. Weissert, chap. 3. Jocelyn Johnston, “Second Report Form: Kansas” (Albany: Rockefeller Institute of Government, 2000). Massachusetts Urban Institute state. Gretchen G. Kirby, LaDonna A. Pavetti, Karen E. Maguire, and Rebecca L. Clark, Income Support and Social Services for Low-Income People in Massachusetts (Washington, DC: Urban Institute, December 1997). Laura Kaye, Demetra Nightingale, Jodi Sandfort, and Lynne Fender, Recent Changes in Massachusetts Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, July 2001). M. Anne Hill and Thomas J. Main, Is Welfare Working? The Massachusetts Reforms Three Years Later (Boston: Pioneer Institute for Public Policy Research, 1998). Melissa Buis, “Taking States Seriously: Welfare Reform Implementation in Massachusetts and Oregon”, Ph.D. Dissertation, Brandeis University, 2001). Michigan Urban Institute state, Rockefeller Institute state. Kristin S. Seefeldt, LaDonna Pavetti, Karen Maguire, and Gretchen Kirby, Income Support and Social Services for Low-Income People in Michigan (Washington, DC: Urban Institute, July 1998). Kristin S. Seefeldt, Jacob Leos-Urbel, Patricia McMahon, and Kathleen Snyder, Recent Changes in Michigan Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, July 2001). Carol S. Weissert, “Michigan’s Welfare Reform: Generous But Tough,” in Learning from Leaders, ed. Weissert, chap. 6. Minnesota Urban Institute state, Rockefeller Institute state. Martha R. Burt, Rob Geen, and Amy-Ellen Duke, Income Support and Social Services for Low-Income People in Minnesota (Washington, DC: Urban Institute, November 1997).

26 Kathryn Tout, Karin Martinson, Robin Koralek, and Jennifer Ehrle, Recent Changes in Minnesota Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, July 2001). Thomas F. Luce, Jr., “Minnesota’s Balancing Act: Boosting Work Incentives and Job Readiness While Controlling Costs” in Learning from Leaders, ed. Weissert, chap. 5. Tom Luce, “First Report Form: Minnesota” (Albany: Rockefeller Institute of Government, 1998). Mississippi Urban Institute state, Rockefeller Institute state. Gretchen G. Kirby, L. Jerome Gallagher, LaDonna Pavetti, Milda Saunders, and Tennille Smith, Income Support and Social Services for Low-Income People in Mississippi (Washington, DC: Urban Institute, December 1998). David A. Breaux, Christopher M. Duncan, C. Denise Keller, and John C. Morris, “To Privatization and Back: Welfare Reform Implementation in Mississippi,” in Managing Welfare Reform, ed. Liebschutz, chap. 3. Missouri Rockefeller Institute state. Peter Meuser and Lael Kaiser, “Second Report Form: Missouri” (Albany: Rockefeller Institute of Government, 2000). New Jersey Urban Institute state, Rockefeller Institute state. Susan A. Riedinger, Amy-Ellen Duke, Robin E. Smith, and Karen C. Tumlin, Income Support and Social Services for Low-Income People in New Jersey (Washington, DC: Urban Institute, September 1998). Robin Koralek, Nancy Pindus, Jeffrey Capizzano, and Roseana Bess, Recent Changes in New Jersey Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, August 2001). Richard Roper, “First Report Form: New Jersey” (Albany: Rockefeller Institute of Government, n.d.). Richard Roper, “Second Report Form: New Jersey” (Albany: Rockefeller Institute of Government, 2000). New York Urban Institute state, Rockefeller Institute state.

27 Susan A. Riedinger, Laudan Y. Aron, Pamela J. Loprest, and Carolyn T. O’Brien, Income Support and Social Services for Low-Income People in New York (Washington, DC: February 1999). Sarah F. Liebschutz, “Welfare Reform in New York: A Mixed Laboratory for Change,” in Managing Welfare Reform, ed. Liebschutz, chap. 4. Sarah F. Liebschutz, “Second Report Form: New York” (Albany: Rockefeller Institute of Government, June 2000). North Carolina Rockefeller Institute state. Deil S. Wright and Kimberly Cartron, “First Report Reform: North Carolina” (Albany: Rockefeller Institute of Government, 1997). Ohio Rockefeller Institute state. Charles F. Adams and Miriam S. Wilson, “Welfare Reform Meets the Devolution Revolution in Ohio,” in Learning from Leaders, ed. Weissert, chap. 2. Charles Adams, “Second Report Form: Ohio” (Albany: Rockefeller Institute of Government, 2001). Oregon Melissa Buis, “Taking States Seriously: Welfare Reform Implementation in Massachusetts and Oregon”, Ph.D. Dissertation (Waltham, MA: Brandeis University, Department of Politics, 2001). Susan Scrivener, Gayle Hamilton, Mary Farrell, Stephen Freedman, Daniel Friedlander, Marisa Mitchell, Jodi Nudelman, and Christine Schwartz, National Evaluation of Welfare-to-Work Strategies: Implementation, Participation Patterns, Costs, and TwoYear Impacts of the Portland (Oregon) Welfare-to-Work Program (Washington, DC: U.S. Department of Health and Human Services and U.S. Department of Education, May 1998). Rhode Island Rockefeller Institute state. Thomas J. Anton, “First Report Form: Rhode Island” (Albany: Rockefeller Institute of Government, 1997). Thomas J. Anton, Jack D. Combs, Richard M. Francis, and Patrick T. Sharkey, “Against the Tide: Welfare Reform in Rhode Island” (Providence: Brown University, Taubman Center for Public Policy and American Institutions, January 2001).

28 Tennessee Rockefeller Institute state. John E. Gnuschke, “First Report Form: Tennessee” (Albany: Rockefeller Institute of Government, 1997). John Gnuschke, “Second Report Form: Tennessee” (Albany: Rockefeller Institute of Government, 2000). Texas Urban Institute state, Rockefeller Institute state. Nancy Pindus, Randy Capps, Jerome Gallagher, Linda Giannarelli, Milda Saunders, and Robin Smith, Income Support and Social Services for Low-Income People in Texas (Washington, DC: Urban Institute, January 1998). Randy Capps, Nancy Pindus, Kathleen Snyder, and Jacob Leos-Urbel, Recent Changes in Texas Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, June 2001). Christopher T. King, Daniel P. O’Shea, and Shao Chee Sim, “First Report Form: Texas” (Albany: Rockefeller Institute of Government, 1998). Christopher King and Dan O’Shea, “Second Report Form: Texas” (Albany: Rockefeller Institute of Government, 2000). Utah Rockefeller Institute state. Gary C. Bryner, “Welfare Reform in Utah” (Albany: Rockefeller Institute of Government, n.d.). Gary Bryner, “Second Report Form: Utah” (Albany: Rockefeller Institute of Government, 2000). Washington Urban Institute state, Rockefeller Institute state. Sharon K. Long, Sandra J. Clark, Caroline Ratcliffe, and Krista Olson, Income Support and Social Services for Low-Income People in Washington (Washington, DC: Urban Institute, March 1998). Terri S. Thompson, Kathleen Snyder, Karin Malm, and Carolyn O’Brien, Recent Changes in Washington Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, August 2001).

29 Janet Looney and Betty Jane Narver, “Washington’s WorkFirst Program: Key Policy Challenges,” in Managing Welfare Reform, ed. Liebschutz, chap. 5. Betty Jane Narver, Steven Smith, and Janet Looney, “Second Report Form: Washington” (Albany: Rockefeller Institute of Government, 2000). West Virginia Rockefeller Institute state. L. Christopher Plein and David G. Williams, “First Report Form: West Virginia” (Albany: Rockefeller Institute of Government, 1997). Christopher Plein and David Williams, “Second Report Form: West Virginia” (Albany: Rockefeller Institute of Government, 2000). Christopher Plein, “Welfare Reform in a Hard Place: The Case of West Virginia” (Albany: Rockefeller Institute of Government, November 2000). Wisconsin Urban Institute state, Rockefeller Institute state. Kristin S. Seefeldt, Laura K. Kaye, Christopher Botsko, Pamela A. Holcomb, Kimura Flores, Carla Herbig, and Karen C. Tumlin, Income Support and Social Services for Low-Income People in Wisconsin (Washington, DC: Urban Institute, December 1998). Jennifer Ehrle, Kristin Seefeldt, Kathleen Snyder, and Pat McMahon, Recent Changes in Wisconsin Welfare and Work, Child Care, and Child Welfare Systems (Washington, DC: Urban Institute, September 2001). Thomas Kaplan, “Wisconsin Works,” in Managing Welfare Reform, ed. Liebschutz, chap. 6. Thomas Kaplan, “Wisconsin’s W-2 Program: Welfare as We Might Come to Know It?” in Learning from Leaders, ed. Weissert, chap. 4. Thomas Kaplan, Thomas Corbett, and Elizabeth Boehnen, “First Report Form: Wisconsin” (Albany: Rockefeller Institute of Government, July 1, 1997). Thomas Kaplan and Thomas Corbett, “Second Report Form: Wisconsin” (Albany: Rockefeller Institute of Government, 2000).

30 References Barrett, Katherine, and Richard Greene. 1999. “Grading the States: A Management Report Card.” Governing 12 (February): 17-90. Barrett, Katherine, and Richard Greene. 2001. “Grading the States 2001: A Management Report Card.” Governing14 (February): 20-108. Barrileaux, Charles, Richard Feiock, and Robert E. Crew, Jr. 1992. “Measuring and Comparing American State Administrative Characteristics.” State and Local Government Review 24 (Winter): 12-18. Blank, Rebecca M., and Ron Haskins, eds. 2001. The New World of Welfare: An Agenda for Reauthorization and Beyond. Washington, DC: Brookings, 2001. Bowman, Ann O’M., and Richard C. Kearney. 1988. “Dimensions of State Government Capability.” Western Political Quarterly 41 (June): 341-62. Chadwin, Mark Lincoln, John J. Mitchell, and Demetra Smith Nightingale. 1981. “Reforming Welfare: Lessons from the WIN Experience.” Public Administration Review 41 (May/June): 372-80. Citizen’s Conference on State Legislatures. 1971. State Legislatures: An Evaluation of Their Effectiveness. New York: Praeger. Cnudde, Charles F., and Donald J. McCone. 1969. “Party Competition and Welfare Policies in the American States.” American Political Science Review 63 (September): 858-66. Cottingham, Phoebe H., and David T. Ellwood, eds. 1989. Welfare Policy for the 1990s. Cambridge, MA: Harvard University Press. Dawson, Richard E., and James A. Robinson. 1963. “Inter-Party Competition, Economic Variables, and Welfare Policies in the American States.” Journal of Politics 25 (May): 265-89.

31 Eckstein, Harry. 1988. “A Culturalist Theory of Political Change.” American Political Science Review, 82 (September): 789-804. Elazar, Daniel J. 1984. American Federalism: A View from the States. 3rd ed. New York: Harper & Row. Erikson, Robert S., Gerald C. Wright, and John P. McIver. 1993. Statehouse Democracy: Public Opinion and Policy in the American States. Cambridge, England: Cambridge University Press. Fischer, David Hackett. 1989. Albion’s Seed: Four British Folkways in America. New York: Oxford University Press. Fitzpatrick, Jody L., and Rodney E. Hero. 1988. “Political Culture and Political Characteristics of the American States: A Consideration of Some Old and New Questions.” Western Political Quarterly 41 (March): 145-53. Friedlander, Daniel, Stephen Freedman, Gayle Hamilton, and Janet Quint. 1987. Final Report on Job Search and Work Experience in Cook County. New York: Manpower Demonstration Research Corporation. Gais, Thomas, and R. Kent Weaver. 2002. State Policy Choices Under Welfare Reform. Washington, DC: Brookings. Gilens, Martin. 1999. Why Americans Hate Welfare: Race, Media, and the Politics of Antipoverty Policy. Chicago: University of Chicago Press. Gray, Virginia. 1973. “Innovation in the States: A Diffusion Study.” American Political Science Review 67 (December): 1174-85. Grumm, John G. 1970. “Structural Determinants of Legislative Output.” In Legislatures in Developmental Perspective, ed. Allan Kornberg and Lloyd D. Musolf. Durham, NC: Duke University Press.

32 Gueron, Judith M., and Edward Pauly. 1991. From Welfare to Work. New York: Russell Sage Foundation. Hagen, Jan L., and Irene Lurie. 1994. Implementing JOBS: Progress and Promise. Albany: State University of New York, Rockefeller Institute of Government. Hanson, Russell L. 1983. “The ‘Content’ of Welfare Policy: The States and Aid to Families with Dependent Children.” Journal of Politics 45 (August): 771-85. Hero, Rodney E., and Caroline J. Tolbert. 1996. “A Racial/Ethnic Diversity Interpretation of Politics and Policy in the States of the U.S.” American Journal of Political Science 40 (August): 851-71. Hill, M. Anne, and Thomas J. Main. 1998. Is Welfare Working? The Massachusetts Reforms Three Years Later. Boston: Pioneer Institute. Johnson, Charles A. 1976. “Political Culture in the American States: Elazar’s Formulation Examined.” American Journal of Political Science 20 (August): 491-9. Kemple, James J., Daniel Friedlander, and Veronica Fellerath. 1995. Florida's Project Independence: Benefits, Costs, and Two-Year Impacts of Florida's JOBS Program. New York: Manpower Demonstration Research Corporation. Key, V.O, Jr. 1949. Southern Politics in State and Nation. New York: Vintage Books, 1949. Kincaid, John. 1980. “Political Cultures of the American Compound Republic.” Publius10 (Spring): 1-15. Lieske, Joel. 1993. Regional Subcultures in the United States. Journal of Politics 55 (November): 888-913. Mead, Lawrence M. 1985. “Expectations and Welfare Work: WIN in New York State.” Polity 18 (Winter): 224-52.

33 Mead, Lawrence M. 1997. “Optimizing JOBS: Evaluation versus Administration.” Public Administration Review 57 (March/April): 113-23. Mead, Lawrence M. 2000. “The Politics of Welfare Reform in Wisconsin.” Polity 32 (Summer): 533-59. Morgan, David R., and Sheilah Watson. 1991. “Political Culture, Political Systems Characteristics, and Public Policies among the American States.” Publius 21 (Spring): 31-43. Moynihan, Daniel P. 1973. The Politics of a Guaranteed Income: The Nixon Administration and the Family Assistance Plan. New York: Random House. Nightingale, Demetra Smith, Douglas A. Wissoker, Lynn C. Burbridge, D. Lee Bawden, and Neal Jeffries. 1991. Evaluation of the Massachusetts Employment and Training (ET) Program. Washington, DC: Urban Institute Press. Norris, Donald F., and Lyke Thompson, eds. 1995. The Politics of Welfare Reform. Thousand Oaks, CA: Sage. Patterson, Samuel C. 1968. “The Political Culture of the American States.” Journal of Politics 30 (February): 187-209. Pavetti, LaDonna A., Pamela Holcomb, and Amy-Ellen Duke. 1995. Increasing Participation in Work and Work-Related Activities: Lessons from Five State Welfare Reform Demonstration Projects, Final Report. Washington, DC: Urban Institute. Plotnick, Robert D., and Richard F. Winters. 1985. “A Politico-Economic Theory of Income Redistribution.” American Political Science Review 79 (June): 458-73. Putnam, Robert D., with Robert Leonardi and Raffaella Y. Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.

34 Rice, Tom W., and Alexander F. Sumberg. 1997. “Civic Culture and Government Performance in the American States.” Publius 27 (Winter): 99-114. Savage, Robert L. 1978. “Policy Innovativeness as a Trait of the American States.” Journal of Politics 4 (February): 212-24. Scrivener, Susan, Gayle Hamilton, Mary Farrell, Stephen Freedman, Daniel Friedlander, Marisa Mitchell, Jodi Nudelman, and Christine Schwartz. 1998. National Evaluation of Welfare-to-Work Strategies: Implementation, Participation Patterns, Costs, and Two-Year Impacts of the Portland (Oregon) Welfare-to-Work Program. Washington, DC: U.S. Department of Health and Human Services and U.S. Department of Education. Sharkansky, Ira. 1969. “The Utility of Elazar’s Political Culture: A Research Note.” Polity 2 (Fall): 66-83. Sigelman, Lee. 1976. “The Quality of Administration: An Exploration in the American States.” Administration and Society 8 (May): 107-44. Tweedie, Jack. 1994. “Resources Rather than Needs: A State-Centered Model of Welfare Policymaking.” American Journal of Political Science 38 (August): 651-72. Urban Institute. 2000. Snapshots of America’s Families II: A View of the Nation and 13 States from the National Survey of America’s Families. Washington, DC: Urban Institute. Uslaner, Eric M., and Ronald E. Weber. 1975. “The ‘Politics’ of Redistribution: Toward a Model of the Policy-Making Process in the American States.” American Politics Quarterly 3 (April): 130-70. Walker, Jack L. 1969. “The Diffusion of Innovations Among the American States.” American Political Science Review 63 (September): 980-99. Weaver, R. Kent. 2000. Ending Welfare as We Know It. Washington, DC: Brookings.

35 Welch, Susan, and John G. Peters. 1980. “State Political Culture and the Attitudes of State Senators Toward Social, Economic Welfare, and Corruption Issues.” Publius 10 (Spring): 59-67. Wiseman, Michael, ed. 1991. “Research and Policy: A Symposium on the Family Support Act of 1988.” Journal of Policy Analysis and Management 10 (Fall): 588666.

36 Table 1: Assignment of states to Elazar’s political cultures. Moralistic (17) Moralistic Colorado #& Maine Michigan #& Minnesota #& North Dakota Oregon Utah & Vermont Wisconsin #& Moralistic/individualistic California #& Idaho Iowa Kansas & Montana New Hampshire South Dakota Washington #&

Individualistic (17) Individualistic/moralistic Connecticut Illinois Massachusetts # Nebraska New York #& Ohio & Rhode Island & Wyoming Individualistic Alaska Indiana New Jersey #& Nevada Pennsylvania Individualistic/traditionalistic Delaware Hawaii Maryland Missouri &

Traditionalistic (16) Traditionalistic/individualistic Alabama # Arkansas Florida #& Georgia & Kentucky Louisiana New Mexico Oklahoma Texas #& West Virginia & Traditionalistic Mississippi #& South Carolina Tennessee & Virginia Traditionalistic/moralistic Arizona & North Carolina &

Note: States included in the case study analysis are shown in bold. # Indicates states from the Urban Institute Assessing the New Federalism study. & Indicates states from the Rockefeller Institute State Capacity Study. Source: Daniel J. Elazar, American Federalism: A View from the States, 3rd ed. (New York: Harper & Row, 1984), pp. 134-7.

37 Table 2: Performance of case study states on good government criteria in welfare reform. No. of docs1

Culture2

Policy

Consen

Resour

Minnesota Oregon Utah Wisconsin

4 2 2 6

M M M M

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1 1

6 6 6 6

Kansas Michigan Tennessee Washington

2 3 2 4

MI M T MI

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1 1

0 0 1 1

1 1 0 0

5 5 5 5

Missouri New Jersey New York Ohio Rhode Island

1 4 3 2 2

IT I IM IM IM

0 1 0 1 1

1 1 0 0 1

1 1 1 1 1

1 0 1 1 1

0 0 1 0 0

1 1 1 1 0

4 4 4 4 4

Arizona California Colorado Massachusetts North Carolina West Virginia

2 3 3 4 1 3

TM MI M IM TM TI

1 1 1 1 0 1

0 0 1 0 1 1

1 1 0 1 0 0

0 0 0 1 1 1

0 0 0 0 0 0

1 1 1 0 1 0

3 3 3 3 3 3

Alabama Mississippi Texas

2 2 4

TI T TI

0 1 0

0 0 1

0 0 0

1 1 0

0 0 0

1 0 1

2 2 2

Florida Georgia

4 2

TI TI

0 0

1 0

0 1

0 0

0 0

0 0

1 1

State

1Indicates

Political performance3

Admin performance3 Commit Coord Capab

number of documents reviewed to rate the state.

2Indicates

state’s political culture. “M” indicates moralistic, “I” individualistic, “T” traditionalistic. See Table 12.1

3In

state ratings, “1” indicates that a state satisfied a criterion, “0” that it did not.

Total

38 Figure 1: Association of culture with performance in welfare reform. Overall score 6 5

Moralistic MN OR UT WI

Individualistic

KS MI WA

4

Traditionalistic

No. of states 4 4

TN

5

MO NJ NY OH RI AZ NC WV

6

2

AL MS TX

3

1

FL GA

2

3

CA CO

No. of states

MA

9

6

9

X2 = 33.1, p = 0.000.

Note: For political performance alone, X2 = 13.6, p = .035. For administrative performance alone, X2 = 8.6, p = .198. Results are weaker than for both dimensions because these performance scales are only 3-point and there is less variation across the sample.

24

39 Figure 2: Association of political with administrative performance. Political performance

0

Administrative performance 1 2

3

NJ RI

KS MI TN WA

2

AZ CA CO MA WV

MO OH

MS TX

NC

1

FL GA

0 No. of states

3 MN OR UT WI

9

8

10 7

NY

AL 2

No. of states

6 1

5

Gamma = .52, S.E. = .22, p = .019; Tau b = .38, S.E. = .17, p = .025. Note: The above tests use asymptotic standard errors. Using a more conservative upper-bound estimate, Gamma’s S.E. = .248, p = .036. See Leonard A. Marascuilo and Maryellen McSweeney, Nonparametric and Distribution Free Methods for the Social Sciences (Monterey, CA: Brooks/Cole, 1977), pp. 469-70.

24