Expertise Meets Politics

Expertise Meets Politics
Efforts to Work with States By Beth A. Stevens and Lawrence D. Brown

Over the past twenty years, state governments have emerged as critical players in health care. They have an important regulatory role and, as payers for medical care under Medicaid and other programs offering services for needy people, can affect the structure of the delivery system.

As states develop health policies, foundations can assist by providing resources for technical assistance, research, and analysis, and demonstration of new ideas. Foundations can also bring together individuals interested in state health policy to share ideas and experiences. The Robert Wood Johnson Foundation has supported more than fifteen national programs to help states improve health policy, health financing, and the delivery of health care services.

Beth Stevens, a senior program officer at The Robert Wood Johnson Foundation, has managed evaluations of some of the key state programs supported by the Foundation. Lawrence Brown is a political scientist and professor of public health at Columbia University. He has followed and evaluated the Foundation's state policy activities for some time. In this chapter, they explore the question of what effect the Foundation's investments have had on health policy at the state level. It is a difficult question to answer, given complex state political environments and the indirect roles--such as improving information and understanding of options--that foundations play. The authors describe various Foundation-supported programs, examine the problems of working on health care improvement at the state level, and offer suggestions for future approaches to state health policy development. At its core, however, Chapter Four is a case study of how the State Initiatives in Health Policy Program, which attempted to assist states in insurance and financing reform, unfolded in the periods just before and after the efforts at national health care reform during the first Clinton administration.

Since the failure of health reform, programs with states have come under attack by people who assert that foundations are imposing their own agendas and are controlling, rather than aiding, state governments. We believe these assertions are wrong, but they underscore the points made by Stevens and Brown that the process of policy development is highly political and that foundations can play only limited roles. Moreover, even in these limited roles, foundations are subject to attack.

Health reform comes hard in the United States because the policy issues on which it turns are complex and controversial. Increasing access and containing costs are policies that arouse conflicts over concepts of equity versus efficiency, social justice versus autonomy, and the polity versus the market. Less obvious, and less noticed, however, is the fact that health reform also triggers disagreements about institutions, especially about which level of government should settle policy conflicts. At times, policy science and public sensibilities favor a powerful national role; at other times, states inspire greater confidence. Those who tread federalist minefields in hopes of advancing health policy debates aim at moving targets: the national government, the fifty states, some subset of the states, or some combination of central and subnational jurisdictions. In analyzing success in health policy, the "what" and the "who" generally intertwine, creating complications above and beyond the substantive issues.

The ever-shifting dynamic of federal versus state reform is an unsettling fact of life for all actors intent on reforming the health care system. Philanthropic foundations, whose mission to improve health care draws them into working with federal, state, and private-sector policy makers, have constantly been forced to place bets on which level of government is the most promising venue for change at any particular time. Should a foundation work alongside the federal government in efforts to expand health care to underserved areas, for example, or should it rely on state government enthusiasm for such efforts?

The Robert Wood Johnson Foundation has often worked with government at each level in pursuit of its mission to improve the health and health care of the American people. Its programs range from research funded jointly with the federal government to assess the feasibility of measuring the quality of home health care to programs that give states funds to try insurance reforms. Some of these have borne fruit, with the adoption of innovations and further refinement of policies; others yield few tangible products but may generate useful knowledge about why options do not work. Unfortunately, there has been little careful exploration of the intriguing questions these projects raise: When should a foundation work with governments to further its goals? How does it decide which governments to work with? What lessons emerge from Foundation-government relations?

The two-and-a-half-decade history of Foundation interaction with national and state governments is long and complex. No one paper can fully explain the origins and outcomes of such interplay between philanthropy and government or delineate its myriad patterns. Therefore, this chapter takes on part of this complex task by limiting its focus to The Robert Wood Johnson Foundation's involvement with state governments. Are there any patterns that can be drawn from Foundation and state government efforts to tell us when, where, and how philanthropic activities can most effectively contribute to improvements in health care and health? Are there any lessons to tell us about the limits philanthropies face in their efforts to improve the public good?

American culture and politics have long entrusted many important functions to the states, both to manage better government activities in a large nation and to balance the power of the central government. In health care, the states license physicians, nurses, hospitals, and health plans; run the Medicaid program; regulate capital spending (if they choose); and sponsor and subsidize various public health activities, such as immunizations and control of contagious diseases. The states often exceed the call of such duties by launching ambitious policy experiments to expand insurance coverage and control costs. Before President Bill Clinton offered his national plan in 1993, an impressive number of states--Hawaii, Washington, Massachusetts, Oregon, Minnesota, Vermont, Florida, and others--had made significant steps toward expanding coverage.

Issuing licenses and running public health programs logically fall to state governments, which after all are responsible for managing or monitoring local conditions. And health care, at least until recently, has been predominantly local, with most Americans receiving their care from community-based hospitals and physicians in private practice. But pursuing ambitious experiments in financing health care is a less obvious task for state governments to undertake. Such reforms often tread on national-level issues and dynamics; they can involve reallocation of federal funds and regulation of interstate commerce. So why do states take on these issues? What are the advantages of pursuing health care reform at the state level rather than the federal?

One reason is that states can test reform strategies and discover the advantages and limitations on a smaller scale. States can study each other's handiwork and emulate strategies if and as they choose. If innovations in one or more states look especially promising, they might even find their way into national policy--policy built on subnational learning instead of (or as well as) sophisticated theorizing. Second, states can tailor programs to their own conditions. The vagaries of disease, equity in access, medical science, the laws of supply and demand, and other staples of health reform may not respect state boundaries, but as the melodrama of 1992-1994 shows, differences among and within the fifty states vastly complicate the task of forming a consensus on national health policies. Liberals and conservatives differ as to how far variations among states should be respected in public policy, but all recognize that a diverse society requires variations on themes as well as common, integrative leitmotifs that pull the country together. Thus, reforms based on states can more closely fit the particular prevailing circumstances of geography, ethnic predominance, economic infrastructure, and political culture.

A third virtue of state control over health reforms is the states' superior "closeness to the people." The recent federal reform proposal collapsed in part because its creators did not adequately gauge the public's fear that Washington would impose alien arrangements on local institutions. State governments often appear to be more in tune with grass-roots values than is distant Washington. Leaving health reform to the states may or may not make it easier to resolve the major conflicts of value and interest on which health politics turn, but closeness to the public may at least make state endeavors less threatening than a full-tilt national push.

Although such arguments can be adduced to promote states as leaders in health reform, a host of practicalities complicates that ambitious mission. Fifty state systems compete with one another for such economic advantages as business investment, strong tax bases, and job growth. These economic goals are in tension with the increased fiscal extractions (higher taxes and employer mandates) and sizable redistributive measures that major health reforms demand. States that get too ambitious in reforming health care risk losing ground to competitor states. Leadership in health care reform can be a costly eccentricity.

Another practicality that constrains state innovation is federal policy. Medicare and Medicaid rules limit the power of the states to reroute roughly half the dollars in the system, and provisions of the Employee Retirement and Income Security Act (ERISA) preclude state regulation of employer health plans. But even if these encumbrances were set aside, state health reform efforts, like their national counterparts, might still founder on the facts of political life. Change (let alone "fundamental" change) in health affairs is intrinsically contentious, and downsizing these battles from Washington to the states is no panacea for settling them. In fact, the states' closeness to the people may merely intensify the conflicts or smother them. Thus, state government assumption of the role of reformer is fraught with difficulties. Sometimes states can generate meaningful reforms, and sometimes they cannot.

In assessing the prospects for working with states to improve health and health care, foundations face the same perplexing package of advantages and disadvantages as do other agents for change. The key question is, what are the conditions that promote effective state-oriented foundation programs? Three factors are central to a foundation's ability to answer this question: the goals it wants to achieve, the strategies it uses to pursue them, and the fit between them.

Since 1991, the Foundation has tried to improve health and health care by aiming at three concrete goals: to ensure that Americans of all ages have access to affordable basic health care, to improve the way services are organized and provided to people with chronic health conditions, and to promote health and prevent disease by reducing the harm caused by substance abuse. The Foundation has sponsored programs that work with state governments in pursuit of each of these goals.

To improve access to care, the Foundation has launched programs to expand insurance coverage for the uninsured. In the program State Initiatives in Health Care Reform, the Foundation helped states plan and develop significant innovations in financing, including insurance markets and Medicaid. The Health Care for the Uninsured Program supported the development and implementation of state and local initiatives to ensure the availability of health care services for those who lack insurance. It funded projects that designed and marketed insurance products, as well as those that sought to reduce various barriers (such as lack of information) that prevented markets from functioning effectively. The Making the Grade program encourages state governments to finance health services for school-age children by funding school-based clinics. States in this program try to forge links between clinics and new financing initiatives by reorganizing state and local funding policies. The Healthy Kids Program is an effort to assist selected states in design and development of insurance for school children. It uses Foundation funds to provide expert assistance to states so they can test and market these new products. Finally, the Foundation funded the Medicaid Managed Care Program, which works with state governments to maintain and improve access for vulnerable populations now covered under Medicaid managed care. Its intent is to build capacity among state governments, consumers, providers, and managed care plans to make managed care work for Medicaid enrollees.

Another set of programs helps state governments improve services for chronically ill people. The Program to Promote Long-Term Care Insurance for the Elderly stimulated public/private state-level partnerships to develop insurance covering long-term care services. The Foundation followed that program with State Initiatives in Long-Term Care, which offers funds to update long-term care financing and delivery systems. These might include integration of acute and long-term care, increased choice among financing options, and expanded coverage of long-term care services under Medicaid. Recently, the Foundation created the Medicare/Medicaid Integration Program to sponsor a ten-state demonstration of the dual-eligibility managed care model. This model integrates long-term and acute care services for elderly patients under combined Medicaid and Medicare capitated payments. The Foundation also funds work to improve the health care provided to other types of chronically ill Americans. In a project cosponsored with the Pew Charitable Trusts, it has funded experts from the Medicaid Working Group to provide technical assistance to states trying to integrate chronically ill and disabled people into Medicaid managed care systems. Finally, there is the Mental Health Services Program for Youth, which works to improve services for children with serious mental illness by supporting state agency consortia to change the financing, organization, and delivery of services.

In recent years, the Foundation has developed state-oriented programs to help reduce the harmful effects of substance abuse. Two major programs--Smokeless States and Reducing Underage Drinking Through Coalitions--fall into this category. The former provides money to strengthen state-level tobacco prevention and control initiatives aimed at children. The latter supports statewide coalitions of community organizations that publicize the benefits of reducing underage drinking and seek to implement plans to reach that goal.

In addition to pursuing its three specific goals, the Foundation works with states to reduce the costs of health care by focusing on the cost of programs that the states, rather than the federal government, fund. The Workers' Compensation Health Initiative funds demonstrations and evaluations of system reforms, such as inclusion of workers' compensation medical care in managed care systems and development of so-called "twenty-four-hour coverage" that integrates workers' compensation with medical and/or disability insurance.

Cutting across all substantive goals are the Information for State Health Policy program and the Intergovernmental Health Policy Project at the National Conference of State Legislators, both of which seek to build and increase the general policy-making capacity of state governments. The former gives funds to state governments enabling them to improve their ability to collect and report data on the health and health care of their citizens, while the latter enables the conference to organize seminars and prepare educational materials about health policy issues for state legislators.

Four basic strategies constitute the fundamental philanthropic approaches to helping state governments improve their policies affecting health and health care. Each one aims to overcome a key obstacle to effective policy development.

The first philanthropic approach is providing funds to convene parties whose formal agreement or informal influence is essential to solving the health care problems in question. "Bringing the parties together" is a necessary condition of action; such activities can range from informal working groups to formal, multiyear state commissions. The Mental Health Services Program for Youth, for example, promoted such "convening" by sponsoring meetings of officials of disparate, competitive state agencies and advocacy groups so that common procedures and funding could be negotiated. This was an integral part of the program's work to break down categorical barriers to comprehensive services for young people.

The second technique is building the capacity of state government officials to analyze problems and craft solutions. State government officials have complex duties in dozens of policy arenas. Securing the full-time services of talented staff people who focus on limited issues increases the likelihood that coherent and practical health reform plans will emerge and that the proposals will be well-represented, even championed, by the governor's office or legislative leaders. The Program to Promote Long-Term Care Insurance, for example, provided funds so that state officials could devote their time to the task of designing and promoting this new form of insurance--a process that required coordination among underwriters, demographers, marketing experts, agencies for the aging, and claims processors.

The third approach is providing technical assistance to state officials to enhance their ability to address complicated policy issues. Here the Foundation seeks to eliminate factual or analytic roadblocks to both policy development and political consensus. If, for example, the governor's office or key legislators want to clarify an issue, they may request funding of a special study or the services of expert consultants. The Foundation gave state grantees funds to hire experts to construct state expenditure accounts in the State Initiatives in Health Care Reform Program and paid for actuarial analyses that underlie state-sponsored insurance programs in the Healthy Kids and Partnerships in Long-Term Care Insurance programs.

The fourth and final approach is promoting interchange among reformers and policy makers in different states. By holding annual meetings for all its grantees in state-oriented programs, funding publication of newsletters and technical documents, and centralizing administration of its programs in national program offices headed by experts, the Foundation facilitates diffusion of ideas from one state to another. The networks of state reformers have the potential to provide mutual support, share possible solutions to common problems, and create a cadre of states pressing for the resolution of particular health problems. This technique has been particularly prominent in the Mental Health Services Program for Youth and Smokeless States programs.

All these programs differ widely, not only in their policy targets and their strategies for creating social change but also in their ambition. Some are aimed at a relatively narrow target, such as helping states design health insurance products for children; others work for broader changes, such as restructured long-term care financing. Similarly, some programs use a few of the philanthropic tools, while others use a mixture of all four. One of the most ambitious programs the Foundation has developed over the past twenty years is State Initiatives in Health Care Reform. The program offers the opportunity to analyze a program that sets complex goals and uses most of the strategies that foundations have available: convening, technical assistance, staff support, and communications among states. The program shows how generic philanthropic ends and means can help improve health care--as well as how they can be twisted into unexpected outcomes by refractory state political dynamics. It offers an exceptionally instructive case study of the ups and downs of Foundation efforts to encourage changes in health policy.

The State Initiatives in Health Care Reform Program was authorized in 1991, before Bill Clinton was elected and made national health care reform a national preoccupation. In 1991, the prospects for achieving access to health care for all Americans (one of the Foundation's major goals) were fairly dim. The Bush Administration, like the Reagan Administration before it, proposed only limited changes in the way most Americans received insurance coverage for their health care. If access were to be expanded, the Foundation reckoned, it was more likely to come through state-level action. Twenty-eight states had already convened commissions to study proposed reforms in health care financing; a smaller number had received waivers from the federal government to expand their Medicaid programs. In many states, governors and legislators were talking up reforms and debating authorizing legislation. The political will to make major reforms in health care financing appeared to be in place or en route. These considerations moved the Foundation to authorize up to $25.5 million for awards to as many as fifteen state governments to advance the development of significant reforms in health care financing on the state level. (The program was subsequently reauthorized on two occasions. In 1993, the original program, which had ultimately funded twelve states, was expanded to eight more. In 1996, the Foundation decided to maintain its support of state-level reform by providing fifteen states--both recipients and new applicants--with $7.5 million more in grants, to run until the year 2001.)

The explicit goal of the program has been to help states plan and develop insurance market and Medicaid reforms that expand health insurance coverage for the uninsured while slowing the rise in health care costs. Grantees have examined a variety of options, ranging from New Mexico's exploration of universal health insurance coverage through a state-sponsored program to Oregon's employer mandate (enacted and then deferred) and Florida's innovative insurance purchasing pools.

The Foundation deliberately used most of the tools at its disposal. Recognizing that few states had the personnel and financial resources needed to develop and implement health care financing reform, it authorized funds for states to hire or reassign employees who would work on the issue.

Internal capacity is only part of the story, however. Without expertise to guide them, states might stall or resort to unworkable quick fixes. The Foundation's program designers therefore secured technical expertise from a central pool of nationally recognized research organizations. The Urban Institute was funded to help states track trends in eligibility and coverage under state programs and to simulate the effects of policy changes on the uninsured and Medicaid populations. The Rand Corporation helped states analyze the consequences of specific policy options for the state budget, total spending on health, and different sectors of the health care market. The National Governor's Association worked with states to understand policy options for, and barriers to, health financing reform. It, too, produced reports, such as an analysis of the impact of ERISA on state options, and also provided a forum for states to share and discuss policy issues.

The staff time and the technical assistance were devoted to helping states one-by-one. Although such help is obviously important, reforms seldom occur in isolation. Reformers benefit by comparing notes on options that look promising or possibilities that became dead ends. The Foundation therefore funded the Alpha Center so that states could help one another create a wave of reform. Alpha convened conferences on technical and strategic issues; issued newsletters, reports, and other documents; and publicized efforts of individual grantees to a wider audience.

Although no state has yet entered the promised land of affordable universal coverage, the State Initiatives program has helped states in several ways to map, and sometimes to navigate, the twisting road toward it.

The first philanthropic strategy--providing funds to convene stakeholders in health reform--was embraced by most state grantees. Many, including Vermont, Montana, and West Virginia, used Foundation funds to establish special commissions or task forces charged with developing and discussing options for financing care. To be sure, these commissions did not lead to wholesale system reform, but Foundation-funded task forces put such changes high on the agendas of many key public and private stakeholders. Would these discussions have taken place without philanthropic funds? Financing reform was pursued by states outside the program, such as Hawaii, Tennessee, and Massachusetts; and some state grantees had initiated substantial changes in health financing policy before the program began (Minnesota, Washington, Florida, and Oregon, to name the most prominent). One can never control completely for the many intersecting variables that shape innovations and therefore cannot isolate Foundation influence more than impressionistically, but in some grantee states the issue of coverage for the uninsured probably gained more prominence than if the program had not existed.

The second Foundation strategy--providing funds for states to hire staff that would be devoted to the development of policy options--sustained a concentrated and penetrating exploration of promising, albeit complicated, ideas. In Colorado, a five-person policy office designed state-licensed purchasing pools and developed a way to adjust Medicaid payments for risk. New York staff analyzed whether it would make sense to extend the states' all-payer regulations to individual physicians (evidently not) and whether to expand community rating to insured groups larger than fifty. Similar analyses proceeded in other states, which gained a surer grasp of the pros and cons of policy proposals.

Program-funded technical assistance to state grantees reinforced the internal abilities enhanced by the other strategies. The Urban Institute produced reports such as the "State Level Data Book" and "The Distributional Effects of Employer Mandates." These offered information and analyses available nowhere else because they used statistical models that few could match. Researchers from the Rand Corporation designed and analyzed a survey of employers and households in the ten original program states in order to provide much needed state-based estimates of insurance benefits and coverage, and thus a more accurate picture of the state's uninsured population. Rand experts also helped states develop State Health Expenditure Accounts, which let states see where their health care dollars went. The Urban Institute's simulation model, "Trim 2," let states test by how much various policy options increased coverage for uninsured populations. Both the Urban Institute and the Rand researchers worked as consultants to the states as well, for example by organizing educational retreats for state legislators who were eager to explore the technical aspects of different policy options. The Alpha Center encouraged new policy options by holding conferences on such topics as innovative state/Medicaid purchasing strategies; monitoring risk-bearing entities; and the implications of the Health Insurance Portability and Accountability Act of 1996 (the "Kassebaum-Kennedy" legislation). Alpha's reports on issues such as the utility of subsidies to low-income individuals for the purchase of insurance and the lessons of state efforts to develop standardized benefits packages helped build a knowledge base across the states.

Finally, the program advanced communication of ideas and experiences among state grantees and from the states as a group to national deliberations. Conferences, working groups, and technical documents built a sense of community among state reformers. Minnesota representatives shared its incremental approaches to expanded coverage with numerous states; New York explained its electronic claims clearinghouses to Maryland; North Dakota learned from Vermont how to collect data on health expenditures; and Nebraska looked to Iowa for ideas about the design of private insurance purchasing cooperatives. Such sharing helped states avoid policy pitfalls and dead ends and enhanced their understanding of the strategies and stakes of innovation. State leaders and other policy experts now know more about how to subsidize the working poor to help them buy insurance without also encouraging employers to cut their own contributions to workers' coverage in hope of getting a subsidy. States are learning how to pool insurance purchasers to gain economies of scale, as well as market leverage, as Florida, Minnesota, Colorado, and Iowa have done. These insights are significant contributions to future policy debates.

The improvements in policy design and deliberation that foundation strategies produce are noteworthy contributions, but they fall short of achieving broad-ranging (and often even narrow-gauged) reform. No foundation program, including The Robert Wood Johnson Foundation's State Initiatives, has made a major dent in the problems of the uninsured in the states. By 1997, after almost a decade of intermittent but often intense deliberations, no state had achieved universal coverage, none (save Hawaii) had implemented an employer mandate, none had comprehensively reformed its health care system, and most watched the number of uninsured rise steadily despite relatively good economic times. These realities are not so much a verdict of failure, however, as a spotlight on our central point: foundation programs such as State Initiatives may be necessary (or, at any rate, highly useful) to help states down the road to reform, but they are far from sufficient. In a democracy, foundations cannot and should not be expected to engineer reform because conflicts of values and interests in the states themselves demand, and deserve, free play. To an account of goals and strategies, then, one should add the third piece of the policy puzzle, namely, the political characteristics of the states that shape and transform the work of foundations.

Whether foundation programs, such as State Initiatives, end up "working" or not depends largely on the funder's sagacity in reading the capacity and political personality of state applicants. Unfortunately, foundation appraisers usually have little pertinent information on which to rely. A large research literature maps correlates of state liberalism and conservatism in social policy, but these studies have many conceptual and methodological limits and, in any case, shed little light on the disparate factors that might predict the capacity of states to tackle particular reforms. Moreover, states that seem to be quite similar on objective measures (such as the level of economic development) can vary dramatically in policy behavior. Because innovation is, by definition, a significant departure from past practice, it is anyone's guess what information would be needed by funders to predict the probability that innovations will occur and succeed. Because off-the-shelf intimations of state capacity are few and poor, foundations must resort to rough-and-ready indicators such as the stature and commitment of key players representing the state in its quest for money, the state's reputation within policy networks for competence, track records in previous programs, the prospects for public and private-sector collaboration, and the copiousness and cogency of the documents submitted. Meanwhile, prediction becomes all the more imponderable because foundations are typically hoping to nudge states to "higher" levels of policy consciousness, eventually crowned by consensus on major reform.

Even if states were equally positioned for reform, the strategies that philanthropies use have distinct limitations. Convening the parties interested in financing reform, although necessary to begin moving toward consensus, is rarely without risk. One never knows in advance where such deliberations will lead. Sometimes dialogue breeds trust and a sense of collective will, but sometimes it reminds participants why they dodged each other in the past and why they can at best agree to disagree. Sometimes it encourages solutions with a lowest common denominator, whose virtue is that they threaten no one at the table. Making agreements stick, moreover, is complicated by turnover in key positions. Governors and legislators, such as those in Washington and Kentucky, who set and pushed the reform agenda are not around to help guide it into law or, alternatively, to help implement laws they passed. Elected state leaders regularly lose office or influence by means of retirement, electoral defeat, shifts from majority to minority party, and other misfortunes that remove them from the reform game. Frequently, reforming states end up entrusting complex proposals and fledgling laws to the ministrations of public and private leaders who may know little (and care less) about how they emerged and what they meant, and who may doubt their wisdom or even oppose them outright. Shifts in political fortunes in Massachusetts in the late 1980s, and Florida in 1994, for example, display this type of change in painful detail. A foundation has trouble encouraging the convening of stakeholders and thus beginning the process of reform when those who ought to be convened and the issues they need to discuss keep changing.

Technical assistance, the most favored tool of foundations because it is easily produced, runs into contradictions. Because reform is a significant departure from past practice, it is difficult to guess what type of information will be needed to create solutions to complicated and value-laden problems. Moreover, particular state conditions demand particularized solutions, for which the standardized national-level technical assistance that foundations provide might be inappropriate. Finally, immersion in "the data" can multiply and diffuse issues and options, instead of narrowing or focusing them. In 1993, for example, the Foundation-funded project in Minnesota developed higher estimates of the number of uninsured than the state had been using. This created some disarray in the state political establishment and disrupted progress toward a political solution.

In the end, technical assistance cannot completely resolve the complex debates over health reform because it runs into political reality. Ultimately, the health reform plans devised by the best and the brightest experts must pass tests of political acceptability. Foundation aid that expands data and analyses that make the development of proposals easier (and doubtless better) may also deflect attention from political realities. Some otherwise excellent plans (in the technical sense) devised by state technocrats affronted the sensibilities of an uncomprehending public and of an all-too-comprehending phalanx of interest groups whose political interests conflicted with the plans. In places like Vermont, Colorado, and Kentucky, such conflict provoked a backlash that caused the downfall of the "technically superior" plan. It is relatively easy for foundations to help set agendas; it is beyond their mandate to bring reform to legislative life.

Convening, providing technical expertise, offering state staff the time and the resources to pursue reform, and even the camaraderie of membership in a program with other states working toward the same purposes can enhance political deliberations but cannot settle political disagreement. Expanding health coverage is a complicated exercise that requires cooperation from purchasers (public, private, individual), beneficiaries (who may or may not be purchasers and who may contribute to the costs of "reform" in the coin of premiums, copayments, forgone wages, and otherwise), subsidy sources, providers (physicians, hospitals, nursing homes), and of course insurance firms, too. These players, caught in the dynamics of managed care and consolidation even as foundation programs unfold, are not always captains of their fate and may not be reliable program partners. Moreover, such projects can involve a number of state agencies--departments of health, insurance, social services, aging--whose preferences may differ from those of private-sector stakeholders (and of other public agencies). Resolution of conflicts is more often a matter of political compromise and difference splitting than of analytical enlightenment. Politics, however messy, is the sole practical means of resolving the conflict among values and interests.

Foundations and those who evaluate their work should recognize that discussion, better staffing, technical aid, and diffusion of knowledge can tidy up the messiness of health politics only so far. Some conflicts are, so to speak, "hardwired" into complex policy problems by the very nature of decision making in a democratic society. One enduring complication is the range of stakeholders in state health policy and their divergent interests. When seeking to promote subnational change, foundations are not simply dealing with a coherent, unitary entity called "the state"; rather, they must work with many independent organizations, public and private. Insurance reform, for example, may engage the departments of insurance and health as well as the governor and staff and various legislators and committees. How these public bodies behave will also partly reflect preferences of private actors: the insurance industry, managed care organizations, provider associations, employer groups. Lucky is the foundation program that minimizes the number of stakeholders and thus shortens the chain of potentially contentious political clearances. For instance, Medicaid reforms for the disabled do not much implicate employers and insurers, and innovations in long-term care coverage have little direct impact on providers (other than nursing homes). Generally, however, reform requires the support (or at least the acquiescence) of numerous public agencies and private groups who want to know why and how it is in their interests to cooperate. Mandating or convincing them to participate, however, lies beyond the foundation's reach.

A second persistent fact of political life is the intensity of the players' preferences and of conflicts among them. If health policy problems admitted straightforward moral or empirical answers, they would presumably have been resolved long ago. Reform comes hard because health affairs evoke deep disagreement about "the facts," their moral meaning, and their implications for stakeholders.

These intrinsic tensions, however, limit the efficacy of foundations, for whom politics must remain a spectator sport. Innovative insurance offerings, perhaps accompanied by public subsidies, look like a plausible strategy for enticing small business firms to start buying health insurance for their workers--or so a foundation may suppose. When such voluntary efforts fail, however, a foundation cannot overcome business opposition to an employer mandate. Foundation-funded programs only flourish with the invention of win-win designs that benefit all sides. Foundations cannot mandate change in the face of opposition. Win-win programs are not impossible--public-private partnerships in long-term care insurance seem to fill the bill, though critics question whether consumers who pay hefty annual premiums are true winners--but they seldom come easily.

Foundations are condemned to falter in pursuit of health reform because their goals are high, the means available to them are limited, and health reform combines complicated policy problems with acute political conflicts. These constraints on activism mean that The Robert Wood Johnson Foundation's state-based programs must often make do with leading to the waters of policy wisdom horses it cannot compel to drink.

This is not to say, however, that the Foundation's efforts to improve state health policy are ineffective or inefficient. The interposition of foundation and other "third force" institutions between the public and private sectors carries its frustrations but also unique opportunities for innovation. Insulated from both long chains of clearance and electoral sensibilities in government, and from demands for a quick, high return from shareholders of corporations, a foundation can conceive and field innovations limited only by its imagination, funds, and sense of what is workable and in keeping with its mission. It is not to be expected that such gambits will "work" well--or at all--every time, everywhere. Talking and analyzing may not much stir the blood of those who would have the states adopt affordable universal coverage immediately, but interventions that bridge the stages of the stammering social conversation that is health reform are no small feat, especially given the glacial progress of consensus in the states.

The true strategic challenge for a foundation is to avoid either ignoring or indulging the realities of health politics in the states, and instead to "read" these fifty polities for signs indicating how best to advance and accelerate reform. For example, programs tend to run more smoothly and achieve better outcomes when they do not implicate the interests of many groups, when group conflicts over ends or means are not deep and intense, and when the technology of problem solving is relatively straightforward. The Foundation's Program to Promote Long-Term Care Insurance for the Elderly seems to meet these criteria. The key players are departments of health (and/or insurance) and the insurance industry. Obligations and costs can be divided in ways that make economic sense for budget makers, Medicaid programs, insurers, consumers, and nursing homes. Writing and selling such insurance policies (even with appropriate protections for consumers) is not rocket science. Programs to expand coverage for the uninsured, by contrast, have needed support from employers, workers, providers, insurers, and public subsidy sources. None of these groups warmed to the effort, and most saw at least as many costs as benefits to themselves in participating. Writing coverage that was both attractive and affordable was complex indeed.

The Robert Wood Johnson Foundation may want to retain ambitious reform goals, but it may also seek means to assess risks more precisely in advance and to target interventions to more particular political conditions. Doing so requires, first, a systematic reading of the political complexion of the states (are they at the stage of talking, analyzing, or legislating, and how likely are they to master that stage and move successfully to the next?) and second, close contemplation of the main players in the reform "game," in terms of the depth of conflicts among their values and interests and the prospects that they might endorse an intervention that they can implement effectively. Program strategies must then be designed to fit the many realities of different states.

We have come full circle. Health reform comes hard in the United States because the issues are complex and controversial. Both expertise--to devise feasible solutions--and politics--to achieve consensus on solutions--are necessary to achieve true change in state health policy. Foundations can initiate the process--setting the challenge, providing resources to explore the possible answers, and in general enhancing the quality of the debate. But foundations cannot settle that debate. In the final analysis, it is the states and politics that sort out the differing preferences and values that shape true health care reform.