The Democratic election landslide of 1964 made congressional passage of Medicare virtually inevitable. But the struggle over the program's entrenchment continued. To guarantee that physicians were on board, newly-elected President Lyndon Johnson called the leaders of the American Medical Association into the White House. The organization's representatives joined the president for what they thought was a press conference about their help supporting the Vietnam War. When a reporter asked if the AMA would boycott Medicare, Johnson said, “Medicare is the law of the land. Of course, they'll support the law of the land.” Looking at the highest-ranking member of the organization, Johnson said, “Tell him.” The doctor, realizing that he had been put on the spot, said the AMA had “every intention of obeying the new law.”Footnote 1
Johnson's move was a product of his confidence that seniors would mobilize to protect their new health benefits and that the public would vilify the medical profession if it refused to serve a sympathetic constituency.Footnote 2 And Medicare has indeed been a tremendous political success. While Medicare's enactment did not lead to the expansion of coverage to other beneficiary groups, as some of its sponsors hoped, the program has generated consistent support in Congress and ingrained a deep sense of entitlement among recipients.Footnote 3 Medicare's postenactment development is thus an example of positive “policy feedback.” Even though Medicare financing has become contentious as health care costs have placed enormous pressure on the federal budget, the program remains extraordinarily popular with citizens.
Will President Obama enjoy the same degree of success as LBJ in achieving durable policy legacies that transform politics? Obama's two signature first-term legislative victories—the Affordable Care Act (ACA) and the Dodd-Frank Wall Street Reform and Consumer Protection Act—seek to tame the excesses of unregulated markets, redistribute resources to the needy, and promote economic security for all citizens. While Obama's 2012 reelection victory ensures that the ACA and Dodd-Frank will not be repealed before he leaves office, the political struggle over the consolidation of these landmark reforms continues. The measures have aroused opposition from powerful clienteles, significant implementation work remains to be done, and many of the benefits the reforms promise to deliver are not yet apparent to ordinary Americans. Liberals clearly will fail to alter the trajectory of American politics without considerable energy devoted to the embedding of these ambitious reform projects.
We develop here an argument about the obstacles to durable liberal reform in the modern American state. Our argument synthesizes and extends several schools of thought in contemporary U.S. political science. While existing scholarship provides many insights, there has been insufficient probing of the conditions under which policy breakthroughs can refashion the political context in ways that actually entrench and deepen the reforms themselves. One might think that research on the long-term consequences of policy choices would be at the center of the study of American politics. But this topic has not received the attention it deserves. One reason is the complexity of the topic. As David R. Mayhew observes, there are multiple factors that shape whether a policy “takes,” among them “the level of support an initiative enjoyed at the start,” “the level of commitment written into the original enactment,” “whether times have changed,” “whether a new program builds a clientele,” and “perhaps most of all, whether a new initiative once set on the ground seems to work.”Footnote 4
Clearly any effort to account for policy embedding on the basis of a single variable will be unsuccessful. What we need are mid-range theories that draw attention to key sources of policy resilience without stripping issues out of their broader contexts. A second reason for the inattention to policy sustainability is that the American politics subfield tends to be organized around the study of discrete institutions—Congress, the presidency, judicial behavior, interest groups, elections—even though governance occurs across multiple sites of political activity.Footnote 5 The compartmentalization of the American politics field creates artificial separations between politics and policy, lawmaking and implementation, and the study of elite and mass behavior. Fortunately, in recent years, these lacunae have begun to be addressed as empirically-oriented institutional scholars have examined the durability of laws,Footnote 6 and political behavior scholars have probed the linkages between feedback and mass participation.Footnote 7 We highlight lessons from these exciting new lines of work, and identify opportunities to move the study of policy feedback and sustainability forward.
Our central claim is that the capacity of public policies to remake politics is contingent, conditional, and contested. Positive feedbacks are shaped not only by the internal attributes of policies, such as the magnitude and timing of their resource flows, but also by the interaction between such policy-specific characteristics, the strategic goals of officeholders and groups, and the forces arising from an uncertain and contentious political environment. Actors who seek to influence the sustainability of a given policy may not accept feedback processes as they find them.Footnote 8 Instead, supporters may seek to deepen the line of development set in motion by an inherited reform, while opponents, especially economically privileged clienteles who stand to lose governmental subsidies and benefits, may seek to undermine it. By exploring the conditions under which policy feedback production is stymied, we hope to shed light on the predicament of liberal reform in the American polity.
In doing so, we seek to be deliberately programmatic. We trace the evolution of reforms, not to describe past politics for its own sake, but to understand how major policy initiatives create, or fail to create, new political relationships and structures on the ground. The aim is to understand these developments conceptually and to identify recurrent patterns of policy entrenchment and reversal. While we draw heavily upon existing research conducted by policy scholars, historical-institutionalists, and rational choice theorists, and while we present two brief case studies, our goal is be synthetic rather than to rigorously test hypotheses. We draw together theoretical insights and empirical findings that have been scattered in the literature and frame new questions about the trajectories of policy reforms that can be tackled using a variety of research methodologies.
Our argument proceeds in five stages. We begin with a discussion of the challenges facing liberal reform in the contemporary American state. We then review the literature in order to situate our methodological approach and arguments in the context of research on lawmaking, agenda setting, and policy feedback. We then examine the reasons why policies may fail to reconfigure politics, focusing on the influence of coalitional and institutional factors during both the enactment and postenactment phases. In the fourth section, we use our theoretical perspective to illuminate the forces shaping the trajectories of two major reform enactments from the Obama presidency: the ACA and Dodd-Frank laws. We conclude by discussing the role of agency and structure in reform sustainability in an era of partisan polarization. We also offer suggestions for future research, including the need for more work on the role of Congress as policy feedback producer, the conditions under which efforts to mislead citizens about policy effects succeed or fail, and the normative implications of policy entrenchment.
The Challenge of Sustainable Reform in the American State
The fragmented and supermajoritarian structure of the American state creates numerous obstacles to the enactment of new laws. Given the high transaction costs to modifying an existing statute, it might seem that a policy reform's entrenchment is assured once it overcomes the procedural hurdles to adoption.Footnote 9 Indeed, the emphasis of scholarship on the difficulty of dismantling existing programs has diminished interest in what actually results in sustainable reforms. In his 1976 book Are Government Organizations Immortal? Herbert Kaufman found that even when government agencies are eliminated, their programs are simply moved to other parts of the bureaucracy, and that “governmental activities therefore tend to go on indefinitely.”Footnote 10 Based on Kaufman's finding, the assumption in the literature has been that program termination is rare, and that there are “simply not enough cases upon which one could begin to generalize.”Footnote 11 More recent empirical research demonstrates, however, that a variety of policy outputs are regularly subject to modification and even elimination.Footnote 12 To be sure, policies are backed by the coercive power of the state, and many governmental commitments are quite resistant to change. Yet recent scholarship demonstrates that policies once thought to be “locked in” can be subject to subtle processes of erosion.Footnote 13 In American democracy, policy entrenchment has limits. Not only may opponents of newly enacted laws retain office, but new actors may gain power and seek to undo the accomplishments of their predecessors by formally repealing laws, starving programs for resources, or enacting new policies that render the earlier measures a dead letter. Whether a new policy establishes facts on the ground and creates pressure for its own maintenance is an empirical question.
The literature on historical institutionalism in political science offers a good starting point for such an inquiry, because policies evolve in particular institutional and temporal contexts. There is nothing new about political resistance to efforts to strengthen the role of government as a guardian of vulnerable or diffuse interests. Improving policy and promoting social justice have always been works-in-progress in American politics. In recent decades, however, the possibilities for durable liberal reform have been constrained by the thickening of the policy infrastructure. As a result, a president who wants to advance reform in any area of policy must contend with a vast number of other policies that will be affected by the change and interest groups and bureaucracies that are invested in the current status quo.
Stephen Skowronek rightly observes that we are living in the “Policy State.”Footnote 14 The central story line of political development over the past century, according to Skowronek, has been the elevation of policy as the main instrument through which the American state governs. During earlier periods, governance was ordered by “structural assumptions and legal proscriptions.”Footnote 15 There were prescribed limits on discretionary state action, and matters as diverse as race and property relations “were ‘locked out’ of the nation's democratic deliberations.”Footnote 16 The question was not what discretionary policy choices the American state should make, but whether it should make any such choices at all. Over time, reformers succeeded in breaking down the impediments to an activist state, and the outpouring of legislation produced a policy explosion that crossed what James Q. Wilson called “the legitimacy barrier.”Footnote 17 Any subject could be a “legitimate topic on the national policy agenda.“Footnote 18
Yet if the American state is far larger and more extensive than it once was, its operations remain contentious. Indeed, the consolidation of liberal reform has become more precarious in the contemporary Policy State, because preexisting configurations of governance militate against the sustainability of reform commitments. As Karen Orren and Skowronek write, “the new American state makes ever-more promises but offers steadily fewer guarantees. Successive displacement has expanded the commitments of government, but those commitments are more susceptible than ever before to the contingencies of political combat and the ambitions of institutional actors.”Footnote 19 As Lawrence Jacobs and Theda Skocpol observe, reformers today do not have the luxury of designing ideal policies on a blank slate, but rather must craft solutions atop a “huge, pervasive, and complex” infrastructure of preexisting policies. In contrast to the New Deal era, almost every reform now requires groups who already enjoy government subsidies or regulatory protections “to accept less.”Footnote 20 In sum, while the challenge of consolidating reform projects is longstanding, it has particular resonance today.
What Scholars Have Written about Policy Sustainability
While the politics of reform entrenchment has received insufficient attention, several literatures provide insight into this topic. While each has added significantly to our knowledge, each also has limitations worth noting. In considering these insights and limits, our goal is to develop a more synthetic framework for understanding the limits of political reform in U.S. politics.
One relevant stream of scholarship is the literature on policy implementation, which explores what happens after a bill becomes a law and shows that the struggle over a law's legitimacy continues in the post-enactment phase.Footnote 21 All reforms require rules and regulations to translate general legislative language into detailed prescriptions, and if agencies fail to carry out their tasks, the measures will eventually decay. A strength of implementation scholarship is that it recognizes that laws only provide the legal frameworks for exercising public authority, and do not determine how programs work on the ground. Implementation scholars, however, tend to focus on internal bureaucratic factors, and neglect the ways in which the actions of elected officials shape a law's durability. Reforms are subject to a commitment problem: today's officeholders may change their minds about the desirability of maintaining a policy, and even if they do not, they will eventually be replaced by officials with different preferences.Footnote 22
Rational choice arguments about “deck stacking” (as developed by scholars such as Mathew McCubbins, Roger Noll, and Barry Weingast) suggest a solution to this commitment problem: reformers who won the enactment battle can strategically create procedures to establish a structural context that endures long after the enacting coalition has frayed, ensuring that the coalitions that generated the program's adoption will also hold sway during the post-enactment phase.Footnote 23 The deck-stacking argument captures an important dimension of policy durability. Research demonstrates that the institutional designs established at enactment can serve to promote policy durability.Footnote 24 Yet there are limits to what deck stacking can accomplish. While reforms are shaped by their structural environments, future policy outcomes rarely can be “locked in.” The deck-stacking perspective overstates the ability of “winners” to impose their wishes on “losers,” who often have the opportunity to embed some of their structural preferences into laws as well, as our discussion of Dodd-Frank will suggest.Footnote 25 Indeed, as Eric Schickler points out, institutional reforms are frequently little more than “common carriers” of multiple interests. Embedded into each reform are a variety of objectives—some of which are at cross-purposes—from the various actors involved in any major debate over institutional change.Footnote 26 An even more important limitation of the deck-stacking hypothesis is that it is static. It fails to recognize that contexts and coalitions mutate over time, and that design-stage choices can be nullified by decisions made during the postenactment stage. As William N. Eskridge and John Ferejohn argue, the deck-stacking model treats the preferences of actors as exogenous, but such preferences are also “a product of deliberation and feedback, not anterior to it.”Footnote 27 The issue, then, is not only whether the interests that originally prompted a reform will endure, but whether a reform sustains the coalitions that brought it about or causes new coalitions to emerge after enactment.Footnote 28
Structural design would matter little if the composition of Congress and the executive never changed.Footnote 29 But elections, retirements, and demographic shifts are continuously remaking the lineup of political leaders. It is critical, then, to look at the partisan structure of governing coalitions, both at enactment and over time as reforms are carried out. In recent years, scholars have produced compelling statistical evidence that coalitional variables offer significant purchase on a range of outcomes. In an important study, Forrest Maltzman and Charles R. Shipan have shown that laws enacted under unified government are more likely to persist in their original form than laws enacted under divided government.Footnote 30 In another insightful study, Christopher R. Berry, Barry C. Burden, and William G. Howell have found that changes in the partisan composition of coalitions affect programmatic durability.Footnote 31These studies substantially advance the research frontier, but they also have limitations for understanding policy entrenchment. Macro-level coalitional factors may not always yield their predicted effects, because these factors are mediated by the preferences of individual leaders, the goals of clienteles, and the attributes of preexisting policy commitments. Consider the unraveling of the Tax Reform Act of 1986. One might have expected George H. W. Bush's election in 1988 to have entrenched tax reform, since the Republicans kept the White House and Bush had been vice president two years earlier. Yet Bush pushed for the creation of new corporate tax loopholes, undermining the principles of tax neutrality and simplification upon which the reform was based.Footnote 32 Attention to macro-level coalitional variables provides insight into the likelihood of changes in outcomes, but such data must be enriched by a more detailed investigation of the evolution of particular policy issues.
The vulnerability of even long-standing policies to sudden bursts of rapid change is a theme of the literature on punctuated equilibrium.Footnote 33 Preexisting patterns of governance can be shattered when political entrepreneurs thrust new issue definitions onto the agenda. As developed by Frank R. Baumgartner and Bryan D. Jones, punctuated-equilibrium models seek to explain both stability and change in policymaking by highlighting the limited attention spans of boundedly rational actors. Punctuated-equilibrium models draw attention to the role of “mobilizations of criticism” in shaking up existing arrangements, and to the role of countermobilizations in maintaining stable outcomes in the face of changing external pressures. The latter process is homeostatic, and helps explain why politicians often seek to balance out the favors given to competing groups. As Baumgartner and Jones persuasively argue, “without countermobilization, political interests would gather ever-increasing powers until they overwhelmed the entire political system.“Footnote 34 Punctuated-equilibrium research suggests that every policy is subject to possible renegotiation at some point, and that at any given moment, some punctuations are under way. These are powerful insights into the fluidity of the American polity, which complement research on the politics of policy development, but they have limits for interrogating the failure of specific governmental commitments to become self-reinforcing.Footnote 35 Punctuated-equilibrium theory helps us detect patterns of governance at the macro level, but the focus on the aggregate distribution of outcomes has a cost in terms of understanding how a specific policy creates demands for its own expansion. As Baumgartner and Jones, writing with James True, state, “Punctuated-equilibrium theory predicts a form of system-level stability, but it will not help us to make point-specific predictions for particular policy issues.”Footnote 36 This is not a problem for the analyst who seeks to understand how political institutions process information, but it is a limitation for scholars interested in understanding increasing returns processes in a given policy arena. Finally, punctuated-equilibrium research focuses on changes in actors' attention, rather than on changes in their preferences, identities, and interests.
The latter shifts are our primary concern, and they are explored in the literature on policy feedback.Footnote 37 Building on E.E. Schattschneider's claim that “new policies create a new politics,” the core argument contained in this literature is that policies are not merely the products of politics but also shape the political landscape by altering state capacities, changing incentives for collective action, and encouraging social adaptations that may become difficult to reverse.Footnote 38 One of the most exciting recent developments in the study of feedbacks has been the emergence of research systematically linking policy attributes with individual-level political behavior. Andrea Campbell has shown how the Social Security program helped transform senior citizens into the most active participatory age group in American politics by giving them resources and enhancing their political efficacy. And Suzanne Mettler has demonstrated that the G.I. Bill increased the civic engagement of veterans, especially those who were economically disadvantaged and less likely to attend college otherwise.Footnote 39 These empirical studies provide a detailed examination of how individuals respond to the incentives contained in specific governmental programs.
Our analysis builds on these important behavioral studies, but broadens the scope of inquiry to examine how policymaking processes evolve over extended periods of time.Footnote 40 Policy feedback is a crucial locus of inquiry in the study of policy development, because preexisting policies not only stimulate reactions and adaptations among individual citizens, but shape the larger material and ideational context in which partisan and ideological struggles over the role of activist government takes place. Some thoughtful scholars who study the politics of specific welfare state programs have expressed skepticism about the utility of policy feedback arguments. For example, in his excellent account The Political Life of Medicare, Jonathan Oberlander argues that it is not easy to identify all the feedback effects relevant to Medicare policymaking that should be measured, and that feedback effects cannot be “deduced automatically from a particular set of institutional arrangements.”Footnote 41 We agree with Oberlander that programs as complex as Medicare have a large number of potential feedbacks and we share his skepticism of monocausal explanations. Yet we believe that it is possible to develop mid-level theoretical propositions about feedback effects that can illuminate patterns of governance across space and time.Footnote 42
In a penetrating 1993 essay, Paul Pierson provided an influential statement about the major feedback processes.Footnote 43 He argued that policies have both “resource” and “interpretive” effects: policies distribute material resources, mediate access to public authority, shape actors' awareness of government activities, and construct the political identities of constituency groups. Despite many studies that have uncovered evidence of these impacts, we still know too little about the conditions under which feedback arises. As Pierson observes in a more recent assessment, while scholars have made major strides in providing “demonstrations” of the significance of feedback, there has been limited progress in translating such work into a “coherent and cumulative research program” focused on the evaluation of propositions about when we should expect to see what kinds of effects.Footnote 44 We share these concerns. As Campbell observes in a thoughtful review essay, “in many respects, the feedback literature has imitated the social movement literature in selecting on the dependent variable by analyzing cases where the phenomenon of interest appeared.”Footnote 45 Yet recent studies argue that there are clear limits to how much policies can remake politics, and show that even landmark laws can be unceremoniously abandoned after enactment.Footnote 46 In sum, we know what feedback can do—but not as much about when feedback will or won't happen.
The Limits of Policy Feedback
In what follows we focus on two key stages of policy development: the enactment phase (when policies are designed, feedback potential emerges, and preexisting governance structures are rearranged or left untouched) and the postenactment phase (when implementation begins, public authority is engaged, and processes of self-reinforcement take off or not). Building on Pierson's analysis, we examine three mechanisms: resource effects, interpretive effects, and institutional supports (refer to Figure 1). Our analysis suggests that the absence of strong feedback is often a product of initial policy design features, divisive enactment, and the failure of reformers to uproot the institutional bases of opponents when a law is enacted. But even after a reform is signed into law, the strength and character of feedback can be shaped by the strategic choices of actors during the postenactment stage.
Figure 1 Factors undermining positive feedback
The Enactment Phase
Enactment phase decisions can have a large influence on whether a policy generates feedback. As we stress below, it is certainly possible for a fragile, poorly conceived policy to overcome the infirmities of its birth and eventually generate strong feedback. But the odds are against it. Three enactment phase outcomes can undermine the prospects for feedback generation: the failure of the initial policy design to create incentives for constituency mobilization, the failure of reformers to undercut the institutional bases of support of the policy's opponents, and the adoption of the policy by narrow margins.
When Congress creates a policy that provides meager resources, group feedbacks will likely be feeble. One might think this self-defeating outcome would occur only as a result of some kind of policy design blunder, but the reality is that Congress often crafts laws that are symbolic in nature.Footnote 47 Such laws are not designed to change society but simply to give reelection-minded lawmakers the opportunity to display their support for popular causes before largely uniformed public audiences, without having to confront the costs substantive reform would entail. Symbolic policies do not make a new politics; they refract an old one. We wish to be clear. The passage of a symbolic policy is not a failure of feedback per se but rather a reminder that feedback production is frequently not a priority for reelection-oriented lawmakers.
Even when enacting coalitions genuinely intend to reconfigure politics, the coalition-building imperatives of the legislative process can lead to the failure to build supportive constituencies. Leaders seeking to build a winning coalition have an incentive to sprinkle the benefits of a bill widely across jurisdictions so that every member gains something. This maximizes credit-claiming opportunities for members of Congress, but it also means that the benefits to any one area may be too small to generate support. In the Model Cities program during the Great Society, for example, the aim was to target resources at a dozen distressed cities to test whether intensive urban renewal assistance could help break the cycle of poverty. But by the time Congress passed the law, 227 congressional districts were eligible for assistance even though the total level of funding did not increase.Footnote 48 No constituency was enthusiastic about maintaining the program because the resources were so meager, and the program was folded into another funding stream a few years later.
The problem of building a constituency often arises when policymakers enact a reform that withdraws economic “rents” from a well-organized clientele, so as to produce diffuse benefits for the general public. After the high enthusiasm that surrounds the reform's passage fades, the interest of ordinary citizens may wane, just as rent-seekers are marshaling their resources for the next battle. This is not to suggest that diffuse-benefit reforms can never become self-reinforcing. They can, as was the case with airline deregulation and the Clean Air Act. The key point is that the direct beneficiaries of such policies will rarely develop into an effective organizational force, because the per capita stakes are too small. When such policies do generate positive feedback, it is typically because they bring about institutional shifts that privilege common interests, or because they produce substantial resources for service providers that obtain an incentive to protect their “spoils.”Footnote 49
Policy designs that delay benefits can also undercut constituency building. Some of the most successful programs had a fast roll-out in the months that followed congressional enactment. The Medicare program, for example, was passed in 1965 and up and running within a year—a remarkably fast launch that Obamacare (as we note later) never attempted to match. But when the benefits flows of a new policy are not immediately forthcoming, its entrenchment is threatened. This was a problem with the original Social Security program of 1935, which delayed the start of retirement benefits until 1942 even though the payroll tax began in 1937. While FDR's intention was to accumulate a reserve fund to reduce the program's burden on future taxpayers, this policy design was untenable. Not only did the prospect of a reserve buildup invite scorn from conservatives (who said it would tempt the government into profligate spending), it failed to give senior citizens any near-term benefits from the program. By the mid 1940s, program advocates were concerned about Social Security's survival. Of course, Social Security ultimately became America's most beloved social program, but (as discussed later) only after Congress reworked the program to strengthen its base of support.
Reforms are more likely to become entrenched when enactment-phase decisions undercut or destroy the institutional bases of support of the reform's opponents. When this does not occur, opponents may be able to prevent the new policy from taking root. As Karen Orren and Stephen Skowronek suggest, there is a hierarchy of institutional shifts.Footnote 50 The most robust changes involve the termination of an existing institutional structure whose continued operation would undermine the potential for the reform's self-enforcement (dismantling). When Congress deregulated the airlines in 1978, for example, it included a sunset clause to eliminate the Civilian Aeronautics Board, which had previously protected the industry from market competition.Footnote 51 Much lower in the institutional recalibration hierarchy, but quite common in the Policy State, are layerings—the placement of new rules and institutions atop preexisting ones, which can generate conflicts among programs and agencies that undermine policy sustainability. When the Tax Reform Act of 1986 was passed, for example, advocates hoped the reform would reconfigure the politics of tax policy-making. But the reform left in place the existing rules for passing tax bills in Congress, and failed to impose structural changes (such as campaign finance reforms or new budget institutions) to protect the reform from postenactment lobbying. Congress's commitment to maintaining the reform's principles fell apart, and the IRS code is today once again riddled with countless particularistic tax loopholes.Footnote 52
Finally, the adoption of a policy by slim margins may both signal and cause durability trouble. Maltzman and Shipan have shown that the greater the roll call opposition when a bill is passed, the more likely the law is to be amended in future years by a subsequent Congress.Footnote 53 While the amendments may not weaken a law (some amendments are friendly), they do change it. Divisive enactment may undermine the credibility of the government's promise to stick with the new policy, which discourages the organizational adaptations needed to make the policy effective.Footnote 54 At a minimum, divisive enactment raises eyebrows. Ordinary citizens who do not pay close attention to public affairs could easily conclude that there must be something wrong with a new law if a near-majority opposes it. A related issue is whether partisanship exacerbates the problems of divisive enactment. While systematic data are lacking, it seems plausible that the prospects for a law's consolidation will be even dicier if a reform was passed over the unanimous or near unanimous objections of the opposition party. As Mayhew suggests, while a cross-party opposition to a policy might fade after its enactment, a party is “an organization built exactly to generate messages and mobilize voters. A party that loses on a congressional issue and stays angry may have an incentive to keep the conflict going.”Footnote 55 In the current polarized environment, in which laws often pass on narrow, party-line votes, and lawmakers have an incentive to keep fighting old partisan fights to stave off primary challenges, enactment victories have arguably become less authoritative and final; the battle can always be rejoined another day. This is not to suggest that passage by a bipartisan majority guarantees that a reform will last, but rather that the adoption of a major law on the basis of a narrow partisan majority is a risky proposition, especially if the policy lacks other entrenching elements.
The Postenactment Phase
The struggle to remake politics does not end when reforms are enacted. New polices are trial-and-error affairs, and many policies don't pan out. According to Berry, Burden, and Howell, a spending program has a one-percent chance of death every year in its first ten years of life.Footnote 56 After a program has survived for a decade or so, the probability of termination slowly begins to decline. The probability of death increases when the Congress that inherits a program differs in partisan terms from the Congress that created it.Footnote 57 This makes sense, and scholars have paid insufficient attention to influence of elections outcomes and partisan forces on the prospects for policy entrenchment.
Yet lawmakers' incentives to modify a preexisting policy are clearly influenced by factors operating at both the electoral and governing levels. The question, then, is not whether politics causes policy, or policy causes politics, but rather how politicians' strategic preferences interact with the attributes of existing policies and the resources of the broader political environment.Footnote 58 Beyond unfavorable electoral tides, there are four ways that policy consolidation and feedback production can become arrested during the postenactment phase: advocates may fail to repair defects in the original policy design; changing perceptions may cause actors to become skeptical of the success or value of a policy; a policy may fail to cause groups to alter their social identities in ways that strengthen constituency support; and finally, a policy may be unable to call upon needed intuitional supports.
While the partisan context constrains what is achievable, it is critical for reformers to strengthen the sustainability of policies once design flaws surface. Social Security's early struggles are instructive. Recognizing that Social Security was losing a popularity and legitimacy “race” with means-tested old age assistance, which had unexpectedly strong feedback effects since it started paying benefits immediately, a unified Democratic Congress in 1939 moved up the starting date of Social Security benefits to 1940 from 1942, increased the generosity of retirement payments so that senior citizens would receive a more favorable rate of return on their contributions, and extended coverage to a variety of groups including surviving dependents.Footnote 59 In 1950, the Democratic Congress passed further amendments that strengthened the pay-as-you-go system and liberalized benefits to make the program much stronger than old-age assistance. These moves helped put Social Security on a trajectory of self-reinforcement during the post-war era.
While postenactment changes to the magnitude and timing of the resources received by constituent groups can shape whether a fragile policy begins to “take,” perceptions of material stakes are also important. Strategic actors do not simply accept actors' understandings of a preexisting policy but may engage in deliberate attempts to alter the way citizens and groups view the policy's benefits and costs. There can be intentional efforts by the opponents of a policy to suppress information or even spread falsehoods with the hope that mass confusion will prevent policy embedding.
The enactment and repeal of the Medicare Catastrophic Coverage Act illustrates how actors can use misleading information as a political weapon in the postenactment phase. In 1988, Congress passed the MCCA to provide prescription drug coverage. President Reagan agreed to support the program, but only if senior citizens financed its costs. Congress raised the Medicare B voluntary premium to pay for a third of the program. To pay for the remainder, Congress imposed a premium on upper-income elderly beneficiaries. The premium was structured to climb with individual income tax liabilities, with the richest seniors required to pay an extra $800 (an amount lower than the insurance value of their benefits). The bill passed overwhelmingly but congressional support quickly unraveled. Opponents claimed, falsely, that all seniors would be required to pay the maximum $800 premium, though only a fraction of seniors would owe this amount. Public support for the bill plummeted, even among low-income seniors who stood to gain the most from the program. About a year later, Congress voted by large margins to repeal the legislation.Footnote 60
Besides changes in views about whether a policy is “working,” beliefs about the legitimacy of a policy can evolve during the postenactment phase in response to debate. Many terminated policies were seen, by some actors at least, as the products of particular historical circumstances. When their moments passed, the ability of the policies to reproduce themselves diminished. For example, the Civilian Conservation Corps, which was created in 1933 to provide work opportunities for unemployed young people, was treated as a temporary government intervention in the labor market, with short-term arrangements for its administration and funding.Footnote 61 Without a solid institutional home, the program lacked a political base. With the military demands that resulted from World War II, politicians no longer saw a need for it as young people were drafted into the army or into wartime production.Footnote 62 Of course, the termination of a policy that was never intended to be permanent may be appropriate. But there may be no consensus among policymakers on when the conditions that legitimated a policy's adoption have changed, especially in an era characterized by elite polarization. If the stock market is booming in ten years, will the Dodd-Frank regulations be seen as a response to a long-ago financial crash, and thus no longer necessary? Most liberals would argue otherwise, but the reform's crisis origins will make it easier for opponents to argue it is no longer needed.
Another postenactment question is whether a reform changes how people construct their identities and perceive their status within the polity. Mettler offers a compelling account of how the GI Bill imbued veterans with a deep sense of civic obligation.Footnote 63 Program recipients saw themselves as citizens seeking to help their families and better the nation. This reframing of personal identity was easy to embrace, and caused little tension with veterans' self-conception as soldiers. Yet not all policies shape identities in such positive and linear ways. Writing about the emergence of new opportunities for Americans to choose more than one race on census forms, Jennifer Hochschild and Vesla Weaver argue that it is too soon to know what impact such a multiracial classification will have on the racial order. It is possible that multiracialism will dissolve racial prejudices, but it is also possible that “an increase in multiracial recognition and identity may dangerously challenge group loyalties that remain essential in a polity where races and ethnicities are still arrayed hierarchically.”Footnote 64
A final reason why policies may wither after enactment is the inability to call upon needed state capacities. Christopher Bonastia shows how institutional weaknesses prevented the Department of Housing and Urban Development from attacking residential segregation in the late 1960s and early 1970s, despite support from the courts and advocacy groups. Unlike the Equal Employment Opportunity Commission, which successfully challenged racial discrimination in the labor market, HUD lacked bureaucratic autonomy, leaving it more vulnerable to a backlash.
The most tragic example of how the lack of adequate state authority can lead to the reversal of a major initiative is post-Civil War Reconstruction. African Americans made extraordinary use of the opportunities afforded them by Reconstruction. They won elections, and established dense organizational networks at the local level.Footnote 65 Yet African Americans were still left in a vulnerable condition without durable federal support. When white southerners moved to disenfranchise African Americans through violence and laws prohibiting political participation, the freedmen were unable to defend their interests.Footnote 66 Congress relied on two institutions to overcome the resistance of southern elites to Reconstruction: federal courts and the military. Both were inadequate to the task. The federal bench was unsympathetic to many aspects of Reconstruction, and there was unease with granting soldiers authority to enforce federal laws outside of a wartime crisis. Northerners hoped to reconfigure power relations through the Thirteenth, Fourteenth, and Fifteenth Amendments and the Enforcement Acts of 1870–1872.Footnote 67 But the effort collapsed following the removal of federal troops from the South, unfavorable court rulings, and the filibustering of the 1890 Federal Elections Bill, which would have established federal supervision of elections.
Will the Obama Administration's Reforms Remake Politics?
We turn now to a brief examination of the entrenchment prospects of Obama's two big first-term victories, the ACA and Dodd-Frank. Will these reforms not only endure but transform the way that politics unfolds? Will they cultivate clienteles, reconfigure authority, increase democratic accountability, and alter citizens' expectations of government? It is hazardous to make predictions about how policies will evolve, especially about those with as many moving parts as the ACA and Dodd-Frank. While President Obama's 2012 reelection victory ensures that efforts will be made to implement the laws, it does not ensure that such attempts will be successful.Footnote 68 The consolidation of both reforms remains incomplete, due to the divisive manner in which the reforms were enacted, the unwillingness of opponents to drop their resistance and work constructively on fixes to problematic provisions, and the reforms' incomplete recasting of state authority. Yet, the two reforms also contain elements that—if allowed to realize their potential —could generate positive feedbacks and thereby promote the measures' long-term success.
The Affordable Care Act
The passage of the ACA was a remarkable achievement given the odds stacked against it.Footnote 69 From the moment the law passed, however, the question has been not just whether the ACA can work, but whether it will last. As Brookings economist Henry Aaron argues, President Obama's reelection victory settled the immediate fate of the ACA, but a massive amount of administrative work must be done, including enrolling millions of Americans in the exchanges. While Democrats thwarted efforts to defund Obamacare in the debt ceiling battle, the opposition has not vanished, and there is “every prospect” that health reform will be a contested issue in the 2016 election.Footnote 70 An analysis of the three key mechanisms of feedback identified above—interpretive effects, resource effects, and institutional supports—offers insights into why the ACA still faces obstacles to its entrenchment.
One challenge is to boost support among citizens who currently lack comprehensive insurance coverage so they begin to see themselves as beneficiaries of the ACA in the same way that seniors identify themselves as valued recipients of Social Security. The benefit standards contained in the ACA will compel many younger workers to purchase more insurance than they have been accustomed to. The political blowback when millions learned their existing plans would be cancelled forced the Administration to apologize and allow the “substandard” plans to stand for another year. Yet such workers must ultimately join the exchanges to avoid an insurance “death spiral.” Also, the complicated nature of the ACA makes positive interpretive effects harder to generate than is the case with entitlements that base eligibility on straightforward criteria like age. As Oberlander points out,
unlike Medicare and Social Security, Obamacare does not have a well-defined population of beneficiaries, and its benefits are diffuse. In fact, the ACA is not so much a program as a series of programs, regulations, subsidies, and mandates …. It treats different groups of Americans in different ways at different times, which complicates efforts to explain the law, enroll eligible populations into new benefits, and mobilize public support.”Footnote 71
The ACA's reliance on tax credits and regulations (such as rules prohibiting health insurance exchanges from discriminating against people with preexisting medical conditions), in lieu of more direct forms of social provision, illustrates what Suzanne Mettler calls “the submerged state.”Footnote 72 It remains to be seen whether these tax credits and regulations will be sufficiently visible (and linked closely enough in the public mind to government) to produce strong feedback effects. Surveys conducted in 2013 showed that many low to middle income Americans remained unaware of the law's key features and were skeptical that it will actually help them.Footnote 73 The Obama administration has launched a major outreach campaign to help citizens understand how the program works and how eligible persons can obtain subsidies for insurance coverage.
This outreach campaign will be critical to building support for the program against the backdrop of the continuing effort of opponents to generate negative interpretive feedbacks by defining the ACA in destructive ways. The divisive debate over Obamacare in Congress, marked as it was by explosive charges that the law would lead to the creation of “death panels,” could not help but create a negative impression among many citizens.Footnote 74 Public opinion about the ACA was divided when President Obama signed the law in 2010, and it remains so today.Footnote 75 Democratic voters are much more supportive of the law than are Republican voters, which reflects the elite polarization over the measure's enactment. Unlike many other major domestic policy breakthroughs, the ACA did not pass by wide bipartisan margins in Congress. Indeed, the ACA did not attract a single GOP vote in either chamber. In contrast, the Social Security Act of 1935 was supported by majorities of Republicans in the House and Senate on final passage. The final votes on Medicare in 1965 were closer, but that law still won the support of a narrow majority of House Republicans and almost half the GOP Senators. The ACA's partisan enactment would matter less to its political sustainability if the program was assured of a fast, error-free roll-out, but it isn't. When something (inevitably) goes wrong somewhere, the GOP will have a strong incentive to tell voters, “We told you so.”Footnote 76 This was exactly what Republicans said when the administration announced it would delay the requirement that employers with more than 50 employees provide coverage to their workers until 2015. The GOP pointed to the change as an example of administrative problems, the unfair treatment of ordinary citizens who would still be subject to the individual mandate, and evidence that President Obama was willing to use illegitimate executive power to make the program work.
The ACA's resource effects is a second dimension on which feedback production is problematic. Policies are more likely to become embedded when they deliver substantial, early-order resources to program recipients, because politicians are reluctant to withdraw tangible benefits from citizens once they are given. However, the ACA's individual-level resource effects will take time to generate. In order to satisfy budget constraints and provide time to get the program going, the Administration designed the ACA with a slow phase-in. The major consumer subsidies, Medicaid expansions, and insurance reforms (with some exceptions such as the provision allowing young adults to stay on their parents' plans) do not take effect until 2014. This protracted implementation schedule has dampened the constituency-building that might have begun immediately upon enactment if the benefits had kicked in right away. Obamacare is still an abstraction for most Americans. It has not yet been woven into the fabric of people's lives, although this could change greatly in the years ahead if the consumer subsidies are not watered down. Making good on the law's benefit promises will not be easy, however, given the pressure to reduce the long-term federal budget deficit. The ACA was designed to be a far more redistributive reform than Social Security, imposing taxes on the affluent to finance insurance subsidies for lower income citizens. Deficit hawks are already calling for cuts in the subsidies for the needy. We agree with Jacob Hacker that “maintaining the redistributive aspects of health reform will thus be a formidable challenge.”Footnote 77
The third challenge involves the institutional supports needed to achieve the law's goal of making health care more affordable and accessible. By international standards, the American health care system is expensive and wasteful. If resources could be better allocated, coverage expansions would be easier to pay for and maintain. One of the ironies of the conservative resistance to the ACA is that it is in certain respects a moderate law. In regulatory ambition, the reform pales in comparison to what liberals had proposed in the 1970s and even under President Clinton. A streamlined, single payer system was never on the table, and the voluntary “public option” that many progressives desired as a check on the power of insurance companies was stripped out to keep the Democratic coalition together and avoid a Republican filibuster. The fact that ACA was built atop our patchwork private-public insurance and delivery system hinders cost control efforts, creates tensions with preexisting health care arrangements, and frustrates the reform's embedding.
Obamacare is financed in part by $716 billion in cuts to Medicare providers and insurance companies. The Obama administration has argued that these cuts will not harm seniors. However, the Medicare actuary has warned that the cuts could lead some providers to drop out of the program.Footnote 78 If seniors believe (or are led to believe) that the ACA has harmed the quality of their care, the politics of the program could remain contentious.
The most creative institution-building move of the ACA is the establishment of the Independent Payment Advisory Board, a commission charged with recommending changes to Medicare if projected per-beneficiary spending growth exceeds specified rates. Congress knew the IPAB would be unpopular, so it used a deck-stacking strategy to protect it from attack: the Secretary of Health and Human Services is required to implement the IPAB's recommendations unless Congress passes an alternative plan to reduce Medicare costs. Because of political compromise, the IPAB lacks the authority to change Medicare benefits, yet critics still charged that the agency is a “rationing board.” The Board has not yet been convened because Medicare spending growth has moderated the last few years. Weakening the IPAB remains a priority of industry groups as well as GOP lawmakers, however, and the House has already amended its own rules to overturn a provision of the ACA that sought to expedite congressional debate on the agency's recommendations.Footnote 79
However, with the exception of the IPAB, and the creation of a new entity designed to promote evidence-based medicine,Footnote 80 the ACA did not attempt to build new institutional capacity to rationalize the health care system and alter the behavior of physicians, providers, and payers. To be sure, the Act contains pilot programs to test innovative delivery and payment arrangements, but history suggests that most such demonstration projects are abandoned.
Most importantly, the ACA's failure to displace preexisting health care arrangements means that the program will rely on existing sites of public authority to make the policy workable. While the national government has grown dramatically in the Policy State, the US retains a federal system, and the Obama administration was compelled to rely on state governments to handle two vital program-building tasks: expand Medicaid to cover 17 million uninsured Americans and build new public marketplaces, called exchanges, where individuals will go to learn what insurance subsidies they are eligible for and purchase a health plan that satisfies their needs. The states have thus far proven to be ambivalent institutional partners in these efforts.Footnote 81 While the Supreme Court's decision to uphold the individual insurance mandate was a huge victory, the ruling also permits states to opt out of the Medicaid expansion without risk of losing their existing federal Medicaid funding. Initially a larger number of Republican governors said they would decline to expand their Medicaid programs, thereby rejecting the federal funds. Under intense pressure from hospitals in their states, some of these governors later reversed course. But while the Obama administration has been wooing states to participate, there is still a fair number of holdouts. As of September 2013, 22 states have announced they won't expand Medicaid, and debate is ongoing in four other states.Footnote 82 The Obama administration has faced an even greater struggle in getting states to set up the exchanges, which are needed not only to expand coverage but to keep costs affordable. As of September 2013, only 17 states and the District of Columbia have agreed to assume full responsibility for creating their own exchanges, many fewer than anticipated when the law was passed. As a result, the federal government will be forced to run exchanges in many states. It is possible that the larger than expected federal role in managing the exchanges will ultimately strengthen the reform, but the initial launch of the federal exchange website was nothing short of disastrous.Footnote 83 Additionally, Republican states have been finding other ways to undermine implementation by refusing to enforce regulations to protect consumers in the health care marketplace. Some states have also been imposing licensing rules to restrict the activities of the thousands of federally-funded “navigators” that the administration is depending on to assist citizens in enrolling in insurance coverage.Footnote 84
In sum, neither the mass public nor state governments are yet fully incorporated into the ACA's organizational routines, and the feedback production potential of the reform remains largely unrealized. But this does not mean that the ACA will fail. Key stakeholders, including insurers and hospitals, would prefer to move forward with Obamacare than return to the pre-reform system.Footnote 85 The evolving position of Republican governors on the Medicaid expansion signals that state-level lawmakers are starting to feel political pressure to embrace the program. If the exchanges begin to work smoothly and Obamacare supporters can point to citizens who were able to obtain coverage at reasonable cost, and most of the hold-out states eventually embrace the Medicaid expansions, the ACA may yet reconfigure American politics in deeply transformational ways. But it will take several years of arduous administrative work, and most likely a Democratic victory in the 2016 presidential election, before Obamacare's achievements are consolidated.
Dodd-Frank Wall Street Reform and Consumer Protection Act
In the aftermath of the 2008 financial collapse, Congress passed, largely along partisan lines, the Dodd-Frank Wall Street Reform and Consumer Protection Act, a 2,300 page law that its Democratic supporters hoped would prevent future financial crashes from occurring. Those who believe that Dodd-Frank will stick and fundamentally change the way that the financial sector does business point out that when the law passed, an overwhelming majority of voters, including six in ten Republicans, supported tighter regulation of banks and financial institutions. Skeptics counter public support for financial regulation has diminished somewhat since Dodd-Frank's passage and become more polarized over time.Footnote 86 The more general claim is that financial reforms like Dodd-Frank are bound to collapse as politicians and the media inevitably turn their collective attention to other issues and as powerful business interest groups reestablish their “usual dominance” over the implementation and regulatory enforcement process.Footnote 87 In our view, neither an optimistic nor a pessimistic assessment is fully justified based on analysis of Dodd-Frank's feedback generation potential.
We share the pessimists' view that the support of the mass public cannot be counted on to entrench Dodd-Frank, and that Wall Street firms will continue to pursue every advantage. Voters may say they favor tough regulatory measures, yet as Nolan McCarty, Keith T. Poole, and Howard Rosenthal observe in their important book Political Bubbles, public support for reform has been “tempered by concerns about whether government could regulate the financial sector in ways that promote the public interest.”Footnote 88 There are good reasons for this public skepticism. While Dodd-Frank offers Americans the promise of a safer, more resilient financial system—a huge social gain, if realized—its benefits are too diffuse and long-term for voters to develop a reliance interest in the reform's maintenance. More importantly, the reform remains vulnerable to interest-group capture, and on two vital conditions for reform sustainability, Dodd-Frank falls short. The reform failed upon enactment to undercut the institutional base of its opponents nor did it provide regulators with strong state capacity to achieve their objectives. Dodd-Frank is not a detailed policy blueprint, but rather a skeletal framework for crafting new rules to govern a complex sector of the American economy. As Daniel P. Carpenter observes, this pattern of delegation is as much a reflection of the influence of highly trained regulators (who were eager to shape the reform's content) as it was due to a strategic decision by Congress to sacrifice control over policy in exchange for greater administrative expertise.Footnote 89 But the effect of leaving the law flexible was to preserve the power base of the financial industry, which naturally opposed new regulations that would constrain its ability to make potentially profitable, though risky, investments, inviting, as one Washington journalist writes, “continued lobbying from industry groups as they work to bend regulators' interpretation of the law to their preferences.”Footnote 90 Just as the Tax Reform Act of 1986 has been the victim of “death by a thousand cuts,” so Wall Street interests have been seeking to carve out special exceptions to render the overall regulatory framework impotent, with the assistance of sympathetic legislators and courts. Some 488 companies, trade associations, and other groups reported lobbying on Dodd-Frank during the first quarter of 2011.Footnote 91 The result has been to slow the rule-making process to a crawl. A case in point has been the struggle over the long-delayed Volcker rule, which bans federally insured banks from proprietary trading. As of September 3, 2013, a total of 280 Dodd-Frank rulemaking requirement deadlines have passed; 61 percent have been missed and only 39 percent have been met with finalized rules.Footnote 92
As we noted, new institutions are often layered atop old ones in the Policy State. Congress gave regulatory authority to a myriad of agencies rather than to a centralized financial-sector oversight authority. The inevitable result has been turf battles between bureaucracies with conflicting organizational values and missions, such as the ongoing tension between the Securities and Exchange Commission and the Commodity Futures Trading Commission. Barney Frank (D-Massachusetts) had wanted to consolidate the two agencies to create a more effective regulatory structure, but he was forced to compromise, because “the farm lobby's congressional allies were eager to maintain the CFTC's historic role overseeing agricultural commodities.”Footnote 93 Meanwhile, other financial watchdog agencies are struggling to carry out their greatly expanded responsibilities under Dodd-Frank because Republicans in Congress have refused to significantly boost the agencies' appropriations, a strategy for weakening enforcement capacity that is largely invisible to the public.Footnote 94
Dodd-Frank does seek to build new state capacity through the creation of the Consumer Financial Protection Bureau. The brainchild of (now Senator) Elizabeth Warren, the CFPB's mission is to protect consumers from deceptive financial services. To promote the agency's autonomy, the bureau receives its budget directly from the Federal Reserve, and has only a single Senate-confirmable director. Republicans succeeded in blocking Warren from becoming director, and continue to look for ways to weaken the agency's power and independence. Partisan skirmishing over its operations will continue.Footnote 95 Still, if the CFPB can become institutionalized, it has the chance to build an organizational reputation as a trusted consumer advocate, and gain power over time.Footnote 96
What of the financial industry's stake in preserving Dodd-Frank? The industry has already spent huge amounts on legal compliance costs. The key from a policy feedback perspective, however, is whether the reform has forced Wall Street firms to organizationally adapt during this post-enactment stage in ways that will be costly for them to reverse, giving the industry a stake in the financial regulations.Footnote 97 Successful economic reforms typically become self-reinforcing by weeding out firms unable to adjust to the new system. For example, airline deregulation led carriers to alter their internal governance structures and business strategies; legacy carriers that couldn't adapt to the harsh realities of market competition either merged with other airlines or went under.Footnote 98 Will Dodd-Frank stimulate similar adaptations in the financial industry by destroying firms that cannot play by the new set of rules and making it impossible for the industry to revert to its pre-reform ways? On the one hand, big banks are developing strategies for separating trading offices and derivative operations from their core banking services in order to comply with the reform. Yet on Capitol Hill, bills have been introduced to roll back the requirement for such separately capitalized subsidiaries, and there is little evidence that Dodd-Frank is transforming the core organizational culture of the financial services industry. In a 2013 survey of 250 financial industry insiders, 23 percent reported that they had seen or had direct knowledge of wrongdoing in the workplace, and 24 percent admitted they would engage in insider trading if they could get away with it.Footnote 99 While Dodd-Frank may remain on the statue books, it remains unclear whether it will reconfigure firms' organizational identities or business models in the same way that airline deregulation accomplished. What the financial services industry appears to want is neither the complete preservation or complete repeal of Dodd-Frank (which would create a more uncertain political environment), but rather “a diluted financial reform law that would relax restrictions on some of its most profitable—and riskiest—investments but maintain enough government oversight to give the banks cover.”Footnote 100 Wall Street may yet get its wish.
Lessons and Questions for Future Research
To shed substantive and theoretical light on the predicament of contemporary liberalism, we have examined the hurdles to remaking politics through policy reform. If Obamacare and Dodd-Frank are to realize their full transformative potential, President Obama will need to devote a significant amount of his second term political capital to safeguarding the legislative accomplishments of his first. The burden of our argument has been that President Obama's difficult leadership situation reflects not just the snags and setbacks that are to be expected whenever an administration seeks to implement an ambitious agenda, but also the forbidding terrain of the Policy State. Given the dense infrastructure of existing public policies, agencies, and organized interest groups in Washington, the project of sustaining reforms that can deliver enduring value to citizens has become enormously challenging.
Heavily implicated in the struggles over governance in the Policy State is the polarization of the two parties. Partisan polarization has made it extraordinarily difficult to achieve bipartisan consensus on any major reform. When major legislation passes, it is often through total party discipline. To be sure, partisan enactment cuts two ways. The majority party might be more able to use deck-stacking strategies to immunize its victory from future administrative assault. Yet partisan enactments also come at the cost of achieving a durable elite consensus on the direction of policy change. Gone are the days when a Republican like Senator Everett Dirksen and a Democratic President like Lyndon Johnson would make a deal that both were willing to defend. What may be even more important than party-line roll call votes has been the growing polarization of the implementation process. Opponents are dug into their positions, harder to move even when new programs benefit their constituents. The fights over governance intensify, the risks to sustainable reform increase. As we saw, Social Security's early programmatic weaknesses were repaired by a unified Democratic Congress in 1939 and 1950. Yet what allowed these changes to stick was that President Dwight Eisenhower gave the program his blessing in the early 1950s. A key question, as Sidney M. Milkis observes, is whether party politics have been so thoroughly polarized that Obama will be deprived of a successor—like Eisenhower—who “bestows bipartisan legitimacy on the changes he brings about.”Footnote 101
At a more theoretical level, our intention has not been to discredit arguments based on positive feedback, but rather to place such arguments on a more solid foundation. In their fine recent book A Republic of Statutes, Eskridge and Ferejohn demonstrate that laws can become so deeply rooted as to become constitutional in nature.Footnote 102 Our analysis begins to problematize, and open the black box around, the entrenchment process. By probing the conditions under which feedback and self-reinforcement do not occur, we gain more insight into the special circumstances under which they do. Our analysis deliberately steered away from the language of “necessary and sufficient” conditions for entrenchment, because we assume probabilistic causation. The configuration of feedbacks influences the susceptibility of a reform to downstream reversal. It loads the dice. But it does not determine outcomes because the fate of reforms is never completely settled, even if some pathways are largely foreclosed.
And yet, identifiable factors clearly shape reform trajectories in discernible ways. Institutions matter because they shape the credibility of the political commitments underlying reforms. They also provide the state capacities needed to make reforms work. Resource and interpretive effects matter because they affect constituency support for the extension of a reform's line of policy development. Rather than asking whether institutions or interests are primary, we believe it is more productive to examine their interaction effects, particularly over the key postenactment phase.
We also highlight the role of both agency and structure in the feedback process.Footnote 103 Supporters and opponents may not take the feedback effects of preexisting policies as given, but may instead actively seek to amplify or suppress such effects, to the extent feasible within institutional constraints. Yet while battles over feedback have become central to democratic deliberation, state-building, and party politics, more research is needed. One issue that requires greater study is Congress's role as feedback-shaper. To what extent are lawmakers and congressional staff aware of the feedback implications of various bills? When do oversight hearings engage the feedback effects of legislation, and how often does a concern about anemic feedback influence debates over reauthorization bills and other amendments? A second issue is the conditions under which opponents can successfully poison citizens' perceptions and beliefs about the effects of a new policy through misinformation campaigns. When are fraudulent efforts to discredit a policy persuasive to the public, and under what conditions are such attempts unable to prevent a policy from building a constituency? Is the impact of misinformation a function of the complexity of the policy or the resources of the opposition? Finally, the cross-national implications of our arguments merit attention. Does the fragmentation of the American state make public policy in the United States particularly prone to weak feedback? How important are wider forces such as globalization and rising income inequality in undermining the capacity of a given state to enact policies that reconfigure power relations?
A final issue concerns the normative implications of policy entrenchment. Here we have taken it for granted that entrenchment is a legitimate goal in a democratic polity, especially when the entrenched policy would serve a diffuse or disadvantaged constituency. The crafting of such policies is central to the liberal reform project and the case for generating positive feedback in order to help insulate reform commitments from the pressures of selfish interests and ravaging majorities seems obvious. And yet, tensions can emerge between the need to make inherited policies harder to undo, and the need to maintain the flexibility required for open debate about the role of the state in light of changing conditions. The struggle to remake politics through policy entrenchment is a struggle not only over the appropriate size and scope of government, but over the legitimate boundaries of democratic deliberation itself.
