Abstract
A large body of research has explored how policies, once enacted, reshape public opinion, governing institutions, and political organizations—a process known as “policy feedback.” Yet this productive research agenda has yet to be translated into practical recommendations of the sort regularly provided by other social science research. This volume of The ANNALS presents the findings of a major collective effort to do just this. The Policy Feedback Project (PFP) is an effort to develop research-backed arguments about how policy feedback might be harnessed to address collective problems in today’s age of partisan polarization and economic inequality. This article orients readers to our collective approach and summarizes some of the contributing authors’ findings. In particular, we show how the feedback effects of policies could be used to (1) tackle long-standing public problems that have resisted effective responses, (2) increase the long-term durability of policy initiatives designed to address these problems, and (3) build political momentum and power to facilitate the adaptation and expansion of these initiatives over time.
The articles in this special issue all emerged out of the Policy Feedback Project (PFP), a unique effort to assist policy-makers seeking to revitalize American democracy and improve public policy. The project was founded with the goal of deepening our understanding of the ways in which policies, once enacted, reshape public opinion, governing institutions, and political organizations—a major area of research in political science that has rarely been translated into practical recommendations or policy advice.
In particular, we asked participants in the PFP—all social scientists with extensive policy expertise—to develop research-backed arguments about the ways in which such “policy feedback” might be harnessed to (1) tackle long-standing public problems that have resisted effective responses; (2) increase the long-term durability of major policy initiatives designed to address these problems; and (3) create the potential to build political momentum and power to facilitate continuing efforts to adapt, expand, and improve these initiatives over time. In short, we asked authors to think about how policies might reshape not just American economic and social outcomes but also American politics—in ways that encourage continued efforts to address pressing public problems.
In this introductory article, we first outline the core ideas and aims guiding the PFP. We then discuss how the insights of prior research needed to be updated, given the current political context. Finally, we use this updated understanding to articulate some basic guideposts for policy design, drawing mainly on the articles in this volume that provide general lessons. We also highlight, by way of illustration, a few of the insights that come from the volume’s more policy-oriented articles, which offer tailored recommendations for generating positive feedback in key policy areas.
The Policy Feedback Project
Leaders and experts have always considered the political consequences of their policy choices. Yet this thinking has too often been an afterthought. Faced by a pressing problem, the questions usually asked are “Can a law be enacted?” and “Will it effectively address the problem?” As crucial as these questions are, however, another may be just as important: “Will this policy create positive political effects—that is, will it encourage ongoing and, ideally, increasing efforts to address the problem?”
As already mentioned, a substantial and growing body of research has addressed this question, albeit in general and analytic ways. For more than a generation, political scientists and other social scientists have intensively studied how major public policies reshape politics. As a result, we know far more than we once did about the systematic political consequences that policies have and what features of those policies and their political context tend to create one type of effect versus another. Nonetheless, we still have limited ability to translate these findings into informed advice of the sort regularly on offer in economics, decision science, public finance, and the like. This is both because these findings have not generally been translated into applied knowledge and because they have not been updated to reflect important recent changes in American politics. Indeed, only a handful of articles or books that examine the feedback effects of public policies say anything at all about their implications for contemporary policy design, and those that do provide very limited recommendations. For example, Suzanne Mettler’s agenda-setting 2018 book, The Government-Citizen Disconnect, devotes five chapters to exploring how American public opinion is shaped by the “submerged” design of U.S. social and educational policies—that is, their reliance on tax expenditures and other indirect forms of public support, rather than direct spending. By comparison, she spends just two pages exploring possible responses, which mostly center on the general recommendation to make policies more visible to citizens.
The need for such application and updating is more pressing than ever. In a highly complex and interdependent society experiencing often rapid and disruptive economic and social change, there is a need to harness public authority to meet collective demands and to adapt prior policy responses to changing circumstances. Yet the American political system is unusually prone to gridlock, and the emergence of intense partisan polarization has made that gridlock even more pervasive. Moreover, the rise of “negative” polarization—an orientation of hostility toward the other side in partisan debates—means not only that it is harder to reach initial consensus, but also that initiatives championed by one side are vulnerable to dismantling or sabotage by the other. Given these circumstances, understanding what allows initial policies to become firmly rooted and then adapt and improve over time is a pressing public matter.
The goal of the PFP is to provide that understanding. Launched in 2017, under the auspices of the Scholars Strategy Network—a national organization of scholars dedicated to offering research-backed advice to policy-makers—the project brought together two groups: social scientists who have studied policy feedback in the past and policy professionals who would like to improve their capacity to make public policies with their potential political consequences in mind. We asked the participating scholars to distill prior research on policy feedback into basic lessons, design principles, and policy recommendations that might help to guide the structure and sequencing of new policy initiatives. In turn, the policy professionals provided their reactions to these arguments and proposals, with an eye toward explaining the general organizational, electoral, and political challenges of contemporary policymaking that the authors may not have considered.
Thus, the authors were asked to focus specifically on how policy might remake American politics, not on proposals to change political institutions directly (for example, revision of legislative procedure, election rules, or campaign finance law). The political effects of new policies are the particular focus of research on policy feedback, and such effects have received much less attention in current debates than have proposals for political reform. As all the authors show, moreover, policies are themselves constitutive of the current political landscape, and hence policy changes can themselves be powerful sources of political reform.
From the outset, the project was designed to involve scholars with diverse viewpoints. However, it began with two premises that inevitably shape the prescriptions on offer. First, the purpose was to think about how government could actively address a set of fundamental challenges, including climate change and the strains facing our health care and criminal justice systems. These policy areas were chosen not only because they are associated with significant costs and risks to our society, but also because they are areas where we have a body of prior research on the political effects of policy that can guide analysis. We make no claim that these are the most important policy challenges we face, though we believe all are extremely pressing. Nor did we choose them without reference to the character of the proposed responses that currently dominate debate. We chose them because they are areas where we have a significant body of research that can help guide the design of active government policies.
Second, a motivating observation was that critics of such active policies have themselves proved adept at using policy to achieve their own preferred ends—often at the state level. For example, a significant body of research has shown that state policies to ban collective bargaining have had profound effects, not just on levels of unionization, but also on politics within the affected states. A similar story can be told with regard to efforts to cut taxes aggressively even in the face of significant revenue needs—policies that have changed the politics of public finance in ways that make raising taxes or undertaking new initiatives not just economically difficult but politically challenging. A starting goal of the project, therefore, was to think about whether policies with more broadly “progressive” goals might also create such feedback effects.
Policy Feedback and American Governance
Over the last generation, social scientists have developed a sophisticated portrait of a reality that skilled politicians have always recognized: major public policies, such as Social Security, can have substantial political impacts, engendering support that helps those policies to endure. Well-crafted policies do not just address societal problems; they create supportive political dynamics, whether by empowering backers, weakening opponents, making citizens’ views of government more favorable, or reducing obstacles to further reform efforts. Drawing on the political scientist E.E. Schattschneider’s (1935) seminal insight that “a new policy creates a new politics,” social scientists refer to the ways in which policy changes politics as “policy feedback” (Weir and Skocpol 1985; Pierson 1993; Campbell 2002).
Politicians, of course, have long crafted policies with their political consequences in mind. Perhaps the most famous example is FDR’s defense of the system of contributory payroll taxes that was a foundation for Social Security. Challenged by an economist on this arrangement, FDR acknowledged,
They are politics all the way through. We put those payroll contributions there so as to give the contributors a legal, moral, and political right to collect their pensions and their unemployment benefits. With those taxes in there, no damn politician can ever scrap my social security program. Those taxes aren’t a matter of economics, they’re straight politics. (Schlesinger 1958, 308)
This type of political engineering was at the heart of the New Deal. “Tax and spend” eventually became an epithet, but it was originally FDR advisor Harry Hopkins’s well-justified boast. Indeed, the full quote was “tax and tax, spend and spend, elect and elect.”
Many of the Democratic achievements of the New Deal, Fair Deal, and Great Society bore out Hopkins’s formula. Policies that began as ambitious and risky undertakings evolved to achieve the status of political “third rails.” Social Security expanded from a small program covering less than half the population into a universal guarantee of retirement security, adding survivors’ and disability insurance along the way. Condemned as “socialized medicine” in the early 1960s, Medicare was untouchable—if not a platform for universal insurance, as its designers hoped—by the early 1970s. In area after area, federal policies that once sparked controversy came to be taken for granted. Social policies expanded, almost never experiencing major retrenchment. New levers of national economic policy, once deployed, were rarely if ever abandoned. Environmental and consumer protections grew, if often through fits and starts. The long arc of history appeared to bend toward successful policy feedback that supported an active “Big Government.”
In recent years, however, advocates of active government have seemed less capable of generating such self-reinforcing effects. President Obama, for example, declared in his second inaugural address, “Progress does not compel us to settle centuries-long debates about the role of government for all time, but it does require us to act in our time” (Obama 2013). Yet while the Obama administration and its allies achieved notable policy victories in 2009 and 2010, their record at designing initiatives that met political as well as policy goals was uneven. Some policies never made it to implementation, such as the long-term insurance program embedded in the 2010 Affordable Care Act (ACA; aka “Obamacare”)—a provision that was abandoned as unworkable without political complaint. But even when policies were successful, the hand of government was often invisible, sometimes intentionally so, undercutting the prospects for assembling and sustaining strong support coalitions.
Consider the large cut in payroll taxes in the 2009 American Recovery and Reinvestment Act. To boost consumer spending most effectively, it was carefully designed to attract no notice (Cooper 2010). Similarly, the Dodd-Frank Wall Street Reform and Consumer Protection Act has not become accepted within corporate circles, and few Americans appear to have grasped its favorable consequences for them. And while the ACA is now producing modestly positive feedback effects, these were perilously slow in coming. Early on, ACA beneficiaries neither reliably perceived the benefits they received nor consistently voted or acted to protect the law (L. Jacobs and Mettler 2016). Worse, the ACA catalyzed a political backlash that not only exacted considerable electoral costs but also left the program struggling to survive.
Setbacks of this sort have prompted much handwringing among these policies’ supporters. To date, however, there has been insufficient systematic thinking about which design choices were fundamental and whether alternative choices would have made a difference. More important, there has been little effort to think about how future efforts to address these and similar problems—through legislation, budgets, and executive powers, as well as actions at the state and local levels—could be designed not just to advance key policy goals but to do so in a way that enhances the prospects that reforms, once in place, will endure and expand. Such thinking requires much more than the use of polls or focus groups to assess the immediate popularity of policies, and it requires much more than the cursory consideration of add-ons that might marginally boost the endurance of policies designed solely around technocratic goals. It entails an early and intensive consideration of the multiple ways in which “a new policy makes a new politics.”
The articles in this special issue of The ANNALS offer such consideration. They review past research on the political consequences of public policies—including, in all cases, research done by the authors themselves—and distill these findings into actionable, straightforward advice. In addition, each article seeks to update prior findings to reflect America’s changing politics—above all, the emergence of intense polarization. Feedback research has focused on historical periods in which partisan cleavages were far less prominent than they are today. Absent careful reconsideration, these traditional assessments will provide a poor guide for today’s policy-makers.
The articles in this issue can be divided into two broad types. Those that lead the issue focus on general recommendations—for example, how to reduce the political backlash that new initiatives inevitably generate. The articles that follow focus on policy-specific recommendations. As noted, these policy-oriented articles look at important areas of policy where there has been ample research about policy feedback in the past: health care, climate change, criminal justice reform, and the promotion of better jobs (with a particular focus on regional inequalities and the challenge of ensuring adequate employment in all communities). For each of these areas, there are two articles that present complementary perspectives, whether covering different levels of government, different specific policies, or different aspects of the relevant policies and their politics.
The remainder of this article proceeds as follows. We first discuss the main ways in which policy feedback research must be updated in light of changes in American politics. We then examine the implications of an updated perspective—first for the general problem of harnessing government to address pressing public challenges in the twenty-first century, and then for policy design in specific domains.
Updating Theories of Policy Feedback
Partisan polarization has upended many of our understandings of how American politics works, and processes of policy feedback are no exception. Indeed, it is not too much of an exaggeration to say that polarization and its accompanying changes have shifted us from one political world to another.
In the old world, party attachments were relatively weak, and both parties were cleaved by cross-cutting issues and commitments. In this context, a well-designed program could build support among the extensive groups of voters, organized interests, and opposition party politicians who were politically “available.” In turn, this support could ensure a policy’s survival irrespective of a party’s immediate electoral fortunes and eventually provide a political foundation for future program growth. For example, Medicare passed in 1965 after a decade of coalition building by its advocates, slowly picking up organized supporters and legislative sponsors as it moved closer to enactment. In its final form, it not only included elements recommended by Republicans, but also gained substantial bipartisan support, with seventeen Republicans backing it in the Senate. Once adopted, the program quickly consolidated support: Medicare’s popularity rapidly induced most who had initially opposed it to climb on board.
In the new world of hyperpolarization, none of these assumptions can be made. Growing tribalism—the tendency for voters and groups to strongly and almost automatically back one party or the other, coupled with increased animosity toward the other party—has meant that many constituencies once up for grabs are no longer free-floating. It is not just that voters cannot be convinced to vote for the other party; increasingly, it is hard to persuade them to view the other party’s proposals sympathetically on any prominent issue. This has made it much more difficult to gain broad acceptance for new initiatives that might traditionally have been expected to have strong positive feedback effects. In addition, there is now abundant evidence that general public opinion only weakly constrains opposition policy-makers. Their greatest risk is offending their “base,” rather than swing voters or the public as a whole. For example, not a single House Republican voted for the American Recovery and Reinvestment Act of 2009 (aka the “stimulus”) at a time that the U.S. economy was in freefall and even though President Obama had made the centerpiece of the law a payroll tax cut.
The new world of American politics is not only politically divided; it is also much more economically unequal (Hacker and Pierson 2010). Economic inequality has compounded political polarization in at least two ways. First, it has weakened the sway of ordinary voters, as opposed to the very affluent and deep-pocketed political organizations. This has made general public opinion even less important as a source of policy durability and expansion. Second, these inequalities have tipped the scales further against new initiatives in many important areas of policy by heightening the blocking power of concentrated interests. Although we focus mostly on polarization in this section, the compounding effects of rising inequality must be kept in mind.
Polarization changes everything
Polarization generates problems for collective problem-solving at all levels of our politics. Parties as organizations increasingly appear to be locked in a zero-sum contest for political dominance. This gives everyone involved—elected and appointed public officials, allied interest groups, activists, and voters—increased incentive to resist initiatives launched by the other side. Case studies of recent policy battles yield a consistent and troubling message: growing polarization has made the task of addressing collective problems vastly more difficult.
Jon Oberlander and Kent Weaver’s (2015, 56–57) assessment of the ACA’s reception is representative:
Partisan polarization explains much of the ACA’s enduring political problems, as partisans framed debate around ACA’s negative impacts, real and imagined, emphasized loss-imposing policy consequences rather than the law’s beneficial impacts, and resisted the ACA’s implementation at the state and federal level.
Theda Skocpol’s (2013) postmortem of the Democrats’ failed 2009 “cap-and-trade” legislation to address climate change carries a similar message. Recognizing the Senate as the biggest obstacle to reform, strategists relied on a traditional policymaking logic: leverage the diversity of views and interests within the Republican caucus to build a strong “middle-out” coalition. The goal was to win the support of perhaps a half-dozen GOP senators, providing both needed votes and the coveted branding of bipartisanship.
Needless to say, the gambit failed. Pursuing a tactic that has become increasingly prevalent and effective, powerful interests within and outside the GOP (including right-wing media) countered by intensifying partisan divisions. They defined climate legislation as a Democratic project and effectively marketed this framing to Republican voters. Republican leaders recognized that any bipartisan deal would be politically costly. Along with allied interests, they worked to deter Republican senators from backing anything with Democratic support, no matter how riddled with concessions those Democratic initiatives might be.
Under current conditions, then, the “old world” strategy of middle-out coalitions that rest on bipartisan buy-in holds limited promise, at least at the national level. Even policy designs that contain strong incentives to climb aboard (such as the ACA), and which might have found consolidation relatively easy a generation ago, have encountered fierce partisan resistance. Zero-sum partisan contestation threatens to swamp everything.
It is worth specifying precisely what has changed. The backdrop for the old strategic playbook was a “Madisonian” system that promoted a proliferation of distinct political contexts and interests, which in turn induced cross-cutting cleavages and pluralism. Adages like “where you stand depends on where you sit” or Tip O’Neil’s “all politics is local” were broadly accurate portrayals of a political system in which coalitions were loose and fluid, reflecting both national partisan battles and the pull of geographic, sectoral, and cultural constituencies that divided as well as unified partisans.
Over the past few decades, however, this fluidity has given way to an increasingly entrenched partisan divide. In part because of the massive growth of policy activity in Washington, major aspects of American politics—from media environments and interest group structures to party organizations and even state-specific policy battles—have nationalized (Pierson and Schickler 2019). These shifts have created a new kind of partisanship: more cohesive, more consistent, more homogenous, more all-encompassing, and much, much more intense.
These trends have been largely driven by the shifting incentives and activities of political elites. (By “elites,” we mean those in and around government who have substantial capacity to shape politics and policy: elected and appointed public officials, interest group and activist leaders, prominent figures in the media, and so on.) Individual politicians, beholden to national organizations, are less free to defect on issues based on local conditions. Interest groups have also increasingly chosen sides. They too operate on a national level, often see one party or the other as far more responsive to their demands, and face incentives to offer political loyalty in return for policy favors. But even voters—relatively late and sometimes reluctant enlistees in the polarization wars—are increasingly attached to national partisan identities and outlooks. Crucially for discussions of feedback, they are also increasingly antagonistic toward politicians, groups, and citizens on the other side of the aisle.
Mass and elite polarization generally work in tandem—for example, the polarization of voters makes it harder for politicians to moderate, and the strategies of elite organization often hinge on creating clear “litmus tests” for pivotal constituencies. Still, it is worth separately taking up the mass and elite sides of the equation before returning to how they simultaneously shape feedback dynamics.
Mass polarization
A growing body of research demonstrates that polarization makes it increasingly difficult to win over voters through policy initiatives if they are not already aligned with a party. Sadly, this remains true even where these voters might benefit significantly from proposed initiatives. This development is especially important to recognize, since analyses of policy feedback have often placed emphasis on the prospects for building support among ordinary citizens who, over time, benefit from and thus become attached to a policy regardless of its partisan origins.
Voters, we know, often have limited awareness of programs. A 1994 poll found, for example, that voters believed the two largest items in the federal budget were “welfare” and “foreign aid”—the former relatively small in budgetary terms (depending on exactly how it was defined), the latter tiny. By contrast, only 15 percent of respondents correctly named Social Security (Kaiser Family Foundation 1994). Yet in our sharply polarized politics, signals or incentives connected to the programs themselves have a much harder time breaking through. Party labels and identities play more and more of a role in how voters interpret even relatively unmistakable realties. Specific policies seem to matter less and less in shaping voters’ attitudes and political behavior. While policies do send signals, how those policies are interpreted seems to have changed. Analysts today speak of “negative” or “affective” partisanship, and generally find it is stronger on the GOP side of the aisle (see Abramowitz and Webster 2018). All it may take for “red” Republican voters to recoil is awareness that something is a priority of “blue” Democratic elites. For instance, there is now strong evidence that Republican voters who are primed to link the program to Obama are less likely to sign up for insurance through the ACA (Lerman, Sadin, and Trachtman 2017).
Moreover, many of these voters live in cultural environments where conservative politicians and media dominate political discourse. We know that voters rely heavily on elite cues to form judgments about policies (Zaller 1992; Gilens and Murakawa 2002). Unfortunately, in our polarized climate Republican elites and conservative media have powerful incentives to demonize policy initiatives offered by Democrats. Even if mass polarization is less deep than many assume, it is increasingly overlaid onto geographic divisions between cities and rural areas, thriving metro centers and declining industrial towns, and red and blue states. Taking their cues from oppositional elites and plied with hostile framing in the media, today’s more partisan voter may be highly resistant to new initiatives—before and even after the actual effects of those programs materialize.
Elite polarization
At the elite level, growing separation between the parties makes it much more difficult to pursue a standard bandwagon strategy, passing and then consolidating policy by peeling off the more moderate or cross-pressured politicians on the other side. There is ample evidence, moreover, that the problem is particularly acute with regard to the Republican Party. First, most measures of ideological divergence suggest that Republican candidates and officeholders have moved farther right than Democratic elites have moved left (Hacker and Pierson 2015). As a result, there are few if any truly moderate Republicans currently in national elected office—a stark change from a generation ago, when the party was marked by influential northeastern, western, and midwestern moderate factions.
Second, the GOP has been distinctly shaped by the party’s increasingly strong performance in rural areas and sparsely populated states, even as Democrats have come to dominate urban regions. Not only does this nonurban edge give Republicans an advantage in the Senate, with its granting of two senators for every state (an advantage that partly carries over to the Electoral College). In addition, it provides a substantial advantage in House elections, since a large share of aggregate Democratic votes are “wasted” in overwhelmingly blue urban districts. The result is that Republicans can capture a larger share of House seats than their share of the overall vote in House elections (an advantage accentuated by partisan gerrymandering, which Republicans, with their outsized strength at the state level, have had greater opportunity to pursue.)
Republicans’ sharp movement to the right and their tendency to run for office in safe states or districts—as well as the nature of the interest-group environment discussed next—create very distinct incentives for GOP officeholders. Almost everywhere, GOP politicians find the potential of challenges from their right flank to be their greatest electoral danger. Persuading even potentially reachable politicians to compromise, always difficult, is nearly impossible when they fear well-funded retaliation from their own copartisans.
Another elite-level aspect of polarization concerns interest groups. Achieving positive policy feedback is partly a matter of building a supportive network of organized interests and weakening those groups that constitute powerful bases of resistance. Just as with voters and politicians, however, fewer and fewer groups are free-floating and thus readily available for coalition building. Instead, more and more interests are closely aligned with a party and see their long-term interests as wedded to these alliances.
Major business groups, for example, have become more aligned with the GOP, despite the diversity of perspectives among business leaders themselves (Grumbach and Pierson 2019). Since the 1990s, the biggest corporate organizations—including the Chamber of Commerce and the network of organizations created by billionaire business leaders Charles and David Koch—have simultaneously grown in size and resources and become sharply aligned with the Republican Party (on the Kochs, see Skocpol and Hertel-Fernandez 2016). Since organization is usually essential for influence, this increased partisan tribalism exacts a high potential toll on government initiatives that once would have been expected to elicit active support (as opposed to tacit acceptance) from at least segments of the business community.
Again, the tightening bonds of partisanship make it more difficult to find allies—even for initiatives those groups might be willing or even eager to support in a different partisan climate. Like moderate politicians, interest groups are likely to fear retaliation if they do not stick with their team. They may also calculate that in the long run they suffer more if they do anything that hurts the electoral prospects of their side even when backing specific policies might be beneficial. As with voters and politicians, the search for “middle-out” coalitions based on shared interests may prove fruitless.
These new polarizing dynamics are game-changers for all the elements of American politics that traditionally created opportunities for policy feedback, from the basic foundations of mass political behavior to the commanding heights of elite influence. In the next section, we consider their major implications.
Implications and a Guide to Reading the Articles in This Volume
The contributors to this special volume were asked to focus on three main aspects of policy feedback. These concern what Jacob Hacker, in his article on Medicare expansion in this issue, calls the “three E’s”: “establishment,” “entrenchment,” and “expansion.”
Establishment. The central issue here is how to minimize and manage what Eric Patashnik, in his article, calls “backlash”—that is, mobilization against a policy during and after its passage that undermines the power of its backers and limits the ability to embed and expand it. Overcoming backlash is not just essential for enactment, since backlash may emerge even before a policy’s full effects are felt. It is also a precondition for the establishment of policies as going concerns—political “facts on the ground” that opponents must confront, even if they wish those facts were otherwise.
Entrenchment. The question here is what policy designs are most likely to cause policies to “stick,” making backsliding less likely. Entrenchment is a process, not a fixed quantity. It occurs when policies strengthen supporters; weaken opponents; create investments by citizens and interest groups; and ultimately swing elites, groups, and the public toward higher and higher levels of support for a policy over time.
Expansion. In addition to establishment and entrenchment, those who design policies often seek their future expansion. The question here is how a policy can create self-reinforcing expansionary dynamics, in which various political actors, groups, and constituencies come to see the updating and expansion of a policy as in their interest, helping to create momentum for further rounds of policy improvement. Entrenchment can occur without expansion—witness the difficulty in expanding Medicare beyond its initial target populations. But since policies require adaptation to changing circumstances, those who design policies hope to generate feedback effects that do not just protect against backsliding, but also facilitate ongoing upgrades.
As noted, these questions were approached by two sets of authors, with each pursuing related but distinct goals. The first set was asked to focus on broad lessons. The second was asked to look at design issues specific to key policy areas. In the remainder of this section, we briefly summarize some of the key findings and recommendations of these complementary analyses, emphasizing the first set’s broader conclusions.
The difficulty of mobilizing public opinion through policy success
We begin with a cautionary point. Perhaps the dominant thread of past research on policy feedback traces the capacity of programs to build mass support over time (Campbell 2012). The argument is straightforward: once beneficiaries see concrete benefits from programs, they become a bulwark against backlash or retrenchment. There is no doubt that this has often been a critical dynamic in the past (it lies behind the famous Harry Hopkins quote), and it certainly remains relevant today. Yet while policy designers obviously should do what they can to build such support, a number of contemporary factors point to the difficulty of anchoring policy sustainability in this way.
To begin with, it is simply hard to build popular support across party lines in a context of negative partisanship. Voters outside a party’s initial coalition are less likely to be open to positive interpretations of new policies. Moreover, even when a policy benefits them, they are likely to receive very negative cues from conservative elites and media.
An additional difficulty of relying on public opinion is that even if efforts to build mass support have some success, this may have only limited impact on the depth of opposition. As noted, mass opinion seems to have a much more modest effect on the political behavior of elected officials than commonly believed. Legislators may be more influenced by powerful concentrated interests—a point we discuss in more detail below.
Based on the partisan asymmetries that we have just discussed, this may be particularly true for the Republican Party. GOP elites may make the reasonable assessment that they can count on the continuing force of negative partisanship to ensure the unwavering support of their “base” irrespective of the party’s position on many issues. Both these dynamics were on stark display in 2017 as the Republican majority in Congress came up just short of passing legislation to repeal the ACA (along with introducing other sharp cuts in Medicaid) despite abysmal polling numbers for their proposals. As we have argued elsewhere (Hacker and Pierson 2018), the fact that such historically unpopular repeal legislation could come within a hair’s breadth of passing is indicative of the diminishing hold of public opinion on legislators.
Another obstacle to relying on a wave of supportive public opinion to entrench policies is the sharp drop of trust in government over the past generation. Many of the classic examples of positive feedback, such as Social Security and the GI Bill, date from an era when government and public programs were held in much higher esteem. Today, policy-makers face a dilemma, painfully documented in the work of Suzanne Mettler (Mettler 2011, 2018): more and more voters are loath to give government credit for even highly valued benefits. One consequence is that policy designers often “submerge” the role of government to circumvent distrust. In doing so, however, they undermine opportunities to demonstrate the value of government initiatives for citizens, feeding a vicious cycle. In her contribution to this issue, Mettler offers some guidance on how to limit this dynamic. Still, relying on public opinion to institutionalize a new program is unlikely to be a sufficient strategy in most domains.
The need to focus on the interest-group environment
If strategies focused on public opinion now seem less promising, the domain of interest groups may represent a more fertile terrain for positive policy feedback. Organized groups with the capacity to deliver money and votes are critical political players. Policies, along with their political advocates, rise and fall in part because of the ways in which they bolster organized supporters and undermine organized opposition. As Alex Hertel-Fernandez’s contribution to this volume stresses, the most important recent conservative experiments with policy feedback reflect this understanding. Conservative policy elites have rarely been focused on increasing public support through broadly popular initiatives. Instead, they have aimed directly at shifting the balance of organized power and electoral strength in American politics, often focusing on domains of political contestation (state governments, the courts, regulatory policy), where the independent capacities of public authorities and opposition to conservative activities are both relatively weak.
As Hertel-Fernandez’s (2019) work highlights, recent trends have heightened the significance of organized interests in policy development. Rising economic inequality has fueled rising resource inequality among political organizations. In particular, the organized power of the wealthy and large corporations has grown, while that of labor unions has declined dramatically. In many policy areas, including health care and the environment, the entrenched position of concentrated interests who derive big benefits from the status quo has become a formidable obstacle to durable reform.
Policy-makers should pay very careful attention to the likely impact of design choices on interest group environments. Groups are the political actors most intensely focused on, and responsive to, policy. And while polarization has increased the tendency of groups to “pick a team,” they are still the political actors most likely to realign their efforts in response to policy developments. Because policies deliver substantial benefits and often powerfully shift incentives, groups can potentially be brought into a coalition—or detached from the other side’s coalition.
Thus, for any significant initiative, policy-makers need to consider whether design choices will alienate or incorporate politically important groups. Although it may be tempting to think that mobilizing group support only really matters for getting a policy passed, the research on policy feedback suggests it is vital to approach this question in a dynamic fashion—to think about establishment, entrenchment, and expansion, as well as enactment (Pierson 1993; Patashnik 2008; A. Jacobs and Weaver 2015). Even if a group may not initially support an initiative, it is crucial to think about how to design the policy to encourage agnostic and sometimes even antagonistic groups to “invest” in and adapt to the new program once it is enacted.
At least as important, well-designed policies can have a substantial effect on which kinds of groups thrive or fail. That is, they can change the balance and character of groups contesting a policy. Indeed, policies can even shape what these contesting groups will want in the future. Large-scale government initiatives encourage specific investments by groups—for example, in particular types of technology or the provision of goods and services financed by the public sector—that reshape not just their sense of the possible, but also their priorities.
In short, those seeking to harness government to address collective problems need to be attentive to opportunities to design policy so that it strengthens already supportive groups or induces previously neutral or skeptical groups to reassess their interests once policies are enacted. This type of advance thinking is especially important in the (many) issue areas, such as health care and climate policy, where there are powerful concentrated interests with a large stake in policy design. As noted, concentrated groups who benefit from the status quo have grown stronger in recent decades as rising economic inequality has spilled over into politics.
The need to take such groups seriously is the major message, for instance, of the two articles on climate change by Jonas Meckling and Samuel Trachtman. Both emphasize that even though science suggests bold action by governments is required now, advocates of such action cannot assume away the fierce opposition of extractive and carbon-intensive industries (including labor unions within these sectors). Instead, they will need to design policy in ways that divide and sideline opponents while fostering and bolstering increasingly powerful entities that can challenge them, such as alternative energy producers and consumers. Raising the cost of carbon emissions will not, by itself, do this—or at least not quickly enough to forestall backlash or backsliding. Changing interests and alliances through policy has to be a concerted policy strategy, deployed at both the state and federal levels and designed with political as well economic effects in mind.
A parallel and vital line of thinking urges a greater focus on the impact of policy designs on the economic viability of continued political resistance. Analysts should consider whether powerful opponents can be economically marginalized, so their buy-in is not required. As Eric Patashnik argues in his article, some market-utilizing policy reforms (such as airline deregulation) succeeded in rendering the business models of opponents obsolete, leading to a quite rapid and favorable transformation of the interest-group environment. In other areas with large market stakes (health care, climate policies, financial regulation, antitrust), similar strategies may be possible. This market-shaping logic argues for making “structural” change a bigger priority than it often is within expert circles. For example, financial reform has generally focused on changing the incentives of existing firms. Sometimes, however, policies that simply outlaw certain practices or enforce key institutional structures (for example, requiring separation of traditional banking from aggressive financial engineering) may be less subject to backsliding, because they restructure private economic actors along lines that divide or eliminate interest-group opposition.
Of course, the greater threat to these interests posed by structural reforms will also intensify initial resistance. Yet even when resistance is fierce, policies might include backdoor structural changes—provisions that do not pose an existential threat to powerful interests but instead induce them to accommodate rapidly to (that is, make specific investments in) a new policy regime. This is a key feature of various Medicare expansion proposals that seek to move toward, but not all the way to, “Medicare for All,” as discussed in Jacob Hacker’s article. The idea is to get providers, insurers, and employers to reorient themselves around flows of benefits coming from Medicare, soliciting their buy-in while also increasing the capacity of federal authorities to restructure consolidated insurance and provider markets.
Finally, there is a flip side to the advantages enjoyed by powerful concentrated interests in an era of high inequality. A highly concentrated set of current beneficiaries implies that there is often a very large circle of potential winners from reform. When preexisting policy arrangements are intensely inegalitarian, there may be opportunities to build broad coalitions through reforms that overturn them. In the past, for instance, policy-makers have generally eschewed highly progressive revenue sources because they fear the backlash of (influential) affluent taxpayers. Yet with inequality more and more stark, high-end taxes may sometimes be the path of least political resistance for financing new initiatives. In particular, policies grounded in such financing may well be more durable than broader levies, simply because—in an age of extreme inequality—targeted progressive taxes can finance substantial benefits for a large number of voters and groups. Similarly, as Robert Manduca argues in his article on the benefits of antitrust enforcement as a means of revitalizing declining regions, as corporate power concentrates in fewer and fewer hands, companies may gain political clout but they also become much more politically attractive targets for reform, as the recent travails of Facebook suggest (Frenkel et al. 2018). Moreover, in such a context, changing the behavior or weakening the market power of these dominant companies can have an outsized effect, too.
In the end, no simple, broadly applicable logic dictates when to compromise with a group and when to try to weaken it. Much will depend on the specific design choices available and the particular political context. Yet the fact that our polarized political environment offers diminishing prospects for successful compromise may strengthen the case for a more aggressive, structurally minded approach that seeks to reconfigure the balance of power among organized groups. Where achieving buy-in may be difficult, the case for using policy windows to alter the balance of power in fundamental ways may grow.
Limiting backlash
Patashnik’s contribution to our volume focuses on the increasingly evident threat of backlash as a potential (negative) feedback effect. Although such an outcome is always possible for a poorly designed initiative, the risk intensifies in our hyperpolarized context. Partisan opponents and their allies have powerful incentives to make new initiatives a focal point for countermobilization. Opponents of policies (along with opponents of the party enacting them) are certain to exploit features of the contemporary political climate to undercut entrenchment. They will draw on perceived fiscal constraints, low trust in government, and heightened polarization (with its attendant partisan media, culturally insulated voter blocs, and team-oriented politician and interest group alignments) to generate backlash.
Managing backlash is about more than minimizing initial resistance to a policy. (Indeed, as just argued, policies that spark greater initial opposition because they bring about big structural changes may well, if successful, reduce the chance of future backlash.) It does, however, require thinking about how postenactment political dynamics could compromise either the power of a policy’s supporters (for example, through electoral losses and investment in counterlobbying) or the durability of a policy (for example, through postenactment sabotage). Put another way, managing backlash is a problem of balancing present and future opposition—of reducing not only the risk that a policy will fail to pass, but also the prospect of countermobilization against it in the future.
In particular, two policy characteristics tend to provoke backlash: very visible financing linked to relatively low-visibility or long-delayed benefits (think of the backlash that led to repeal of the Medicare Catastrophic Coverage Act in the 1980s, discussed by Patashnik) and very direct challenges to the identities or status of people highly reliant on or attached to prereform policy arrangements (think of the backlash of older voters against the ACA, which was effectively demonized as threatening Medicare). The point is not that new initiatives should be deficit-financed, or that reformers should not build on policies that already have claimants; most of the research on feedback argues for seeking a predictable stream of financing and building on existing institutions. The point is that policy design must be mindful of the risk of backlash.
As with a policy’s effects on powerful groups, no precise formula dictates how backlash can best be minimized. To lessen backlash from current policy claimants, for example, Patashnik recommends careful attention to their perceptions of the program (for example, as an earned benefit) and how those might be threatened by new initiatives. But Patashnik acknowledges that the exact sources and responses to backlash vary by policy and context. The key point is that policy designers need to put sustainability considerations front and center: to think not just about the exigencies of enactment, but also where organized opposition might come from in the future and how those opponents can be mollified, divided, or undermined. Thus, prescriptions for preventing backlash dovetail with our larger message about organized groups. Often, their preferences and influence—not the public’s—are the factor most crucial to success.
Our two articles on criminal justice policies both put considerations of backlash front and center. Both Weaver and Geller, on one hand, and Lerman and Harney, on the other, emphasize the distinctive feedback effects that operate in this domain and make reform extremely difficult. As Weaver and Geller put it, “Criminal justice policies created a new political constituency, strengthening police unions, correctional organizations, prosecutors, and companies attached to the business of managing and housing inmates and cementing their coordination with one another”—a “constituency” that is itself a source of backlash but that also is adept at mobilizing white home owners outside city centers who are pivotal to the formation of criminal justice policies. The two articles offer alternative innovative ideas for reducing backlash: Weaver and Geller call for a new approach that focuses on reducing the toxic interaction between young people and the criminal justice system; Lerman and Harney show how those who work in the correctional system might be harnessed to the cause of criminal justice reform. In both cases, the near inevitability of backlash is taken as a given; the question is how to defuse and redirect it in ways that allow a new set of feedback dynamics to take root.
Utilizing federalism
Much fresh thinking about durable policy design will necessarily concern national policy. Nonetheless, we see a compelling rationale for integrating careful reflection about federalism into these efforts. In our federated political system, national, state, and local policies are deeply interrelated. Those who seek to use government to address collective problems must be attentive to how new initiatives can harness multilevel governance so that it might best contribute to, rather than undercut, long-term success—both in addressing those problems and in fostering a supportive political environment.
A number of the contributors to this volume point out that, in our polarized era, states are increasingly important sites of policymaking. This reflects at least three developments (Grumbach 2018). First, states are much less likely to be gridlocked. Unified partisan control unencumbered by minority vetoes is a situation that national policy-makers almost never face. But it is common in the states, which typically lack supermajority requirements and are increasingly likely to tilt heavily toward one party or the other. Second, states are becoming more economically and demographically diverse, as a century-long trend toward economic convergence has reversed and immigrants have moved into a few key states. As a result, specific states will often provide much more favorable environments than the national government for implementing ambitious initiatives. Finally, and partly for the same reason, states are likely to be sites of countercyclical political action, with at least some vigorously objecting to whatever political constellation is ascendant nationally. Minimizing backlash and building support will almost always rest in part on thinking through how states both allied with and opposed to policy goals will respond.
What are the implications of this increasingly prominent feature of the policy landscape for feedback dynamics? If federalism creates new obstacles to feedback, as the ACA’s rocky rollout suggests, how can these obstacles best be managed? Or does federalism create a new opportunity structure in some areas? For example, can policies be advanced and consolidated in some states without needing immediately to surmount the obstacles to national policymaking associated with polarization and gridlock? Again, the recent conservative focus on policy at the state level highlighted by Hertel-Fernandez is instructive. National policy-makers may want to pay close attention to whether their initiatives encourage the right kinds of experimentation at the state level.
In a scenario of state-initiated reform, the medium-run effects of such territorially uneven implementation of new policies also needs to be considered. Hertel-Fernandez’s article details how conservative success in particular states spills over to shift the balance of political resources in other states, since organized political actors in one set of states often rely on transfers from better resourced allies elsewhere. Samuel Trachtman’s article on climate policies argues that those seeking to enact energy and climate reforms can harness these spillovers. He observes that well-designed state initiatives can create administrative capacity and expand supportive constituencies in ways that improve the opportunity structures for further reforms in other states as well as at the national level. So too does Jamila Michener, in her careful consideration of how momentum for Medicaid expansion and improvement can be maintained at the state level, with more ambitious states not only setting the national agenda but also encouraging diffusion of innovative approaches across the states.
Sequencing
In addition to highlighting the balance of state and federal action, both Trachtman and Jonas Meckling (in his complementary article on national climate policies) argue that thinking through policy feedback means thinking about policy sequence. If you cannot get from A to Z in one great leap, which policies will get you to, say, G and create political pressures for additional steps that eventually get you to Z? To take an example offered by Meckling, international and cross-state experience suggest that significant new taxes on carbon typically come toward the end of a sequence of reforms that encourage clean energy—precisely because these initial policy departures increase the receptivity of key interests to introducing a price on carbon.
Attention to sequence does not always counsel caution. Even ambitious policy efforts that do not succeed may eventually break through if they plant seeds for the future, increasing the mobilization and power of key groups while making clear who stands in the way of popular initiatives. Nonetheless, policy failures can be extremely costly (as the failure of the Clinton health plan in 1994 suggests), and being on the “right side” of a battle is far better when it is also the winning side. Moreover, the kinds of broad openings for increases in government activism that characterized the 1930s and 1960s—especially the outsized Democratic Party margins that facilitated them—seem highly unlikely to materialize in the near term.
If policy ambitions outstrip political opportunities, however, the best response is usually not retreat but step-by-step advance. Reformers should not only ask what “fallback” reforms are politically feasible and how far toward an ultimate goal they will go. They should also consider how much pressure these reforms will create for future steps in that direction, based on the kinds of feedback effects under discussion. In short, they should analyze how the political effects of policies in round one might affect possibilities in rounds two, three, and beyond.
This kind of thinking, for example, is a central element of Andrew Schrank’s proposal for rebuilding worker representation and encouraging better-paid jobs. Schrank argues that the current system of labor law is ill-suited to the present economic context, creating a vicious cycle of policy inaction—punctuated by occasional dramatic retrenchment, as described by Hertel-Fernandez—and the inexorable erosion of traditional labor groups. To create a virtuous cycle, he calls for changing the basic incentives of the labor market through two key regulatory shifts: a higher minimum wage and a new approach to workplace regulation that dramatically improves enforcement. Together, these early steps would create a virtuous, rather than vicious, cycle—in which workers felt freer to join labor organizations and new types of worker organizations gained greater ability to respond to this increased demand. In other words, Schrank argues that relatively modest changes, if done early and well, could create the opportunity for bigger changes down the road, because of the feedback effects they generate. That message, indeed, is conveyed in all the articles in this volume, which remind us that big transformations do not necessarily require big initial policy changes, as long as the feedback effects of those changes are substantial and self-reinforcing.
Conclusion
Efforts to think through issues of policy feedback in the contemporary context will involve a complex range of considerations, and no simple set of principles will be broadly applicable across the range of challenges that policy-makers face. In this article, we have sought to identify and clarify some of the most important considerations that have emerged so far through the work of the PFP. We recognize, however, that this article—and, indeed, this entire special issue—is only a preliminary statement. Although the scholarly literature on policy feedback is now extensive, the examination of how feedback dynamics play out in a climate of high polarization and high inequality is at an early stage. More important, the PFP is the first effort—of hopefully many efforts—to translate an updated understanding of feedback into applied knowledge for contemporary policy-makers.
Still, scholars who have researched how “a new policy makes a new politics” provide ample grounds for thinking that policy design matters for whether policies become established and entrenched and expand over time. Policy-makers who wish to use government to address our nation’s serious collective problems should take these lessons to heart.
Footnotes
Note:
We thank everyone who provided feedback and guidance at two project workshops: a September 2018 meeting in Cambridge, MA, hosted by the Scholars Strategy Network (SSN); and a February 2019 meeting in Washington, DC, funded by the Washington Center for Equitable Growth. At SSN, Avi Green, Pao Maynard-Moll, Hannah Reuter, and Theda Skocpol were of invaluable assistance. Even before our first meeting, we benefited from the wise counsel of Heather Boushey, Paul Glastris, Jacob Leibenluft, Tara McGuiness, Ellen Nissenbaum, and Felicia Wong. At the American Academy of Political and Social Science, Tom Kecskemethy and Emily Babson made this special issue possible. Thanks also to Yale College student Neeraj Patel for helping to assemble the final papers. Finally, we thank Aaron Goldzhimer for inspiring this project in the first place.
Jacob S. Hacker is Stanley Resor Professor of Political Science and director of the Institution for Social and Policy Studies at Yale University. He is the author or coauthor of five books and numerous articles on American politics and policy and a fellow of the American Academy of Arts and Sciences.
Paul Pierson is John Gross Professor of Political Science at the University of California, Berkeley, and codirector of the Successful Societies Program of the Canadian Institute for Advanced Research. He is the author or coauthor of many articles and five books on American and comparative politics and is a fellow of the American Academy of Arts and Sciences.