1Thinking and writing about public policy in the United States at the present time is both difficult and painful. It is difficult because there is no clarity about what the policies are from day to day, or sometimes hour to hour. Anyone attempting to describe in any detail, much less predict, what is happening with American policy risks appearing very foolish very quickly. It is painful to write about policy now because perhaps the only thing that is clear is that many policies and programs that have been developed over decades or even centuries are being essentially dismantled. The American welfare state has never been well-developed, but the progress that has been made is being lost rapidly.
2I was asked to write this Forum essay as someone who have been observing and writing about American public policy for over 55 years. This essay may therefore appear to be primarily of antiquarian interest to contemporary students of public policy. It is important, however, to understand the roots of contemporary policy debates and policymaking models (Peters & Zittoun, 2026). Without understanding those roots, it is difficult to understand the dynamics of the contemporary world of public policy. Further, this essay is in part autobiographical, and in part an attempt to link the models of policy we all are familiar with to the intellectual and policy milieu in which they emerged.
- 1 Supplemental Nutrition Assistance Program.
3I began investigating public policy as an academic “professional” during what was the heyday of public policy in the United States. The assassination of President John F. Kennedy had created a window of opportunity in John Kingdon’s words, and President Lyndon Johnson used that opportunity to launch his Great Society programs (Zelizer, 2015). These programs represented a major attempt to rectify the inequality and poverty that afflicted (and continues to afflict) the United States. A number of social programs that had been taken for granted for a number of years–Medicare, Medicaid, Food Stamps (now SNAP1), Head Start and others–were initiated during the Johnson administration.
4As well as being the heyday of substantive policy development, this period was the beginning of the heyday of public policy studies. At the same time that a host of new policies were being developed, new models of policy were being developed, and old ones elaborated. This was true even though public policy, public administration, and institutional analysis were rather marginal in political science at the time, given the dominance of the methodological individualism of behavioralism and rational choice. But policy studies was also having its own methodological revolution, beginning to use quantitative analysis for questions such as public expenditures and the impacts of social programs. My own dissertation in the late 1960s, for example, was a quantitative study of the development of social policy expenditures in three European countries.
5As a newly minted assistant professor in 1969, I also worked as a consultant for the Model Cities Program in Atlanta, Georgia and as a consequence wrote my first paper on policy coordination. While the individuals involved in this and other components of the War on Poverty were not naive, there was still a strong sense of optimism and hope. The sense among the advocates was that government could be used to address long-standing social problems and, if that were done long enough and with enough resources, that a “Great Society” might indeed be possible. That feeling was far from universal throughout the population of course and these programs were condemned by the Republicans and some Southern Democrats. Given my experiences with these programs, and their importance for understanding policymaking in the United States, much of this essay will focus on social policy. The same general pattern of increasing federal government involvement in the economy and society can be found in other policy domains (Peters & Hoornbeek, 2026).
- 2 Daniel Patrick Moynihan, Nixon’s advisor and later senator from New York, was the author of the pla (...)
6Although certainly not solving the problems of poverty, these programs did work to some degree. More importantly, the basic idea that the Federal government could and should do something about poverty became institutionalized. Although still questioned by conservatives, this idea was largely accepted in the policymaking community in Washington, Indeed, even an arch conservative like Richard Nixon, who followed Johnson in the White House, would go even further than the War on Poverty and propose a minimum income plan – the Family Assistance Plan– for all Americans (Kornbluh, 2008).2 This plan was not adopted but the very fact that it was proposed represented a sharp deviation from the limited Welfare State in the United States.
7For scholars of public policy, the Great Society programs were important because of the amount of money that was devoted to evaluation. That funding supported evaluation “shops” within government, but it also created an evaluation industry in universities, consulting firms, and think tanks. The emphasis on evaluation also helped promote the development of new methodologies, and the borrowing of old methodologies from other disciplines. Public policy studies continue to have a strongly institutional foundation, but also could compete in the use of rigorous methods with other components of political science.
8The programs of the War on Poverty, and attempts at even more significant reforms, demonstrated the willingness to engage in major policy interventions. Three other pieces of legislation since that time demonstrated the same willingness. The first two were the Child Health Insurance Program (CHIP) and the Temporary Assistance to Needy Families (“workfare”) passed during the Clinton Administration. These acts demonstrated two important points about policymaking in the United States during that period that was not as innovative as the Johnson years, but still capable of major reforms.
9The CHIP program was notable for its bipartisan nature, something that seems almost impossible today. In the Senate it was sponsored by one of the most liberal (Edward Kennedy) and one of the most conservative (Orren Hatch) senators and had broad bipartisan support. While CHIP demonstrated bipartisanship in support of a liberal reform, TANF had an essentially liberal administration adopting many conservative principles about social policy, e.g. “workfare” in order to have a major reform of the social assistance program adopted.
10The final example of major social policy reform was the Affordable Care Act – Obamacare – passed in 2010. While now reviled by Republicans who have made dozens of efforts to repeal the act, this was in many ways a Republican idea. The basic structure of the program was advocated by Republicans as an alternative to the health policy initiatives of the Clinton administration, and it was adopted in Massachusetts under a Republican governor before being adopted at the national level. Thus, many major social policy reforms in the US have been the product of sharing ideas, if not always sharing political support.
11This (all too) brief history of social policy can be used to point to several characteristics of policymaking in the United States that were both successful and conducted within the policymaking framework contained within the Constitution. This succession of policies can also be linked to ideas about policymaking contained within the academic literature on public policy. Unfortunately, this tale of slow progress does seem to be almost entirely history, and policymaking now is much less hopeful, and certainly less contained within the democratic framework of the Constitution.
12The case of Obamacare and the continuing attempts to repeal it, leads rather naturally to a discussion of American policymaking at the quarter-century mark of the 21st century. A number of factors that defined policymaking prior to 2010 have vanished and with them the capacity to govern effectively. I pick 2010 as the key date because this was the election in which the Tea Party (Arceneaux & Nicholson, 2012) came to prominence, and with it the increased attacks on the “administrative state” and the Washington establishment. These changes have been important for the substance of policymaking, but they also have relevance for the body of theory used in the study of public policy.
13As discussed above, when American policymaking has been successful it has been so when operating in a bipartisan manner. The extreme political polarization of the population and of the elites has meant that such co-operation is no longer possible. Using Lijphart’s (1999) familiar terminology, American government is a consensual system, disguised as a majoritarian system. The multiple institutional actors and the checks and balances require cooperation even when all three major policymaking institutions are controlled by the same party. The extended shut-down of government occurring as I write this is evidence of the need for cooperation across parties.
14Since 2010, a few major pieces of legislation, notably Obamacare, have been passed when all three institutions were controlled by the same party, but historically divided government was as successful in legislation as unified government (see Mayhew, 1991). With the current level of division between the parties, doing anything with a divided government is difficult to contemplate. The idea that the task of members of Congress is to govern, rather than to protect political and ideological positions in order to be reelected, appears to have been lost in the conflicts between MAGA Republicans and increasingly progressive Democrats.
- 3 Some early work by Chicago School of political scientists had foretold this type of research in the (...)
15Not only are the public, political parties, and political institutions more polarized, there have been attempts (some successful) to politicize the other major institutions of government – the courts and the bureaucracy. The courts, and especially the Supreme Court, had been able for most of their existence to maintain the useful fiction that they were impartial, legal arbiters and that fiction was essential to their maintaining their legitimacy. Political scientists have, however, for over 60 years, argued that judges have identifiable ideological and political leanings that shape their votes on cases (Schubert, 1958).3 This politicization of the courts has been more apparent in the Roberts court, and ordinary citizens have recognized that many decisions are highly political. That recognition has in turn caused the legitimacy of the Supreme Court to decline – only about ten percent of the American public now say they have a great deal of confidence in the Supreme Court.
16In addition to the politicization of the courts, the rest of the legal system has also become more overtly political. The Trump administration has attempted to “weaponize” the criminal justice system in two ways. One is through indicting individuals who have crossed swords with the president in the past, e. g. James Comey and Letitia James. The other has been to use the presidential pardon power to free friends of the president, with the most egregious example being the rioters who stormed the Capitol on January 6, 2021. This politicization of the legal system is not directly policymaking, but it does reinforce the loss of objectivity in the system.
- 4 Attempts on the part of democratic backsliding regimes to politicize the bureaucracy have by no mea (...)
17The politicization of the public bureaucracy has been even more obvious. Although the public bureaucracy has often been blamed for many of the failings of government (Hinterleitner, 2025), it has become a more important target for political leaders, especially in the Trump administration (Moynihan, 2025),4 The claims of the bureaucracy constituting at least part of the “deep state” that had been doing a disservice to ordinary Americans, has helped justify the attacks on the federal workforce and agencies. In addition to the immediate damage to public programs, the destruction of a highly skilled federal workforce built over decades will produce long term problems for American policymaking.
18The differences among the parties are heightened by the utilization of “Zombie” ideas (Peters & Nagel, 2022) about policy, and the ways to improve the economy and society. A Zombie idea is a policy idea that has been tried (often many times) and has been found to have little or no validity. Despite the lack of evidence, these ideas continue to reappear on the agenda and continue to be adopted. These ideas tend to be firmly held, at least by one side or another in the political divide.
- 5 We know how well that worked.
19In contemporary American policymaking, several Zombie ideas appear to be dominating economic policy. First, there is the commonly held belief among Republicans that tax cuts, especially for the more affluent, will produce economic growth. This idea was prominent in the first Trump administration and has reappeared in the second. As has been true with previous tax cuts, including those by Democrats, there is little demonstrable effect on rates of economic growth (Campbell, 2025). In addition, the widely disproven idea that tariffs will enhance economic growth by promoting investment in domestic industries has been resurrected and implemented. The last time the United States enacted major tariffs was in 1930 with the Smoot-Hawley tariffs, which were intended to end the Great Depression.5
20It is not only in economic policy, however, that Zombie ideas have been appearing. In social policy the familiar notion of work incentives for social programs is being reintroduced for health and nutrition programs. The advocates have failed to notice that the vast majority of the recipients of these programs are already working, or are caregivers for children, or are disabled. There is little to be gained in terms of employment from these requirements, although the increased administrative burdens may cause some potential recipients to abandon even trying to get benefits (Herd & Moynihan, 2025).
21As is argued by some standard policymaking approaches (see below) ideas can have an independent impact on policy choice. But even familiar ideas that tend to persist across time may not be able to be readopted without some political assistance from policy entrepreneurs (Petridou & Mintrom, 2021). In addition to President Trump himself, the White House and the rest of the administration is now filled with individuals promoting the above ideas that lack any evidential base but which are part of the doctrine of populist politics.
22The initial discussion of policymaking in the United States emphasized some major pieces of legislation that constituted landmarks in the development of the American welfare state. That emphasis on major bills to some extent ignores the major feature of American policymaking, which has been Incrementalism. This concept has been used both theoretically (Lindblom, 1959; Hayes, 2023) and empirically (Davis, Dempster & Wildavsky, 1966) to describe policymaking in the United States and is considered by its advocates as a superior method for making policy wherever that may be occurring.
23That discussion of major pieces of legislation somewhat paradoxically confirms the incremental nature of American policymaking. Rather than passing a piece of comprehensive legislation for the entire population, health coverage was implemented first for the elderly and the medically indigent, then children, then the “working poor” and self-employed. A significant portion of the population continues to receive health insurance from their employers, and approximately 14 million people still lack coverage. This piecemeal approach not only leaves gaps in coverage but increases the administrative costs for health care significantly. Yet these pieces of legislation were possible with short-term coalitions that could not agree on more comprehensive change. Pressures to create a comprehensive system, e.g. “Medicare for All”, persist but face very strong political resistance (Hulse, 2025).
24The incremental model is closely linked to the ideas of bounded rationality (Simon, 1947; Jones, 2010) and the belief that governments (and private actors as well) cannot make comprehensively rational decisions because of the absence of information, especially information about the future and wide diversity of preferences. Incremental policymaking is also a product of the extreme institutional complexity of American policymaking, and the need to bargain within and across institutions as well as parties to accomplish policy goals. The attempts to pass an extension of subsidies for Affordable Care Act policies has been indicative of that institutional complexity.
25As has happened in recent years, decrementalism appears much easier than incrementalism. In some cases, American policymaking has gone beyond decrementalism to policy and organizational termination during the Trump years. This decrementalism and termination has attempted to bypass institutional complexity by relying entirely on executive action. I say “attempted to bypass” because the courts have been blocking some of the attempts to cut staff and terminate programs.
26Although not explicitly linked, bounded rationality and incrementalism are to some extent allied with pluralist thought that was central to American political theory. Pluralist theory assumed that the political arena was an open marketplace in which interests and ideas competed more or less equally (Dahl, 2006). The social heterogeneity of the United States was assumed to produce the political heterogeneity central to that contestation of ideas. Policymaking in this theory tended not to produce big victories, but rather generated gradual change that, over time, could move in one direction, but took its own sweet time getting there. However, this movement in one direction could easily be reversed with the creation of a new coalition of forces.
27There were, of course, theorists who took a more elitist position (Bachrach & Baratz, 1962; Schnattschneider, 1960). They argued that American policymaking was controlled by socio-economic elites that may not always be visible, but which do shape policy. That said, there was an assumption that the political elites involved in the contestation of ideas had a common acceptance of the ideas of liberal democracy and would not stray too far from those core values, an assumption which was also rejected by some critics of the dominant strand of political theory at the time (Walker, 1966).
28The major models of policymaking that emerged in the United States may not have been explicitly pluralist, but they certainly did not have the deterministic nature that one would expect from elitist models. For example, the several versions of the Multiple Streams Framework (Kingdon, 2003; Cohen, March & Olsen, 1972) are based on random confluence of forces (albeit perhaps with the assistance of a policy entrepreneur) rather than exercise of power. The heterogeneity central to pluralist thinking produces an environment populated by numerous interests and groups representing those interests, so that it may be good fortune (aided by that entrepreneur) that one may be able to assemble a sufficient coalition to enact legislation.
29The Advocacy-Coalition Framework (Jenkins-Smith & Sabatier, 1993) does assume the existence of a dominant model at the start of a policymaking episode, but the model is centered on other actors and policy ideas challenging that domination and perhaps winning. As implied above for pluralism and indeed for almost any form of policymaking in democratic politics, a fundamental goal is to form a coalition of actors that can alter the status quo (or defeat attempts at change). And again, ideas constitute a major component of the armamentarium of those attempting to form coalitions (Béland & Cox, 2011).
30Both Multiple Streams approaches and the Advocacy-Coalition Framework depend, less or more explicitly, on building coalitions. That need for coalition formation may emphasize some of the pluralist aspects of policymaking. There are multiple interests and some means must be found to assemble a sufficient number to adopt a policy. That coalition is based on the self-interest of the groups and individuals involved, possibly connecting pluralist thought with rational choice ideas. But the very ideas that undergird the policies may be the glue that binds the coalition, even if that coalition will be fleeting.
31The Punctuated Equilibrium Theory (Baumgartner & Jones, 1993) shares some pluralist assumptions about policymaking but also has some more elitist elements. The latter components of the approach are that once government acts in a policy domain, the result tends to be perpetuated for a significant period of time, with both policy elites and dominant groups in society preserving that solution. On the other hand, however, the model is open to change through the mobilization of other interests and the upsetting of the status quo. This approach to policymaking is more open to mass politics than the two models mentioned previously, as it focuses on how public opinion and other political forces move items onto the agenda, and then move on to other matters of concern.
32The open and democratic assumptions of pluralism appeared rather optimistic and naive to many people even when pluralism was a dominant pattern of thinking in US political science, and appears even more so in the present period (see below). It seemed to ignore the institutionalization of power within the political system and barriers to access by numerous segments of society. At the same time, however, pluralists could point to policy changes produced by shifting group coalitions and the ways in which apparent elite domination could be thwarted (McFarland, 2004).
33Considering the range of ideas about policymaking in those formative years of public policy studies, the one approach that was alleged to have elitist and/or technocratic foundations was policy design (Linder & Peters, 1984; Bobrow & Dryzek, 1987). While most other approaches to policymaking assumed that responses to policy problems would arise by actions driven from the bottom up, the design approach assumed that they were and should be the product of purposive design. Further, that design would be largely the product of experts considering the nature of the policy problem and the range of available policy instruments.
34That elitist and technocratic conception of policy design has been challenged by more participatory conceptions of policymaking. These challenges have been in part about the process of designing. The more traditional pattern of decision-making was “downstream”, considering a limited range of options and coming to a decision in a relatively rapid manner. “Upstream” conceptions of design seek to broaden the range of options considered, producing slower but perhaps more creative solutions.
35In addition to emphasizing the development of a wide range of alternatives, contemporary approaches to design also seek to include a wider range of participants in the design process. Rather than relying on experts in the bureaucracy or think tanks to design policies, there have been attempts to bring more groups and more ordinary citizens into the process. As with upstreaming, this practice may slow the process but it is intended to be another means of broadening the range of policy alternatives and the values which are applied to choices of policy designs.
36The concept of the administrative state is discussed more frequently by scholars of public administration than by policy scholars (see Waldo, 1948), but the concept has a great deal of relevance for policymaking as well. This characterization of patterns of governing is clearly antithetical to pluralist ideas, it has been important for the development of political theory in the United States and can be linked to ideas about policy design. The relevance of the administrative state has become even greater with the reemergence of populist politics.
37The notion of the administrative state is that as government grew during the Great Depression and then World War II, the volume of decisions that had to be made increased dramatically, so that Congers and political executives could not manage the demands. The bureaucracy expanded to fill that breach, but this raised concerns about the loss of democratic control over policy. The Administrative Procedures Act (1946) was passed to address those concerns, creating a system for administrative rule-making that is more open to public influence than others in democratic political systems (Kerwin & Furlong, 2025). That act was still not sufficient to assuage the fears of the critics of administrative rule-making.
38One of the most important critics of the administrative state, not so much as an intellectual construct but as a governing reality, was Theodore J. Lowi. The power of the administrative agencies, and their linkages to interest groups, had led to Theodore Lowi’s (1969) book, The End of Liberalism, which was a description of the emergence of a “second republic” with a much weakened Congress and powerful administrative agencies and interest groups. That description might be seen as a precursor to the “administrative state” arguments that have become common in the 21st century, although his interest was much more about restoring greater democratic control than facilitating control by authoritarian executives.
39While Lowi expected the courts to be the guardians of democracy, the US Supreme Court to some extent increased the power of the administrative agencies with Chevron v. Natural Resources Defense Council (467 U.S. 837 (1984)). The basic finding of the Court was that the courts should in general defer to the expertise of the agencies when examining whether a regulation made by an agency was reasonable. Two decades later, a Republican dominated Supreme Court in Loper Bright Enterprises v. Raimondo, (603 U.S. 369 (2024)) found that Chevron gave too much power to the agencies and reasserted the power of the courts to determine the reasonableness of regulations, thus weakening the administrative state.
40The past several presidencies, and especially the second Trump administration, gives the lie to anyone who still believes in pluralist politics. The presidency has increasingly become the dominant institution in policymaking, often utilizing executive orders rather than legislation in Congress to make policy (Moe & Howell, 2025). If Donald Trump continues to issue executive orders at the current rate, he will issue over one thousand orders, while Joe Biden issued 162 even while coping with COVID and its economic aftermath. The access of many groups within society to the policymaking process is also being reduced by eliminating advisory bodies and public hearings.
- 6 The members of the Congressional committees and subcommittees typically would come from districts w (...)
41The arguments about the administrative state can easily be related to the idea of the “iron triangle” that was prominent in American political science for some years. The vertices of the iron triangle were an administrative agency, an interest group, and a congressional committee or subcommittee. This model of politics and policy argued that these triangles were composed of groups with relatively common interests who cooperated to gain as much resources as possible from the budget process and exclude other actors from influencing “their policy”.6 The iron triangle had some elements of corporate pluralism and networks discussed in European political science, but it was more closed than those systems.
42This “iron triangle” model of American policymaking has been declared defunct and argued to have become “big, sloppy hexagons” (see Jones, 1982). The legacy of the triangles endures, however. The part of the iron triangle that is defunct is the one-on-one relationship between an interest group and an agency, given the proliferation of interest groups and changes in laws governing access to the policy process. That access has tended to produce more complex laws (Ehrlich, 2011), but it has not significantly diminished the autonomy of many agencies. Further, as already noted, there are continuing pressures to reduce the decision-making powers of the administrative agencies, with the question being whether the President or Congress is to fill the void that is being left.
43The events of the last year, foreshadowed to some extent by the previous eight years, must lead one to the conclusion that, if pluralism was ever a viable explanation, it is now moribund. The political system has become more presidentialized than ever (see Poguntke & Webb, 2005), and most of the structures that once linked the masses and the elite have been dismembered (see Kornhauser, 1959). Current politics are often described as populist, but they do not appear to be driven by the wishes of most of the public, but rather by a small group of wealthy individuals. Populism is now more a rhetorical style than a reality.
- 7 Again, that was something of an elite project that was not supported by much of the working class. (...)
44Clearly the policy choices being made are anything but populist in the original sense of protecting the least affluent and least influential from the power of the elite. They further differ from original versions of populism in making government the principal enemy of the people rather than big business. In addition to dismantling social programs, the current administration is placing a great deal of pressure on organizations in and out of government to eliminate diversity-enhancing programs. Those programs had been a major emphasis of the previous Biden administration.7
45The relevance of apparent elite dominance in much contemporary policymaking is not just about the authoritarian tendencies of some leaders. It is also that, as noted above, most of the major approaches to policy tend to have pluralist assumptions and to ignore the role of elite power. Numerous scholars have spent countless hours attempting to refine those approaches, but given contemporary political change, they may as well have been rearranging the deck chairs on the Titanic.
- 8 For an earlier recognition of these differences see Mills (1956).
46One problem with elitist theories of the State is that they tend to assume that the elite has relatively homogenous preferences. This does not appear to be true for contemporary populism in the United States, despite the domination by Donald Trump8. Further, Trump’s policy preferences tend to vary hour by hour, if not minute by minute, so that there is a substantial inconsistency across time. While there may be an underlying theme of supporting business and wealth accumulation, there are marked differences across policy domains and across time.
- 9 This effort may be in the narcissistic pursuit of a Nobel Peace Prize.
47One thing one might reasonably expect from an elitist policymaking system is some coherence and coordination. Among the many astounding features of the Trump administration has been the number of contradictory policies. For example, the Department of Health and Human Services is pledged to reduce obesity, especially among children, but cuts to SNAP mean that families will be forced to rely on cheaper, calorie-laden processed foods, The administration was to withdraw the United States from its international role, yet the president has invested an increasing amount of time on foreign policy9, including threats of armed conflict with several countries and threats of territorial expansion.
48American government has had strong tendencies toward incoherence for most of its history, in part because of the fragmented nature of the executive branch and the strong links of agencies with clientele groups. The checks and balances within the system had helped to produce some coordination of actions, but those have been largely swept away by the increasing power of the president and the rest of the executive branch. The difference now is that the groups to which the executive agencies are linked have changed and become even narrower than in the past. For example, health agencies are now linked to anti-vaxxers and disregard the established medical organizations and practitioners.
49This all too brief descriptions of the development of public policy studies in the United States and its relationship to the “real world” of policy has excluded a number of things. Perhaps the most obvious is the absence of more discussion of the quantitative approaches that now appear to dominate the field. Also missing are discursive and constructivist approaches, as well as the important contribution of Elinor Ostrom and her colleagues at Indiana University.
50This was meant to be both a personal essay and an attempt to link changing policy theory with changes in the political and intellectual environment. My own “adventures” as a student of public policy have to some extent tracked the development of our discipline, even with the omissions mentioned above. This essay has also documented the extent to which pluralist thought in the United States has influenced thinking about policymaking. This is not imperialistic, but rather reflects the realities of policy theory. Unfortunately, thinking about policymaking continues to have implicit pluralist assumptions, even while the policymaking world becomes more elitist. This leaves us all with the challenge of rethinking the ways in which we address the academic challenges of policy theory and the real challenge of greater elite control.
51What will the future hold for policy studies? The most obvious answer is more of the same. We are all trained in certain ways and committed to certain approaches and will therefore have difficulties in moving beyond them. We will also need to be careful that artificial intelligence does not supplant human intelligence, especially when we play the role of policy advisor. I do hope that as a discipline we will begin to pay more attention to a wider range of theoretical approaches, as well as think about theory more carefully. And last but certainly not least, I hope that there will be more concern for comparative analysis, and that public policy will become the truly international field of study that we in the International Public Policy Association are attempting to make it.