Escaping Policy Traps: Strategic Options for Overcoming Entrenchment

Abstract

Policy entrenchment per se is a neutral concept; both good and bad policies may become entrenched. A policy trap, however, is entrenchment's pathological form: a self-reinforcing array of policies that simultaneously creates (1) well-established, often widely recognized failures in society, and (2) high barriers to change. A familiar type of policy trap arises when the benefits of a policy are concentrated while the costs, albeit greater, are widely diffused, opaque to many who bear them, or seemingly remote. But policy traps are not necessarily permanent; they may persist only as long as reformers are unable to identify their vulnerabilities and seize moments of political opportunity. Reconstituting a domain of policy ultimately requires formulating an alternative. Without presuming to be exhaustive, this article outlines four general strategies for overcoming policy entrenchment: Schumpeterian innovation, globally oriented innovation, institutional conversion, and social creativity (the nonmarket analog to Schumpeterian change). Focusing on three areas, the article examines how policy traps have arisen and might be overcome in fossil fuels, the internet economy, and the US health care system.

To say that an institution or policy is entrenched is not to say that it is eternal—only that it is tenaciously resistant to change at its foundations and that the threshold for the adoption of alternatives is exceptionally high. For those interested in foundational change, whether as scholars or as reformers, entrenchment poses related analytical and practical questions: How do breakthroughs take place? What kinds of strategic options are available for overcoming entrenchment?

In much historical-institutionalist work, “critical junctures” are the conceptual pivot in explaining large-scale institutional change. A critical juncture typically results from a disruptive event such as a war, depression, pandemic, natural disaster, or major social upheaval like a revolution or the collapse of an empire. Capoccia and Keleman (2007: 343) define critical junctures as moments when structural constraints are “significantly relaxed for a relatively short period, with two main consequences: the range of plausible choices open to powerful political actors expands substantially and the consequences of their decisions for the outcome of interest are potentially much more momentous.” Such moments do indeed create windows of opportunity for structural change, but the concept of critical juncture has two limitations. While critical junctures are retrospectively identifiable, it is often not clear during a disruptive event, or even in its immediate aftermath, whether it will become a historical inflection point. Some disruptive events have no long-term effect; some bursts of attempted change fizzle out or even reinforce the status quo. A critical juncture may only interrupt or marginally redirect institutional development without yielding a constitutive moment—that is, a moment when institutions are fundamentally and durably redesigned. The outcome may depend on, among other things, the strategies that political actors adopt and their ideas about alternatives. In other words, agency matters, not just circumstances.

The second limitation of the concept of critical junctures is that it does not apply to institutional change that takes place in slow-moving processes. As important as large-scale disruptions are, much institutional change grows from small beginnings in a snowballing process, sometimes as a result of generative reforms—that is, reforms that trigger further change. Reformers do not only take advantage of opportunities created by exogenous shocks and other events beyond their control; they sometimes make their own opportunities, or at least they prepare to make the most of whatever opportunities come their way. “Just wait for a critical juncture” is not helpful advice.

These abstract considerations take on heightened interest when the entrenched institutions and policies in question are deeply objectionable from the standpoint of human well-being. Entrenchment per se is a neutral concept; institutions with good and bad effects may become entrenched, and, as a subtype, policy entrenchment also has both potentialities. A policy trap, however, is policy entrenchment's pathological form: a self-reinforcing array of policies that simultaneously creates (1) well-established, often widely recognized failures in society, and (2) high barriers to change. By “self-reinforcing,” I mean that the forces sustaining a policy are greater after its enactment than before: enactment raises the bar against alternatives. As a result, a policy trap not only blocks collective gains but also affords no mechanism of self-correction.

Not all policy problems are the result of policy traps. Some are amenable to reforms without changing institutions at their foundations or confronting the resistance of those who have a stake in those structures. Other problems may be inherently difficult or even impossible to solve, given limitations of available knowledge or human nature. The notion of a policy trap assumes the feasibility of improvements that the political system is persistently unable to make. It is this combination of possibility and blockage that makes policy traps of such great interest.

The reader may object that policies and institutions in a democracy cannot create “well-established, often widely recognized failures” while having enough support to be sustained over the long run—but indeed they can, and for a great variety of reasons. By “well established” I mean according to the best evidence available about the impact of policies on human well-being and flourishing. By “widely recognized” I mean that the failures of those policies are the subject of extensive public debate, not just technical discussion, even if opinion is not unanimous. Failing policies do not persist only because of concentrated interests prevailing over diffuse ones. The responsiveness of democratic governments to popular opinion is subject to numerous confounding influences (Achen and Bartels 2016). A plethora of veto points in a government, as in the United States, may make large-scale institutional change difficult. At a practical level, people may recognize a need for alternatives, while they may resist change for ideological reasons, as in the famous contrast between operational liberalism and symbolic conservatism in the American electorate (Stimson 2004). Electoral systems of different kinds, such as the single-member district system (Rodden 2019), may systematically underrepresent some constituencies and overrepresent others. Oligarchical power may overwhelm other interests by constraining taxation and thereby confining government to limited measures (Winters 2011).

Although the identification of policy traps inevitably involves moral and political judgment, some instances should be recognizable even to many who disagree about their causes and remedies. Three policy traps serve as the basis for this analysis:

• * An array of policies in energy, transportation, and related fields has built up the dependence on fossil fuels that is responsible for global warming. Despite an overwhelming international scientific consensus on the threat and some important but still insufficient measures, the United States and other countries have been unable to take the political steps that would be necessary to avert catastrophic climate change.

• * The internet, originally expected to decentralize economic systems and improve the functioning of democracies, has instead led to the rise of online platform monopolies and threats to democratic values and personal autonomy. Although still a new phenomenon, online platforms benefit from the same mechanisms of entrenchment that have made monopoly a recurrent pattern in the history of communications networks.

• * The American health care system, with its distinctive combination of exceptionally high costs and inferior performance, has distinguished itself as an outlier among health care systems of high-income countries. A half century of reform has been unable to remedy the system's underlying problems.

This article begins with a theoretical discussion of entrenchment and some general options for overcoming policy traps. Then I turn to the two contemporary examples of carbon energy and online platform monopolies. The final section examines the evolution of strategies for escaping from the American health-policy trap.

I. Policy Entrenchment and Policy Traps

Entrenchment is not simply a condition. It is also a process: the development of stress resistance, the capacity to withstand pressure for change.¹ As circumstances change, that capacity typically requires renewal and adaptation. Intervening in that process—overcoming impediments to change created early in an institution's development as well as those that subsequently emerged—is the challenge facing those who want to break through entrenchment and escape a policy trap.

Two general types of mechanisms of entrenchment may be usefully distinguished. Strategic entrenchment is the active pursuit of irreversibility. Setting aside the use of brute force, strategic entrenchment typically involves intentionally designing features of an institution to make them hard to undo and to constrain and channel subsequent change. A paradigmatic example is “constitutional entrenchment”: cementing a rule or policy in a constitution, perhaps even designating it as unamendable, in an effort to prevent successors from reversing it. Other laws and regulations may be arranged in a hierarchy according to the rules for altering them (i.e., the “rules of change”). The level of entrenchment will then depend on the number of veto players (institutions or parties that can block change) and the thresholds of support required to overcome each potential veto. The more rigorous the rules of change—that is, the greater the number of veto players and the higher the thresholds needed to overcome them—the more entrenched an institutional arrangement will generally be. The designers of an institution may also attempt to entrench it by empowering those who benefit from it or by directing benefits to those who already hold power. Other mechanisms include insulating policies from opposing political forces by creating semiautonomous institutions such as central banks or by entering into international treaties or organizations that impose requirements from which a future government cannot readily extricate itself. When a political party or interest group reliably holds sway over a higher-level body—for example, a state government over a locality—it may be able to entrench a policy by preempting any alternative at the lower level (see, e.g., Michener in this special issue).

In addition, entrenchment can result from an emergent process that the originators of an institution or policy regime may or may not have had in mind. A paradigmatic example is lock-in from increasing-returns path dependence (Pierson 2004), which over time raises the cost of switching to an alternative (i.e., the costs of change). Those switching costs may arise from several mechanisms: high setup costs, gains from learning by doing and learning by using, network externalities, and adaptive expectations. In cases of political lock-in, the crucial question about switching costs is not necessarily the net social cost of an institution or policy, but rather their distribution—on whom the costs and benefits fall. The distributive impact is particularly relevant to policy traps, where the benefits of an institution or policy may flow to the powerful even though the net social costs far exceed the benefits. Like strategic entrenchment, lock-in from an emergent process always implies the exclusion or marginalization of alternatives. Lock-in implies lock-out.

A rational-actor analysis strictly in terms of who wins and who loses, however, would be an oversimplification. Lock-in can result from an emergent cultural process that naturalizes and objectifies institutions, making them appear to be inexorable facts rather than human creations. Dominant institutions are also sustained through ideologies that help recruit others to their defense besides those who benefit materially. Psychological and institutional status quo bias may add to the obstacles to reform. Taking advantage of the uncertainties and anxieties of change, the defenders of dominant institutions may try to magnify suspicions about alternatives and the motivations of those who advocate them. In democracies, some even find high principles to defend institutions that are patently undemocratic. In the United States today, the Electoral College and the Senate filibuster come to mind, but there have been far worse examples—slavery, for instance.

Although an institutional or policy trap creates barriers to change, the social failures for which it is responsible create a potential political base for change—and that base may grow under new conditions even if those conditions emerge only gradually. When changed conditions prevent institutions from working as they once did, or when basic understandings of the facts and standards of moral judgment have shifted historically (as they did as a result of religious and moral change in the case of slavery), even some of those who have defended an institution or policy may be open to abandoning it. But the sense that things no longer work or have become morally intolerable does not just appear by itself. Opponents must convey it. “Entrenchment,” Teles (2008: 270) points out, “is the joint product of the structure of the incumbent regime and the failures of rival agents.” The task of rival agents is recognizing where social failures create political vulnerabilities and using those vulnerabilities as a basis for shifting public opinion and mobilizing opposition from counterelites, popular movements, or both.

Although the initial imperative for opponents may be to discredit the status quo, reconstituting an institution or a domain of policy ultimately requires formulating an alternative. Without pretending to be exhaustive, here are four general options for breaking through entrenchment that are potentially usable individually or in combination:

1. Schumpeterian innovation: leveraging new technology or organization to bring about institutional change. Depending on the nature of a policy trap, disruptive technological and organizational innovation may be a means of change or even a necessary element of it (as it is in the case of the carbon-energy trap). The conservative version of this idea relies on market-driven technological development, perhaps aided by government in the form of exemption from taxes and regulations and procompetition policies. Change then comes about through the superior efficiency of alternatives that displace existing forms of production and organization—the Schumpeterian gale of creative destruction in its classic form (Schumpeter [1942] 1950: 81–86). Progressive policies may also seek to use new technology and organization to bring about large-scale change, albeit usually with more of an active role for public investment and regulation, possibly in the form of industrial policy. Direct government involvement in new technologies, on both the supply and the demand sides of the market, can break through obstacles such as the reluctance of firms to make long-term investments in high-risk innovations.

2. Globally oriented innovation: leveraging of international institutions. Instead of looking only to the market and technological innovation, reformers may look abroad to import or learn from other countries' policies or to use international negotiations to effect change both domestically and globally. The more global the policy trap—as, for example, in the case of the entrenchment of nuclear weapons in major-power arsenals—the more logical and necessary a move to the international level is. International connections in the form of the transfer of ideas and strategies for reform may also develop at the subnational level, contributing to institutional and policy diffusion.

3. Institutional conversion (repurposing): leveraging existing domestic structures. The models and means for breaking through a policy trap may already exist within a society. An institutional framework or policy platform may be converted to different ends from its earlier ones (Streeck and Thelen 2005). Alternatively, if developed only to a limited degree or only locally, it may be taken to a higher level, scaled up, and built on to break out of a policy trap.

4. Social creativity: the introduction of new institutions in the public sector and civil society. Social creativity is the nonmarket parallel to Schumpeterian creative destruction. Generative reform in government may begin by planting a seed through a policy experiment. Civil society may be especially important as a nursery for infant reforms, even if those can ultimately succeed only if adopted on a large scale commercially or governmentally. Social creativity does not necessarily mean that all the elements are new; much innovation involves the recombination of elements that may have previously been seen as unrelated or incompatible or the transposition of institutional models or ideas from one realm of activity to another (Padgett and Powell 2012).

Instead of just extending path dependence, all four of these strategies can lead to path departure—breaking out of a policy trap. A constitutive moment is a time when power looks to new ideas, and new ideas find their way to power; that is, when the politics of institutional change and strategically chosen policy alternatives come together. This convergence requires not just that political actors accept the case for change; they must also give change high priority to overcome the inherent difficulties of escaping a policy trap.

II. Two Contemporary Cases

Two contemporary cases—the tenacious hold of fossil fuels in energy production, and the domination of platform monopolies in online communication and markets—illustrate both the origins of policy traps and the strategic options for escaping them.

The Tenacity of Fossil Fuels

The carbon-energy trap has been long in the making. From the 19th century onward, industrial societies developed on the basis of fossil fuels, and governments adopted a host of complementary policies to promote economic growth on that basis. It was only after dependence on fossil fuels was literally built into energy, industrial, housing, and transportation infrastructures that scientists demonstrated that the resulting greenhouse gases were accumulating in the atmosphere and threatening catastrophic changes in climate. By the early 1990s, international institutions recognized that threat, but for the past 30 years emissions have continued to rise.

The barriers to shifting from carbon energy to clean energy exemplify some of the classic mechanisms of entrenchment. The built infrastructures for carbon energy provide a massive base for increasing returns, which keeps down prices for fossil fuels and the vehicles, machinery, and heating systems that make use of them. Without the benefit of installed systems paid for in the past, clean energy technologies have been at an inherent price disadvantage. Moreover, thanks to the sunk investments in existing infrastructures, the incentives for energy producers have for decades favored investing in further improving fossil-fuel production—for example, through fracking to increase the yield from oil and gas fields—rather than investing in new clean-energy alternatives. Carbon taxes might have corrected for the failure of the market to take environmental externalities into account, but governments have been unwilling to adopt them at the levels necessary for them to be effective.

The political barriers to needed changes in policy also reflect familiar mechanisms of entrenchment. The benefits of a transition to clean energy are diffuse, lie in the seemingly remote future, and would accrue in significant part to low-income, politically weak populations who will not have the means to protect themselves in the regions most endangered by rising temperatures and sea levels. In contrast, the costs of reduced emissions are immediate and fall on wealthy and powerful interests. The fossil fuel industry is only the most obvious of those interests. Its position is analogous to that of slaveholders in 1860 in the sense that its principal asset—the basis of its wealth and power—is in danger. If we are to avoid a climate catastrophe, much of the industry's oil and gas reserves must be left in the ground. The stranded assets of the energy industry potentially also include coal mines, drilling rigs, pipelines, tankers, processing terminals, and power plants. In view of what the industry has at stake, it is hardly surprising it has tried to entrench itself by financing efforts to deny the realities of climate change and lobbying against policies for a green transition. Public doubt and institutional inertia have become two of the industry's most important products.

Nonetheless, clean energy technologies have made progress and, in some sectors of the economy, are fully competitive with fossil fuels. From 2009 to 2020 in the United States, the costs of electricity produced by large solar farms fell by 90%, while wind energy prices dropped by 70% (Azevedo 2020). Although challenges remain—such as integrating clean energy into the grid, siting long-distance power lines, and providing for long-duration grid storage—supportive policies can accelerate the shift from carbon to clean energy, not only in energy production but also in major areas of energy consumption such as automotive transport. More challenging are hard-to-decarbonize sectors of the economy, such as aviation and the production of steel and cement. Change, however, can create a virtuous circle through generative reforms. The more that clean energy and related industries develop—the greater scale they reach, the more political influence they acquire—the more they close the gap with fossil-fuel producers in both increasing returns and political power. Successful innovation also raises public confidence in the feasibility of change. Although some have compared the economic challenge to World War II, the needed investments in the United States—that is, the additional energy expenditures above previous levels—amount to a projected 1%–2% of GDP through 2050, whereas federal outlays during World War II jumped to 43% of GDP from the prewar level of 10% (Sachs 2019).

But while the United States and other countries have the potential to transform their energy infrastructures, three decades of delay have made the challenge more daunting. According to the Global Carbon Project (2018), if decarbonization had begun globally in 2000, a reduction of 2% a year in emissions would have been sufficient to keep global warming below 2 degrees Celsius (the objective of the Paris Agreement). By 2020, the needed emission reduction had increased to 5% a year because of the continuing accumulation of greenhouse gases. By 2030 it is projected to be 9%. The popular resistance to such an abrupt transition and the repercussions it would bring may be more than the United States and other governments can bear.

Until Congress adopted a climate bill in August 2022, the American political system seemed incapable of enacting even modest changes to reduce greenhouse emissions. The 2022 legislation, passed entirely on Democratic votes and with the narrowest of margins (51 to 50 in the Senate), relies on tax incentives to promote electric vehicles and other clean-energy technologies. Except for a tax on methane gas emissions, it has no measures to penalize or limit dirty energy. The legislation asks for no sacrifices from consumers or industry; by using all carrots and no sticks, Democrats avoided energy-industry opposition. Nonetheless, the law is expected to have significant effects. Before its passage, US net greenhouse emissions were projected to fall 27% below their 2005 levels by 2030; with the legislation, they are projected to fall 42% (Jenkins et al. 2022).

The carbon-energy trap differs from other cases, however, in one crucial dimension: time, that is, how quickly entrenched obstacles are overcome. The speed of conversion to clean energy is critical. The 2022 legislation would still fall short of the reduction necessary to avert disastrous climate change, and it is by no means clear that the United States—or, for that matter, other countries—can move fast enough to close the gap. Republicans are unlikely to support those steps; indeed, in the past they have reversed domestic climate policies and international agreements put in place by Democrats, and they may do so again. This start-and-stop pattern has clouded the investment environment for alternative technologies and undercut US climate leadership.

Although breaking out of the carbon-energy trap is feasible, entrenched obstacles in several dimensions—technological, economic, political, even cultural—stand in the way. I return to the four general strategies described earlier. Schumpeterian innovation has been the central hope of many of those involved in climate reform. Radical transformations of technologies and markets are a necessity for achieving net-zero carbon emissions. Market-oriented strategies such as cap-and-trade, tax incentives, and carbon taxes have the aim of stimulating private investment in clean-energy breakthroughs and creating rising long-term incentives to adopt them. The idea of public funding for R and D in disruptive innovation lay behind ARPA-E, an agency established in the US Department of Energy in 2009 on the model of the Pentagon's Advanced Research Projects Agency, which financed the creation of the internet. Globally oriented innovation, such as the 1997 Kyoto Protocol, which sought to institutionalize cap-and-trade internationally, has had the same general purpose of indirectly promoting change in technology and markets. So have some efforts to convert or repurpose existing policy frameworks, such as the German use of industrial policy to promote solar energy. Since tax benefits have long been used to promote fossil-fuel production, tax incentives for clean energy might also be seen as a case of repurposing. Social creativity has often taken the form of both public and philanthropic funding of experimental clean-energy technologies and programs.

Despite the hopes invested in them, however, market-oriented reforms and voluntary international agreements have relied on what are, in effect, only nudges, and it is unclear whether nudges will push conversion fast enough. In the United States, the political right has effectively blocked what might be thought of as radical centrism, the use of neoliberal measures such as carbon taxes at levels high enough to break out of the carbon-energy trap. With the takeover of the Republican Party by Trumpian right-wingers, the center-right—the natural base of support for market-oriented climate reforms—has collapsed. Any carbon taxes that might be enacted are likely to be too small and gradual to bring about the accelerated, deep decarbonization needed. The two major international agreements, Kyoto in 1997 and Paris in 2015, have also amounted only to nudges because they included no penalties for noncompliance and no provisions for coordination. International coordination to meet the climate threat may, in fact, be impossible in an era without a global hegemon capable of enforcing compliance—a task for which the United States is peculiarly unsuited when one of its two major parties is the world's leading force opposed to climate reform.

In light of these realities, many progressives have concluded that the only climate reform equal to the challenge must be simultaneously politically transformative. While this is not to say it needs to be enacted all at once, the progressive breakout strategy is a comprehensive clean-energy industrial policy aimed at bringing about an accelerated turn toward net-zero emissions, providing aid to displaced workers and communities, increasing employment opportunities, and addressing inequities in environmental justice. This approach has come to be known as the Green New Deal, a fundamental rejection of both nudging and neoliberalism in favor of a more concerted and urgent use of governmental power. The 2022 legislation may be a first step in that direction—or an example of reforms that are too little, too late at a moment when we have come to realize that climate change is accelerating and we have less time to adjust than we thought.

The Entrenchment of Online Platforms

In some respects, the rise of online platform monopolies could not be more different from the persistent hold of fossil fuels. Dependence on oil and gas is a legacy of the industrial age, whereas dependence on online platforms is only a recent development, a phenomenon of a postindustrial information economy. But information is “the oil of our time” in more than one sense. Like oil, information technology has been the basis not only of individual fortunes and economic growth but also of entrenched structures of power and policy. The policy decisions and nondecisions made about both of them share the same tendencies toward lock-in, ensnaring contemporary societies in problems that are exceptionally difficult to escape.²

The internet was the great hope of the late 20th century for overcoming the legacies of industrial capitalism. As online communication and markets developed from the 1980s to the early 2000s, many analysts saw them as breaking up concentrations of power and wealth that they identified with industrial-age systems of mass production and mass media. So convinced were policy makers of the benign effects of new technologies that the only role they saw for government in the digital revolution was to promote it. In several critical areas of policy, they either refrained from taking action or relieved internet intermediaries of any liability. The three main kinds of policy that might have limited aggregations of private power—antitrust, regulation, and public ownership—were largely abandoned, in line with the general neoliberal premises of the era. In 1996, as part of telecommunications legislation, Congress immunized internet intermediaries from liability for user-generated content—a protection that would enable later social media platforms to publish content without the liabilities of a publisher (this in addition to judicial doctrines that since the 1960s have made it nearly impossible to sue for libel). Finally, the United States also failed to enact any legislation to protect privacy online, thereby allowing internet intermediaries to sweep up personal data without restraint and enabling the formation of enterprises whose entire basis is accumulating data about individuals and targeting messages to them aimed at changing their consumption choices and political behavior—the system of “surveillance capitalism” that underlies Google, Facebook, and other major tech companies (Zuboff 2019).

The net result of these policy decisions and nondecisions is diametrically opposed to the intentions of the internet's original visionaries. “The web is so popular, powerful, and omnipresent,” Hindman (2018: 3) writes in The Internet Trap, “that we forget just how badly it has failed at its original goal.” As Hindman points out, the web's original goal, as Tim Berners-Lee conceived it, was to create a system of communication with no centralized control; more than any medium in the past, the internet would foster democracy and free expression. But instead of decentralizing control, the technology has recentralized it. Instead of promoting democracy, it has facilitated the spread of disinformation, vilification, and extremism, undermining democratic values. And instead of advancing personal autonomy, it has become the basis of the most comprehensive, granular surveillance the world has ever known.

The barriers to reform are steep. Like energy infrastructures, communications infrastructures have tendencies toward entrenchment that make it difficult to reverse patterns of development. Lock-in from increasing returns has been a persistent pattern in communications, and for good reason. Communications networks often have sharply increasing returns because of high setup costs and network externalities; that is, they require large up-front capital investments and become more valuable as they connect more people. As a result, a network that gains an edge over its rivals often gains sufficient power to dominate markets and extinguish its competition. From the late 1840s to the mid-1860s, the telegraph industry in the United States went from wide-open competition to a national monopoly controlled by Western Union. The telephone industry went from a highly competitive phase beginning in the mid-1890s to a monopoly in the 1910s as AT&T took over effective control. Radio broadcasting went from pluralism and competition in the early 1920s to a network duopoly by the end of the decade with the rise of NBC and CBS.

During the mid-20th century, however, antitrust law and extensive telecommunications and broadcast regulation limited the power of network incumbents. AT&T, for example, was subject to rate regulation, requirements to connect its network with local independents, and mandatory licensing of its patents. NBC was forced to divest itself of one of its networks (which became ABC), and all broadcast networks were subject to rules limiting the number of stations they could acquire, the control they could exercise over affiliates, and aspects of their programming.

When the internet developed, many people believed it changed everything. But one thing it did not change was the tendency toward monopoly in communications. Today Amazon, Facebook, Google, Apple, and Microsoft dominate whole ecosystems of the digital world, controlling key choke points for commerce and news. But they dominate those ecosystems in the absence of any of the countervailing policies that checked their counterparts in the 20th century.

Antitrust and regulatory policies might have limited the growth of monopolies and abuses of market power. But since the 1980s, the federal government has greatly relaxed antitrust enforcement, thanks to the influence in the judiciary of theories associated with Robert Bork and the Chicago School. These theories hold that corporate dominance of a market is no problem if it improves “consumer welfare,” interpreted largely to mean lower consumer prices. That criterion has made it difficult to prosecute antitrust cases against the behemoths of surveillance capitalism such as Facebook and Google, which rely on advertising and user-generated content and charge consumers nothing, or against Amazon, which has sacrificed short-term profits for market domination.

During the struggles against monopoly power in the railroads and other industries in the late 19th and early 20th centuries, Congress and the courts took steps not only against horizontal mergers but also against predatory pricing, price discrimination, and vertical integration. In recent decades, however, the courts have tended to dismiss these concerns, failing to anticipate the new potential for monopoly power in digital platforms that benefit from network effects. As a network grows, the cost of being excluded from it may increase; for example, the potential cost to a seller from being excluded from Amazon Marketplace may be so high as to enable Amazon to set one-sided terms. In the digital world, scale also affords the capacity to extract data from users to train systems of machine intelligence; only the largest companies can compete effectively.

The role of social media platforms in the spread of falsehoods, defamation, and extremism has not been the result merely of lax oversight. It reflects the companies' underlying incentives to maximize profits by building scale and user engagement (“stickiness”). The algorithms that influence what users see on Facebook or YouTube (a Google subsidiary) have been optimized for those purposes. Unlike newspapers, the platforms developed without any tradition of editorial accountability; if the content favored by their algorithms was false and defamatory, it was not their concern. Furthermore, while the platforms were optimizing for high-voltage, sensational content, they were gobbling up the bulk of online advertising revenue and contributing to the collapse of regional and local journalism—in effect, a double assault on the underpinnings of democracy.

Since 2016, the dominant platforms, particularly Facebook, have elicited a backlash as concerns have risen about their market power, accumulation of personal data, and role in augmenting disinformation and extremism. The online spread of disinformation about COVID-19 and the 2020 election has heightened those concerns, and the companies have adjusted their practices, though it is unclear whether they can control the forces they have unleashed. How much of a change will be forthcoming in government policy is also uncertain. Neither the critics nor political leaders have a consensus about what to do, and partisan divisions stand in the way of action.

The online platforms are new enough that some may doubt the barriers to change are so formidable as to justify conceptualizing the policies that enabled their rise to monopoly positions as a policy trap. Perhaps as-yet-unanticipated changes in technology and markets will disrupt them. Perhaps the platforms will be able to reform themselves. Perhaps relatively modest changes in law will suffice.

Turnover in platforms is certainly possible, and some corporate self-regulation and modest legal changes could help limit the spread of disinformation. But the tendency toward monopoly in online platforms, their role in spreading false and extremist content, and their surveillance of users all stem from structural features of the industry and of viral communication. Political barriers to change arise from the familiar pattern of concentrated benefits and diffuse and opaque costs. In this case, the benefits are concentrated in the tech companies and their investors, and the costs are not only diffused among the public at large but also largely invisible. To many people—perhaps most—Google, Facebook, and other platforms seem to have no cost whatsoever because their services are free of charge. Consumers seem unaware that their lives, their movements, and even their emotions are being monitored and that they are being given cues (for example, in their news feed) aimed at modifying their purchasing or political decisions. The collapse of local journalism does not register as a cost at all.

For all these reasons, the policies that have structured online communications have the defining features of a trap: well-established failures and high barriers to change. Breaking out of the online-tech trap, unlike the carbon-energy trap, will probably not focus, at least in the near term, on either Schumpeterian technological breakthroughs or global negotiations. The more promising general strategies for overcoming entrenchment are likely to be the conversion of existing policy frameworks, particularly antitrust and regulation, and social creativity. Antitrust reform is not just a matter of winning a case against one of the platforms, though that would help; it requires overturning the Chicago School premises that have come to dominate judicial antitrust doctrine. And regulatory reform is not just a matter of tweaking the 1996 telecommunications framework; it requires adopting a new framework for digital platform regulation. Social creativity will be needed, too, in developing new public and nonprofit means of financing journalism without jeopardizing journalistic independence and integrity. Overcoming the online tech trap is going to be as difficult a journey as breaking out of the carbon-energy trap.

III. Escaping the American Health Policy Trap

In its 2021 crossnational analysis of the performance of health care systems, the Commonwealth Fund shows the United States to be an outlier, with a drastically lower level of overall performance and a higher level of spending than other high-income countries (Schneider et al. 2021). Since previous studies have found the same pattern, the findings are not a surprise, but they testify to the persistent hold of the American health policy trap. That trap has four principal elements (Starr 2013: 11), rooted in the system of health care finance:

1. Most obviously, the system is enormously profitable for interests in the health care industry. Since by definition health care expenditures equal health care incomes, inflated expenditures have inflated the incomes that are drawn from health care, and those who benefit are well prepared to defend their position.

2. The system has also generated support from the protected public, that is, large and well-organized groups such as seniors, veterans, and employees with good benefits.

3. While offering concentrated benefits to powerful constituencies, the system obscures much of its costs. In particular, it reduces the visibility of costs to many of those it protects, not only at the point of service but also at the point of insurance payment. Employees are not hit directly with the full cost of their coverage because their employer is supposedly picking up a large share, even though the costs of that coverage detract from wages and other compensation.

4. Major elements of the system—veterans' health care, employer health benefits, and Medicare—convey the message that health care goes to those who earn it, consequently inviting the interpretation that those who have not earned it do not have a right to it.

Taken together, these features of the system create a powerful combination of interest-group, popular, and ideological barriers to change, despite the well-established failures of the system.

The origins of the health-policy trap are historical. In particular, they lie in the series of defeats of tax-financed, universal health coverage, beginning in the Progressive era and culminating in the defeat of national health insurance in the late 1940s. The last of these defeats, coinciding with the start of the Cold War and the general collapse of the left, led union leaders and other advocates of universal coverage to retreat to two second-best alternatives: collective bargaining for health benefits for workers and a proposal for a program based on Social Security for hospital coverage for seniors, to be called Medicare.

Collective bargaining helped entrench employment-based coverage, which had already begun in the late 1920s and 1930s. In explaining the rise of employment-based health insurance, some accounts suggest it was just an accidental result of the exclusion of health benefits from wage controls during World War II and the subsequent codification in 1954 of the exclusion of those benefits from taxable income. But this account fails to explain what nonemployment-based method of spreading health costs might have developed after the defeat of a tax-financed public program. Compared to voluntary associations like fraternal groups, employer coverage has major advantages in controlling risk (because employees are a relatively healthy population) and processing insurance payments (because employers can deduct them from wages). Other forms of group health insurance, such as that of fraternal lodges, had been notoriously unstable, while individual coverage suffered from inherent problems of adverse selection and high administrative costs. Once national health insurance failed, employment-based coverage was an overwhelmingly likely outcome regardless of its tax treatment, especially given other conditions, such as the relatively high level of unionization in the early postwar period.

The second consequence of the defeat of national health insurance—the turn toward Medicare as second best—led to the establishment of a series of incremental coverage platforms, beginning with Medicare and Medicaid in 1965. The 1965 legislation was both generative and entrenching. It was generative in the sense that it created two platforms for coverage that could later be expanded; it also established a model for other incremental reforms filling in voids of coverage, such as the 1997 Children's Health Insurance Program and the 2010 Affordable Care Act (ACA). What the 1965 legislation entrenched, however, was a fragmented, complex payment system, and the subsequent incremental expansions made the system even more fragmented and complex.

The resulting complexity of the US system has had three dimensions: (1) an institutional dimension, requiring additional bureaucratic overhead; (2) a cognitive dimension, making the system's financing and organization hard to navigate, especially for consumers; and (3) a political dimension, as a source of anxieties about change. The additional overhead and consumer frustrations could be a basis for disrupting the status quo, but the fog of health care encourages Americans to defend whatever protection they have and to look anxiously on alternatives. Opponents of large-scale change perennially take advantage of that anxiety.

Some of the reforms proposed in the period from Medicare to the ACA could have worked out differently and become bases for escaping the policy trap; at least some reformers at the time certainly thought so. Several of these strategies are noteworthy from the standpoint of the general strategic options I have discussed.

Schumpeterian disruptive innovation was the general idea behind the promarket policies that originally focused in the 1970s on the creation of health maintenance organizations (HMOs) as an alternative to fee-for-service payment and independent medical practice. As in other areas, the hope was a radical transformation through the displacement of inefficient traditional structures by a less costly and higher-quality form of organization. The Kaiser Permanente plans and other prototype HMOs provided empirical evidence for that expectation, with savings chiefly resulting from sharply lower hospitalization rates. The same period also saw technological changes in medicine permitting the substitution of ambulatory care for in-hospital procedures; changes in insurance were intended to incentivize this and other economizing shifts. By the mid-1980s, alternatives to traditional insurance proliferated under the general rubric “managed care” but then brought about two reactions in the following decade: a consumer backlash against restrictions on choice and a consolidation among providers, as local hospital systems and other provider organizations merged or were acquired by larger systems, increasing their leverage against insurers and managed care. Despite a parade of market innovations in this period, the results in cost containment were not impressive in comparative perspective; indeed, the divergence in costs between the United States and other high-income countries grew more rapidly after 1980 than before (Schneider et al. 2021: 5).

International influence and the repurposing of domestic policy frameworks are confounded in the case of a second escape strategy—a universal, national health insurance program. Beginning in the early 20th century, the advocates of universal programs of social protection looked to European models of reform; Social Security and Medicare reflected that influence. The legitimacy gained by Social Security passed over to Medicare, enabling reformers to frame universal proposals as based on a proven American institution, even as they held up European experience in criticizing US performance on access to care, costs, and population health. In the late 1960s, Senator Ted Kennedy and other Democrats sought to extend Medicare universally, and in 1972, when many believed national health insurance was imminent, Congress did extend Medicare to the disabled and patients with end-stage renal disease. But the movement for national health insurance stalled in the mid-1970s and then collapsed with Reagan's election in 1980; in the following decades, incremental expansions took place through Medicaid and other means-tested programs, throwing obstacles into the path of a universal program organized along the lines of Medicare. Since the adoption of the ACA, the movement for a universal, federally structured program, now called “single payer” or “Medicare for all,” has taken off again, only to be caught once more in the health-policy trap.

A third strategy for escaping the trap might best be thought of as an example of social creativity, a recombination of elements previously thought incompatible. Under President Clinton in 1993, Democrats called for a system of universal coverage in which private HMOs, preferred-provider organizations, and fee-for-service plans competed for members through governmentally organized “regional health alliances” operating under a federally imposed cap on average premiums. Since the “alliances” were conceived as (near) single payers of plans in their regions, they would have effectively enforced regional global health budgets, an idea modeled on European examples. Many elements of the Clinton plan found their way into the ACA, but not the shift of employer coverage into the regional health alliances or the global caps on expenditures, which would have subjected the system to comprehensive financial control. Instead, the ACA's health care marketplaces became a supplementary platform, another complexity-increasing increment to the system.

The differences between the rejected 1993 Clinton plan and the enacted 2010 ACA illustrate a more general pattern: the conversion of potentially disruptive innovations into sustaining innovations. The groups with a stake in the existing financing system can sometimes be persuaded not to block reforms if the reforms support the financing status quo instead of threatening it. Their priority is preserving fragmentation on the payer side; a miscellany of uncoordinated public interventions to expand access to coverage is not necessarily a problem for them.

As the foregoing analysis suggests, while the health-policy trap has some elements in common with the carbon-energy and online-tech traps, it also deviates from them. The common element is the political imbalance created by concentrated gains and diffuse, opaque costs. But the carbon-energy and communication infrastructure are sources of increasing returns, whereas the health care finance system is not. If reforms in health care appear to raise the costs of change, it is chiefly because they increase the visibility of health care costs by shifting them to the public budget. Objectively, costs would probably fall—and it is precisely because expenditures would fall that the health care industry is resolutely determined to block what would, in effect, be a reduction in its incomes.

Although there are other examples of policy traps, I have not picked the carbon-energy, online-tech, and health-policy cases at random. They pose central challenges of our time. All three are examples of political failure of tragic dimensions.

At the beginning, I suggested that the concept of “critical juncture” has two limitations: a disruptive event does not necessarily yield a constitutive moment, and many changes emerge from slow-moving processes. Disruptive events from climate change—extreme weather events, droughts, and fires—are already increasing, but despite some belated progress, the United States has not yet undertaken the kind of transformation of policy needed for deep decarbonization. The COVID-19 pandemic has revealed severe problems of inequity in the US health system and population health, and the accompanying “infodemic” has shown how poorly online communication serves basic social interests in civility and survival. But it does not appear that the pandemic will turn into a constitutive moment of a constructive kind for either health care or online communication. The response to the pandemic has been an ominous signal about future responses to climate change: even immediate threats to life have not overcome ideologically driven denial and disinformation. While the pandemic has raised alarms about the health care and tech industries, the firms in both fields are doing splendidly.

In all three of these areas, reform is urgent precisely because we want to avoid critical junctures: devastating shifts in climate, and breakdowns in democracy and health care. Perhaps ultimately only the most powerful exogenous shocks will open the way to an escape from the carbon-energy, online-tech, and health-policy traps. But if that proves to be necessary, our generation will have failed.

Notes

References