Jump to ContentJump to Main Navigation
The Delegated Welfare StateMedicare, Markets, and the Governance of Social Policy$

Kimberly J. Morgan and Andrea Louise Campbell

Print publication date: 2011

Print ISBN-13: 9780199730346

Published to Oxford Scholarship Online: January 2012

DOI: 10.1093/acprof:oso/9780199730346.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 21 June 2018

Exploring the Delegated Welfare State

Exploring the Delegated Welfare State

(p.18) 2 Exploring the Delegated Welfare State
The Delegated Welfare State

Kimberly J. Morgan

Andrea Louise Campbell

Oxford University Press

Abstract and Keywords

This chapter develops the concept of delegated governance and applies it to the American welfare state. The chapter also places the American practice in historical, comparative, and cross-policy perspective, analyzing how the US differs from other countries, but also how delegated governance in the welfare state has evolved over time and differs across policy areas. Of particular interest is the shift over time from relying on non-profits to increasingly using for-profit actors and social welfare marketplaces to deliver publicly-funded benefits and services. The chapter also discusses the implications of these administrative arrangements for scholarly debates about how best to characterize bureaucratic power and authority in the United States.

Keywords:   governance, contracting out, welfare state, medicare, federal government

In recent years, scholarship on the American welfare state has increasingly focused on what some have called the “private welfare state” of employer-provided benefits and individual self-servicing that are subsidized through the tax code. This work has opened our eyes to the complex mix of public and private action that constitutes the American approach to social welfare. There is, however, another way in which public and private action is entwined. Even in seemingly public social programs, private actors often play a critical role in the delivery of benefits and services, and sometimes in the administration of programs themselves. In the case of Medicare, for instance, responsibility for administering the program has always been largely in the hands of private insurers and other nongovernmental entities, overseen by only a small number of federal civil servants. Similarly, in most states, managed care firms bear much of the responsibility for delivering Medicaid benefits.1 And since the 1996 welfare reform, a number of localities and one state government have contracted out the administration of their welfare programs to for-profit firms.2 By looking closely at the actual governance of the welfare state, we can see that much responsibility for social provision is delegated from federal authorities to state and local ones, and—of chief interest here—from public authorities to the private sector.

How extensive are these arrangements, and does it matter whether programs are administered by public or private agencies? This chapter develops the concept of delegated governance, maps the different forms it has taken over time and in different policy areas, and provides a comparative perspective on the phenomenon. As this chapter will show, the United States is not alone in relying on private actors in the delivery of social programs, but is distinctive in the degree to which it does so and in its enthusiasm for profit-making welfare providers—particularly in the reliance upon for-profit firms in publicly funded health insurance programs. The nature of delegated governance in the United States has varied over time: in its earliest phases the emphasis was on delegating responsibilities to nonprofit organizations and trusted professionals, both of which were viewed as public-regarding in their motives. Since the 1980s, however, there has been a policy shift toward (p.19) encouraging for-profit actors, market competition in social welfare provision, and consumer choice. Along with this has come a change in the locus of decision-making and risk. Whereas previously, governments made decisions about contracting out and bore risk for cost overruns, more recent forms of delegated governance put decision-making responsibility on individuals and shift risk from government to profit-making firms and individual consumers.

The chapter also raises some of the implications of delegated governance, a topic to which we will return in depth later in the book. The way programs are actually administered is no mere technical factor, but affects the real-world functioning and effectiveness of social programs, their redistributive effects, and their political consequences (or “policy feedbacks”). In addition, for academics less concerned with the nitty-gritty of social policy regimes, delegated governance has importance for how we conceptualize the nature of the American state—an issue that has received growing attention in recent years. We argue that scholars need to get away from abstract conceptualization of “the state” and look more concretely both at what states actually do and how they do it. An examination of the delegated welfare state takes us away from a view of the state as a unified entity—and away from conceptualizations of “strong” versus “weak” states—and instead reveals the multiplicity of functions in which states engage and how power relationships vary across governing structures. Being attentive to different forms of governance and their consequences for effectiveness, redistribution, and accountability helps us to better grasp the real-world meaning of governmental action.

Conceptualizing the Delegated Welfare State

Delegated governance is the delegation of responsibility for publicly funded social welfare provision to non-state actors. Rather than set up bureaucratic agencies to directly achieve a particular set of objectives, such as distributing benefits or providing public services, collective goals are realized through private entities that may be profit-making or nonprofit organizations. The government role is one of financing, regulation, and oversight, but not direct provision. By contrast, in a situation of direct governance, government agencies directly provide benefits or services. Some examples of direct governance in the United States include Social Security, which is administered by a federal agency, the Veterans Administration (VA) health care system—a domestic example of “socialized medicine”3—and, at the local level, public schools.

Why do we use the term “delegated governance”? “Delegation” is employed to highlight the deliberate act that is involved in assigning responsibility for social welfare to non-state actors. This is not something that simply evolves in the private sector; rather, a decision is taken to shift the provision of a social welfare benefit or service to the private realm.4 The term also reflects the notion that governing involves delegation from the legislative branch to the administrative (p.20) authorities that implement policy, as well as the legal “non-delegation doctrine” that limits the ability of Congress to delegate its lawmaking power to other public or private entities.5 Although often invoked around the delegation of power to executive agencies, some courts have imposed greater scrutiny on the delegation of governing authority to private actors whose self-interest may conflict with public objectives.6 In practice, however, delegation to private actors has occurred with great frequency, leading some scholars to conclude that the non-delegation doctrine cannot be used to limit the granting of authority to private actors.7 Regardless of where one stands on the legal question, the non-delegation doctrine embodies a similar concept to that which we are studying—the notion that when public responsibilities are shifted to non-state actors, this represents a delegation of governing authority to those entities.8

The term “governance” has many uses and thus also requires some explication. “Governance” has become a fashionable term in the scholarly literature on public administration, yet it has many meanings.9 For some scholars, the word “governance” conveys the idea that governing power is not confined to public authorities but is often exercised through private firms, international organizations, and other non-state entities. Scholars writing from this perspective often emphasize the decline of hierarchical, command-and-control governing relationships and their replacement by networks and other forms of horizontal relationships.10 Another approach is to use governance in a narrower and more traditional sense—the act of governing—and leave the site and nature of this governing to be specified, e.g., public sector governance, corporate governance, networked governance, etc. We would like to reclaim the original use of the term, which can then be qualified and made more precise. Rather than assume that governance involves diffused authority, extensive private involvement, and negotiated contracts, we assume only that governance involves governing and then examine how this governing takes place.

Why not simply describe delegated governance as other scholars frequently have, as “privatization” or “contracting out”? There are several reasons that we think delegated governance is an improvement on these two terms. Privatization has been used to describe quite distinct phenomena, ranging from the complete shedding of state responsibility for social provision, the selling-off of state-owned enterprises, or the contracting out of the provision of public services.11 Given these disparate meanings, we find that the concept has been stretched to the point that it tends to obscure the object of study rather than clarify it. In addition, privatization connotes the withdrawal of public responsibility in a particular area, such as complete load-shedding by public authorities. This more aptly describes phenomena such as the sale of state-owned enterprises, but fails to capture that which most commonly occurs in countries such as the United States—the intertwining of public and private responsibility.12 Moreover, administering public programs through private actors often entails substantial government involvement through financing, oversight, and regulation, contrary to what the term (p.21) “privatization” implies. Finally, privatization has acquired ideological connotations related to the free-market movement that emerged in the 1970s and 1980s in many Western countries.13 This renders the term less useful to describe the larger phenomenon of delegated governance, which predates the market reform movement and lacked that ideological orientation.

A term that is closer to that of “delegated governance” is the idea of “contracting out”—contracting with private providers to deliver public benefits or services.14 Although this more accurately captures the phenomenon of interest to us, and we use it at times, we still believe that delegated governance adds something to the existing social policy lexicon. In the arena of social provision, the government is doing more than simply contracting out service provision, as a municipal government might do for garbage collection, or the Defense Department might do in the procurement of weaponry. When responsibility for administering social welfare programs is contracted out, this effectively shifts power and authority over the lives of program beneficiaries—some of whom are deeply dependent upon these programs—from the state to non-state actors.15 Governance, and governing, is ultimately about power: the ability to make people do what they otherwise might not be able or inclined to do. Delegating governance is about moving this power from public to private actors.16 Although contracting out may technically describe the practice, it fails to convey the wider implications of it in the way that delegated governance does.

We see delegated governance as a subset of a larger phenomenon in the United States—the reliance upon indirect mechanisms and private actors to provide for collective welfare. By now, this phenomenon is well-known: in the 1980s, scholars began noting that private firms bear much of the responsibility for social welfare provision in the United States through employee benefits systems.17 This “hidden welfare state” is subsidized by features of the tax code, as is individual self-servicing in private markets for, among other things, housing (mortgage interest deduction), child care (dependent care tax credit), and health care (tax break for medical expenditures).18 In more recent years, scholars have more fully explored the origins and evolution of the public-private divide in social welfare, focusing on employer-provided pensions and health insurance benefits.19 More recent research has examined the system of subsidies via government-sponsored enterprises (e.g., Freddie Mac) and tax credits for the provision of higher education and housing.20

The result is what might be called, adapting a term developed by Elisabeth Clemens, the Rube Goldberg welfare state. In looking more broadly at the American state, Clemens argues that it functions less like the Weberian ideal-type and more like one of the complex machines dreamed up by Rube Goldberg.21 Thus, instead of directly exerting authority through centralized, hierarchically organized public bureaucracies, the American state has frequently relied upon non-state actors to achieve its objectives. This blurs the boundaries between public and private while obfuscating lines of authority (p.22) and accountability. The same is clearly true of the American welfare state, as evidenced by the above-mentioned means by which social policy goals are attained. Tax expenditures, government-sponsored enterprises, tax-subsidized employer-provided benefits—all involve indirect governmental involvement to secure social welfare aims.

We view delegated governance as one subset of this Rube Goldberg welfare state that has been neglected thus far by students of social policy. One reason for the research lacuna is that, in general, political scientists often end their analysis with the passage of policy, paying only limited attention to the way in which programs are actually implemented and administered. An important exception is the literature on policy feedback effects, which explores the ways in which public policies shape the consequent politics around an issue. As several scholars have argued, program design is of paramount importance, as it can influence the interest groups that subsequently mobilize, beneficiary self-perceptions of political efficacy or worth, and mass perceptions of the role of government.22 Outside of this literature, however, political science generally focuses more on explaining the passage of policy than on tracing its implementation and lasting effects.

On the other hand, scholars of public administration have been highly attentive to program administration and have long been charting the development of what they have called “third party government,” “government by proxy,” or “the hollow state.”23 Most of these scholars have not looked specifically at social policy, however, and many have been less concerned with analyzing the political origins and dynamics of delegating governance than with probing its implications.24

What Does the Delegated Welfare State Look Like?

In developing the idea of delegated governance, we do not intend to establish it as a “one-size-fits-all” concept. To the contrary, the way in which delegated governance has been executed in practice varies by policy area, and also over time. For that reason, we categorize in Table 2.1 a number of social programs into a spectrum that reflects both the degree and nature of delegation in various arenas and the spread of delegated forms of governance over time. Some programs have always had delegated elements, such as Medicare, which since its inception in 1965 delegated much of its administration to the nonprofit Blue Cross/Blue Shield plans and the actual delivery of health care to private-sector doctors and hospitals (as opposed to the Veterans Health Administration, which has its own system of hospitals and salaried physicians). A second category of programs includes those to which delegated elements have been added over time: welfare benefits, social services, the JOBS program, child support enforcement, and Medicaid (specifically the use of HMOs). These are programs that began with traditional government provision (although Medicaid, like Medicare, (p.23) always used private doctors and hospitals), but that increasingly have delegated administration and/or service delivery to private actors over time. The third category consists of programs that have been characterized by delegated governance from the outset: Medicare Advantage, Part D drug plans, Section 8 housing vouchers, school vouchers, and private prisons. Temporally, these programs tend to come later than the other programs as policymakers have increasingly embraced delegation as a mode of governance.

Table 2.1. Forms of Delegated Governance

Programs with DG Elements

Programs with DG Added Over Time

DG Programs

FFS Medicare

Welfare benefits (pre- post-TANF)

Welfare services


Child support enforcement

Medicaid (HMOs)

Medicare Advantage

Part D drug plans

Section 8 housing vouchers

School vouchers

Private prisons

Who administers program

Federal govt.; nonprofits; for-profits (esp. in Part B)

Pre-TANF: state/local governments Post-TANF: some nonprofits, for-profits

Pre-TANF: state/local governmentsPost-TANF: some nonprofits, for-profits.

Federal government

State and local governments

State governments; for-profits

Federal govt.; for-profits

Federal govt.; for-profits


Public, private lottery programs


Who delivers service

Private providers– nonprofits or perceived as non-market actors

Local governments, nonprofits, for-profits.

Mostly nonprofits, for-profits.

Mix of government, nonprofit, for-profit centers.

Local governments, for-profits




Private landlords

Private schools (nonprofits)


Locus of decision making

Government contract with intermediaries; individual chooses provider.

State, local government contract

State/local government contract; individuals w/ vouchers

Government contract

Government contract

Government contract

Individual chooses provider

Individual chooses provider

Individual chooses provider

Individual chooses provider

Government contract

Who bears risk

Government; providers.


Government; individual for costs above voucher amount.



Provider; individual for poor choice of plan.

Individual for poor choice of plan; provider bears market risk

Individual for poor choice provider; provider bears market risk

Individual for poor choice provider; provider bears market risk

Individual for poor choice of provider; provider bears market risk

Provider bears market risk

Can services be denied?

Yes, but rare (doctors refuse Medicare patients)

Yes: sanctions imposed.

Yes: sanctions imposed.

Care is ‘managed’: services may be denied.

Care is ‘managed’: services may be denied.

Drug use is “managed”: therapies denied (or more expensive)



Program Administration and Service Delivery

A quick glance at the first two rows of Table 2.1 reveals how heterogeneous delegated governance is in practice. We differentiate between program administration and service delivery and find that different entities—governments, self-governing professionals, nonprofits, for-profits—do each of these functions for different programs, and that for a given program different entities may perform each function. We also find that for-profits are more common in service delivery than in program administration, and are more common for both service delivery and administration among the programs that were created later and that have always been delegated.

Hence we have mostly nonprofits administering fee-for-service Medicare, welfare, and social services—government administering housing vouchers—and a mix of government and private actors administering the small number of extant school voucher programs.25 Even though the actual administration of social programs by market-oriented for-profits is less common, we do see it with Medicaid HMOs, Medicare Advantage, and private prisons. Service delivery is performed in some cases by a mix of government, nonprofits, and for-profits (as in social services and JOBS programs), but mostly government stays out of service delivery in the other areas, delegating responsibility either to private professionals and organizations (as with Medicare, housing, and school vouchers) or to for-profit firms (child-support services, Medicaid HMOs, Medicare Advantage, Part D plans, and private prisons).

There are two main points to take away, besides an appreciation for the complexity of these arrangements. The first is that there has been a movement toward for-profit providers over time, both because older programs, whose benefits were originally provided by government or nonprofits, are now using for-profits, and because newer programs have utilized for-profits from the outset. The second is that for-profits are particularly common in service delivery, with potential implications for the quality of provision. Profit-making entities may have incentives to meet consumer demand, but they also face incentives for cost-cutting and cream-skimming (serving only the “easiest” or “cheapest” customers). This is an empirical question, however, and some have argued that growing competition with for-profits and other voluntary organizations has spurred many nonprofits to behave more like proprietary actors.26

(p.24) (p.25) (p.26) Locus of Decision-making

An important aspect of delegated governance is the locus of decision-making: Who makes the decision to engage a certain provider, the government or the beneficiary? Traditionally, the delegated welfare state was set up such that the government would contract with a provider. The government solicits bids, prospective providers compete to gain the contract, and government selects a contractor, with that decision in place until the next contracting period. This is the pattern we see in Medicare administration, the JOBS program, child support enforcement, Medicaid HMOs, and private prisons. This differs from programs in which individuals, typically holding vouchers for services or an entitlement to benefits, select a provider on their own. Individuals select their own Medicare Advantage and Part D plans, their own physician under Medicare, their own apartment under Section 8, their own private school under the education voucher programs, and sometimes their own social service providers. Individual choice of provider is more common among programs that come later, when the concept of delegated governance—especially the consumer choice variant—was adopted more broadly, although it was present in Medicare and the initial incarnation of state Medicaid programs from the beginning.27

Shifting the decision-making locus from government to individuals has consequences for the nature, quality, and comprehensiveness of the information brought to the decision-making process. On the one hand, when government chooses a service provider, clients are stuck with that provider, which may not be to their liking. But in theory the government choice is a considered one, with much data and relevant criteria brought to bear. With individual choice, beneficiaries may be pleased with their autonomy, but they also may be paralyzed by too much choice or fail to use the most salient considerations to inform their choices. For example, in selecting Part D prescription plans, many beneficiaries are overwhelmed by too much choice—studies show they are more likely to select an optimal plan when faced with ten or fewer choices rather than the 40-plus they typically encounter28—and they tend to use the wrong criteria, weighting monthly premiums more than overall out-of-pocket costs, or choosing plans that cover the “doughnut hole” coverage gap even though their annual drug costs are well below the doughnut threshold.29 Unconstrained individual choice has its downsides as well as its appeal.

Risk Bearing and Service Denial

Another major issue concerns who bears risk under these various governance scenarios and who determines whether services can be denied. In most of the “older” programs, the government ultimately bears the financial risk of provision—if services cost more than projected, then the government absorbs the cost, with traditional Medicare being a chief example. Contrast that with the newer programs, in (p.27) which the provider bears the brunt of market risk: if Medicaid or Medicare Advantage capitated plans fail to deliver care under the budget they are allotted for each patient, they must absorb some of the cost. Similarly, private prisons must run their facilities under the contracted budget from the state or their profit margin will suffer. Private landlords and private schools bear the risks of taking in Section 8 or education voucher holders.30

In the most recently developed forms of delegated governance, individual clients themselves bear some risk. These programs offer their clients choice, but they also impose upon the clients the risks of poor choices. If a prescription drug that a senior citizen needs is not covered by the Part D plan they chose, or if their plan imposes unanticipated costs, that person has to remain in the plan until the annual open enrollment period rolls around at the end of the year. Individuals must similarly bear the risks of poor choice of a Medicare Advantage plan, an apartment, or a private school.

In addition, services can be denied by private actors in some forms of delegated governance. Doctors can refuse to take Medicare patients (few do, but more threaten to every time reimbursement levels are cut), and private landlords and schools can refuse voucher holders. Under Medicaid HMOs, Medicare Advantage, and Part D drug plans, care is supposed to be “managed” and channeled in the most cost effective way for the provider, which may be experienced as service denial by patients who wish to get a test or see a specialist or get a brand name rather than a generic drug but are prevented from doing so. The constraining effects of this are somewhat limited for Medicare recipients because they can exit a Medicare Advantage plan and return to the fee-for-service government program, and can choose among a wide array of Part D plans, although in general they can do this only once a year when the open enrollment period comes around. Medicaid recipients often lack those kinds of choices, as do welfare recipients who face the prospect of service denial determinations that are made by private entities.31

In sum, the delegated welfare state is extremely heterogeneous in form, involving a plethora of actors and administrative arrangements that have changed over time. One key change has been the shift in the locus of decision-making and risk from government to private actors and, in some instances, to clients themselves. The nature of delegation also has changed over time, as has the array of actors to whom responsibilities are delegated, with nonprofits and self-governing professionals joined later by for-profit firms. Although we use a single term to cover these multitudinous administrative arrangements, the implications and consequences of delegation do change. In some instances, what is delegated is the processing of claims so that a non-state entity merely implements rules determined by a government agency. Altogether different is delegation of the power to determine eligibility or benefit levels—in which a non-state entity exerts decision-making power over clients’ outcomes. In each case, one must think through the implications of shifting governing authority to private entities for accountability, distribution, and program effectiveness.

(p.28) A Cross-National Perspective on Delegated Governance

How unique is the governance of the American welfare state when viewed in cross-national perspective? Clearly, the United States is not alone in having “Rube Goldbergesque” arrangements for the delivery of social programs. The Weberian ideal-type of bureaucratic governance is just that—an ideal-type that has been frequently violated in reality by the concrete practices of social welfare regimes across the advanced industrialized world. In many countries, we find complex mixes of public and private social provision and blurring of boundaries between these two sectors. Although in some countries this has long been the case through heavy reliance on the voluntary sector, it has become more so with the adoption of marketizing reforms since the 1980s, which have further broken down state monopolies in health, education, and welfare; increased the role of private actors in the delivery of social programs; and promoted consumer choice in social welfare markets.

At the same time, however, it is important not to lose sight of the ways in which the United States has been, and remains, distinctive. Although a number of European countries long relied on voluntary associations to deliver social programs, these entities are often highly coordinated and serve essentially as state appendages. In the United States, the voluntary sector has always been less structured and more decentralized than that in western Europe. Although marketizing reforms and other forces have challenged voluntary association monopolies in western Europe, the same tendencies have done even more to fracture the sector of social welfare providers in the United States. In addition, although scholars have noted a trend in a number of western European countries toward increased reliance on private delivery of services since the 1980s, it remains a relatively recent phenomenon. In the United States, by contrast, we argue that virtually the entire construction of the American welfare state, outside of Social Security, occurred through systems of delegated governance and that this pre-dated the marketization fervor that took hold in a number of countries in the past three decades. Delegated governance is particularly pronounced in the health care arena, where the United States is distinctive in the degree to which it has delegated governance to non-state—and particularly for-profit—actors.

A final, fundamental difference lies in the continued existence of effective, Weberian bureaucracies in western European countries, despite reliance in some upon private actors for the delivery of services or administration of public programs. Outside of the Social Security program, the federal bureaucratic infrastructure around health and welfare programs in the United States is distinctively fragmented, uncoordinated, and in some cases quite small. This is especially pronounced in the field of health.32

(p.29) Delegated Welfare States Abroad

In the American popular press, European welfare states are often lumped together and tagged as state-run socialism. In fact, direct governance of social programs—in which the state not only funds but also directly provides social benefits and services—is relatively rare and most comprehensively found in the Nordic countries, where social benefits and services are funded through general taxation and provided by public bureaucracies. For example, day care centers in Nordic countries are largely publicly run, and these countries have integrated health care systems in which government not only funds health care but also provides it.33 A number of other countries, including the United Kingdom, New Zealand, Australia, and several southern European countries utilize direct governance in health care, having created integrated health systems—often called national health services—in which public authorities play a major role in the delivery of care.34 In these systems many health care personnel are public-sector employees—although some may be private contractors receiving public funds, a practice that is particularly common in southern Europe35—and the hospital sectors are largely public.36 Until the 1980s, the British public housing system was, in addition to being quite extensive, directly administered by local authorities.37 Outside of these exceptions, however, direct state provision of social services did not extend to most other areas of social welfare, and these responsibilities were effectively left to families and voluntary organizations.

Non-state actors have long played a vital role in the administration of health, education, and welfare programs in much of continental Europe. In countries such as Germany and the Netherlands, nonprofit welfare associations have for decades been the preeminent providers of educational and social welfare programs.38 In the Netherlands, for instance, 70 percent of primary and secondary school students are in publicly funded, privately run schools.39 Moreover, the “Bismarckian” welfare state model, found in countries such as Austria, Belgium, France, Germany, and the Netherlands, employs non-state insurance funds to distribute benefits such as pensions, health insurance, and family allowances.40 Usually governed by employers’ associations and labor unions, these insurance funds effectively “share public space” with state actors.41 In Germany, there are hundreds of nonprofit sickness insurance funds that are governed by representatives of business and labor and provide coverage to all. Similarly, in France, people receive their health coverage through private health insurance funds that collect payroll taxes and pay providers, and family allowances are distributed through non-state entities—family allowance funds.42 The actual power of the social partners in managing social insurance funds does vary across continental Europe; in some cases, parliamentary decisions set the main policy parameters, whereas in some other countries more policy-making power has been delegated to the social partners.43

(p.30) In these countries, as well as in Canada, the national health service model was rejected in favor of a system of publicly funded insurance coverage and privately delivered care. Thus, medical facilities are largely privately run and physicians are not salaried public employees, as they would be in a national health service, although hospitals may be publicly run. Individuals choose their own health care provider whose costs are reimbursed by the insurance fund. Each country differs in its public-private mix, however; France, for instance, has both an extensive public hospital system as well as the largest private hospital sector in Europe.44 France also has a large sector of private, supplemental health insurance that covers gaps in public insurance plans, and in Germany, formally private insurance funds serve higher income clientele. In short, in continental Europe, the governance of social programs has long been distributed across a wide array of state and non-state actors.

Even so, the delegation of governance in the American welfare state does differ from that found in other countries. In much of continental Europe, the non-state welfare and educational sectors have been fairly centralized and have had an institutionalized function as welfare provider and partner in corporatist governing arrangements. Thus, nonprofit welfare associations in Germany have long been members of centralized peak associations that, since the 1960s, have been “functional equivalents of public sector institutions”45 or a “branch of government”46 that are part of a corporatist regime.47 Similarly, Dutch nonprofit social service providers originally were independent of the state but gradually became “quasi-public” entities in the post–World War II period48 for which, in some cases, government payments covered nearly 100 percent of staff and program costs.49 Social insurance funds in continental Europe are semi-public entities whose actions have been increasingly subject to state constraints, such as global budget caps and greater parliamentary controls.50 As will be discussed later in this chapter, the system of delegated governance to nonprofits in the United States is considerably more decentralized, which undercuts the coherence of policy interventions.

Perhaps a closer parallel to the United States is the adoption of market-based reforms in many Organisation for Economic Co-operation and Development (OECD) countries since the 1980s, which altered the governance of some social programs by increasing the role of non-state (and often for-profit) actors in service delivery.51 Probably the most-publicized reforms along these lines were in the United Kingdom under both Conservative and Labour governments. For instance, starting in the 1980s, public housing was sold to tenants and the remaining system was converted from a largely state-run affair to one in which nonprofit housing associations play a growing role.52 The long-term care sector in the United Kingdom was also transformed, with a significant increase in the role of for-profit care providers. By 2004, private institutions accounted for 88 percent of users of residential facilities, and the public provider share of home health care was just 31 percent, down from 95 percent in 1993.53 Between 1997 and 2010, New Labour governments dedicated greater resources to social welfare programs but often did (p.31) so by subsidizing consumers and non-state providers of social services, thereby moving “the center of gravity of governance away from the bureaucratic state.”54

Similarly, the child care and old-age care sectors in many countries have been shaped by a shifting mix of government, nonprofit, and for-profit providers and a growing emphasis on consumer choice and voucher-type support. In Germany, the long-term care insurance program created in 1994 allowed for-profit providers into the system, and they now make up about half of long-term care providers—a clear departure from the past practice of directing subsidies to voluntary organizations.55 In France, rather than continue to develop expensive, municipally run child care centers, successive governments have subsidized parents who purchase care (and other domestic services) on their own. In 2006, the Netherlands converted its child care subsidy system into one that reimburses parents through the tax code for the care they purchase in private markets.56

Perhaps more startling are the reforms that have taken place in Sweden, where there was always a strong commitment to direct state provision of services in order to assure equal access to high-quality programs. Since the 1980s, the role of non-state (and often for-profit) actors has grown in health care, old-age care, child care, and education.57 The public school system has been converted to a voucher-type system that allows private schools to receive public funds and compete for pupils, and old age care has been transformed from an almost entirely state-operated sector to one in which for-profits play a significant role in running residential facilities and home care services.58 The child care sector also has been altered somewhat since the 1990s, as private providers now account for about 20 percent of centers, most of these being parental cooperatives or other nonprofit associations. Finally, since 2003 when a major reform to the pension system was fully implemented, a portion of an individual’s pension contributions goes into an individual investment account, with individuals able to choose among a large array of funds.59

How significant are these kinds of changes? Although there have been some clear shifts in the governance of welfare programs in a number of countries, in most places they come nowhere near the kinds of reforms that took place in the United Kingdom. Neo-liberal reform movements have held much less sway in continental Europe,60 and thus have only minimally altered the governance of welfare regimes in Austria, Belgium, France, Germany, or Italy, for instance. In many countries, the nature and role of the voluntary sector has changed, but this is not necessarily due to a drive toward marketization rooted in neo-liberalism. In France, for example, a leftist critique of bureaucracy spurred the drive to decentralize governance from Paris to regional and local governments, and to augment the role of the nonprofit sector.61 In Germany, there also have been changes in the nature of the German voluntary sector, but not due to marketizing reforms. Instead, new organizations have successfully challenged the privileged status of the traditional voluntary groups, thus opening up equal funding opportunities for these challengers and rendering the voluntary sector more pluralistic.62

(p.32) In addition, those countries that have adopted marketizing reforms have done so in the past three decades, with some initial reforms in the 1980s and an accelerated marketizing trend in the 1990s or 2000s. In other words, these are relatively recent phenomena for welfare states that scholars long characterized as having hierarchical, bureaucratic, and centralized forms of governance. By contrast, the U.S. welfare state has always relied heavily upon delegated governance as its mode of administration and service delivery, and this way of structuring social programs predated the marketization movement. Explaining the governance of the American welfare state thus requires one to dig deeper than the international privatization movement that began to take hold in the 1980s.

Finally, a basic difference between the United States and western European welfare states concerns the role of bureaucratic authority in structuring public-private relationships. In France, for instance, where many elements of the “old” American health care system—including fee-for-service medicine and independent physicians—have been maintained, and health insurance is paid through para-public funds, the central state weighs heavily upon the entire medical sector. From the state’s control over medical schools to its role in setting reimbursement rates for pharmaceuticals, bureaucrats in France play a vital role in shaping the organization of the medical field and the prices charged.63 In the Netherlands, despite reliance upon private organizations to deliver many social and educational services, the expansion of the welfare state post-1945 spurred the growth of administrative authority, producing a “strongly centralized Dutch state” that intervened considerably in how non-state actors delivered these programs.64 We can find this across western Europe—the presence of welfare state bureaucracies that, while one should not exaggerate their power, have considerably greater authority in managing social programs in their societies than is, by the large, the case in the United States (with the exception of the Social Security Administration).

In sum, there have been changes in how government programs are actually delivered, but in many countries these changes have been limited, at least so far.65 Although the importance of marketizing and privatizing reforms in recent years should not be downplayed, in most cases these developments have not altered the basic architecture of social welfare systems along American lines of not just delegated, but highly decentralized, governance that is often ineffectively overseen by bureaucratic authority.

The Limits on Market-Based Reforms in Health Care

Another notable observation that emerges from a comparative perspective on governance is the limits of marketizing or privatizing reforms in the health care arena, the area where marketization has been pursued the most in the U.S. context. Instead, the rhetoric surrounding privatization and marketization of health care generally exceeds the realities of what has taken place, as few countries have (p.33) enacted (and stuck with) reforms that significantly increase the role of private actors in this arena.66 In some countries with national health services, such as New Zealand, Sweden, the United Kingdom, and much of southern Europe, governments created an internal market by dividing purchasers from providers, so that the governmental entities that previously were responsible for both are now choosing services from an array of competing providers. Although this introduces some degree of competition into these systems, it has not led to the empowerment of private actors in the delivery or purchasing of health care; instead, most actors in these health care systems remain public. Even in countries that are said to have enacted major reforms to their public sectors (Czech Republic, Netherlands, New Zealand, Sweden, UK), the public sector remains, by far, the largest payer of health care.

In the Bismarckian systems, in which health insurance funds have always been non-state actors, a number of countries have required those funds to compete for beneficiaries’ business. In general, this did not spur the growth in for-profit insurance entities or generate strong competitive pressures that fundamentally changed the architecture of these systems. In Germany, the insurance funds increasingly offer special benefits in order to compete for beneficiaries and have gained some power as health care purchasers. At the same time, however, state intervention has steadily increased in what was once a self-governing insurance system, with the state setting sectoral budgets and mandating a fixed contribution rate for insurance coverage.67 In Belgium, reforms sought to change the incentives facing the insurance funds toward more cost control but also limited the degree of competitive pressures on the funds.68 Sidestepping the pro-competition movement altogether, France has experienced a steady étatisation of its health care system that has diminished the power of the para-public insurance funds and the social partners in managing the system.69

Moreover, resistance against market-oriented reforms in health care has impeded the drive to delegate governance to private actors. In New Zealand, for instance, policy-makers in the early 1990s sought to transform their integrated health care system such that public and private purchasers (insurers) would compete for business—not unlike some of the market-based reforms that have been proposed for Medicare. After strong opposition was voiced, including fears that the “Americanization” of the health care system would result (by which was meant a system based in markets that is both wasteful and exclusionary), the plan was dropped.70 Thus, after all the hoopla surrounding the reform proposals, the New Zealand health care system ended up being only “marginally more private” than it was before the reforms.71 In the United Kingdom, resistance to turning more of the health care sector over to private forces actually came from the Conservative government in power, which ultimately prioritized fiscal austerity over market-based reforms. Allowing consumers more choice and/or decentralizing decision-making to non-state actors threatened to increase costs, not reduce them, in the (p.34) immediate term. Alan Jacobs sums up the UK experience well, concluding, “Consumers would have a voice in the new NHS. But it was to be a faint whisper, muffled by the thunderous engine of public administration.”72 As Doorslaer and Schut conclude more generally about the drive to shift more authority to non-state purchasers of health care:

Are the respective governments willing to hand over some of their traditional supply-side regulatory tools (contracting, fee setting, quality control) to individual insurers? In the logic of the managed competition model, the answer self-evidently has to be “yes,” but in practice there appears to be substantial reluctance on the part of governments and regulatory agencies to give up these [regulatory] instruments and “jump in[to] the competitive dark.” This reluctance seems to be fueled primarily by the fear that insurers individually may be less successful in resisting new demands by patients and providers for more resources, and that such a weaker negotiation position may result in cost increases.73

Ultimately, cost control would be the driving force shaping health care reforms, which often produced more government intervention through regulatory and spending controls, not less.74

Two countries outside the United States that have promoted consumer choice in health insurance markets are the Netherlands and Switzerland.75 Both systems require individuals to buy insurance from an array of private plans. These plans are heavily regulated to prevent cream-skimming: companies must offer a basic plan of benefits; cannot turn any applicants away; must charge community-rated premiums; and cannot make a profit on the basic plan (in both countries, insurers can make profits from supplementary benefits they offer to consumers). Both countries also have a risk equalization scheme to compensate funds that end up with a sicker patient load. Finally, individuals with incomes below a certain level receive income-adjusted subsidies to help them cover the cost of health insurance—an arrangement that is considerably more generous in the Netherlands, where more than 50 percent of households received such a subsidy in 2008, than in Switzerland.76 In short, the Dutch and Swiss health insurance models resemble the premium-support model advocated in the United States for Medicare, in which beneficiaries would receive subsidies to purchase care from a regulated, private insurance market of plans.

Despite these apparent resonances with the American health care system, the insurance markets in the Netherlands and Switzerland are still tightly regulated and the state plays a large role in controlling costs. Okma argues that the Dutch health care system is less a model of market competition with only modest state interference than a system with a “modest amount of competition, with a large dose of government regulation.”77 The Dutch Health Ministry not only heavily regulates insurers but uses global budgets and price controls to keep down health (p.35) care costs.78 Similarly, according to Reinhardt, there is “pervasive government regulation that guides the Swiss health care system,”79 including federal setting of formularies and maximum prices for pharmaceuticals, federal determinations over which lab tests and medical devices are covered by the mandatory insurance plan, and canton-level planning of hospital and nursing home capacity. Although insurance companies negotiate with health care providers over the cost of care, they do so “under the watchful eye and heavy hand of government.”80

Thus, even in countries that have sought to introduce more competition into their health care systems, public bureaucracies still weigh heavily upon these systems, regulating insurance plan competition and shaping the price of health care services and technologies. Non-state actors may play important roles in the provision of health and welfare services and/or the administration of benefits programs, but the influence of the state remains significant.

Why Did the Delegated Welfare State Emerge in the United States?

Why has the United States developed a “Rube Goldberg welfare state” of delegated and diffused authority? We argue that the interplay of public opinion, political institutions, and organized interests shaped choices about the way social programs would be delivered. Delegation has proven a way for policy-makers to address a basic dilemma at the heart of American public opinion: that Americans simultaneously want small government but also social protections from life’s risks and perils. Delegating social welfare functions to non-state actors has been a way to satisfy the public’s desire for an apparently small governmental sector while also securing a broad array of social welfare programs that might have been blocked if administered directly through public bureaucracies. In addition, decentralized administration of programs emerged as a way to appease Congressional moderates and conservatives—such as Southern Democrats—who opposed the growth of the federal government, and to secure the support of powerful interests, especially in the health care sector. Such groups were often opposed to and successful in blocking direct governance but were quite happy to provide the same services themselves with government paying much of the bill. Delegation has thus been a way to build coalitions and negotiate the shoals of a veto-ridden, status-quo oriented set of policy-making institutions. Over time, the rationale for delegated governance changed, as market advocates pushed for delegating authority to risk-bearing, profit-making actors in the hope of improving government efficiency and performance. However, beneath these new rationales lay the same political calculus: delegated authority as a way for anti-statists to respond to public demands for social support without increasing the size of government or antagonizing private interests.

(p.36) Less Persuasive Reasons for Delegated Governance

The large body of research on both the welfare state and administration of government programs offers some potential explanations for delegated governance. Perhaps the first explanation that might jump to mind is simply that delegated governance makes sense for technical-administrative reasons: it may be an efficient and effective way to administer programs or deliver services. For instance, in contrast to a relatively straightforward program like Social Security, Medicare may be too large and complicated to directly administer. Thus, delegating those administrative responsibilities to private insurers and accreditation organizations, and delivery of care to private providers, could have been a logical choice on the part of policy-makers. Moreover, in more recent decades, some have argued for administrative and delivery arrangements that stimulate market competition in the belief that this will improve the cost-effectiveness and quality of public programs. For these and other reasons, it could be that policy-makers have been choosing the best possible means of administering public programs.

It is less than obvious, however, that delegated governance in any of its forms has been the more effective or efficient way to administer programs and deliver services. As chapter 6 discusses, there is a large body of research questioning whether contracting out improves program delivery, and local governments have been moving away from such arrangements after promised savings and efficiency gains did not materialize. Beyond the fact that the merits of this approach are contested, such an explanation is functionalist and apolitical, positing that when a policy is needed or makes sense, it is adopted. Yet much decision-making in the political sphere fails to follow an efficiency-maximizing logic.81 As political processes determine what kinds of programs are adopted and how they are administered, we must look carefully at those processes to uncover the forces driving policy change.

Research comparing the welfare states of rich democracies offers another potential explanation—the weakness of left-wing political power in the United States. Not only have left-wing political parties often been strong advocates of income redistribution but they have typically favored centralized and direct state administration of programs and delivery of services so as to prevent inequalities in access to programs of equal quality.82 Certainly, the relative weakness of labor unions in the United States, and the long-time fracture within the Democratic Party between Southern conservatives and Northern liberals, impaired the growth of the welfare state.83 Instead, business groups have been highly influential in American politics, and while they have not always been antipathetic toward the growth of social welfare provision, they have had distinct visions of what the welfare state should look like.84

We view these political factors as a crucial backdrop for thinking about the politics of the welfare state in the United States. Yet they beg many questions about why, given ample public support for social programs, a more unified Left (p.37) did not emerge to represent these views. Attention to the institutional structures of American politics, as well as to the power of organized interests in the policy-making process, certainly offers some answers (treated below). Moreover, as Christopher Howard has persuasively shown, the welfare state has often significantly expanded under Republican political control, or otherwise been vitally shaped by powerful Southern Democrats.85 These works suggest that we should not view conservatives or centrists solely as a blocking force against the welfare state, but rather one that favored particular forms of social provision. Why then were conservatives sometimes willing to countenance, or even actively pursue, the growth of social programs? The answer to that question lies, in part, in the structure of American public opinion.

The Basic Dilemma at the Heart of American Public Opinion

Observers of public opinion in the United States have long noted a fundamental ambiguity: that many members of the public hold anti-government views in the abstract but also very much like what government does when it comes to specific instances and programs. They are, to use the terminology of the first researchers to describe this phenomenon, “ideological conservatives” and “operational liberals.”86 Large majorities hew to an anti-government line when considering the role of government in general. However, when asked to consider specific government functions or programs, the very same individuals who described themselves as conservatives or who distrust government respond with much enthusiasm when asked whether government spending should be increased in various concrete areas.

Public opinion pioneers Lloyd Free and Hadley Cantril first reported this phenomenon during the 1960s. Their survey data showed that large proportions of Americans described themselves as conservative or said government has too much power or that “we should rely more on individual initiative and not so much on governmental welfare programs.” But what they also discovered was that similarly large proportions of respondents—from two-thirds to three-quarters—also said that spending by the federal government on a number of programs should be kept at the present level or increased: Head Start, job retraining, federal aid to education, Medicare, grants to build low-rent public housing, unemployment reduction, and so on.87 Subsequent researchers demonstrated that these two sides of public opinion were not merely a phenomenon arising from a particular epoch in American politics. Analyzing survey data collected 40 years later, Benjamin Page and Lawrence Jacobs uncover the very same patterns: skepticism about government and an ideological commitment to free enterprise remain widespread, but so too does considerable support for government help in areas from nursery schools to health insurance to retirement.88 Analysis of survey respondents’ reasoning about their beliefs shows that these seemingly contradictory viewpoints arise from the diverse values embedded in American political culture, values that (p.38) are themselves contradictory: individualism, liberty, and opposition to big government on the one hand, but humanitarianism, equality, and a recognition that some need help, on the other.89

Analysis of the 2008 American National Election Study (ANES) illustrates these points and shows that it is the very same individuals who often profess these seemingly contradictory preferences. Significant proportions of Americans identify as conservatives in questions that tap either their ideological self-identification or their general orientations toward government. In the 2008 ANES study, 32 percent of respondents identified themselves as conservative, compared to just 21 percent who identified as liberal (the plurality of respondents, 47 percent, either identified as “middle of the road” or said that they had not thought much about it). Thirty-seven percent said they were Republicans or were independents who leaned toward the GOP. Even larger proportions professed skepticism about government in other ways: 70 percent said they trusted the government to do the right thing only some of the time or not at all (up from 54 percent just four years earlier), 73 percent said the government wastes “a lot” of the money they pay in taxes, and 70 percent said that they thought the government in Washington was too strong.90

However, although many Americans profess to being conservatives in the abstract, when we examine preferences about specific government aims and programs, Americans turn out to be operational liberals. Means-tested programs are an exception: in recent ANES surveys, just 17 percent of all respondents wanted spending on food stamps increased and 27 percent favored higher spending on welfare (Table 2.2, column 1). But when asked about spending on vulnerable groups rather than reviled welfare programs, support is much higher—61 percent want increased spending on the poor and 58 percent want more spending on the homeless. Programs that would benefit middle class Americans garner even higher levels of support, with 53 percent desiring increased spending on financial aid for college students, 58 percent for child care, 66 percent for Social Security, and 75 percent for public schools.91

Table 2.2. Desire for Increased Government Spending among All Respondents, Conservatives, Republicans, and Government Skeptics, 2008

(1) All Respondents

(2) Conservatives

(3) Republicans

(4) Trust Government None/Some of Time

(5) Government Wastes a Lot of Tax Dollars

(6) Government in Wash. too Strong

Food stamps



8% c










12 c

The poor







The homeless



43 b




Financial aid for college



43 b




Child care







Social Security







Public schools







Note: Cells show percentage of respondents of each description who want program spending increased (as opposed to decreased or kept the same). Data from 2008 except a1992; b1996; c2000.

Source: American National Election Study cumulative file.

Where ambiguity in American public opinion becomes most apparent is in the programmatic spending preferences of those who say they are conservative, who identify with the Republican Party, or who otherwise hold negative views of government (Table 2.2, columns 2 through 6). Not only do Americans overall want increased spending on many social welfare programs, so do self-described conservatives and government skeptics. Two-fifths of conservatives and Republicans support increased spending on the poor, the homeless, child care, and financial aid for college, and one-half to three-fifths want increased spending on Social Security and public schools. The proportions desiring increased spending in these areas are the same or even greater among those who trust government none or only some of the time, think government wastes a lot of the money it collects in taxes, and believe the government in Washington is too strong—majorities of respondents who are skeptical about government by these measures nonetheless want (p.39) increased government spending in nearly all of these areas (Table 2.2, columns 4 through 6). Contradictory strands in American public opinion thus appear not just at the aggregate level but are characteristics of individual-level preferences.

Indeed, this attitudinal ambivalence is widespread among the public. Figures 2.1 and 2.2 take two of the issue areas discussed above, Social Security and spending on the poor, and analyze spending preferences among different population subgroups. These figures show the percentage of the total sample and of each subgroup that falls into one of five categories (arrayed from left to right in the figures): liberals; moderates who support increased spending in that area; conservatives who support increased spending in that area; moderates who want spending kept the same or decreased; and conservatives who want spending kept the same or decreased. The second and third groups—moderates and conservatives who want spending increased—are shaded gray (p.40) (p.41) (p.42) and represent the most Janus-faced members of the public: ideologically they state that they are moderate or conservative, but operationally they are liberal, desiring increased federal government spending on Social Security or on the poor (a very small proportion of liberals are ambivalent in the opposite direction, desiring a decrease in spending; we do not break them out separately). Note that ideological and operational ambivalence is widespread and not confined to certain demographic or political subgroups: we see large proportions of individuals by gender, race, income, party identification, and turnout—often majorities—holding ambivalent views. Also note that the rightmost two groups—moderates and conservatives who want spending in these areas kept the same or decreased—that is, those who are consistent in both ideologically and operationally embracing small government—are in the minority, both overall and within every subpopulation (the one exception being that a small majority of Republicans are both ideological and operational conservatives when it comes to spending for the poor). Far more common overall and among most subgroups is ambivalence rather than ideological and operational consistency.

                   Exploring the Delegated Welfare State

Figure 2.1 Ideological and Operational Ambivalence: Social Security.

Source: American National Election study cumulative file.

Note: Figure shows, from left to right, the proportion of respondents overall and of each demographic/political category who are liberals, ideological moderates who want federal government spending on Social Security increased (ambivalent group, in gray), ideological conservatives who want spending on Social Security increased (ambivalent group, in gray), ideological moderates who want spending on Social Security kept the same or decreased, and ideological conservatives who want spending on Social Security kept the same or decreased. Income is in terciles. The Republican and Democratic categories include leaners. Data are from 2008.

                   Exploring the Delegated Welfare State

Figure 2.2 Ideological and Operational Ambivalence: The Poor.

Source: American National Election study cumulative file.

Note: Figure shows, from left to right, the proportion of respondents overall and of each demographic/political category who are liberals, ideological moderates who want federal government spending on the poor increased (ambivalent group, in gray), ideological conservatives who want spending on the poor increased (ambivalent group, in gray), ideological moderates who want spending on the poor kept the same or decreased, and ideological conservatives who want spending on the poor kept the same or decreased. Income is in terciles. The Republican and Democratic categories include leaners. Data are from 2008.

Moreover, the findings from Free and Cantril and Page and Jacobs across the decades suggest that not only is this ambiguity in American public opinion long-standing, but also it has not changed much over time. To be sure, the proportion of Americans holding conservative, anti-government stances has varied over time: Figure 2.3 shows that while the proportion of self-identified (p.43) conservatives increased modestly, from 26 percent in 1972 to 32 percent in 2008, and the proportion of Republicans increased from 34 to 37 percent over the same period, the percentage of ANES respondents with low trust in government, who say the government wastes a lot of tax money, and who say the government in Washington is too strong has varied over time, with government skepticism peaking in 1980, 1994, and 2008, and diminishing in 1984 and 2002, after the 9/11 terrorist attacks. However, the relationships revealed in Table 2.2—the strong support of government skeptics for government spending outside of food stamps and welfare—are very similar whether we examine a high-trust or a low-trust year. Table 2.3 shows that in both 1994 (low trust) and in 2002 (high trust), large proportions of conservatives, Republicans, those trusting government to do the right thing only some or none of the time, and those saying government wastes a lot of tax money, also desire increased spending on the poor, child care, Social Security, and public schools. The proportion of respondents desiring increased spending is a little higher in 2002, the high-trust year, and openness to increased spending even on welfare is higher in the high-trust year, but overall the same relationships hold. Americans embrace small government but strongly support government programs—even those Americans who are the most skeptical about government, and even in years when overall trust in government is the lowest.

                   Exploring the Delegated Welfare State

Figure 2.3 Proportions of Government Skeptics over Time.

Source: American National Election study cumulative file.

Table 2.3. Government Spending Preferences among Government Skeptics in Low and High Government Trust Years

(1) All Respondents

(2) Conservatives

(3) Republicans

(4) Trust Govt None/Some of Time

(5) Govt Wastes a Lot of Tax Dollars

a. Low Trust Year: 1994

Food stamps












The poor

Child care






Social Security






Public schools






b. High Trust Year: 2002

Food stamps







The poor






Child care






Social Security






Public schools






Note: Cells show percentage of respondents of each description who want program spending increased (as opposed to decreased or kept the same). The item on the poor was not asked in 1994; the food stamp item was not included in the 2002 survey.

Source: American National Election Study cumulative file.

The abstract conservative-operational liberal dichotomy is also seen in public opinion surrounding the Obama health care reform. Support for the reform overall was lukewarm, hovering between 35 and 47 percent in the major polls in the months before passage,92 likely reflective of Americans’ antipathy to large government efforts. However, when asked about specific features of the legislation, support was far higher. Just as Americans embrace individual programs or spending areas, as shown above, so too did they support particular pieces of the health care law: tax breaks for small businesses to make coverage more affordable (90 percent in favor); new health exchanges (81 percent); requiring insurers to cover those with preexisting conditions (80 percent); filling in the Medicare prescription drug doughnut hole (79 percent); the employer mandate to offer health insurance (69 percent), federal and state government review of insurance premium increases (66 percent); increased taxes on high-income households (64 percent); and preventing insurers from dropping coverage for those who are sick (62 percent).93 The micro-level data from these surveys were not available as of this writing to allow us to examine how many individuals who objected to the health care reform overall nonetheless supported these specific features. However, given the large proportions that supported many of the individual provisions, there was certainly considerable overlap, squaring with the patterns observed in other social policy areas.

In short, American public opinion about government is fundamentally ambiguous. Americans—often the very same individuals—hold contradictory views, (p.44) professing an abstract preference for small government but also robust support for increased spending on many specific government programs. Survey questions asking people whether they are conservative or liberal, how much they trust government, and whether they think the government wastes tax dollars seem to prime one set of values that constitute what has been called the “American ethos”94—the aspect of Americans’ underlying beliefs that are skeptical about government. But questions about specific programs seem to invoke the other side of American beliefs—and people express support because the program names prime positive experiences with those programs, or thoughts of friends and family who have been helped by them, or the equality of opportunity they may provide.95

(p.45) How is it that people can hold these contradictory views simultaneously? In the course of everyday life we are seldom called upon to reconcile such contradictions.96 Instead, we profess on the spot the particular attitude triggered by our proximate environment at that moment, due to the previous question on the survey, recent events in one’s own life, or the nature of political discourse at that time. The crucial point is that because people hold competing considerations simultaneously, public opinion becomes a resource that can be tapped by political actors seeking either to expand or to contract the government role.97 Either set of attitudes—skepticism of government in the abstract, or support for specific programs and government action—can be primed; both kinds of attitudes are pervasive among the public, waiting to be brought to the fore.

Precisely because the public is antipathetic toward a large federal government staffed by armies of bureaucrats, but also supports the development of redistributive programs, this Janus-faced nature of public sentiment affects the parameters of social policy. Contracting out to private actors responds to public demand for help with their economic needs but does not appear to increase the government’s reach. Delegated governance thus poses a convenient solution to the inherent ambivalence in American public opinion.

Interest Group Influence

A second factor in the rise of the delegated welfare state is the role of interest groups, although sorting out the nature and degree of their influence is more complicated than one might think. Journalistic and other popular accounts of policy-making often assume that these groups dominate the policy-making process, using their financial and lobbying power to muscle their way into congressional offices and dictate the kinds of programs they want to see or prevent.98 Yet the political science literature on organized interests has yielded conflicting findings. On the one hand, some have persuasively shown how the rising influence of business groups since the 1970s shifted the terrain upon which political battles are fought, in part by injecting new ideas into the political system through think tanks and other intellectual venues.99 Yet political scientists have struggled to find consistent evidence that, when controlling for ideology and other factors, campaign donations translate into votes in Congress or that lobbyists systematically impact decision-making.100 Part of the problem stems from the difficulty of decisively showing influence. Another is the complicated nature of the relationship between interest groups and politicians. It could be that politicians are employing these groups for their own purposes, and not the reverse. In the case of delegated governance, for instance, using private agents to deliver programs may help build allies in the private sector, creating dependencies that can then be exploited by politicians seeking reelection or bureaucrats seeking political support for the programs they oversee.

(p.46) We use a method of detailed process-tracing to try to uncover the power dynamic between advocates of public programs and organized interests. We find that the great degree of delegation in the health care sector is in part due to the particularly dense thicket of preexisting interest groups and their access in the political system. Medical interest groups, such as those representing physicians, hospitals, insurers, and the pharmaceutical industry, have long fought expansions of the federal role in health care and, when faced with the political inevitability of such expansions, worked to undercut any accompanying growth of federal bureaucratic capacity in this area. They have wielded power through alliances with like-minded legislators, shaping bills that provide the social protections the public wants but that use publicly subsidized private actors to provide the services. Doing so meets the goals both of interest groups, who wish to expand their businesses and are glad to do so on the government nickel, and of lawmakers who wish to satisfy public demands without seeming to increase the reach of government.

Institutional Fragmentation, Congress, and Social Policy

A third factor contributing to delegated governance are the institutional arrangements of American government. A number of scholars have emphasized how the fractured, veto-ridden structure of American politics turns the policy-making process into an obstacle course that is extremely difficult to navigate.101 Fewer have explored the full implications of that which is particularly distinctive about American government: the preeminent role of Congress in making social policy.102 Nowhere in the world of advanced industrialized states is there a legislative branch that is so powerful, so independent of the executive, yet so fragmented in its organizational structure. Added to this are other peculiar features of the Congress that have worked against the growth of a conventional welfare state—the way in which it empowered Southern Democrats, for instance, through a seniority system that rewards those who are consistently reelected.103 With their lock hold over local and state-level politics, Southern Democratic Congressmen and Senators secured control over the levers of institutional power for decades and exerted tremendous influence over redistributive policy.104 While political realignments and institutional reforms since the 1970s eroded the power of Southern Democrats, opponents of policy change have made growing use of the institution’s various veto points—the Senate filibuster, for instance—to block new initiatives.

There are several ways in which the power of Congress over public policy has shaped the use of delegated governance. First, given the strong status quo bias that results from institutional fragmentation, the passage of social programs often requires the assembly of large political coalitions in support. One way to do so has been to draw in supportive interests, cutting them into the deal, for instance, by allowing them to govern publicly funded programs or to otherwise shape program administration in ways to their liking. This, in turn, enables political leaders to create Congressional coalitions in support of new policies, as (p.47) potential opponents retract or muffle their opposition to legislative initiatives. In addition, members of the coalition may have different policy visions, with some viewing delegated governance as a way to achieve marketizing reforms, while others are simply leery of governmental growth and its accompanying fiscal burdens. Still others may embrace delegated governance as a second-best option—a compromise that enables a particular bill to pass. By agglomerating these diverse views, delegated governance has made possible the growth of federal responsibilities but also produced “jerry-built” systems of social policy administration that often sacrifice managerial efficiency to the demands of well-organized groups.105

A second important aspect of congressional decision-making is the aversion of its members to doling out pain to constituents or organized groups.106 Although social programs offer many opportunities for credit-claiming, they also require tough decisions, either now or down the road. Thus, although adding a prescription drug benefit to Medicare might seem to create a win-win situation for its architects, it also requires determinations about how the benefit should be structured, how it will be paid for, and how its use should be limited for reasons of fiscal sustainability. Delegating those responsibilities to private insurers, who in turn engage in contentious negotiations with drug companies over prices and decide what drugs are available to beneficiaries, enables policy-makers to avoid having to make these politically fraught decisions. As Crenson and Rourke sum it up: “A system that allowed national policy to be administered in different ways in different places could also reduce the necessity of achieving agreement in Washington about just what the policy was in the first place.”107

In the case of health care policy, delegating authority to private managed care firms, who are supposed to make tough decisions about cost and coverage, holds particular allure, as it offers politicians a way out of difficult dilemmas while enabling them to reward interest groups knocking on their doors. As Robinson has remarked,

The interest among private insurers in serving public programs is reciprocated by the interest among public programs in outsourcing the management of their coverage benefits, provider networks, and enrollee expectations to private health plans. State and federal coverage sponsors increasingly lack the will to navigate the conflicting claims for resources between beneficiaries, taxpayers, and other stakeholders. They seek someone else to perform that inevitably thankless task.108

In sum, the delegated welfare state results from the confluence of public opinion, political institutions, and interest group power—with the latter an especially important factor in the health care sector, compared to other social policy areas. American political institutions make it particularly difficult to pass redistributive programs, as they have given weight to anti-statist political forces and have (p.48) provided access to organized interests threatened by the expansion of federal authority and power. The mixed and ambiguous character of public opinion matters in that it provides resources to political actors and interest groups seeking to exploit either the support for public programs or the opposition to federal expansion.

Hence liberals have often been constrained in their ability to develop publicly provided programs, but centrists and conservatives have also been unable to entirely ignore the public clamor for new social programs. Attempting to square the circle—to navigate the institutional and political obstacles to new social programs—policy-makers repeatedly have turned to delegated governance as a way to buy off interest group support, tamp down claims that a new program expands oppressive federal authority, and yet respond to public demands for assistance with life’s risks and costs. The result has often been policy arrangements that are Rube Goldbergesque in their structure—think about the complex legislation that is Medicare Part D—in order to satisfy these conflicting demands.

Does It Matter How Social Programs Are Administered?

The question of how social programs are delivered has important real-world and theoretical implications. The administration of social programs impacts the effectiveness of public policies, the form of redistribution that follows, and the extent to which the providers of services are accountable to the mass public. All of these questions are hotly contested and are at the crux of debates about state versus non-state social provision. We do not aim to definitively settle these debates for all policy areas, but in the succeeding section will lay out some of the competing arguments that we test in the case of Medicare in several later chapters. The nature of program administration also has important implications for academic debates in the social sciences, particularly those concerning the nature of the American state. Political scientists have paid insufficient attention to what public policies actually look like, on the ground, as most analyses tend to stop with a policy’s passage. By looking concretely at the implementation and effects of policies, we can gain insight into the nature and functioning of the American state.

Real-World Consequences of Delegating Governance for Social Provision

Does it matter how social programs are administered? One could argue that whether one administers programs through local governments, private actors, or federal bureaucracies is less important than the existence of a program, its eligibility parameters, and its generosity. Perhaps social policy ends justify whatever administrative means are necessary to make a program work. Yet, for many programs, the devil lies in the details: how programs are actually administered and (p.49) experienced on the ground by stakeholders and beneficiaries can be tremendously consequential for program effectiveness, equity, and democratic accountability. Given the importance of these questions, much of the political heat generated around particular social policy reforms concerns modes of governance.

One major question about program design and implementation concerns effectiveness: Are program goals met in a relatively efficient and effective manner? Or do governing arrangements impede the effective delivery of promised benefits and services? This issue has featured prominently in debates about privatization, with advocates arguing that government is an intrinsically poor provider of most benefits and services. Some argue that significant delegation problems arise between legislators and the bureaucrats who administer public programs, as the latter have incentives to expand spending on their agencies and possess the information needed to outmaneuver legislative efforts at control.109 Moreover, government employees receive salaries that are not tied to their performance in delivering services to the public, and bureaucratic organizations are funded through budgetary appropriations instead of having to prove their worth in a competitive environment. As a result, some argue that there are few organizational incentives for good performance when government has a monopoly over the provision of services.110 Moreover, in many areas, civil servants tend to be paid more than people working in the private sector, in part due to their high degree of unionization. This may make it more expensive to directly administer government programs than to contract them out to private agencies.

Government bureaucracies are also said to be rule-bound and rigid, focused more on meeting legal requirements than innovating in their service provision. Or, they may become too independent of the legislative branch, generating bureaucratic drift as agency officials pursue their own agendas. Either way, directly administered programs may fail to be accountable to public needs. By contrast, advocates of market- or community-based organizations argue that these organizations are more likely to deliver high-quality services in a low-cost manner, thereby reducing public spending and enhancing popular accountability.111 In the case of market actors, as long as the contracting relationship is structured such that their profits are at stake and there is sufficient competition among providers, they should work to survive in a competitive marketplace and thus deliver quality services while holding down costs.112 In the case of community-based actors, some argue that because they are closer to the people they serve, they are better able to respond to the needs of particular communities.113 Voucher programs should generate a similar dynamic, creating competition between suppliers for the business of beneficiaries and thus promoting innovation.114 Finally, contracting with private actors may also enable greater governmental flexibility because contracts can quickly be arranged or terminated as needs wax and wane.

An opposing view challenges whether private sector actors are generally more effective in providing social welfare benefits and services than government agencies. First, some argue that contracted government services rarely represent (p.50) a real market.115 Especially in many social welfare areas, there are only a small number of potential service providers—and sometimes only one in a given area—which limits the degree of competition.116 Providers who successfully procure contracts also tend to have them renewed, perhaps because they gain political influence over the contracting agency, or simply because they are trusted or develop expertise that is valued and cannot immediately be acquired by new providers.117 These realities may vitiate the theoretical effects of market competition on the quality and cost-efficiency of service delivery. In addition, many have noted that relying on private agencies for the delivery of programs and services rarely leads to the elimination of any role for government in this area. To the contrary, there may be the need for a more muscular set of government agencies that can establish, regulate, and monitor the social welfare marketplace.118 The effect may be to multiply the number of people working in a particular social welfare field in both the public and private sector, leading to less efficient production than if there were simply one set of actors involved. Finally, some critics of contracting find that it tends to produce less flexibility, and not more, as agencies get locked into relationships with particular contractors.119

A second question concerns the effect of program administration on democratic accountability.120 This is a concern for any system of bureaucratic administration: civil servants are unelected and only indirectly accountable to the public through the oversight of democratically elected officials. Moreover, their actions can never be perfectly supervised: program execution requires some degree of bureaucratic discretion that cannot be fully overseen by elected politicians, although legislators do have an array of control mechanisms at their grasp.121 In the course of using their discretion, agency officials may not only make inefficient choices about program delivery, but also may prove indifferent to the population they should be serving. In light of these problems, advocates of community- or market-based provision argue that such arrangements improve democratic accountability.122 Community organizations are not elected bodies but may nonetheless contain members of the community who can hold them accountable to the public. In the case of for-profit firms, because their survival is tied to their ability to attract clients, they should scramble to respond to the diverse needs and preferences of their users. Thus, even though the chain of accountability from elected officials to service providers is obscured in a system of delegated governance, this may improve democratic accountability, not weaken it.123

Critics of these arrangements worry about lodging administrative discretion in the hands of non-public institutions. One consequence could be to create an administrative void, a situation in which “… responsibility is broadly shared, but no one is fully in charge.”124 Additional problems may arise in the case of publicly traded firms, whose fiduciary responsibility is to shareholders and not to program beneficiaries. A more general issue concerns whether or not social welfare markets actually work to make firms accountable to their customers. For instance, when beneficiaries have a choice between competing service providers, are they (p.51) able to effectively navigate between options, making optimal (or at least not ruinous) decision? Some argue that consumers are notoriously bad at decision-making in commercial markets, with paradoxically more choices leading to paralysis and decisions based on the wrong criteria (e.g., glossiness of marketing materials). Perhaps even more important is whether or not people leave bad service providers, forcing the kind of market accountability that advocates theorize should exist. If people fail to “exit” bad providers, or use “voice” to contest poor service provision, then markets will fail to convey the right information to firms, thus weakening the accountability mechanisms in private welfare markets.125

The governance of social programs also has important redistributive consequences. The Left has often championed the ideal of direct governance in the belief that when benefits and services are highly heterogeneous, inequalities of access will develop. Particularly when markets deliver services, people of a higher socioeconomic status will likely benefit by having their preferences catered to, while less educated and lower income people will have inferior services. On the other hand, defenders of markets might point to failures in government-provided services to the poor—inner-city school systems in the United States, for instance—and argue that markets can do a better job of responding to these disempowered citizens.126

Yet another question concerns how the administration of public programs affects the politics of social provision. Some critics of contracting out public services argue that it gives power to private organizations, which then use this power to further enrich themselves. One consequence may be corruption, as public officials develop cozy relationships with private firms whose campaign contributions, or outright bribes, ensure their continued access to valued public contracts. A softer version of this dynamic may simply be to increase the voice of private provider groups in the political process. Consumers are a notoriously difficult group to organize, and while consumer groups may work hard to insert the perspective of program beneficiaries into policy debates, they may be outgunned by well-financed private groups who have a strong stake in protecting their interests. How the relative power of these groups plays out in the political process is an empirical question, likely to vary in different social policy arenas. It may be that giving private actors a stake in public programs promotes the expansion of spending on these programs, contrary to what market reformers might desire.127 Such was the case with the 2003 Medicare Modernization Act, which was supported by both insurance companies—who would profit from delivering the benefit—and pharmaceutical companies, who benefited from the increased volumes and promise of no government involvement in determining prices. Thus, delegated governance may be the distinctive way that welfare states grow in societies and polities marked by antipathy toward public officials and “direct” forms of government.

One final question is more clearly normative and concerns the suitability of giving private actors power over the lives of individuals. This is part of what makes the social welfare sphere distinctive from other areas of government contracting. (p.52) There is a difference between employing private firms to provide routine office services or deliver products to federal agencies, and delegating responsibility for programs that affect people’s health, welfare, and livelihood. When firms have the ability to determine eligibility for welfare programs and to mete out sanctions for failure to meet certain requirements, these firms are exerting power over these individuals.128 Similarly, when a private drug plan uses its formulary to “manage” a beneficiary’s use of prescription drugs, this is again a form of power being exerted over that beneficiary’s life. Of course, we all live in a world of public and private institutions that exert power over us, and over which we do not have full control. This fact should not prevent us from probing the differential consequences of public and private power, however. As Grant McConnell argued over four decades ago, “… the persistence and growth of private power have posed an embarrassing problem for all who are involved in exercising it. What justifies the existence of power; by what principle is it rightful?”129

In public provision there is a principle of democratic accountability that exists at least theoretically if not empirically: governing arrangements are structured so that bureaucrats are accountable to elected officials in some way, and elected officials are periodically accountable to the public. Unless markets can be shown to offer improved accountability through their responsiveness to consumers, accountability is diminished in a system of delegated governance to private actors.

Theoretical Implications

This study also has implications for how we should conceptualize the nature of the American state, which has long been a source of scholarly puzzlement. When measured in terms of public sector employees or public spending, the federal government in the United States is smaller than the national states of many other western European countries, and this has been true since the founding of the American Republic. This fact has long inspired claims about the relative “statelessness” of the United States.130 By most objective measures, the nineteenth-century federal administration was miniscule in size, and the dominant actors in the American polity were courts and parties, not federal bureaucrats.131 Although the twentieth century brought the rise of “bureaucratic autonomy” in key domains, the overall size of the federal government remains fairly truncated,132 particularly when compared to other advanced industrialized countries. For instance, measured in terms of total tax receipts, including taxes collected at all levels of government, U.S. tax revenue as a proportion of GDP is nearly eight percentage points lower than the OECD average—28.0 percent compared to 35.9 percent—and is notably lower than countries we tend to think of as similarly market-oriented, such as the United Kingdom (37.1 percent) or Canada (33.3 percent).133

At the same time, however, the twentieth-century American state has been one of the most powerful actors in the world. Even in the nineteenth century it would be misleading to focus excessively on bureaucratic incapacity: the federal (p.53) government created the public land system, propelled the settlement of a vast territory (which involved the forced removal of native peoples), engaged in internal improvements to develop the nation’s public infrastructure, and upheld the institution of slavery.134 In the social policy sphere, the federal government did not develop the social insurance programs that were emerging in Britain and Bismarckian Germany at the time, yet did institute a large-scale system of veterans’ pensions and created a progressive income tax—a tax that the federal government did a far better job collecting during and after World War I than did the allegedly “strong” French state at the time.135 In the twentieth century, the United States fought and won two world wars and a cold war, and is now the dominant superpower in the international arena—hardly what we might expect from a “weak” state.136

How can we reconcile these two images of the American state? Part of the problem lies in the difficulty scholars have in getting away from a dichotomous, “strong versus weak” understanding of states.137 Thus, in the laudable effort to combat images of perennial state weakness, scholars have sought to prove that the American state was in fact stronger than we might think by pointing to pockets of bureaucratic development or effectiveness. Yet this hardly helps us make sense of the many other instances or indicators we can find of bureaucratic failure. The social policy arena offers numerous examples of both strength and weakness. To take the example of health care, analysts from across the political divide agree that the Centers for Medicare and Medicaid Services, which oversees Medicare, Medicaid, and other health policy, is undermanned, under-resourced, and frequently outmatched by powerful organized interests in the health care sector. Yet the same organization has been able to implement a complex system of price-setting in the health care sphere, and somehow, despite its lack of resources, manages to oversee the processing of around one billion health care claims a year.

Instead of asking whether the American state is strong or weak, we need a vocabulary that better captures how the American state actually works in practice. Here, we take inspiration from William Novak’s view that scholars should reclaim the tradition of American pragmatism that developed over a century ago but has been displaced in the social sciences by abstract European theory. Rather than start with an ideal type and see how the phenomenon of interest measures up, we should instead seek to understand how the object of study actually functions, letting the concrete realities that we can observe and describe inform our categories and drive our conclusions about it. In the words of Theodore Lowi, we are interested in the “almost forgotten [issues] of what kind of government, what ends of government, what forms of government, and what consequences of government….”138

Such an approach has often been lacking in political science analyses of public policy, which tend to stop with passage of the law rather than examining how programs and policies are actually implemented on the ground. Yet a lot of the conflict around public policy concerns precisely how services or benefits are going to be delivered. Thus, debates over universal health coverage have not only been (p.54) over philosophical questions of economic redistribution, but have been centrally concerned with who is going to deliver this insurance, who will provide health care services, and precisely how all of these actors will be paid. The answers to these questions, in turn, determine the structure of governing institutions—the nature of the state and how it actually functions. When we are attentive to how, in actuality, abstract programs are turned into concrete realities, we gain a better understanding of the political stakes around many redistributive programs, but also can see more clearly how it is that American governing institutions actually work.

What does such an approach yield in looking at the American state? The American state is better described not as institutionally strong or weak, but as a system of delegated and diffused authority.139 Contrary to the Weberian ideal-type of the centralized, hierarchically organized bureaucratic structure characterized by clear lines of authority and accountability, the contemporary American state is, as Elisabeth Clemens says quite bluntly, “a mess.” This was true in the first decades of the twentieth century when, at a time of expanding governmental responsibility, program administration was achieved through heavy reliance upon private actors for the delivery of social services and the use of intergovernmental grants.140 The tremendous expansion of federal responsibilities in the post–World War II era created ever greater dependence upon non-federal actors for the implementation of federal programs.141 Federally funded programs would be delivered by state and local governments through intergovernmental arrangements, such as grants-in-aid. Many functions are contracted out to private actors: in 2005, the number of contract employees working for the federal government has been estimated to be four times the number of civilian employees, rendering the “true size of government” significantly larger than we might think when looking only at government employees.142 There also are hybrid governing institutions that defy simple categorizations as public or private, including government-sponsored enterprises, such as Fannie Mae and Freddie Mac. As Frederick Mosher once put it:

The use and the extent of all of these tools have grown enormously in recent years—even as federally performed operations virtually stood still. Few new policies and programs failed to rely upon other governments or institutions in the private sector for a major part or all of their execution. The extension of federal interest and intervention into the nooks and crannies of our economic, social, cultural, and even personal lives seems almost unlimited. And most of this is being done through others, not strictly a part of the federal government itself. The growth of federal influence defies precise quantitative measurement, but there can be no question that it has been pervasive….143

Do these governing arrangements produce a “strong” or “weak” state? A better way to frame the question is to explore the implications of delegated governance along the lines we described above: the cost-effectiveness of state action; its (p.55) political consequences; and its effects on redistribution and accountability. The answers to these questions are empirical and hinge on the nature of the governing arrangements developed in different sectors of public policy. Were we able to cumulate studies of a large number of different areas, we could paint a general picture of the American state, how it functions, and the political forces that have shaped its development. We can only do this in one domain—the welfare state, which we winnow down even further to the area of Medicare—but we hope to demonstrate the merits of analyzing how programs are governed, why they are governed that way, and what consequences this has for societies and polities.


The American welfare state intertwines public and private authority in pervasive ways. We develop the concept of delegated governance—the delegation of administrative authority for publicly funded programs to private actors—to describe some of these arrangements. Although the United States is not alone in the world in relying on private actors for the implementation of public programs, it does so to a great extent and in some unusual ways. Historically, the United States often turned to nonprofit organizations to deliver social welfare benefits and services. In more recent decades, however, the emphasis has shifted to building social welfare marketplaces comprised of competing, for-profit actors in the social welfare field. Along with this change has come a shift in the locus of decision-making and risk. Hitherto, the government made decisions about who would provide services and who bore risk for these decisions (often, the government itself). Increasingly, individuals are expected to choose from a menu of competing providers, and both for-profit providers bear risk (their profits are at stake), and individuals bear risk (for their choice of welfare provider). Moreover, the extent of delegation, the embrace of the market form, the relative lack of governmental supervision, and the spread of delegation to health care set the United States apart from peer nations.

The ways in which programs are administered are hardly technical details deserving only of attention in public administration textbooks. Yet political scientists have generally ignored these arrangements, ending their analyses with the passage of policy. This is unfortunate, given that much of the political conflict generated by public policy concerns how programs will actually be put in place, such as who will provide them and how providers will be paid. Can we really understand the political stakes around public policies if we do not explore the implications of policy design and implementation for government bureaucrats, interest group stakeholders, and beneficiaries? We believe that one cannot, and thus argue that scholars of the welfare state should give attention to its governance, which has profound implications for effectiveness, redistribution, and accountability.


(1) . In 46 states, more than half of Medicaid enrollees are in managed care as of 2010. Overall, 70 percent of Medicaid beneficiaries receive some or all of their health care through managed care programs. Kaiser Family Foundation, “Medicaid and Managed Care: Key Data, Trends, and Issues,” February 2010, accessed October 18, 2010, http://www.kff.org/medicaid/upload/8046.pdf.

(2) . See Joe Soss, Richard C. Fording, and Sanford F. Schram, Disciplining the Poor: Neoliberal Paternalism and the Persistent Power of Race, forthcoming, University of Chicago Press.

(3) . Rosemary Stevens, “Can the Government Govern? Lessons from the Formation of the Veterans Administration,” Journal of Health Politics, Policy and Law 16, 2 (1991): 281–305.

(5) . Edgar Kiser, “Comparing Varieties of Agency Theory in Economics, Political Science, and Sociology: An Illustration from State Policy Implementation,” Sociological Theory 17, 2 (July 1999): 146–170; Cass R. Sunstein, “Nondelegation Canons,” University of Chicago Law Review 67, 2 (Spring 2000): 315–343.

(6) . Dru Stevenson, “Privatization of Welfare Services: Delegation by Commercial Contract,” Arizona Law Review 45 (2003): 83–131.

(7) . Jody Freeman, “The Private Role in Public Governance,” New York University Law Review 75 (June 2000): 543. See Stevenson, “Privatization of Welfare Services,” for a contrary view.

(8) . Gillian E. Metzger, “Privatization as Delegation,” Columbia Law Review 103 (2003): 1367–1502.

(9) . R.A.W. Rhodes, “The New Governance: Governing without Government,” Political Studies 44, 3 (1996): 652–667; Mary Daly, “Governance and Social Policy,” Journal of Social Policy 32, 1 (2003): 113–128; Carolyn J. Hill and Laurence E. Lynn Jr., “Is Hierarchical Governance in Decline? Evidence from Empirical Research,” Journal of Public Administration Research and Theory 15, 2 (2005): 173–195; H. George Frederickson, “Whatever Happened to Public Administration? Governance, Governance Everywhere,” in The Oxford Handbook of Public Management, eds. Ewan Ferlie, Laurence E. Lynn, Jr., and Christopher Pollitt (Oxford: Oxford University Press, 2005), 282–304.

(10) . Freeman, “Private Role”; Mark Considine, Enterprising States: The Public Management of Welfare-to-Work (Cambridge: Cambridge University Press, 2001); Janet Newman, “Modernising the State: A New Style of Governance?” in Welfare State Change: Towards a Third Way? eds. Jane Lewis and Rebecca Surrender (Oxford: Oxford University Press, 2004), 69–88; Jon Pierre and B. Guy Peters, Governing Complex Societies: Trajectories and Scenarios (Houndmills: Palgrave, 2005); Frederickson, “Whatever Happened,” 285.

(11) . Ted Kolderie, “The Two Different Concepts of Privatization,” Public Administration Review 46, 4 (July/August 1986): 285–291; Starr, “Privatization”; Paul Seidenstat, “Theory and Practice of Contracting Out in the United States,” in Contracting Out Government Services, ed. Seidenstat (Westport, CT: Praeger, 1999), 3–25.

(12) . Metzger, “Privatization,” 1395.

(p.255) (13) . Starr, “Privatization”; Savas, “Public-Private Partnerships.”

(14) . Still other terms include “purchase-of-service” (Harold W. Demone, Jr., and Margaret Gibelman, eds., Services for Sale: Purchasing Health and Human Services [New Brunswick, NJ: Rutgers University Press, 1989]) and “public-private partnerships” (Pauline Vaillancourt Rosenau, ed., Public-Private Policy Partnerships [Cambridge: MIT Press, 2000]).

(15) . Metzger, “Privatization,” 1395–1396.

(16) . Grant McConnell, Private Power and American Democracy (New York: Alfred A. Knopf, 1966).

(17) . Martin Rein and Lee Rainwater, “From Welfare State to Welfare Society,” in Stagnation and Renewal in Social Policy: The Rise and Fall of Policy Regimes, eds. Rein, Rainwater, and Gøsta Esping-Andersen (Armonk, NY: M. E. Sharpe, 1987), 143–159; Beth Stevens, “Blurring the Boundaries: How the Federal Government Has Influenced Welfare Benefits in the Private Sector,” in The Politics of Social Policy in the United States, eds. Margaret Weir, Ann Shola Orloff, and Theda Skocpol (Princeton: Princeton University Press, 1988), 123–148; Gøsta Esping-Andersen, The Three Worlds of Welfare Capitalism (Princeton: Princeton University Press, 1990).

(18) . Howard, Hidden Welfare State; Howard, Welfare State Nobody Knows.

(19) . Hacker, Divided Welfare State; Klein, For All These Rights.

(20) . Mettler, “The Submerged State”; Dalton Conley and Brian Gifford, “Home Ownership, Social Insurance, and the Welfare State,” Sociological Forum 21, 1 (March 2006): 55–82.

(21) . Clemens, “Rube Goldberg State.”

(22) . Theda Skocpol, Protecting Soldiers and Mothers (Cambridge: Belknap Press, 1992); Joe Soss, Unwanted Claims: The Politics of Participation in the U.S. Welfare System (Ann Arbor: University of Michigan Press, 2000); Andrea Louise Campbell, How Policies Make Citizens: Senior Political Activism and the American Welfare State (Princeton: Princeton University Press, 2003); Mettler, “Submerged State.”

(23) . Salamon, “Rethinking Public Management”; Kettl, (Mis?)Managing Federal Programs; Brinton, “Hollow State”; DiIulio, “Government by Proxy.”

(24) . There has, however, been some important work on the role of nonprofit organizations in delivering social welfare services, including Ralph M. Kramer, Voluntary Agencies in the Welfare State (Berkeley: University of California Press, 1981), Smith and Lipsky, Nonprofits for Hire; Lester M. Salamon, Partners in Public Service: Government-Nonprofit Relations in the Modern Welfare State (Baltimore: Johns Hopkins University Press, 1995); and Andrew Morris, The Limits of Voluntarism: Charity and Welfare from the New Deal to the Great Society (Cambridge: Cambridge University Press, 2008). Much of this research focuses on the consequences of contracting out service provision to non-profits, but Morris explores the origins, evolution, and political dynamics around this process.

(25) . There have been publicly funded voucher programs in a handful of cities and states since the early 1990s, as well as several privately funded voucher initiatives. Selection into either is typically by lottery.

(26) . Byrna M. Sanger, Welfare Marketplace: Privatization and Welfare Reform (Washington, DC: Brookings Institution, 2003); Colleen M. Grogan and Michael K. Gusmano, “Political Strategies of Safety-Net Providers in Response to Medicaid Managed Care Reforms,” Journal of Health Politics, Policy and Law 34 (February 2009): 5–35.

(27) . In traditional fee-for-service Medicare, individuals choose their own doctor and hospital with a kind of virtual voucher from the government. But Medicare differs from a true voucher program in that there is no fixed subsidy within which individuals have to work: they choose a physician, and the physician is reimbursed as much as the government is willing. Risk then is borne to some extent by the physician, who must be satisfied with current reimbursement levels, but ultimately by the federal government, which pays the cost when beneficiaries use more care than expected. This is FFS Medicare in its original incarnation, in contrast with the use of private managed care firms to deliver Medicare benefits, which do not pay on a FFS basis, but instead receive capitated payments and then negotiate with providers over reimbursement, and try to “manage” beneficiary care so as to hold down utilization and cost.

(p.256) (28) . Jason T. Abaluck and Jonathan Gruber, “Choice Inconsistencies among the Elderly: Evidence from Plan Choice in the Medicare Part D Program.” NBER Working Paper 14759 (February 2009), accessed October 18, 2010, www.nber.org/papers/w14759; Yaniv Hanoch, Thomas Rice, Janet Cummings, and Stacey Wood, “How Much Choice is Too Much? The Case of the Medicare Prescription Drug Benefit,” Health Services Research 44, 4 (August 2009): 1157–1168; Thomas Rice, Janet Cummings, and Daniel Kao, “Reducing the Number of Drug Plans for Seniors: A Proposal and Analysis of Three Case Studies,” unpublished manuscript, UCLA.

(30) . Originally, landlords who took one Section 8 voucher holder were mandated by law to accept all others who came along, and there was an “endless lease” provision that prohibited the termination of a lease with a Section 8 client except for good cause. These “risks” to landlords limited the supply of housing, and both were eliminated in 1998 legislation. Morton J. Schussheim, Housing the Poor: An Overview (New York: Novinka Press, 2003).

(31) . Metzger, “Privatization,” 1387.

(32) . Beryl Radin, “When Is a Health Department Not a Health Department? The Case of the US Department of Health and Human Services,” Social Policy & Administration 44, 2 (April 2010): 142–154.

(33) . Teppo Kröger, “The Dilemma of Municipalities: Scandinavian Approaches to Child Day-Care Provision,” Journal of Social Policy 26, 4 (1997): 485–507; Elizabeth Docteur and Howard Oxley, “Health System Reform: Lessons from Experience,” in Towards High-Performing Health Systems, eds. Docteur and Oxley (Paris: OECD 2004). Norway is somewhat exceptional in that the nonprofit sector has long played a significant role in delivering child care and other services. Benjamin Gidron, Ralph M. Kramer, Lester M. Salamon, “Government and the Third Sector in Comparative Perspective: Allies or Adversaries?” in Government and the Third Sector: Emerging Relationships in Welfare States, eds. Gidron, Kramer, and Salamon (San Francisco: Jossey-Bass Publishers, 1992), 22–23; Bente Blanche Nicolaysen, “Voluntary Service Provision in a Strong Welfare State,” Working Paper 35 (Mannheim, Germany: Mannheimer Zentrum für Europäische Sozialforschung, 2001).

(34) . Docteur and Oxley, “Health System Reform.”

(35) . Maurizio Ferrera, “The ‘Southern Model’ of Welfare in Southern Europe,” Journal of European Social Policy 6, 1 (February 1996): 17–37.

(36) . In the original incarnation of these systems, individuals could not even choose their physician or hospital; they would use that which was locally available to them (much as the Veterans Administration health care system works in the United States).

(37) . Hal Pawson, 2006, “Restructuring England’s Social Housing Sector Since 1989: Undermining or Underpinning the Fundamentals of Public Housing?” Housing Studies 21, 5 (September): 767–783.

(38) . Gidron, Salamon, and Kramer, “Third Sector”; Birgit Fix, The Institutionalization of Family Welfare: Division of Labour in the Field of Child Care in Austria and Germany Working Paper 24 (Mannheim, Germany: Mannheimer Zentrum für Europäische Sozialforschung, 1998).

(39) . Paul Dekker, “The Netherlands: From Private Initiatives to Non-Profit Hybrids and Back?” in The Third Sector in Europe, eds. Adalbert Evers and Jean-Louis Laville (Cheltenham, UK: Edward Elgar, 2004), 144–165.

(40) . Esping-Andersen, Three Worlds; Giuliano Bonoli, “Classifying Welfare States: A Two-Dimensional Approach,” Journal of Social Policy 26, 3 (1997): 351–372.

(41) . Colin Crouch, cited in Bernard Ebbinghaus, “Reforming Bismarckian Corporatism: The Changing Role of Social Partnership in Continental Europe,” in A Long Goodbye to Bismarck? The Politics of Welfare Reform in Continental Europe, ed. Bruno Palier (Amsterdam: Amsterdam University Press, 2010), 255–278.

(42) . Paul V. Dutton, Differential Diagnoses: A Comparative History of Health Care Problems and Solutions in the United States and France (Ithaca, NY: Cornell University Press, 2007); Rémi Lenoir, “Family Policy in France since 1938,” in The French Welfare State: Surviving Social and Ideological Change, ed. John S. Ambler (New York: New York University Press 1991), 144–186.

(p.257) (43) . Ebbinghaus, “Bismarckian Corporatism.”

(44) . David Wilsford, “The Continuity of Crisis: Patterns of Health Care Policymaking in France, 1978–1988,” in The French Welfare State: Surviving Social and Ideological Change, ed. John S. Ambler (New York: New York University Press, 1991), 94–143; Dutton, Differential Diagnoses, 23.

(45) . Annette Zimmer, “Corporatism Revisited—The Legacy of History and the German Nonprofit Sector,” Voluntas: International Journal of Voluntary and Nonprofit Organizations 10, 1 (1999): 41.

(46) . Rudolph Bauer, “Voluntary Welfare Associations in Germany and the United States: Theses on the Historical Development of Intermediary Systems,” Voluntas 1, 1 (1990): 97–111.

(47) . Wolfgang Seibel, “Government-Nonprofit Relationships in a Comparative Perspective: The Cases of France and Germany,” in The Nonprofit Sector in the Global Community: Voices from Many Nations, eds. Kathleen D. McCarthy, Virginia A. Hodgkinson, Russy D. Sumariwalla and Associates (San Francisco: Jossey-Bass Publishers, 1992), 205–229.

(48) . Maria Brenton, “Changing Relationships in Dutch Social Services,” Journal of Social Policy 11, 1 (1982): 68,

(49) . Ralph M. Kramer, “The Use of Government Funds by Voluntary Social Service Agencies in Four Welfare States,” in The Nonprofit Sector in International Perspective, ed. Estelle James (New York: Oxford University Press 1989), 218–219.

(50) . Christa Altenstetter, “Insights from Health Care in Germany,” American Journal of Public Health 93, 1 (January 2003): 38–44; Ebbinghaus, “Bismarckian Corporatism”; Bruno Palier, “The Dualization of the French Welfare System,” in A Long Goodbye to Bismarck? The Politics of Welfare Reform in Continental Europe, ed. Palier (Amsterdam: Amsterdam University Press, 2010), 73–99.

(51) . Rhodes, “New Governance”; Bob Jessop, “The Changing Governance of Welfare: Recent Trends in Its Primary Functions, Scale, and Modes of Coordination,” Social Policy and Administration 33, 4 (December 1999): 348–359; Paula Blomqvist, “The Choice Revolution: Privatization of Swedish Welfare Services in the 1990s,” Social Policy & Administration 38, 2 (April 2004): 139–155; Neil Gilbert, Transformation of the Welfare State: The Silent Surrender of Public Responsibility (Oxford: Oxford University Press, 2004).

(52) . Pawson, “England’s Social Housing.”

(53) . Emmanuele Pavolini and Costanzo Ranci, “Restructuring the Welfare State: Reforms in Long-Term Care in Western European Countries,” Journal of European Social Policy 18, 3 (2008): 254.

(54) . Daly, “Governance,” 120,*

(55) . Pavolini and Ranci, “Reforms in Long-Term Care,” 254.

(56) . Kimberly J. Morgan, Working Mothers and the Welfare State: Religion and the Politics of Work-Family Policies in Western Europe and the United States (Palo Alto: Stanford University Press, 2006).

(57) . Gilbert, Transformation of the Welfare State, 112.

(58) . Blomqvist, “Choice Revolution”; Michael Baggesen Klitgaard, “Do Welfare State Regimes Determine Public Sector Reforms? Choice Reforms in American, Swedish and German Schools,” Scandinavian Political Studies 30, 4 (2007): 444–682.

(59) . Karen M. Anderson and Ellen M. Immergut, “Sweden: After Social Democratic Hegemony,” in The Handbook of West European Pension Politics, eds. Immergut, Anderson and Isabelle Schultze (Oxford: Oxford University Press, 2007), 349–395.

(60) . Monica Prasad, The Politics of Free Markets: The Rise of Neoliberal Economic Policies in Britain, France, Germany, and the United States (Chicago: University of Chicago Press, 2006).

(61) . Vivien Schmidt, Democratizing France: The Political and Administrative History of Decentralization (New York: Cambridge University Press 1990); Claire F. Ullman, The Welfare State’s Other Crisis: Explaining the New Partnership between Nonprofit Organizations and the State in France (Bloomington: University of Indiana Press, 1999).

(62) . Zimmer “Corporatism Revisited,” 43–44; Ingo Bode, “Disorganized Welfare Mixes: Voluntary Agencies and New Governance Regimes in Western Europe,” Journal of European Social Policy 19, 4 (2006): 346–359.

(p.258) (63) . Wilsford, “Health Care Policy-Making in France”; Patrick Hassenteufel and Bruno Palier, “Towards Neo-Bismarckian Health Care States? Comparing Health Insurance Reforms in Bismarckian Welfare Systems,” Social Policy and Administration 41, 6 (December 2007): 574–596.

(64) . Walter J.M. Kickert, “Expansion and Diversification of Public Administration in the Postwar Welfare State: The Case of the Netherlands,” Public Administration Review 56, 1 (1996): 89.

(65) . Certainly, some analysts forecast that the privatization and marketization of welfare programs and services will accelerate in the years ahead due to fiscal constraints and the strains brought by demographic aging. See Wolfgang Streeck, “The Fiscal Crisis Continues: From Liberalization to Consolidation,” Comparative European Politics 8, 4 (2010): 505–514.

(66) . James A. Morone, “Citizens or Shoppers? Solidarity under Siege,” Journal of Health Policy, Politics and Law 25, 5 (October 2000): 959–968; Jacob S. Hacker, “Review Article: Dismantling the Health Care State? Political Institutions, Public Policies and the Comparative Politics of Health Reform,” British Journal of Political Science 343 (October 2004): 693–724.

(67) . Hassenteufel and Palier, “Neo-Bismarckian Health Care States,” 591.

(68) . Eddy van Doorslaer and Frederik T. Schut, “Belgium and the Netherlands Revisited,” Journal of Health Politics, Policy and Law 25, 5 (2000): 881–882.

(69) . Hassenteufel and Palier, “Neo-Bismarckian Health Care States,” 591–592.

(70) . Geoff Fougere, “Transforming Health Sectors: New Logics of Organizing in the New Zealand Health System,” Social Science & Medicine 52 (2001): 1236.

(71) . Todd A. Krieble, “New Zealand,” Journal of Health Politics, Policy and Law 25, 5 (2000): 925–930.

(72) . Alan Jacobs, “Seeing Difference: Market Health Reform in Europe,” Journal of Health Politics, Policy and Law 23, 1 (February 1998): 21.

(73) . Doorslaer and Schut, “Belgium and the Netherlands,” 886.

(74) . Hacker, “Dismantling the Health Care State?” 701.

(75) . The following paragraph is based on Yvette Bartholomée and Hans Maarse, “Health Insurance Reform in the Netherlands,” Eurohealth 12, 2 (2006): 7–9; Jan-Kees Helderman, Bringing the Market Back In? Institutional Complementarity and Hierarchy in Dutch Housing and Healthcare, PhD dissertation, Erasmus University Rotterdam 2007; Pauline Vaillancourt Rosenau and Christiaan J. Lako, “An Experiment with Regulated Competition and Individual Mandates for Universal Health Care: The New Dutch Health Insurance System,” Journal of Health Politics, Policy and Law 33, 6 (2008): 1031–1055; Kieke G. H. Okma, “Commentary on Rosenau and Lako,” Journal of Health Politics, Policy and Law 33, 6 (2008): 1057–1071; Uwe E. Reinhardt, “The Swiss Health System: Regulated Competition without Managed Care,” JAMA 292, 10 (2004): 1227–1231.

(76) . Okma, “Commentary,” 1059.

(77) . Okma, “Commentary,” 1060.

(78) . Okma, “Commentary”; Helderman, Bringing the Market Back In?

(79) . Reinhardt, “Swiss Health System,” 1227.

(80) . Reinhardt, “Swiss Health System,” 1230.

(81) . Paul Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94, 2 (2000): 251–267.

(82) . Gøsta Esping-Andersen, “Power and Distributional Regimes,” Politics & Society 14, 2 (1985): 223–256.

(83) . Jill Quadagno, The Color of Welfare: How Racism Undermined the War on Poverty (New York: Oxford University Press 1996).

(84) . Jacob S. Hacker and Paul Pierson, “Business Power and Social Policy: Employers and the Formation of the American Welfare State,” Politics and Society 30 (2002): 277–325; Peter A. Swenson, “Varieties of Capitalist Interests: Power, Institutions, and the Regulatory Welfare State in the United States and Sweden,” Studies in American Political Development 18 (Spring 2004): 1–29; Jill Quadagno, One Nation, Uninsured: Why the U.S. Has No National Health Insurance (New York: Oxford University Press 2005).

(p.259) (85) . Howard, Welfare State Nobody Knows; Howard, Hidden Welfare State; Julian E. Zelizer, Taxing America: Wilbur D. Mills, Congress, and the State, 1945–1975 (Cambridge: Cambridge University Press, 1998); Hacker, Divided Welfare State.

(86) . Lloyd A. Free and Hadley Cantril, The Political Beliefs of Americans (New Brunswick: Rutgers University Press, 1967).

(87) . Free and Cantril, The Political Beliefs of Americans, 30; 12–15.

(88) . Benjamin I. Page and Lawrence R. Jacobs, Class War? What Americans Really Think about Economic Inequality (Chicago: University of Chicago Press, 2009).

(89) . Stanley Feldman and John Zaller, “Political Culture of Ambivalence: Ideological Responses to the Welfare State,” American Journal of Political Science 36 (1992): 268–307; see also Herbert McClosky and John Zaller, The American Ethos: Public Attitudes toward Capitalism and Democracy (Cambridge, MA: Harvard University Press 1984), and Jennifer Hochschild, What’s Fair? (Princeton: Princeton University Press 1981).

(90) . This figure is from 2000, the last time it was included in the ANES.

(91) . These figures are from 2008 except for homeless and financial aid for college (1996) and food stamps (2000).

(92) . Robert J. Blendon and John M. Benson, “Public Opinion at the Time of the Vote on Health Care,” New England Journal of Medicine, e55(2). See also Mollyann Brodie, Drew Altman, Claudia Deane, Sasha Buscho, and Elizabeth Hamel, “Liking the Pieces, Not the Package: Contradictions in Public Opinion During Health Reform,” Health Affairs 29, 6 (2010): 1125–1130.

(93) . Blendon and Benson, “Public Opinion at the Time of the Vote on Health Care,” p. e55(3).

(94) . McClosky and Zaller, The American Ethos.

(95) . Feldman and Zaller, “Political Culture.”

(96) . John Zaller, The Nature and Origins of Mass Opinion (New York: Cambridge University Press, 1992).

(97) . Jacobs, Health of Nations.

(98) . Marcia Angell, The Truth about the Drug Companies (New York: Random House 2004).

(99) . David Vogel, Fluctuating Fortunes: The Political Power of Business in America (New York: Basic Books 1989); Steven M. Teles, The Rise of the Conservative Legal Movement: The Battle for Control of the Law (Princeton: Princeton University Press, 2008).

(100) . Frank R. Baumgartner and Beth Leech, Basic Interests: The Importance of Groups in Politics and in Political Science (Princeton: Princeton University Press, 1998).

(101) . Jacob S. Hacker, “The Historical Logic of National Health Insurance: Structure and Sequence in the Development of British, Canadian, and U.S. Medical Policy,” Studies in American Political Development 12, 1 (April 1998): 57–130; Sven Steinmo and Jon Watts, “It’s the Institutions, Stupid! Why Comprehensive National Health Insurance Always Fails in America,” Journal of Health Politics, Policy and Law 20, 2 (1995): 329–372.

(102) . Some exceptions include Zelizer, Taxing America; Howard, Welfare State Nobody Knows; Howard, Hidden Welfare State; and Jonathan Oberlander, The Political Life of Medicare (Chicago: University of Chicago Press, 2003).

(103) . Quadagno, Color of Welfare.

(104) . Theodore R. Marmor, The Politics of Medicare, 2nd ed. (New York: Aldine de Gruyter, 2002); Zelizer, Taxing America.

(105) . Ellis W. Hawley, “Social Policy and the Liberal State in Twentieth-Century America,” in Federal Social Policy: The Historical Dimension, eds. Donald T. Crichlow and Ellis W. Hawley (University Park: Pennsylvania State University Press, 1988), 125–127; Barry D. Karl, The Uneasy State: The United States from 1915 to 1945 (Chicago: University of Chicago Press, 1983), 236–238.

(106) . R. Kent Weaver, “The Politics of Blame Avoidance,” Journal of Public Policy 6 (October–December 1986): 371–398.

(107) . Matthew A. Crenson and Francis E. Rourke, “By Way of Conclusion: American Bureaucracy since World War II,” in The New American State: Bureaucracies and Policies since World War II, ed. Louis Galambos (Baltimore: Johns Hopkins University Press, 1987), 149; see Theodore J. Lowi, The End of Liberalism (New York: W.W. Norton & Company 1969).

(p.260) (108) . James C. Robinson, “The Commercial Health Insurance Industry in an Era of Eroding Employer Coverage,” Health Affairs 25, 6 (November/December 2006): 1484.

(109) . William Niskanen, Bureaucracy and Representative Government (Chicago: Aldine-Atherton, 1971).

(110) . Anthony H. Pascal, “Clients, Consumers, and Citizens: Using Market Mechanisms for the Delivery of Public Services,” Rand Corporation Research Paper P-4803 (1972); Savas, Privatizing the Public Sector: How to Shrink Government (Chatham, NJ: Chatham House Publishers, 1982).

(111) . Savas, Privatizing the Public Sector; Stephen Moore, “Contracting Out: A Painless Alternative to the Budget Cutter’s Knife,” Proceedings of the Academy of Political Science 36, 3 (1987): 60–73; Simon Domberger and Paul Jensen, “Contracting Out by the Public Sector: Theory, Evidence, Prospects,” Oxford Review of Economic Policy 13, 4 (1997): 67–78.

(112) . Steve H. Hanke, “Privatization: Theory, Evidence, and Implementation,” Proceedings of the Academy of Political Science 35, 4 (1985): 101–113; Seidenstat, “Contracting Out.”

(113) . Julien Le Grand, “Quasi-Markets and Social Policy,” The Economic Journal 101, no. 408 (September 1991): 1256–1267; Michael Lipsky, and Steven Rathgeb Smith, “Nonprofit Organizations, Government, and the Welfare State,” Political Science Quarterly 104, 4 (1989–1990): 633–634.

(114) . Milton Friedman, Capitalism and Freedom (Chicago: University of Chicago Press, 1962); Pascal, “Clients, Consumers and Citizens.”

(115) . Smith and Lipsky, Nonprofits for Hire.

(116) . David M. Van Slyke, “The Mythology of Privatization in Contracting for Social Services,” Public Administration Review 63, 3 (May/June 2003): 296–315; Stevenson, “Privatization of Welfare Services,” 129.

(117) . Steven Rathgeb Smith and Judith Smyth, “Contracting for Services in a Decentralized System,” Journal of Public Administration and Theory 6, 2 (April 1996): 277–297.

(118) . Peter F. Drucker, “The Sickness of Government,” Public Interest 14 (Winter 1969): 3–23; James A. Morone, “Hidden Complications: Why Health Care Competition Needs Regulation,” American Prospect 10 (1992): 40–48; Kettl, Sharing Power; Lawrence D. Brown and Lawrence R. Jacobs, The Private Abuse of the Public Interest: Market Myths and Policy Muddles (Chicago: University of Chicago Press, 2008).

(119) . Smith and Lipsky, Nonprofits for Hire, 243–244.

(120) . Frederick C. Mosher, “The Changing Responsibilities and Tactics of the Federal Government,” Public Administration Review 40, 6 (November/December 1980): 541–548.

(121) . Mathew McCubbins, “The Legislative Design of Regulatory Structure,” American Journal of Political Science 29, 4 (1985): 721–748; Mathew McCubbins, Roger Noll, and Barry Weingast, “Administrative Procedures as Instruments of Political Control,” Journal of Law, Economics, and Organization 3, 2 (1987): 243–277.

(122) . Pascal, “Clients, Consumers, and Citizens.”

(123) . Pascal, “Clients, Consumers, and Citizens.”

(124) . Donald F. Kettl, “The Transformation of Governance: Globalization, Devolution, and the Role of Government,” Public Administration Review 60, 6 (November/December 2000): 494.

(125) . Albert O. Hirschman, Exit, Voice and Loyalty (Cambridge, MA: Harvard University Press, 1970).

(126) . John E. Chubb and Terry M. Moe, Politics, Markets, and America’s Schools (Washington, DC: Brookings, 1990); Le Grand, “Quasi-Markets.”

(127) . Smith and Lipsky, Nonprofits for Hire, 249.

(128) . For a vivid account of the consequences of the delegation of welfare programs to for-profit firms, see Joe Soss, Richard C. Fording, and Sanford F. Schram, Disciplining the Poor: Neoliberal Paternalism and the Persistent Power of Race (forthcoming, University of Chicago Press).

(129) . McConnell, Private Power, 51.

(130) . Novak, “‘Weak’ American State.”

(131) . Louis Galambos, “By Way of Introduction,” in The New American State: Bureaucracies and Policies since World War II, ed. Galambos (Baltimore: Johns Hopkins University Press, 1987), (p.261) 1–20; Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920 (Cambridge: Cambridge University Press, 1982).

(132) . Skocpol and Finegold, “State Capacity”; Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies, 1862–1928 (Princeton: Princeton University Press, 2001).

(133) . Data are from 2006; the American figure is even further from the European Union average, which is 39.8 percent of GDP. OECD in Figures (Paris: OECD, 2009).

(134) . Laura S. Jensen, “Government, the State, and Governance,” Polity 40, 3 (July 2008): 379–385.

(135) . Skocpol, Protecting Soldiers and Mothers; Kimberly J. Morgan and Monica Prasad, “The Origins of Tax Systems: A French-American Comparison,” American Journal of Sociology 114, 5 (March 2009): 1350–1394.

(136) . Aaron L. Friedberg, “American Antistatism and the Founding of the Cold War State,” in Shaped by War and Trade: International Influences on American Political Development, eds. Ira Katznelson and Martin Shefter (Princeton: Princeton University Press, 2002), 239–266.

(137) . Peter Baldwin, “Beyond Weak and Strong: Rethinking the State in Comparative Policy History,” Journal of Policy History 17, 1 (2005): 12–33; Desmond King and Robert C. Lieberman, “Ironies of State Building: A Comparative Perspective on the American State,” World Politics 61, 3 (July 2009): 547–588.

(138) . Lowi, End of Liberalism, xi.

(139) . We are grateful to Ann Orloff for her insights on this.

(140) . Clemens, “Rube Goldberg State”; Kimberley S. Johnson, Governing the American State: Congress and the New Federalism, 1877–1929 (Princeton: Princeton University Press, 2007).

(141) . Mosher, “Tactics of the Federal Government”; DiIulio, “Government by Proxy.”

(142) . Paul C. Light, “The New True Size of Government,” Organizational Performance Initiative Research Brief No. 2 (New York University Wagner School, 2006).

(143) . Mosher, “Tactics of the Federal Government,” 543.