Welfare policy in the United States rests on judgments about poor people's worthiness. According to conventional wisdom, those who are poor through no fault of their own deserve our support, so long as they strive to become more self-reliant. But anyone whose behavior leads them into poverty, or prevents them from escaping it, merits condemnation, not aid. To enforce this distinction we have established a fine mesh of federal and state policies to prevent the undeserving poor from obtaining public assistance.

Of course, political judgments about people's worthiness are themselves open to judgment. It is fair to ask if we have collectively done right by the poor, according to our own lights. Does our current system of governing the poor deserve the approbation of citizens who subscribe to democratic values, including equal treatment under the law? Or are present policies contributing to the persistence of racially disparate outcomes, and perhaps even reinforcing them, under the guise of promoting personal responsibility?

This is the motivating concern of Joe Soss, Richard Fording, and Sanford Schram, who make four important claims in Disciplining the Poor. First, they characterize our current method of governing the poor as a combination of discipline and incarceration. In so doing, they move beyond the argument of Frances Fox Piven and Richard Cloward in Regulating the Poor (1971), which analyzed policies that exclude people from welfare and push them toward low-wage jobs. That is still an important aspect of welfare provision, but Soss, Fording, and Schram persuasively argue that current policies are more ambitious. Temporary Assistance to Needy Families (TANF), which replaced Aid to Families with Dependent Children (AFDC), now involves federal, state, and local agencies in extensive workforce training and development, thereby subsidizing low-wage employers who require a dependable supply of newly disciplined laborers.

Punishment for crimes born of poverty is now part of the regime, too. In fact, Loïc Wacquant suggests that imprisonment has displaced regulation as the dominant form of social control since the late 1960s. Soss, Fording, and Schram reject that exaggeration, but they accept the need to update Piven and Cloward's account of regulatory mechanisms to include incarceration. Indeed, they show that the number of persons receiving AFDC/TANF declined by 50% during the 1990s, even as the number of people incarcerated increased by 50%. This is no coincidence, as “get tough” reforms of welfare and criminal sentencing procedures were simultaneously pursued by several states during this period.

The second, and certainly most arresting, claim advanced by Soss, Fording, and Schram is that the current policy regime produces a racialized pattern of outcomes. They present compelling evidence that African Americans are far more likely to be disciplined and punished than whites (or even other minorities), legal guarantees of equal protection under the law notwithstanding. This is especially true after a Republican Congress—with the approval of Democratic President Bill Clinton—ended “welfare as we knew it” under AFDC.

Instilling discipline became a national goal with passage of the Personal Responsibility and Work Opportunity Act of 1996, which greatly expanded state policymakers' freedom to choose disciplinary methods. The most stringent limits and work requirements were imposed by states with large populations of African Americans. In fact, such states were more than twice as likely to adopt restrictive TANF policies, even when other factors were taken into account. This pattern was accentuated in the 14 states that devolved responsibility for TANF to local governments or regional boards. Locales with large proportions of African Americans were more likely to impose stringent discipline on TANF recipients than areas in the same state with low proportions of African Americans, particularly where African Americans constituted a large proportion of the total population in a state.

The cumulative result of these disciplinary actions was disconcertingly large. By 2001, white families comprised 63% of the TANF caseloads in states with the least restrictive policies. Black families accounted for only 11% of the caseloads there. In states with the most restrictive policies, however, black families were 63% of the TANF caseload, whereas white families represented only 29% of the caseload. All things being equal, blacks in these states were also subjected to higher rates of incarceration. Thus, Soss, Fording, and Schram, looking across the American states, “find a tightly configured relationship consisting of higher black population rates, more stringent and locally controlled TANF regimes, and more aggressive applications of correctional control” (p. 137).

What accounts for such disparate outcomes, now that outright discrimination is illegal and racial prejudice among policymakers and citizens alike seems to be declining? This outcome is frequently attributed to covert racism on the part of white officials, but the authors eschew that explanation. Instead, they present a new way of understanding when, where, and how widely shared assumptions about welfare recipients lead contemporary overseers of the poor to adopt policies that yield racially disparate outcomes across and within states.

Soss, Fording, and Schram contend that policy choices are guided by decision makers' tacit assumptions about the people they target for policy action. These assumptions develop in cultural discourses and operate as shared cognitive frameworks used by policymakers to identify problems and devise solutions to them. Policymakers who understand poverty as a structural problem will choose policies quite different from those favored by policymakers who attribute the problem to welfare recipients' laziness, immorality, or irresponsibility. The former will wage war on poverty; the latter will prescribe various forms of discipline as the appropriate remedy for recipients. The discipline will be imposed on groups that show the highest propensity for inappropriate conduct, that is the target population of welfare or correctional policy.

Racial classifications and reputations unavoidably enter into policymakers' analysis in varying degrees. They are culturally more salient in some places, and at some times, than others. Hence, “the likelihood of racially patterned outcomes will be positively associated with the degree of policy-relevant contrast in policy actors' perceptions of racial groups … which is a function of (a) prevailing cultural stereotypes of racial groups, (b) the extent to which policy actors hold these stereotypes, and (c) the presence or absence of stereotype-consistent cues” (p. 79).

The authors' third major claim, then, is that the observed pattern of racialized outcomes, which is more pronounced in some states than others, reflects differences in policymakers' “mental maps,” which are in turn rooted in their different sociocultural contexts. Officials overseeing large numbers of black recipients will rely on harsh discipline in contexts where blacks' behavior is viewed stereotypically. Officials in states with relatively few blacks on welfare may conclude that whites would suffer unduly from discipline, even if those officials take the same view of blacks' behavior. It all depends on the salience of group stereotypes and the behavioral tendencies attributed to people of different races.

This line of analysis is codified in the authors' Racial Classification Model (RCM), which predicts that racialization of policy outcomes is contingent on time and place. Drawing on cognitive psychology, Soss, Fording, and Schram derive several explicit hypotheses from the RCM. Those hypotheses are subjected to rigorous statistical analyses of rich and relevant data assembled by the authors, which are fully reported in an online appendix at http://www.discipliningthepoor.com. These analyses are buttressed by insights gained from a detailed case study of poverty governance in Florida, which has been lauded as a national model, and where all the elements of “neoliberal paternalism” come into play.

This brings me to the authors' fourth claim, which is that a turn toward “neoliberal paternalism” explains the emergence of a system for governing the poor that relies on discipline and punishment. In a sense, American policy has always been paternalistic, and has leaned in the direction of regulating the poor, if not disciplining them. But Soss, Fording, and Schram contend that a neoliberal form of paternalism triumphed in the 1990s as the federal government created incentives for states to experiment with new ways of instilling discipline among the poor. In the process, the traditional determinants of policy choice at the state-level, as well as their “street level” application, have receded in importance. The racial composition of the target population, that is, the blackness of welfare caseloads, has become more important and is now the dominant explanation for varying policy outcomes across the states.

This new form of paternalism is neoliberal insofar as it stresses the importance of personal responsibility, now seen as an object of public policy, which may increase the efficacy of personal action by training and discipline. Put that way, Soss, Fording, and Schram have no objection to policies aimed at promoting greater self-reliance. What bothers them is that our current definition of what constitutes self-sufficiency penalizes certain groups by ignoring or downplaying barriers to behaviors deemed normal for white, or more broadly, “responsible,” citizens. As they put it, “Neoliberal paternalism is a decentralized, disciplinary mode of poverty governance; it seeks to instill and insist on preferred behaviors among the poor; it weaves penal and welfare operations together as instruments for this project; and its construction and operation are both fundamentally entwined with race” (p. 137).

This is certainly a plausible framing of welfare policy over the past two decades, though it is not the only one. One could, for example, portray these developments as paternalist without a neoliberal qualifier. Somewhat less plausibly, one might describe it as neoliberal without invoking paternalism. But one could follow Wacquant in blaming liberalism for underwriting the current regime of poverty governance, or accept conservative claims that welfare-state liberalism was actually a corruption of liberalism, and that neoliberalism is really the old and true liberalism. Much depends on which social theory is admitted to the discussion of policy outcomes and their determinants.

That said, the rise of neoliberal paternalism is a more plausible explanation for changes in federal and state welfare policy than it is for changes in state penal policy. If disciplining the poor is understood as a form of neoliberal paternalism, the corresponding emphasis in criminal sentencing policy ought to exhibit some tendency toward rehabilitation, that is, the correction of behavior. Instead, sentencing policy in the United States for the past two or three decades has been punitive in orientation. It is neither neoliberal nor paternalist and, as the authors note, it actually aggravates the problems of poverty in minority neighborhoods and communities. Contemporary American penal policy may be an inconsistency in neoliberal paternalism, but it also poses a challenge to the explanatory framework offered by Soss, Fording, and Schram, who treat penal policy as an aspect of neoliberal paternalism.

Disciplining the Poor is a provocative book and a disturbing one, insofar as it renders a harsh judgment of our current governance of the poor. The book is a model of politically engaged research that is theoretically informed and rigorously prosecuted. Indeed, it epitomizes the kind of synthetic, multidisciplinary research now being touted by many social scientists. It is therefore dismaying that no federal agency or private foundation provided major funding for this groundbreaking analysis. Fortunately, the University of Chicago Press and its reviewers saw the value of this powerful research, which surely deserves the attention of scholars, policymakers, and citizens.