You Win Some, You Lose Some...: Explaining the Pattern of Success and Failure in the Second Reconstruction
Jennifer L. Hochschild
February 27, 1998
For publication in Morton Keller and R. Shep Melnick, eds., Taking Stock: Policy and Governance in the Twentieth Century (Cambridge University Press, forthcoming 1998)
NOTE: not quite final version.
My thanks to Michael Danielson, Edwin Dorn, John Goering, Robert Gooding-Williams, M. Gould, Bonnie Honig, Morton Keller, Robert Lieberman, David Mayhew, R. Shep Melnick, Michael Rogin, John Skrentny, Earl Smith, Clarence Stone, J. Philip Thompson, Fred Woodward, Frederick Wirt, and two anonymous readers for very helpful comments on various drafts of this chapter.
White Americans believe on balance that the American racial structure has been successfully transformed since the 1950s, and that overall African Americans have the same political, social, and economic opportunities as other Americans. For example, in a 1995 Washington Post survey, 55% of whites (compared with 29% of blacks) mistakenly agreed that “the average African American” is at least as well off as “the average white person” in terms of jobs and education. Over 40% of whites (and about 20% of blacks) held the same view with regard to housing and income. Between one-tenth and one-third of whites believe that compared with whites, blacks have more opportunities, are less vulnerable to economic upheaval, receive better health care, are treated better in the courts and the media, and are more likely to obtain good jobs and be admitted to good colleges.
Black Americans do not on balance share that belief. In some surveys, up to half claim that the situation of blacks has worsened since some reference point in the past. Although only a handful of whites think that more than half of whites share the attitudes of the Ku Klux Klan, a quarter of blacks see more than half of whites as Klan sympathizers. In that context it is not surprising that over half of blacks but only a quarter to a third of whites think our nation is still moving toward two separate and unequal societies.
How can citizens of the same country, with the same national and shared local histories and with no apparent disparity in perceptual capacity, have such different -- and increasingly different -- views? And how can our nation sustain any kind of shared culture, set of political practices, and social intercourse with a gap so wide, intense, bitter, and growing? I will not answer those questions in this essay, but they provide the setting for the question that I do address. In particular, they suggest the urgency of understanding the historical trajectory of the movement for racial equality in the past few decades; if we can develop a clearer sense of what has and has not changed and why, perhaps we can work more effectively toward further reforms over the next few decades.
This chapter explores the recent history of struggles for African American civil rights, more schematically than most historians would but with the same focus on change over time as the central phenomenon to be understood. It thus uses the tools of social science in the service of both historical explanation and policy prescription. The chapter proceeds through several steps. I briefly provide evidence showing that white racial domination in the United States is declining in some ways but persisting or even growing in others. That pattern explains much of the disparity between white and black Americans’ perceptions of movement toward racial equality. I then propose reasons for this pattern that center on governmental structures and policy processes. Once the pattern is clear and its (proximate) explanations adduced, I conclude by suggesting what is needed for the pattern of racial policy-making to shift in the direction of more successes and fewer failures.
Success, Partial Success, and Failure
Recent policies to transform the American racial structure have had some indisputable and probably permanent successes. These are manifestations of the abolition of state-endorsed racial segregation, discrimination, and domination. The political system no longer creates and supports overt racial hierarchies, and in principle it may no longer ignore or reinforce covert racial hierarchies. Included in that abolition are:
* desegregation of public accommodations, such as public transportation, hotels and restaurants, gas stations, hospitals, recreational facilities, and other locations of transactions that are directly run by the state or serve public functions.
* desegregation of housing, such that communities may no longer exclude residents of a disfavored race, public housing authorities may not discriminate by race, and private sellers or real estate agents may no longer discriminate against potential buyers who are members of a disfavored race.
* illegality of employment discrimination, such that most employers may no longer refuse to hire or promote otherwise qualified workers because they are members of the disfavored race.
*desegregation of the armed services, such that the acceptance of citizens into the armed services, living and occupational conditions of soldiers, and promotion may not be racially structured. Furthermore, some branches of the armed services, notably the Army, have taken substantial steps to ensure the abolition of racially biased behaviors and to promote the success of black soldiers.
* ending of discrimination in political participation, such that citizens may no longer be excluded from voting or seeking office because of their race or justifications that are thinly veiled stand-ins for race. In addition, electoral systems may no longer be structured in order to dilute the effects of black electoral participation.
* ending of discrimination among participants in professional sports and the entertainment industry, such that many of the nation’s most successful, best-paid, and highly adulated athletes, musicians, and actors are African American.
None of these successes are complete; housing segregation, separate schooling, the absence of people of color in the management of sports and entertainment, and employment discrimination all persist. Nevertheless, the policy of racial equality is uncontroversially established in all of these arenas, and many African Americans have benefited from reductions in housing, school, and job discrimination since the 1960s. It is even arguable that recent policy initiatives, such as federal monitoring of discrimination in mortgage decisions, are accelerating the move from official policy to actual practice.
Partly as a consequence of these changes, the United States now enjoys a substantial and thriving black middle class. In the 1960s, blacks enjoyed “a perverse sort of egalitarianism” of almost uniform poverty. But by conventional measures of income, education, and occupation, about one-third of the African American population can now be called middle class (compared to about one-half of the white population). The best-off third of African Americans hold a larger share of their race’s income than do the best-off third of white Americans, and the gaps between well-off and poor blacks are greater than the parallel gaps between well- and badly-off whites with regard to educational attainment, occupational status, likelihood of being a victim of a crime, and even mortality. As far as I know, no other ethnic group in the history of the United States has achieved such a rapid and extensive growth of a middle class in one generation.
This is a long and impressive list of accomplishments. It would be misleading, however, without consideration of a comparable list of even less complete successes and outright failures in the effort to end white racial domination. Partial successes include:
* school desegregation. Schools are no longer formally segregated, and overall the number of students attending school with students of another race has increased since the 1950s and early 1960s. But racial separation in schools, with its attendant disparities in educational resources and social standing, was never eliminated and is now rising. Since mandated school desegregation began, racial separation within each school, either through tracked classes or within classrooms, has always remained higher than racial separation across schools. Few members of either the black or white communities now insist on desegregated schools. Conversely, there is considerable and increasing sentiment -- sometimes backed by court orders -- to eliminate extant school desegregation plans that rely on mechanisms other than voluntary student transfers or that use tax resources to increase the likelihood of voluntary student transfers.
* residential desegregation. Housing is no longer formally segregated, federal loan policies no longer encourage racial homogeneity in neighborhoods, and banks may not legally discriminate in lending for mortgages. But residential segregation is declining in most metropolitan areas at an extremely slow rate, and banking practices are changing slowly and reluctantly at best. These facts obtain for African Americans of all classes, and they obtain more for African Americans than for any other racial or ethnic group.
* political empowerment. The proportion of blacks who voted rose dramatically in the 1950s and 1960s, stayed roughly even during the 1970s and 1980s, and has declined since then. The proportion of the voting population that is black has similarly risen, then fallen. Again, the proportion of elected offices held by African Americans rose dramatically and has recently leveled off. Recent Supreme Court decisions make it unlikely or even impossible for voting districts to be designed in order to increase the number of black elected officials; redistricting in the foreseeable future may in fact reduce that number.
* affirmative action. The federal government committed itself and its contractors and grant recipients to a fairly robust program of affirmative action during the 1970s. The rules remained in place during the 1980s, but enforcement was lax or nonexistent. In the 1990s, many elected officials and some federal courts have backed off from even that low level of commitment, California’s voters approved a referendum to abolish affirmative action in 1996, although a year later Houston’s voters did not. The future of affirmative action policies and practices is uncertain.
* the criminal justice system. Before the 1960s, African Americans in the south (the vast majority) could almost never serve on juries or appear in courtrooms as attorneys. They could be sure, whether as plaintiffs or defendents, not to receive the same treatment as whites. That radical disparity has mostly been eliminated. In addition, legal reforms of the 1960s, including the provision of attorneys to the indigent and the implementation of the Miranda rule, have materially improved the chances in court of the poor, who are disproportionately people of color. Since the 1960s, affirmative action and anti-discrimination policies have made police departments and juries demographically more similar to residents of cities than at any time since the Irish dominated both the immigrant rolls and urban public offices. Nevertheless, African Americans still complain vigorously and often correctly of police brutality in the streets, malfeasance in the courtroom, and racial discrimination at all stages of the criminal justice process from writing laws to executing murderers to supervision of parolees.
Among public policies intended to eliminate black subordination, the set of failures is most easily summarized by the phrase “inner cities.” Components of that failure include:
* an atrocious school system. Although the overall dropout rate for African Americans has declined dramatically over the past few decades, in some large city school systems over half of black teens leave school before graduation. Fully half of black adults living in extreme poverty areas of large cities have not completed high school. In fourteen schools in Hartford, Connecticut, fewer than 5 percent of fourth-graders have achieved grade-level mastery of reading; “in 14 of the District [of Columbia]’s 18 high schools, more than 94 percent of the students tested were functioning below grade level in math.” School boards are sometimes corrupt, often motivated more by political ambition than by educational vision, and not infrequently ignorant about their charges. Inner city schools lack windows and basic plumbing; students and teachers must dodge falling plaster and puddles; coal-fired furnaces labor to provide enough heat to cope with New York winters.
* an overwhelmed social service system: Hospitals in New York City took care of almost 13,000 “boarder babies” during a 31 month period in the late 1980s. Substantiated reports of child abuse rose from 500,000 in 1980 to over 1 million in 1993, and rates of child abuse are higher among the poor and African Americans than the rest of the population (although they are not higher among African Americans once social class is controlled for). The foster care caseload in New York State doubled in the two years after 1987 and grew by 44% in California from 1985 to 1988. Foster children also are disproportionately poor, urban, and black. Social service systems in many big cities confront horror stories of children who are murdered by their parents partly because case workers have no time for proper investigations and record-keeping systems are hopelessly outdated and inadequate.
* an inadequate or possibly counterproductive income support system: Over one-quarter of black children, compared with 6% of white children, lived below half of the poverty line in 1988. During the past twenty years, about one-third of black children remained poor for six or more years, and the number is rising. In the 100 largest central cities, the proportion of poor blacks living in neighborhoods where more than 40% of their neighbors were also poor rose from 28% in 1970 to 42% in 1990. Between 1983 and 1992, the poorest fifth of whites gained only 5% ($570) over their previous income, but the poorest fifth of African Americans lost 12% ($558) of their meager income. Many Americans believe (although the research evidence does not clearly substantiate the view) that dependence on AFDC encouraged teens to bear children with no intention to marry and no capacity to support their children. Many also argue that “the welfare system” provided enough of a financial cushion to inhibit adult parents from seeking or maintaining jobs.
* a deteriorating public infrastructure: Disparities in wealth and amenities between central cities and their suburbs have grown steadily since 1970. Cities struggling to balance their budgets in the face of declining federal aid and growing proportions of poor residents frequently close branch libraries, hospitals, and fire stations, and defer maintenance on bridges and sewer systems. Some cities have “solved” the problems of high-rise public housing projects by dynamiting them into rubble; the waiting list for the relatively adequate public housing system in New York City has a backlog of almost a decade.
*an unevenly distributed criminal justice system: Although victimization from violence, theft, and household crimes steadily declined in the United States over the past few decades, victimization in cities rose as a proportion of all victimization. More precisely, in 1976, 5% of poor blacks suffered from violence, compared with 4% of well-off blacks, 4% of poor whites, and almost 3% of well-off whites. By 1992, the affluent of both races were less likely to be victimized -- violence rates had fallen to 3.5% among well-off blacks and 2% for well-off whites. However, the poor, and especially the black poor, had become more likely targets -- victimization rates from violence had risen to 4.5% for poor whites and to fully 6% for poor blacks. During the 1980s and early 1990s, violent and often fatal attacks by young black men on each other skyrocketed, primarily in large cities.
And so on -- the list could continue. The very cumulation of ills produces a dynamic of its own; each policy failure makes the others more harmful to citizens and harder for public officials to deal with, and the overall effect is “a process of social contagion that generates epidemics of social problems ... [in] America’s inner cities.”
Can these policy failures be legitimately considered a part of American racial policy? Not as obviously as school or housing desegregation, but I would nevertheless argue that they are intrinsic to it. This point is worth dwelling on for a moment, because how one answers that question reveals a lot about how one connects race and class, public and private, overt and covert policy-making. It is true that the cumulative failure of public services in cities is not explicitly a matter of racial policy. After all, some poor European Americans, Latinos, and Asian Americans live in central cities, and almost half of African Americans do not. In addition, one can explain the problems of inner cities in terms of poverty, mismanagement, inadvertence, or the inexorable workings of advanced capitalism rather than in terms of racial domination. Finally, in this view the civil rights movement was aimed at ending official structures of racial domination, not at economic equality or the control of market forces.
That view, however, seems too narrow and excessively formalistic. For one thing, residents of the most severely dysfunctional urban neighborhoods are disproportionately black, and African Americans are disproportionately harmed by the failures of cities to provide decent living conditions. In addition, the very boundaries between city and suburb which help to concentrate poverty, unemployment, and mismanagement as well as people of color within the city are themselves the result of deliberate policy choices which were partly made on racial grounds. Finally, African Americans involved in the drive for racial equality throughout the twentieth century never sharply separated civil and legal rights from political empowerment and social and economic betterment. On the contrary, black leaders have always firmly linked civil rights with social welfare, and fought for both simultaneously. Thus it seems appropriate to include policy failures in large cities in the same domain as policy successes in the courts and military services.
Readers probably know, at least in general outline, all of what I have just rehearsed; I do not, in any case, see much that is controversial up to this point. My reason for this rehearsal is to emphasize the wide range of outcomes in recent efforts to transform the American racial structure. Focusing on that range allows us to do several things. First, it helps us to understand the disparity in trajectories of white and black racial views described in my introduction. Whites legitimately point to the impressive list of transformations, which after all have come at some cost to members of their race. Blacks equally legitimately point to the no less impressive list of partial transformations and failures, which persist at great cost to members of their race.
Second, emphasizing the range of outcomes of efforts over the past four decades to transform the American racial structure leads us to ask why? Why have efforts in some some policy arenas produced substantial and presumably permanent change for the better, whereas other efforts have produced transitory improvements and still others none at all (for the better, at least)? That is the question to which the remainder of this chapter is devoted.
“Power to...”: Explanations that Inhere in the Policy
Some policies are easier to design, promulgate, implement, maintain, and measure than others. As soon as one says that, caveats rush in: policy-makers (and citizens) are poor judges of what will be easy to do; policies that seem easy to implement turn out to be, or are made to be, complicated; unanticipated consequences always abound and sometimes rule the day. “Easy” policies to enable African Americans to become first class citizens took over a century to promulgate after slavery was abolished, and required violence and death to move from promulgation to implementation. Nevertheless, if nothing is easy, some things are easier than others.
What kind of policy initiatives might be easier to promulgate and sustain than others? Put otherwise, what kind of policy initiatives contain within them the power to accomplish their end? One promising kind are policies that offset market distortions created by “inefficient” tastes, such that a market freed by government mandate from such artificial distortions will then automatically and successfully take over the enforcement effort. One inefficient taste is that for racial discrimination. This explanation helps to explain the success of desegregation of public accommodations: once restaurant owners and bus companies were forced to be free of their own and their current customers’ discriminatory preferences, they found it easy and profitable to accept the dollar bills of anyone who cared to proffer them. The same principle partly accounts for the partial success of affirmative action. Once corporate managers and foremen were forced to look beyond their assumptions about who makes a good worker, they found that African American employees could at least enhance their relations with surrounding communities and draw in new customers, and at most provide an almost untapped pool of talent and new perspectives. The same principle accounts for the success of black athletes and singers; many are simply more talented than their white counterparts, and once the market was forced to operate according to criteria of merit and consumer preference, the racial composition of stadiums and concert stages changed dramatically.
But the fact that a policy initiative corrects a market distortion is neither a necessary nor a sufficient condition for that policy to succeed. Market corrections do not help to explain some successes in racial policy such as voting rights, desegregation of the armed forces, and changes in the criminal justice system. And the claim that market corrections are always effective is contradicted by the fact that residential segregation persists even in affluent communities. One would expect a priori that once the government required a free market in home sales and rentals, the incentives for the real estate industry would be to seek out as many clients as possible and to encourage as many buyer-seller matches as possible. But that has not been the case.
A second kind of policy that should be relatively easy to promulgate and sustain is one that is a “single-stage policy” -- colloquially, one that can be created by a single stroke of the pen. A single-stage policy has “no intermediate steps between adoption of the policy and production of the effects;” a classic example is a legislative change in the tax code. A single-stage policy can be distinguished from a multiple-stage policy, which implies a long causal chain from, say, prenatal health care through Head Start through raising teachers’ salaries in order eventually to improve high school graduation rates.
Here is the most straightforward meaning of “power to”: although no policy worth its salt is very easily implemented (think of the number of tax accountants and H. & R. Block employees needed to implement a change in the tax code), the fewer the stages that a new policy must go through the more likely it is to succeed. Most of the clear successes in American racial policy are closer to single- than to multiple-stage policies. Once an end to discrimination against black would-be voters is required, it is fairly easy to figure out how to stop discrimination in registration and polling places. Once the owner of a baseball team decides to try a “racial experiment,” he needs only to find a Jackie Robinson waiting in the wings.
But many imaginable single-stage policies that would affect the racial structure in the United States have not been implemented, so this explanation too is insufficient to explain the full pattern of racial policy success and failure. Examples range from reparations to African Americans whose ancesters lived in the United States before 1865, to redrawing school district or municipal boundary lines, to promulgating electoral mechanisms such as proportional representation or cumulative voting. Conversely, some multiple-stage policies have been successfully implemented, so having only one step in the policy process is also not necessary for success. After all, the armed services’ efforts to move beyond abolishing discrimination to recruiting, training, and promoting black officers have required many steps and much reworking of old policy. They still have not succeeded completely. But they have not been abandoned, as have other complicated efforts such as mandatory school desegregation, and on balance they are effective enough to provide a model for the rest of the nation.
So far I have described policies in terms of their internal characteristics. One can also characterize policies with “power to” (be relatively easily implemented) in terms of the distribution of their intended effects. For example, a policy that diffuses costs among the public and concentrates benefits among the politically active is usually thought to be relatively easy to promulgate and sustain. At first glance, this rule of thumb does little to explain the pattern of successes and failures in American racial policy. On the one hand, many successful transformations imposed clear costs on some citizens who were symbolically or substantively important to the electoral system; examples include desegregation of the armed forces and the abolition of discriminatory structures for voting and office holding. On the other hand, a few successful transformations provided benefits to citizens who were symbolically or substantively unimportant to the electoral system; one example is the provision of attorneys to indigent defendants. More importantly, the failure of policy structures in inner cities is a case of concentrated and highly visible costs to some voters (inner city residents and people who live in suburbs but would prefer to live in the city), and no publicly legitimate benefits to any.
The converse of this principle, however, does help to explain why some apparent successes were only transitory. That is, a policy that imposes high costs on many voters, with unclear or concentrated benefits for a few, will be very hard to sustain absent some form of powerful support. School desegregation was very costly to white citizens in their view, and seems increasingly costly to black citizens. Few whites and not many more blacks see countervailingly clear benefits from it. It has thus lost most of what political constituency it had. Affirmative action similarly seems to impose high costs on many white employees, with few beneficiaries except for “undeserving” blacks (and perhaps corporate managers and stockholders). As the economy appeared to worsen and white workers became more insecure in the mid-1990s, the costs of affirmative action rose in their view, and their general dissatisfaction with the policy hardened in some cases to vehement opposition.
A similar rule of thumb is that the most sustainable policies are those that allow citizens to engage in positive-sum rather than zero-sum or negative-sum games. That is, if most people can gain from the promulgation of a new policy, it is good politics as well as good policy. That is the theory behind the Laffer curve: lowering the rate of taxation on the wealthy will directly benefit them; they will invest more and more productively in the economy, which will produce more jobs and higher tax revenues, thus indirectly benefiting all citizens. High rates of taxation, conversely, illustrate a negative-sum game; they harm the wealthy who withdraw from the economy, thereby harming everyone else.
Racial policies that come to be understood as positive-sum are among those on the list of successes. The best examples are the abolition of discrimination in public accommodations and in sports and entertainment. In parallel fashion, racial policies that come to be understood as zero-sum or even negative-sum are among those on the list of transitory successes or failures. For example, many whites fear that if their neighbors sell or rent their houses to African Americans, the value of their own property will decline or the quality of life in their community will deteriorate. They thus see residential desegregation as a gain to the seller but a loss to the rest of the neighborhood. Their pressure, either directly on their old white friends and new black neighbors, or indirectly through social networks and ties to real estate agencies, is enough to keep the 1968 Housing Act the least well implemented of the three big civil rights acts of the 1960s.
But creating a positive-sum game is neither necessary nor sufficient to explain the pattern of success and failure in American racial policy. On the one hand, some zero-sum games have been sustained: the best examples are assertive policies of promotion of African Americans in the armed services and changes in local voting structures from at-large to single-member districts. White would-be military officers and political office-holders lost in both cases, and one would not have expected them to be without political influence. On the other hand, some positive-sum games have not been sustained or even begun: it would presumably be in everyone’s interest for children in inner-city schools to be able to read and calculate, and yet the policies and practices of inner-city schools contribute to the academic failure of many children. So this policy characteristic, too, does not suffice to answer the puzzle posed at the beginning of the chapter.
So far I have described four characteristics of policies themselves that presumably make them more likely to be promulgated and sustained than policies with the opposite characteristics. All of these characteristics have to do with the “power to” -- the ability to accomplish what one sets out to do because the policy is straightforward, accords with powerful market mechanisms, is politically attractive, or substantively benefits most citizens. Some evidence suggests the validity of each claim of “power to” but other evidence either contradicts each claim or suggests that that claim is irrelevant. So we have only partially explained the puzzle of why some racial policies have mostly succeeded, others partly or temporarily succeeded, and others failed over the past four decades.
“Power Over...”: Explanations that Adhere in Policy-makers
I turn now to potential explanations for the pattern of success and failure that focus on who is making the policy rather than on what the policy consists of. These explanations revolve around the question of “power over...” -- who has the capability and incentive to make and enforce new policy. Here too we have several plausible candidates.
First, policy makers who can create or build on strong institutional structures are more likely to be able to impose their preferences than policy-makers who are flying without much of a net. Even if the origins of a robust institution are controversial, it becomes robust by virtue of the citizenry’s eventual acceptance of its mission and activity. Officials located within such a robust institution typically are committed to the success of a policy initiative, understand that their salaries and professional prestige depend on the policy’s success, and know how to attain the resources and shape the tools needed to implement, enforce, evaluate, and modify a new policy. Once a policy is firmly lodged within a robust institution, it is relatively easy to keep it going and to ensure that modifications are constructive rather than threatening. So, at any rate, goes the basic theory of institutional capacity and path dependency.
The creation of strong institutions designed to change racial policy and to ensure the continuation of those changes does help to explain some racial successes. New laws and new agents of the Justice Department were needed to police voting regulations and practices and to create new boundaries for voting districts. Once those changes were in place, path dependency came into play: African Americans were increasingly elected to office, and they attained the status, means, and opportunities of incumbent in the (usually Democratic) party organization. From that point on, they had many resources besides their race to help them attain re-election, move to a higher office, and promote the prospects of more junior black candidates. Similarly, once the Army reconfigured its barracks and other facilities to accommodate desegregation, created and staffed race relations training programs, and promoted many black officers, this new set of structures, processes, and actors made it more likely that racial equalization would persist and develop.
However, creation of an apparently strong institution is not sufficient to sustain a racial policy. After all, mandates to desegregate schools and engage in affirmative hiring generated complex and elaborated bureaucracies, with the committed personnel, budgets, and new procedures attendant on a robust institution. But school desegregation offices are being dismantled, and affirmative action offices are under attack. There is no institution stronger than the Federal Reserve Board, yet its repeated statements of commitment to the implementation of the Equal Credit Opportunity Act of 1974 have not had much impact on the practices of mortgage-granting bankers.
Nor is the creation of a powerful institution necessary for success. Again public accommodations provide the best example, in this case of a major change in one aspect of racial domination with essentially no institutional penumbra.
Consider next a more direct manifestation of “power over...” -- the ability of one political actor to require another to change its behavior, and to impose the costs of change on the one doing the changing. President Truman in Executive Order 9981 told the armed forces to desegregate; the Supreme Court in Brown v. Board of Education told southern educators and judges to desegregate schools; Northern members of Congress told southern election officials to desegregate voting booths; Judge Garrity remained in Wellesley while he told residents of Boston to desegregate their schools; Branch Rickey told the Brooklyn Dodgers to play ball with their new teammate. Each of these examples manifests “A getting B to do something that B would not otherwise do” without A having to do the same thing. One branch of government gives an order to another; one region of the country imposes a change on another; one policy actor gives a command that he or she need not live under; an owner of a business imposes a new work rule on his employees.
But like the others I have discussed, this characteristic of policy makers does not by itself fully explain the pattern of the United States’ racial successes and failures. Some policies imposed on reluctant recipients have stuck, whereas others have not; again contrast desegregation of the armed forces and sports teams with desegregation of schools, or the persistent, if bumpy, reduction of employment discrimination with the faltering path of affirmative action. Why is that the case? Partly because some policies are easier to impose and implement than others, which takes us back to the “power to” characteristics already discussed. Partly because some policies cost a lot more than others (costs can be financial, organizational, or political), and even powerful policy actors face limits to the costs they can impose on others. And partly because powerful actors only sometimes decide to make other less-powerful actors carry out and live under a new and unpopular policy. After all, many of the policies that have failed most severely in inner cities could be improved if an external authority imposed reform, provided some resources, and mandated that city officials pay the remaining costs. Consider social services for children: Case workers are paid very poorly, they are drastically overworked, their offices are just edging into the computer age, and they are frequently overseen by political appointees with an eye for the next job. It is entirely predictable that under these circumstances some children who could have been helped will instead be injured or killed. Mayors or state legislatures could choose to reallocate funds, effective managers, and equipment into social service agencies, but they seldom do.
One could adduce a further array of explanations for why no one imposes improvement on social service agencies, starting with the fact that poor black children and their parents do not vote. But my point here is that we need that further array to fully understand why powerful actors do or do not impose changes. The claim that policies are likely to succeed when some policy actors have power to make others change their behavior is, in short, only partly true and is only true for some policies.
A third form of “power over” focuses on characteristics of the policy’s recipient rather than promulgator. That is, policy mandates issued to organizations that are tightly hierarchical rather than loosely coupled are more likely to succeed. Some recipients such as school teachers, homeowners, and criminals operate in an environment where their actions cannot or may not easily be controlled by external authorities. They can thus ignore a directive with relative impunity. Other policy recipients such as enlisted soldiers, baseball players, or government contractors operate in an environment where their actions are readily known to, and easily and legitimately controlled by, external authorities. They ignore directives only at their peril.
This characteristic of recipients helps to explain the relative success, on the one hand, of desegregation of the armed forces and affirmative action hiring among government contractors during the 1970s, and the relative failure, on the other hand, of open housing policies, policies against second-generation discrimination in schools, efforts to control police brutality in violent encounters, and affirmative action among noncontractors. It does not explain the success of desegregating public accommodations (gas stations and restaurants are loosely coupled) or the failure to maintain the public infrastructure of cities (oversight agencies could have direct control over public hospitals, branch libraries, roads and bridges, and apartment buildings).
A final form of “power over...” addresses the incentives of the actors, both policy-makers and policy recipients. Policies are most likely to succeed if policy-makers have strong enough incentives for success that they figure out how to overcome the inevitable obstacles to it. In one sense, that observation is circular -- policies succeed when policy-makers work hard enough to make them succeed. Nevertheless, the observation offers an entry point into the vexed question of when policy- makers are prepared to exercise their “power over” and when they are able effectively to do so. The military has worked so hard to desegregate the activities of the soldiers and the composition of the officer corps because it had to. The Army needed black recruits in order to fill out the ranks, and once it had them it needed to ensure that they could fight and win battles. That was possible only if black soldiers were as well trained as white ones, and if members of both races were used to taking orders from members of both races. Hence recruitment, education, aggressive promotions, and training in race relations. Similarly, the 1965 Voting Rights Act was promulgated and aggressively enforced partly because of the Democratic party’s need for new voters to replace the whites defecting to the Republican party as a consequence of the 1964 Civil Rights Act. Once a few African Americans were elected to office, the electoral imperative was strong enough to impel them to devote considerable resources to ensuring that there would continue to be black voters and majority-black districts.
Conversely, school administrators seldom see school desegregation as essential to their core mission, and real estate agents seldom see housing desegregation as contributing to the attainment of what they most desire. Thus, even if administrators and agents desire racial equality, they lack a sufficiently strong incentive to pursue it through seemingly unending difficulties. That logic partly explains the failure of so many policies in inner cities; mayors are usually most responsive to constituencies for whom racial equality is not the first priority, whether they be downtown developers, taxpayers’ organizations, municipal unions, or bondholders. The countervailing forces of social activists and poor African American voters are simply not strong enough to keep even well-intentioned mayors from placing other priorities ahead of racial and class equalization.
Conclusion: The Role of Power in Explaining Racial Policy Success and Failure
We have at this point eight potential explanations, in two clusters, for why some race policies have mostly succeeded, some began to succeed but are now faltering, and some have failed or were never seriously attempted. Each explanation provides some leverage, but none is either necessary or sufficient to explain the pattern. That may be too stringent a criterion to apply to any social science explanation, but it seems equally true that no single combination of the potential explanations is either necessary or sufficient to explain the pattern of racial success and failure.
Instead I propose an alternative, looser, formulation. For a racial policy to succeed, it has required at least one “power to” characteristic to be combined with at least one “power over” characteristic across a period of time long enough to enable the policy to take deep root. This prix fixe menu formulation -- choose two from column A and one from column B -- lacks analytic elegance. Its only justification is that it works better than any alternative.
In short, among the many efforts over the past forty years to overcome the American racial hierarchy, policies thrived only if they were both amenable to implementation in one of the four ways specified and able to invoke strong and persistent backing from powerful policy actors. The first list, of successes, were all created by strong actors -- the Supreme Court, the military brass, the Justice Department, the owner of a baseball team – and they all possessed some feature that allowed them to beat out their competitors in the polity or economy. Policies in the second list, of partial or faltering successes, were promulgated by actors with just as much power over others as had the proponents of the first list -- courts, the Justice Department, the Federal Reserve Board. But they did not enjoy the crucial features that generate success in the polity or economy so they have not been able to sustain their initial gains.
The third list, of failures, lacks both “power to” and “power over” characteristics. Most of these policies are the responsibility of local or state actors, who are too politically dependent on their predominantly wealthy or white constituents to have the freedom of action sometimes enjoyed by the President, administrative agencies, or the Supreme Court. Most of these policies are also messy. They are complicated, they often seek to offset rather than to reinforce market forces, and -- given the nature of city-suburban boundaries -- they appear to benefit the few (poor urban blacks) at the expense of the many (the rest of the population). No wonder they fail.
If this set of explanations for the pattern of success and failure in American racial policy is persuasive, three sets of questions remain. The historical questions are: what institutional and political configuration made it possible during a given period for powerful actors to promulgate policies which proved to be implementable? Conversely, why were other choices not made, other actors not so committed, and other institutional conjunctions and politics less amenable to reform? The predictive questions are: will the supposedly firm successes remain so? Can the faltering successes be bolstered? Can the dismal failures be alleviated? That is, what should we make of the implication contained within the phrase “the second Reconstruction” -- that most if not all of the changes in the American racial structure achieved since the 1950s will be reversed by policies designed to reestablish white domination? The prescriptive questions are: what must we do to create institutions and political contexts that will allow new policies to come to the fore? What new policies are available that combine enough features of “power over” and “power to” to give them a good chance of succeeding in reducing white racial domination? What can be done to create the essential mix of public opinion, social pressure, and political dynamic that will encourage potentially powerful actors to exercise their power in this arena?
Hugh Davis Graham’s chapter in this volume contributes to answering the historical question, and I will not consider it further here. I will instead conclude more like a social scientist than like a historian -- with two observations on whether the United States is facing the end of the second Reconstruction or whether, in contrast, we can design newly successful racial policies.
If my analysis is correct, there is little reason to expect new racial successes similar to those on my first list, or movement out of the second and third lists into the first. The relatively easy policies -- policies that contain within themselves the power to accomplish their goals and that enjoyed powerful and committed sponsors -- have mostly been done. (Let me clarify a final time: they were not at all easy to accomplish – their ease is only relative to that of other racial policies.) The powerful and committed sponsors have, so far as I can predict the foreseeable future, mostly lost their will or are embedded in a political context that is too complicated to allow them to exert control. The few remaining policies that seem relatively easy by my substantive criteria – reparations, changing municipal boundaries, proportional representation, stringent forms of affirmative action – are political nonstarters. Thus the contours of public opinion interact with the nature of the remaining policy problems and the complexities of the American governance structure to set severe limits on future racial transformation. If that argument is correct, the deepest problems of black poverty, alienation, and danger in the inner cities, and the continuing problems of discrimination in jobs, schools, and housing, are likely to persist.
Nevertheless, the racial configuration of the United States in 1997 is not the same as that of 1897, and in that sense the second Reconstruction has been a success. Our nation will not, I predict, return to Jim Crow apartheid. More whites now believe in, practice, and are prepared to work for racial equality than was the case a century ago. Immigrants who are neither white nor black have enough numbers and political and organizational resources to permit significant alliances among people of color. Most importantly, there now exists a substantial black middle class with enough wealth, education, professional standing, political clout, and determination to prevent Jim Crow from returning.
Where does this combination leave us? I see a growing class division within the African American community, not because middle-class blacks will choose in any straightforward way to distance themselves from poor blacks, but because they will have little choice in the matter. The racial successes on my first list will continue to propel some African Americans into the successful mainstream; the racially-disproportionate failures on my third list will continue to prevent other African Americans from having any chance to move in the same direction. The direction in which the partial and faltering policies of the second list will move will be decided by political activity over the next few decades. That direction will largely determine the relative proportions of well-off and poor African Americans. I predict therefore that the second Reconstruction will end in a way that makes the black community resemble the white one at least in the ironic sense that it will add class divisions to already extant religious, regional, ethnic, and gender divisions.
Abernathy, Scott (1996) “Preliminary Capacity Study: Intra-district Vouchers and Sheff v. O’Neill” (Princeton University: Woodrow Wilson School of Public and International Affairs, unpublished paper).
Adarand Constructors Inc. V. Pena, Secretary of Transportation, et al. 115 S. Ct. 2097 (1995)
Ards, Sheila (1989) "Estimating Local Child Abuse," Evaluation Review 13, no. 5: 484-515.
Armor, David (1995) Forced Justice: School Desegregation and the Law (New York: Oxford University Press).
Arnold, Douglas (1990) The Logic of Congressional Action (New Haven: Yale University Press)
Bardach, Eugene (1977) The Implementation Game: What Happens After a Bill Becomes a Law (Cambridge: MIT Press).
Becker, Gary (1957) The Economics of Discrimination (Chicago: University of Chicago Press).
Bell, Derrick (1992) Race, Racism, and American Law, 3rd edition (Boston: Little, Brown and Company)
Besharov, Douglas (1996) “Child Abuse Reporting,” in Irwin Garfinkel, Jennifer Hochschild, and Sara McLanahan, eds., Social Policies for Children (Washington D.C.: Brookings Institution): 257-273.
Binkin, Martin and Mark Eitelberg (1982) Blacks and the Military (Washington D.C.: Brookings Institution).
Boozer, Michael, Alan Krueger, and Shari Wolkon (1992) “Race and School Quality Since Brown v. Board of Education,” Brookings Papers: Microeconomics (Washington D.C.: Brookings Institution): 269-338.
Bradbury, Katharine, Karl Case, and Constance Dunham (1989) "Geographic Patterns of Mortgage Lending in Boston, 1982-1987," New England Economic Review, Sept./Oct.: 3-30 .
Burbridge, Lynn (1991) "The Interaction of Race, Gender, and Socioeconomic Status in Education Outcomes" (Wellesley College, Center for Research on Women).
Bush v. Vera 116 S.Ct. 1941 (1996)
Canner, Glenn (1991) "Home Mortgage Disclosure Act," Federal Reserve Bulletin 77, no. 11: 859-881.
Card, David and Alan Krueger (1992) “School Quality and Black-White Relative Earnings: A Direct Assessment,” Quarterly Journal of Economics, Feb.: 151-200.
Case, Anne and Lawrence Katz (1991) "The Company You Keep: The Effects of Family and Neighborhood on Disadvantaged Youths" (Boston: NBER).
Child Welfare League of America (1992) The Youngest of the Homeless II (Washington D.C.: Child Welfare League of America).
Clinton, William J. (1995) Memorandum on Affirmative Action to Heads of Executive Departments and Agencies, July 19.
Crane, Jonathan (1991) “Effects of Neighborhoods on Dropping Out of School and Teenage Childbearing,” in Christopher Jencks and Paul Peterson, eds., The Urban Underclass (Washington D.C.: Brookings Institution): 299-320.
Dahl, Robert (1963) Modern Political Analysis (Englewood Cliffs NJ: Prentice-Hall).
Daula, Thomas, D. Alton Smith, and Roy Nord (1990) “Inequality in the Military: Fact or Fiction?” American Sociological Review 55, no. 5: 714-718.
Dembo, Richard (1988) “Delinquency Among Black Male Youth,” in Jewell Taylor Gibbs, ed., Young, Black, and Male in America: An Endangered Species (Dover MA: Auburn House): 129-165.
DiIulio, John Jr. (1989) "The Impact of Inner-City Crime," Public Interest 96: 28-46.
Doig, Jameson and Erwin Hargrove, eds. (1990) Leadership and Innovation (Baltimore: Johns Hopkins University Press).
Duncan, Greg and Willard Rodgers (1991) "Has Children's Poverty Become More Persistent?" American Sociological Review 56, no. 4: 538-550.
Eggebeen, David and Daniel Lichter (1991) "Race, Family Structure, and Changing Poverty among American Children," American Sociological Review 56, no. 6: 801-817.
Farley, Reynolds (1992) The Changing Status of Blacks and Whites, Men and Women (University of Michigan, Population Studies Center, manuscript).
Farley, Reynolds and William Frey (1994) "Changes in the Segregation of Whites From Blacks During the 1980s," American Sociological Review 59, no. 1: 23-45.
Freeman v. Pitts 112 S. Ct. 1430 (1992)
Frey, William and Reynolds Farley (1993) "Latino, Asian and Black Segregation in Multiethnic Metropolitan Areas," Paper at the annual meeting of the Population Association of America, Cincinnati.
Frommer, Harvey (1982) Rickey and Robinson: The Men Who Broke Baseball’s Color Barrier (New York: Macmillan).
Giles, Micheal, Douglas Gatlin, and Everett Cataldo (1974) “The Impact of Busing on White Flight,” Social Science Quarterly 55, No. XX: 493-501.
Goering, John and Ron Wienk, eds. (1996) Mortgage Lending, Racial Discrimination, and Federal Policy (Washington D.C.: Urban Institute Press).
Gramlich, Edward, Richard Kasten, and Frank Sammartino (1993) “Growing Inequality in the 1980s: The Role of Federal Taxes and Cash Transfers,” in Sheldon Danziger and Peter Gottschalk, eds., Uneven Tides: Rising Inequality in America (New York: Russell Sage Foundation): 225-249.
Greenwood, Peter (1992) “Reforming California’s Approach to Delinquent and High-Risk Youth,” in James Steinberg, David Lyon, and Mary Vaiana, eds., Urban America: Policy Choices for Los Angeles and the Nation (Santa Monica CA: Rand Corporation): 207-225.
Hamilton, Dona and Charles Hamilton (1997) The Dual Agenda: Race and Social Welfare Policies of Civil Rights Organizations (New York: Columbia University Press).
Hartsock, Nancy (1983) Money, Sex, and Power: Toward a Feminist Historical Materialism (New York: Longman).
Heckman, James and Kenneth Wolpin (1989) “Determining the Impact of Federal Antidiscrimination Policy on the Economic Status of Blacks: A Study of South Carolina,” American Economic Review 79, no. 1: 138-177
Hellerstein, Judith, David Neumark, and Kenneth Troske (1997) “Market Forces and Sex Discrimination” (Cambridge MA: National Bureau of Economic Research, Working Paper # 6321).
Hendrie, Caroline (1997) “Politics of Jobs in City Schools Hinder Reform,” Education Week, March 26: 1, 28.
Hirschman, Albert (1991) The Rhetoric of Reaction: Perversity, Futility, Jeopardy (Cambridge: Harvard University Press).
Hochschild, Jennifer (1995) Facing Up to the American Dream: Race, Class, and the Soul of the Nation (Princeton: Princeton University Press).
Hochschild, Jennifer and Michael Danielson (1998) “Can We Desegregate Public Schools and Subsidized Housing? Lessons from the Sorry History of Yonkers, New York,” in Clarence Stone, ed., Changing Urban Education (Lawrence: University of Kansas Press).
Hogan, Dennis and David Featherman (1978) “Racial Stratification and Socioeconomic Change in the American North and South,” American Journal of Sociology 83, no. 1: 100-26.
Hopwood v. Texas 84 F. 3rd 72 (1996)
Jaynes, Gerald and Robin Williams, eds. (1989) A Common Destiny: Blacks and American Society (Washington D.C.: National Academy Press).
Joint Center for Political and Economic Studies (1994) Black Elected Officials (Washington D.C.: Joint Center for Political and Economic Studies).
Kasarda, John (1993) "Inner-City Concentrated Poverty and Neighborhood Distress: 1970 to 1990," Housing Policy Debate 4, no. 3: 253-302.
Kozol, Jonathan (1991) Savage Inequalities: Children in America’s Schools (New York: Crown).
Ledebur, Larry and William Barnes (1992) City Distress, Metropolitan Disparities, and Economic Growth (Washington D.C.: National League of Cities).
Leonard, Jonathan (1990) “The Impact of Affirmative Action Regulation and Equal Opportunity Law on Black Employment,” Journal of Economic Perspectives 4, no. 4: 47-63.
Massey, Douglas and Nancy Denton (1987) "Trends in the Residential Segregation of Blacks, Hispanics, and Asians: 1970-1980," American Sociological Review 52, no. 6: 802-825.
Massey, Douglas and Nancy Denton (1993) American Apartheid: Segregation and the Making of the Underclass (Cambridge MA: Harvard University Press).
Mazmanian, Daniel and Paul Sabatier (199X) Implementation and Public Policy XXXX
McConahay, John ( 1982) “Self-Interest versus Racial Attitudes as Correlates of Anti-Busing Attitudes in Louisville: Is It the Buses or the Blacks?” Journal of Politics 44, no. XX: 693-717.
McConahay, John (n.d., c. 1978) “It Is Still the Blacks and Not the Buses: Self-Interest vs. Racial Attitudes as Correlates of Opposition to Busing in Louisville, a Replication” (Duke University, Institute of Policy Sciences).
Mead, Lawrence (1992) The New Politics of Poverty: The Nonworking Poor in America (New York: Basic Books).
Meier, Kenneth J., Joseph Stewart Jr., and Robert England (1989) Race, Class, and Education: The Politics of Second-Generation Discrimination (Madison: University of Wisconsin Press).
Merton, Robert (1996 [orig. 1936]) “The Unanticipated Consequences of Social Action,” in Robert K. Merton, On Social Structure and Science (Chicago: University of Chicago Press): 173-182.
Miller v. Johnson 115 S.Ct. 2475 (1995)
Milloy, Courtland (1998) “Low Test Scores Demand a Better Answer,” Washington Post, Jan. 11: B1, B10.
Missouri, et al. v. Kalima Jenkins, et al. 115 S. Ct. 2038 (1995)
Moore, Brenda (1996) “From Underrepresentation to Overrepresentation: African American Women,” in Judith Stiehm, ed., It’s Our Military Too!: Women and the U. S. Military (Philadelphia: Temple University Press): 115-135.
Moskos, Charles and John Butler (1996) All that We Can Be: Black Leadership and Racial Integration the Army Way (New York: Basic Books)
Nathan, Richard and Charles Adams, Jr. (1989) "Four Perspectives on Urban Hardship," Political Science Quarterly 104, no. 3: 483-508.
National Black Child Development Institute (1989) Who Will Care When Parents Can't? (Washington D.C.: National Black Child Development Institute).
National Center for Children in Poverty (1990) Five Million Children: A Statistical Profile of Our Poorest Young Citizens (New York: Columbia U., School of Public Health).
National Center on Child Abuse Prevention Research (1994) Current Trends in Child Abuse Reporting and Facilities (Chicago: National Committee to Prevent Child Abuse).
New York Daily News (1996) “Shocking Report on District School Boards,” May 30: XXX
Nordlinger, Eric (1981) On the Autonomy of the Democratic State (Cambridge: Harvard University Press).
Oakes, Jeannie (1985) Keeping Track: How Schools Structure Inequality (New Haven: Yale University Press).
Orfield, Gary, Mark Bachmeier, David James, and Tamela Eitle (1997) "Deepening Segregation in American Public Schools” (Harvard University, Graduate School of Education).
Petersilia, Joan (1992) “Crime and Punishment in California: Full Cells, Empty Pockets, and Questionable Benefits,” in James Steinberg, David Lyon, and Mary Vaiana, eds., Urban America: Policy Choices for Los Angeles and the Nation (Santa Monica CA: Rand Corporation): 175-205.
Pressman, Jeffrey and Aaron Wildavsky (1973) Implementation (Berkeley: University of California Press).
Purdy, Matthew (1996) “Web of Patronage in Schools Grips Those Who Can Undo It,” New York Times, May 14: A1, B4.
Purdy, Matthew and Maria Newman (1996) “Students Lag in Districts Where Patronage Thrives,” New York Times, May 13: A1, B4.
Rampersad, Arnold (1997) Jackie Robinson: A Biography (New York: Knopf).
Rich, Wilbur (1996) Black Mayors and School Politics: The Failure of Reform in Detroit, Gary, and Newark (New York: Garland).
Richmond VA v. J. A. Croson Co. 488 U.S. 469 (1989).
Rosenberg, Gerald (1991) The Hollow Hope: Can Courts Bring About Social Change? (Chicago: University of Chicago Press).
Rosenstone, Steven and John Hansen (1993) Mobilization, Participation, and Democracy in America (NY: Macmillan).
Shaw v. Reno 113 S. Ct. 2816 (1993).
Skocpol, Theda (1992) Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (Cambridge: Harvard University Press).
Smith, Earl (1996) “Serving Our Country: African American Women and Men in the U.S. Military Service,” in Joyce Tang and Earl Smith eds., Women and Minorities in American Professions (Albany: SUNY Press): 135-156.
Stone, Clarence (1989) Regime Politics: Governing Atlanta (Lawrence: University of Kansas Press).
Stone, Clarence, ed. (1998) Changing Urban Education (Lawrence: University of Kansas Press).
Tenner, Edward (1996) Why Things Bite Back: Technology and the Revenge of Unintended Consequences (New York: Knopf).
Thernstrom, Stephan and Abigail Thernstrom (1997) America in Black and White: One Nation, Indivisible (New York: Simon & Schuster).
Tomasky, Michael (1996), “All Fall Down,” New York Magazine, Feb. 12: 44-49.
Tygiel, Jules (1983) Baseball’s Great Experiment: Jackie Robinson and His Legacy (New York: Oxford University Press).
U.S. Congress (1989) House of Representatives, Select Committee on Children, Youth, and Families, Born Hooked: Confronting the Impact of Perinatal Substance Abuse, Hearing, April 27 (Washington D.C.: U.S. Government Printing Office).
U.S. Department of Health and Human Services (1990) "OIG Management Advisory Report: 'Boarder Babies'," by Richard Kusserow (Washington D.C.: U.S. Department of Health and Human Services)
U.S. Department of Justice (1994) Sourcebook of Criminal Justice Statistics (Washington D.C.: USGPO).
Washington Post, Kaiser Family Foundation, Harvard University (1995) The Four Americas: Government and Social Policy Through the Eyes of America’s Multi-racial and Multi-ethnic Society (Washington D.C.: Washington Post).
Weick, Karl (1970) “The Twigging of Overload,” in H.B. Pepinsky, ed., People and Information (New York: Pergamon): 67-129.
Wilson, James Q. (1973) Political Organizations (New York: Basic Books)
Wulczyn, Fred and Robert Goerge (1992) "Foster Care in New York and Illinois," Social Service Review 66, no. 2: 276-294.
 Hochschild 1995: chap. 3.
 Hochschild 1995: chap. 3.
 Hochschild 1995: chap. 3.
A complete analysis would also consider the content and intensity of public opinion, the amount and kind of resources available to policy implementers, and the level and efficacy of social activism in shaping the array of policy possibilities and the effects of policy implementation. Given the theme of this volume, however, I focus in this chapter on institutional structures and policy processes as explanations for racial success and failure.
 By “recent” I mean since roughly 1950. “Policies” range from overt decisions about race (such as civil rights laws, executive orders on affirmative action, and court decisions about school desegregation) to covert decisions such as choices about the tax structure, staffing and budgetary levels of regulatory agencies, or location of municipal boundaries that are not explicitly about race but have clearly predictable and important racial consequences. Arguably all policies, overt and covert, have racial implications, but I will focus here on those that a reasonable person would agree substantially affect the American racial structure (always recognizing the indeterminacy of those phrases).
By “the American racial structure” I mean all of the ways in which whites, institutionally as well as individually, dominate blacks. I will assume for the sake of simplicity that there is a meaningful social distinction between the white and black races, regardless of whether that distinction has any biological basis. I will also, for the sake of essential simplifying, ignore issues of ethnicity and races other than African Americans and white Americans.
“Success” is the hardest term in this sentence to define. In this chapter, a successful policy must have at least two characteristics: it has been enforced for several decades and shows no sign of being abolished or significantly watered down, and it has had a substantial positive impact on the quality of life, range of opportunities, or resources of many African Americans. A complete definition would have to give a precise meaning to “strongly,” “substantial,” and “many,” but the main point here is to suggest a threshold that is high but not insuperable.
 Binkin and Eitelberg 1982; Moskos and Butler 1996; Daula, Smith, and Nord 1990; Moore 1996. But see Smith (1996) for continued skepticism on this point.
 There are surprisingly few studies that directly link racial policy changes to outcomes for individuals. Among the best are studies by Card and Krueger (1992), Massey and Denton (1993), Heckman and Wolpin (1989) and Leonard (1990); they all show that appropriate policies are effective in narrowing racial gaps.
 The phrase, and evidence, come from Hogan and Featherman 1978.
 Hochschild 1995: chap. 2
Orfield et al.1997; Boozer, Krueger and Wolkon 1992
 Meier, Stewart, and England 1989; Oakes 1985
Freeman v. Pitts 1992; Missouri v. Jenkins 1995
Massey and Denton 1993; Farley and Frey 1994; Bradbury, Case, and Dunham 1989; Canner 1991; Goering and Wienk 1996
Massey and Denton 1987; Frey and Farley 1993
Farley 1992: chap. 7; Rosenstone and Hanson 1993: 219-24; Joint Center for Political and Economic Studies 1994; Thernstrom and Thernstrom 1997: 288-294
 Shaw v. Reno 1993; Miller v. Johnson 1995; Bush v. Vera 1996
Richmond v.Croson 1989; Adarand v. Pena 1995; Clinton 1995; Hopwood v Texas 1996
 Jaynes and Williams eds., 1989: chap. 9; Petersilia 1992; Greenwood 1992; Dembo 1988; Bell 1992: chap. 5
Kasarda 1993: 172-73; Burbridge 1991: 7, 9.
 Abernathy 1996; quotation from Milloy 1998
 In Boston, the School Committee was for decades a standard jumping off point for a mayoral campaign. In New York City, a recent survey showed that over half of the district school board members did not know how many students were in their district, and over three-fourths could not tell the amount of their district budget, even though they had voted on it a few weeks before the survey (New York Daily News 1996; see also Purdy and Newman 1996; Purdy 1996; Hendrie 1997).
 Kozol 1991; Tomasky 1996
 On boarder babies, see U.S. Dept. of Health and Human Services 1990: 2; U.S. Congress 1989: 120-131; Child Welfare League of America 1992. On child abuse, see Besharov 1996; National Center on Child Abuse Prevention Research 1994: 5; Ards 1989. Besharov, like virtually all analysts of child abuse and neglect, cautions about the unreliability of these figures either across race and class or over time.
 Wulczyn and Goerge 1992: 279-280; Besharov 1996; National Center for Children in Poverty 1990: 60; U.S. Congress 1989: 46-88; National Black Child Development Institute 1989: 23-51
Eggebeen and Lichter 1991: 809; Duncan and Rodgers 1991: 543; Kasarda 1993: 266-68.
 During this period, the wealthiest fifth of both whites and blacks gained 15% over their previous income (Hochschild 1995: 49). Considerable evidence suggests that at least during the 1980s, governmental policies deliberately redistributed income upward, so it is perhaps an ironic misnomer to call persistent poverty a failure of governmental policy (see Gramlich, Kasten, and Sammartino 1993: 234-239).
 Mead 1992
Nathan and Adams 1989; Ledebur and Barnes 1992.
 U.S. Department of Justice 1994: 247, 248, 285, 288, 353, 363-64.
 Hochschild 1995: 50.
 Hochschild 1995: 47, 202.
 Crane 1991: 319; Case and Katz 1991; DiIulio 1989.
 Just over 54% of African Americans and 22% of whites live in central cities; 23% of residents of central cities are African American and 54% are white. Roughly speaking, the larger the city, the higher the proportion of people of color among its residents.
 John Goering argues, in response to an earlier draft of this chapter, that “by linking traditional civil rights program outcomes with ‘inner city’ social and policy dynamics you have set up a straw-person.... The broader American policy failure in addressing systematic, place based inequalities gets unfairly placed at the door of the civil rights movement.” I do not wish to blame the civil rights movement for failing to solve a problem that it did not set out to solve. But as I read the history of the movement, social and economic inequality, especially when disproportionately endured by black Americans, was one of the problems that many civil rights activists sought to solve -- unsuccessfully.
 Hamilton and Hamilton 1997
 Except possibly for the inclusion of “inner cities” in the list of racial failures.
 I say “helps us to understand” in response to an argument by John Goering (in a private communication, October 23, 1996) that “many of the deeply held, patterned inconsistencies between black and white views of race relations precede the civil rights era and... [are more] determined by... a broader set of pressures” stemming from centuries of control and subordination. I agree, with two caveats. First, these views are not fixed on either side; in the 1960s, blacks were more optimistic than whites about the current state and future trajectory of American race relations. Second, members of both races point to particular policy successes and failures when expounding and explaining their racial views.
For a more formal and extensive analysis of why white and black Americans hold such disparate views of black racial progress, see Hochschild 1995: chaps. 3, 7
 Stone (1989, chap. 11) provides the best analysis of the distinction between “power over” and “power to.” See also Hartsock 1983, esp. chapter 9
What my mother-in-law, an interior designer, calls the “couldn’t you just...?” syndrome.
Pressman and Wildavsky 1973; Mazmanian and Sabatier 19XX; Bardach 1977
 Merton 1996 (1936); Hirschman 1991; Tenner 1996
 Becker 1957
 Thus railroad owners were not enthusiastic about implementing the principle of “separate but equal” at the turn of the century -- it was expensive, complicated, and reduced ridership and profits -- and officials of large corporations are not enthusiastic about ending federal affirmative action policies. Among firms “with high levels of product market power, those that employ relatively more women are more profitable” (Hellerstein et al. 1997).
 My logic, of considering whether each policy’s or policy maker’s characteristic is either necessary or sufficient for policy success, resembles, though it was not derived from, that of Rosenberg (1991).
 The failure of supposedly free markets to increase the pace of housing desegregation may be explained by the fact that when one buys a home one is buying so much else -- a level and kind of status, self-image, schooling, social life, security, and convenience -- that housing markets are exceptionally “sticky.”
 The quotation and my explanation of the concept come from Arnold (1990: 19-20); the colloquial phrase was the language of President Kennedy’s promise to end discrimination in public housing. For much the same idea, see Mazmanian and Sabatier 19XX: YYY
 The classic demonstration of the problems of multiple stage policies is Pressman and Wildavsky 1973.
 Frommer 1982; Tygiel 1983; Rampersad 1997
Wilson 1973: 330-337; Arnold 1990: 25-28
 See Armor (1995) on “the harm and benefit thesis” in the context of school desegregation.
 Although fewer than 10 percent of whites claim that they or a close connection have lost a job due to affirmative action, up to two-thirds believe it likely that a white will suffer such a loss (Hochschild 1995: 144; see, more generally, chap. 7 for a discussion of blacks’ and whites’ perceptions of the costs of racial change).
 The claim that positive-sum games are more easily sustained than negative-sum ones does not even explain white hostility to school desegregation very well, since opposition to “forced busing” was not closely related to whether one’s own child was involved in the desegregation program (McConahay 1982, n.d.; Giles, Gatlin, and Cataldo 1974). On current school practices that inhibit the success of inner city children, see Stone 1998; Rich 1996.
 As I have noted several times so far, Mazmanian and Sabatier (19XX) structure their argument along somewhat similar lines; their category that comes closest to my “power to” formulation focuses on the “tractability of the problem.” Their lists do not, however, fully address my puzzle in this chapter. For example, one of their four items in the “tractability” category, “target group as a percentage of the population” has no variance in the arena of racial policy, so it does not help to explain different rates of success and failure. In addition, Mazmanian and Sabatier do not take advantage of the analytic power of the power to/power over distinction, and they do not address the issue of necessity and sufficiency, which is central to the logic of my analysis. Thus our analyses overlap, but do not duplicate each other.
 Nordlinger 1981; Skocpol 1992; Mazmanian and Sabatier 19XX
 This point is stated historically, since the clearest case is the increase in black mayors and council members of cities since the 1960s. It also, however, fits the case of the five black members of Congress who gained office in newly-drawn majority black districts in 1994, and were able to retain their seats in even-more-newly-drawn majority white districts in 1996, largely due to the powerful effect of well-managed incumbency.
 Hochschild and Danielson (1998) analyze why an apparently very powerful agency -- New York’s Urban Development Corporation -- could not attain its mission of desegregating public housing.
 Robert Dahl’s (1963: 40) classic formulation of influence
 Weick 1970
 Doig and Hargrove 1990
 Stone (1989) provides one of the best demonstrations of this dynamic.
 Within this framework, the precise causes of public opinion about racial policy – whether racism, ideological individualism, fiscal conservatism, mistrust of government, or whatever – do not much matter. What matters is whether the intersection between achievable policies, public willingness to pursue new policies, and leaders’ will and capacity is or is not a null set.