Losing Ground (10th Anniversary Edition)

A NOTE ON PRESENTATION

THIS BOOK tries to do two things that are not quite mutually exclusive, but almost. It draws upon a varied and sometimes technical body of social science data; and it addresses a general audience. How does one be fair to the data (which often entail complications and ambiguities) and still present the material in a way that an intelligent but not obsessive reader can be asked to follow? This tension has shaped the presentation in important ways.

Part of the solution has been to write what amounts to a subtext in the notes for the chapters in parts II and III. The reader who so wishes may read the book from beginning to end without once referring to the notes and (I trust) come away with an accurate understanding of the argument and the evidence. But every so often the reader is also likely to stop short and want to take a closer look. The notes, some of which amount to small essays, have been written for such occasions.

Another part of the solution has been to rely heavily on basic trendline data—graphs of what happened, using widely understood and accepted measures, from 1950 to 1980. Often such data need to be supplemented with more specialized analyses; occasionally, the trendline by itself will be misleading. But the most straightforward data are generally also the best ones for beginning to understand what happened.

Finally, I have added an appendix that includes much of the raw data. It is intended for readers who want to examine the trends from other perspectives or who want to apply these fundamental indicators to other questions.

PREFACE

LOSING GROUND grew out of sixteen years of watching people who run social programs, and my first debt is to them. Whether they have been counseling inner-city students in Atlanta, trying to keep Chicago delinquents out of jail, or teaching prenatal care to Thai villagers, they have shared an uncommon energy and dedication. Over the years, however, I was struck by two things. First, the people who were doing the helping did not succeed nearly as often as they deserved to. Why, when their help was so obviously needed and competently provided, was it so often futile? In the instances when the help succeeded, what were the conditions that permitted success? Second, the relationship between the ways people were to be helped and the quality of their lives became increasingly confused. Clearly, certain minimums of physical well-being were critical. But once those had been met, it was just as clear that, among the many things that produce satisfaction, dignity, and happiness, few were purely economic. How did the goods that social programs dispense fit in with the noneconomic assets? Two and a half years ago, I set out to pursue these lines of inquiry more systematically.

Some months later, Joan Kennedy Taylor of the Manhattan Institute saw the potential for a book in a monograph I had written. She shepherded the work (and me) through to the end. William Hammett, president of the Manhattan Institute, took a chance and decided to use the foundation’s resources to underwrite the effort. Without them, the book would not have been written.

I relied principally on the collections at the Library of Congress and the Bureau of the Census. The people who helped at both places are too numerous to list individually. At the Bureau of the Census, special thanks go to Bruce Chapman, then director, who took a heartening interest throughout the work. My thanks also go to Gordon Green of the Population Division and Carol Fendler of the Poverty Statistics Section, whose expertise made my job much easier. At the Federal Bureau of Investigation, Ken Candell oversaw the special computer runs I needed. At the Office of Family Assistance, Ken Lee, Laurence Love, Michael deMaar, Howard Rolston, and Jo Anne Ross patiently led me through the complicated history of the Aid to Families with Dependent Children (AFDC) program. Many others whose names I do not know in the libraries of the Bureau of Labor Statistics, the National Institute of Education, the National Institute of Justice, and the National Center for Health Statistics took time to answer my questions and look up elusive documents. The custodians of the federal data bases are extraordinarily ready to go out of their way to help the anonymous researcher, and I am grateful to them all.

Donald A. Cook, who was involved in the earliest planning for the War on Poverty, provided an extremely helpful commentary on my account of that period. Robert Krug and Norman Gold, longtime colleagues and veterans of the Office of Economic Opportunity, brought their experience to bear on the discussions of the job training and educational innovations of the reform period and thereafter. Others who read parts or all of the manuscript and provided invaluable criticisms were Paul Schwarz, mentor and friend for many years at the American Institutes for Research; Irving Kristol, whose encouragement came at just the right moment; and Michael Horowitz, of the Office of Management and Budget, who sees a common purpose in his civil rights work in Mississippi in the 1960s and in his efforts to cut social programs in the 1980s.

In an earlier phase of the work that led to the book, Burton Pines of the Heritage Foundation first gave me the chance to concentrate my thinking on why it is that we could have spent so much money and have bought so little. Sheldon Danziger at the Institute for Research on Poverty was always ready to talk over my questions and provide me with materials from the Institute’s ongoing work. At Basic Books, Martin Kessler understood exactly what I was trying to accomplish and reminded me of it when necessary, and Nina Gunzenhauser was a meticulous, occasionally inspiring, copy editor. Together, they have improved the book immeasurably.

Losing Ground contains some contentious interpretations of recent history. Few of the people I have named share all of them; some of them share virtually none. It should be clear that neither the interpretations nor any factual errors that may remain are their fault. This is especially true of my most thoughtful critic, and dearest one, Catherine Cox. We knew from the beginning that she was unlikely to agree with every conclusion. But it has been essential that she approve of their spirit.

CHARLES MURRAY
Washington, D.C.
15 February 1984

Introduction to the 2015 Edition

YOU HOLD IN YOUR HANDS a book that was written during Ronald Reagan's first term in office. The question naturally arises: Why? Thirty years have elapsed since it was published—thirty years of additional technical analyses of all the issues that Losing Ground discusses and thirty years that have made the United States a radically different place than it was in 1984. What could Losing Ground still have to offer?

In my own view, undoubtedly biased, the best reason for reading this book is its exploration of a problem that surfaced with the earliest attempts to use government to help people in need: How does government help without also doing harm? English thinkers worried about it from the time the Poor Laws were first put in place in the sixteenth century: How can the penalty for not working be lightened for those who are unable to work, without producing more people who decide not to work?

Perverse incentives attend almost any effort to help. Most people recoil from policies that would incarcerate a fourteen-year-old boy who has been caught burglarizing a house, fearing that incarceration will do more harm than good. But not punishing the burglar is problematic too. It is in the nature of adolescent males to be less than fully mature in their appraisal of long-term consequences. In the short term, burgling seems fun and profitable. How are we to be more understanding toward the young burglar without creating more young burglars? Similar problems bedevil changes in school discipline, the enforcement of drug laws, and assistance for unmarried mothers. For that matter, we need not limit the problem to government programs. The process of raising children forces every parent to assess the same difficult tradeoffs.

For me, the most difficult moral issue is how to deal with culpability. In the 1960s and 1970s, "blaming the victim" became a catchall indictment of any policy analysis daring to say that members of a disadvantaged group were behaving in self-destructive ways. You still encounter the phrase today. It has always been an intellectually lazy way to avoid tough issues, but it does have truth to it. Go back to my young burglar. Suppose he has grown up with a single mother in an impoverished inner-city neighborhood. His role model as he grew up could not be his father, so it was the most charismatic adolescent male in his neighborhood—the worst of all possible role models. He is failing in school and has no prospects for getting a good job. According to our usual understanding of moral agency, he is not fully to blame for burglarizing the home. And yet one of the worst possible things we can do is to tell him that. Chapter 14, "The Destruction of Status Rewards," is a long meditation on this issue, and one that I think is as relevant now as it was in 1984. More broadly, parts III and IV of Losing Ground are not time sensitive. They raise fundamental questions that need to be asked of social policy today and too often are not.

Reading Losing Ground in the second decade of the twenty-first century has another value: it gives you a window onto a pivotal era in American life written by someone who came of age during it. I was twenty years old when John F. Kennedy was shot in Dallas (I still have a recording of my voice announcing his assassination on Harvard's student radio station). I wasn't around for the student movements of the 1960s—I was working in villages in rural Thailand—but I returned to the United States as the social policies introduced during the Johnson years were being implemented on a large scale. I spent most of the 1970s as an evaluator of social programs, seeing at firsthand how social policies to help the poor worked in practice.

Being a participant in events doesn't necessarily make you a better judge of them, but it does let you know what things looked and felt like on the ground. That fourteen-year-old incarcerated for burglary? I wrote those lines recalling a boy about that age whom I met while evaluating a program to deinstitutionalize chronic juvenile offenders in Chicago. He had finally been sent to a custodial facility after twenty-six arrests. He was indignant. The actual offense for which he was sentenced was, as I recall, a fairly minor theft. "Why did they send me here for that?" he complained. "They'd picked me up for lots worse than that." To him, the system looked irrational and unjust. I could base the discussions of incentives in part III on real people who were experiencing those incentives at that moment in American history, just as I knew at firsthand what a chaotic inner-city classroom sounded like and looked like when I wrote about changes in the way schools were run.

The value of having been around at the time also shows up in part I, when I describe the logic and ethos of the people who were involved in shaping the Great Society agenda. I returned from Thailand to attend graduate school at M.I.T. in 1970, when the optimism of the early years had faded. Several of the faculty at M.I.T. and nearby Harvard had been key players in social reforms, and many of my texts and classroom discussions resembled after-action debriefings on what had gone wrong.

For someone born in the decades since then, it is hard to realize the degree to which planners of the Great Society were starting from scratch. Americans had been trying to administer to the human needs of their fellow citizens from the nation's founding, but it had been done by relatives helping relatives, neighbors helping neighbors, and voluntary organization (mostly religious) working in slums, workhouses, or prisons. Using the government to do similar kinds of good on a wholesale basis seemed so obviously simple in 1964. The concerns of earlier generations about perverse incentives were brushed aside. The Civilian Conservation Corps had worked during the New Deal, right? The programs of the Great Society would be the same kinds of things, only bigger and better.

The reaction when the programs didn't work was bewilderment and confusion: Why weren't people responding to these wonderful opportunities the way they were supposed to? I hope my narrative conveys how little social policy planners understood what they were up against. They had a limited body of historical examples of government social programs to draw upon, assumptions about the nature of poverty and disadvantage that were compassionate but empirically dubious, and unbounded confidence about their intellectual and managerial ability to fashion good policy. Here is just one indicator of how naive everyone was: Today, it has long since been conventional wisdom that government social programs have unintended outcomes. When Lyndon Johnson took office, nobody appears to have given that possibility a thought. A search for "unintended outcomes" on Google's Ngram shows the phrase first appearing in the last half of the 1960s—and its use soaring thereafter.

I have given you some reasons for reading Losing Ground thirty years after it appeared. But the reason you even know about the book is because of the controversy it triggered when it was published and the role it has played in the social policy debate since then.

Losing Ground was published in September 1984. By October, the book was beginning to be talked about in print. By December, it was being treated as a phenomenon. In February of the new year, a well-known columnist at the time, Meg Greenfield, made me into a verb in Newsweek. Every time she went to a dinner party, she wrote, people were Charles Murrayed if they tried to suggest that social programs had a useful role to play.

About this time, a mythology about Losing Ground began to overshadow the book itself, and it was no longer necessary to read the book before talking about it. All you had to know, according to the mythology, was that Murray said (1) the Great Society was a failure, (2) social programs only make problems worse, and (3) the welfare system ought to be scrapped. The other part of the mythology was that Losing Ground had become "the bible of the Reagan administration," as the New York Times's Leonard Silk put it. Then a dark mythology began to be spread as well: Murray had written the book to order at the behest of right-wing interests. It had been promoted with a slick public relations campaign. Murray had fudged the numbers and ignored data he didn't like. I was a Social Darwinist, John Kenneth Galbraith said. A New York Review of Books cartoon had me in top hat and tails, grinning fiendishly, the silent movie villain who no doubt had just foreclosed on the widow and her children.

None of the mythology was true—not even the good parts, unfortunately. The untold story about Losing Ground is that it was not popular within the Reagan administration, nor did it have any direct impact on policy. When a reporter for the Wall Street Journal tried to write a story about Losing Ground's influence on the White House and asked me for names of people she could interview, I had to tell her that, to my knowledge, no one in a senior position in the administration had read Losing Ground. She called back a few weeks later to tell me that she had spent a lot of time asking around, and that I was right.

What no one had noticed in all the furor was that Losing Ground was not a book congenial to the people who ran the Reagan White House in the second term. The book did not say that the problem with welfare was welfare cheats. It did not say that the problem of the underclass could be solved with minor reforms and a growing economy. If the Reagan administration had taken the message of Losing Ground seriously, it would have been obliged to confront messy, politically costly issues that the White House wanted to sidestep. The funniest example of this occurred when the administration came close to offering me a senior position in the Department of Health and Human Services that I didn't want but felt obligated to accept. I am told that the then secretary of HHS was given a copy of Losing Ground, dipped into it, and promptly scotched my appointment—I was far too radical.

All in all, Losing Ground did not have the hallmarks of a big book. It got a lukewarm-to-critical reception from the mainstream press, was subjected to savage attack from the left, sold only about 30,000 copies in its hardback version, and was largely ignored by the most conservative presidential administration in decades. And yet Losing Ground, by common consent of observers from left and right, had an enormous impact on the social policy debate. Many have argued that it changed the terms of that debate. Why and how? I am probably the worst person to ask, for my perceptions of the reaction to Losing Ground are skewed in many ways. But here are my best guesses:

Part of Losing Ground's effect came from its tone. I had been working for years with people who ran social programs at street level, and knew the overwhelming majority of them to be good people trying hard to help. The first acknowledgment in the preface was to them. When I then wrote in the prologue that the most troubling aspect of social policy was "not how much it costs, but what it has bought," I meant it, and most readers who came to the book as liberals saw that I did. There is no glee in Losing Ground's criticism of social programs, and this made it easier for many readers from left of center to follow my argument.

Losing Ground also liberated people to the right of center. In my experience, principled conservatives tend to be compassionate and generous in their personal lives. Losing Ground gave them intellectual permission to acknowledge those qualities when talking about social policy. One can be grieved by the plight of poor children and yet be opposed to new government social programs intended to relieve their plight—a combination of positions that would have surprised no one throughout most of American history, but had been impermissible since the 1960s.

I would like to think that another reason for Losing Ground's influence was the power of its argument, but cause and effect are always ambiguous in such cases. Did Losing Ground persuade people to change their minds, or did it articulate a view that many people had reached on their own? Some of both, presumably.

For whatever reasons, much of what was controversial when Losing Ground appeared had become conventional wisdom by the time I wrote the introduction for the tenth-anniversary edition in 1994. It was by then accepted that the social programs of the 1960s had broadly failed; that government is clumsy and ineffectual when it intervenes in local life; and that the principles of personal responsibility, penalties for bad behavior, and rewards for good behavior had to be reintroduced into social policy.

Other aspects of the conventional wisdom had changed radically. In 1984, Losing Ground had been attacked for the proposition that, in the short term, a pregnant, low-income single woman is economically better off staying single and going on welfare than marrying a man with a typical low-income job. By 1994, that had become a truism accepted by all sides in the debate over welfare reform. In 1984, at every college speaking engagement, I had to defend the proposition that nonmarital births are a problem for children and for society. By 1994, that proposition had been accepted by most academics from the left who had kept up with the technical literature. In 1984, Losing Ground's argument that a growing number of poor people were engaged in self-destructive personal behavior that would keep them on the bottom of society provoked angry retorts that the system was to blame, not the people who suffered from it. By 1994, no major figure in either academia or public life still argued that such a population did not exist. It had even acquired a label in the intervening ten years: the underclass.

Most astonishing was the evolution in attitude toward the policy proposal for which Losing Ground became most well known: to end welfare altogether. In 1984, I had made the case tentatively, putting it as a thought experiment in the concluding chapter. Twelve years later, Congress passed and President Clinton signed the Personal Responsibility and Work Opportunity Reconciliation Act of 1996. It abolished the Aid to Families with Dependent Children (AFDC) program and substituted for it a cash payment called Temporary Assistance to Needy Families (TANF) that was supposed to be time limited, accompanied by requirements that women receiving the cash payment take steps toward getting a job.

The welfare reform act of 1996 didn't actually get rid of welfare. The package of benefits available to a single woman with a child didn't change much, and the time limits on welfare never played a major role. As far as I could tell, the biggest changes were rhetorical: women were told that welfare was temporary and that they were supposed to look for jobs. I didn't expect much effect.

I was wrong. The welfare rolls plummeted. In the year the bill was passed, 6.3 percent of American families were on AFDC. By 2000, 3.0 percent were receiving TANF, a reduction of more than half in just four years. The welfare rolls continued to drop until the Great Recession. Surprisingly, the rolls rose only fractionally during the Great Recession, and then dropped again. In the most recent numbers, for 2013, the TANF figure was 2.0 percent— the lowest percentage of families getting cash welfare since the Kennedy administration.

No other social welfare legislation has ever produced anything approaching the immediate before/after change that accompanied the 1996 welfare reform. There has of course been extensive academic debate about what caused what, with some arguing that the economic expansion explained much of the drop. But those analyses did not confront the fact that the surge in welfare recipiency during the last half of the 1960s occurred during a red-hot economy, and that another booming economy in the mid-1980s was associated with flat welfare rolls. Something was different about 1996, and the welfare reform act is the parsimonious explanation. And, gratifyingly, Losing Ground got a lot of credit for having made the bill possible. I want to believe that's true, but I have no way of knowing for sure.

The welfare reform act had no effect on the broader deterioration of life in working-class America. Crime started going down after the early 1990s, which led to a major improvement in the quality of daily life in urban America. Almost all of the other indicators I used in Losing Ground showed no improvement or continued to worsen. But now policy analysts can reach these conclusions with much more revealing evidence than I had available in 1984. Then, I had to use statistics for black Americans as a substitute for what I really wanted: a longitudinal sample of disadvantaged Americans. I had no choice, because in 1984 policy analysts were still limited to the tables that the Bureau of Labor Statistics and the Census Bureau published. Today, I have the raw data from all of the Current Population Surveys going back to 1960 and all of the General Social Surveys going back to 1972, plus specialized longitudinal databases such as the National Longitudinal Survey of Youth, sitting on the hard drive of my computer—an unimaginable resource in 1984.

So when I revisited the problems of disadvantaged Americans in a book I published in 2012, Coming Apart, I could analyze the status of working-class Americans over time. The news is not good. The data I presented in Coming Apart