When quotas replace merit, everybody suffers

by Peter Brimelow and Leslie Spencer
Forbes, 1993

Bill Clinton promised that his cabinet would show as much diversity as is seen in the US. A quota policy, however, is unpopular with voters. A number of economic experts give their views of how employment quotas would affect the US economy.

President Clinton complained about "quota games" and "bean counters" when people criticized the composition of his Cabinet. Welcome, Mr. President, to a problem that you now share with millions of your fellow Americans.

"Quota games ... math games ... bean counters!"
 

President-elect Bill Clinton had every reason to lash out at feminist groups at his Dec. 21 news conference. In fact, he had been bean-counting busily himself: According to widespread reports, some of his original Cabinet picks were bumped because they were the wrong sex or race, key constituencies like urban Catholics and supporters of Israel have been crowded out, and his entire appointment process has been seriously slowed. But now mindless feminist pressure was forcing him to admit the ultimate contradiction of all such affirmative action policies: "Diversity" can conflict with merit.

Above all, the President must know the issue is death for the Democrats: His own pollster, Stanley Greenberg, conducted the post-1984 focus group interviews that found opposition of white working-class voters. (The party promptly suppressed Greenberg's report and now uses only happy-talk such as "looking like America." But a quota by any other name is still a quota.)

If quotas are clogging the Clinton transition, what are they doing to the economy? The subject went unmentioned, needless to say, at Clinton's two-day economic summit in Little Rock. In fact, it has gone virtually undiscussed throughout the quarter-century of bureaucratic and judicial decrees that have effectively transformed the color-blind 1964 Civil Rights Act into a pervasive quota system.

Ironically, just as socialism has collapsed across the globe, the leading capitalist power has adopted a peculiarly American neosocialism, putting politics (and lawyers) in command of its workplace, albeit on the pretext of equity rather than efficiency. Says Edward Potter of the Washington, D.C.-based Employment Policy Foundation: "We have, without doubt, the most far-reaching equal employment laws found anywhere in the world."

Before applauding Potter's sweeping statement, stop for a minute and ponder this question: What does the replacement of merit with quotas cost the American people? The answer is: plenty. The impact may easily have already depressed GNP by a staggering four percentage points—about as much as we spend on the entire public school system.

Quotas are not the law of the land, exactly. They are explicitly banned in both the 1964 and 1991 Civil Rights Acts. Nevertheless, corporate America has been terrorized by the legal legerdemain whereby any statistical disparity between work force and population is equated with intentional discrimination. Throughout American business, newly entrenched affirmative action bureaucrats are enforcing discrimination by race and sex—in favor of the "protected classes" (women, minorities and, most recently, the disabled)—as decreed by Washington.

One such bureaucrat, Xerox Manager of Corporate Employment Theodore Payne, puts it bluntly: "We have a process that we call 'balanced work force' in Xerox, everybody understands that, and it's measurable, it's goals.... Relative numbers. Relative numbers. That's the hard business, that's what most people don't like to deal with, but we do that all the time."

"Balanced work force" is, of course, yet another euphemism for quotas. Payne is apparently saying that Xerox discriminates against white males in favor of the "protected classes." He says it without apology. But, if anyone cares, white males have feelings (and families to support), too.

"To cut whites out of the entire process is racism pure and simple," laments a white male reporter for the San Antonio Light, which is due to close down any day. He says Gannett and other major news organizations are showing interest in his Hispanic colleagues exclusively. But he adds: "I don't want to be quoted. I'll never find another job if I am."

In a blistering 1987 article in Society magazine, the late Professor William Beer of Brooklyn College described his fellow social scientists' attitude to affirmative action as one of "resolute ignorance." FORBES' search of academic journals and Ph.D. theses confirms that ignorance has remained resolute. What little work has been done tends to focus only on whether affirmative action policies have benefited the "protected classes." (Have they? For an answer, see below.)

Corporate America contributes to this resolute ignorance by declining to disclose its costs. "Our members would never say," the National Association of Manufacturers' Diane Generous predicted (rightly). "They would be concerned they might be accused of complaining about how much money they had to spend on this."

Another big business lobby, the Business Roundtable, did publish a study by accountants Arthur Andersen more than a decade ago on how much its members spent to comply with federal regulation, including specifically the Equal Employment Opportunity Commission (EEOC). But today a Roundtable spokeswoman says the organization has no plans to update the study—and that it no longer even possesses any copies.

Sure, measuring the costs of regulation is difficult. But it can be done. For example, the Environmental Protection Agency is required by executive order to make regular estimates of its economic impact.

So here is a rough but reasonable try at figuring the cost of quotas. That funny noise you hear from now on is economists gritting their teeth. Our response to them: Go make your own estimates. And remember—the truth shall set you free.

Two points about quotas emerge immediately: Quotas are a very big deal. All employers with more than 15 staff, public, private or nonprofit, come under the EEOC's Uniform Guidelines on Employee Selection Procedures. All can be sued by the EEOC for "discrimination" if the racial, ethnic and sex mix of new hires diverges sufficiently from that of all other qualified applicants—for example, if the percentage of blacks hired is lower than the percentage of blacks applying. That covers 86% of the entire non-farm private-sector work force.

Additionally, more than 400,000 corporations doing business with the federal government, covering about 42% of the private sector work force, have to file with the Office of Federal Contract Compliance Programs (OFCCP). This process is so onerous that the OFCCP's explanatory manual is about 700 pages long. Corporations with contracts of $50,000 or more must develop an "affirmative action plan" aimed at achieving staffing at all levels that is proportionate to the composition of the qualified work force.

Many colleges and universities are subject to no fewer than three federal agencies: EEOC, OFCCP and the Department of Education's Civil Rights Office. And finally, there are federal, state and local governments. Here a racial and gender spoils system has effectively subverted the merit hiring rules so painfully established by Progressive Era reformers at the beginning of the century.

All of which means that the 1984 poll that found one in ten white males reporting they had lost a promotion because of quotas was quite possibly accurate. Indeed, it could be an underestimate. Quotas have been implemented with extraordinary secrecy and deceptiveness, in part because of their dubious legal status.

"Word comes down, but it does not go out," says Claremont McKenna College sociologist Frederick Lynch, author of the only study on the impact of quotas on white males, Invisible Victims. He cites a Los Angeles manufacturer whose receptionist was instructed to accept but quietly shelve employment applications from whites and Hispanics—after they had left the room—because the plant did not have "enough" blacks.

Typical of the secrecy and scale of quotas: the "race-norming" saga. EEOC Vice Chairman R. Gaull Silberman—a Reagan appointee—says that until she read it in a newspaper in 1990, she and EEOC Chairman Evan Kemp had "absolutely no idea" that their own agency was pressing for aptitude tests to be race-normed. This bit of bureaucrat-speak refers to the practice of radically adjusting scores to compensate for minorities' systematically lower results. Yet race-norming had been going on throughout the 1980s. It reportedly subjected at least 16 million test-takers to a quota system they knew nothing about.

After public outcry, race-norming was banned in the 1991 Civil Rights Act. But quotas, like vampires, have proved virtually impossible to kill. Now they seem to be rising from the grave in the shape of a new test-twisting technique called "banding"—concealing differences in performance by lumping ranges of scores together. The second point about quotas: Quotas are very expensive. There's surprising denial about this. University of Chicago free market economist Gary Becker, a 1992 Nobel laureate, wrote the standard analysis, The Economics of Discrimination (1957). But Becker recently shrugged off affirmative action in a Business Week column. He argued that although affirmative action "does hurt some individuals, as it caters to minorities with political clout," it "probably causes less harm than many other programs," such as farm supports.[ HOW IS AFFIRMATIVE ACTION LIKE CROP SUBSIDIES?, April 27, 1992]

Strangely, however, Becker tells FORBES that in fact he has no idea what quotas cost ("I think it's an important subject for research"). But we do know what farm supports cost: about $9.7 billion in 1992, which is substantially lower than our estimate of $16 billion to $19 billion for private-sector and education compliance costs alone (see p. 82).

Nevertheless, Becker's analysis of discrimination remains the best framework for assessing the economic impact of quotas: In a free market, Becker argued, there is an inexorable tendency for everyone to receive the marginal value of his or her labor. This means that ultimately, you are likely to be paid something like what your work is worth. If you belong to an unpopular group, employers may pay you less. But that means that they will make more money off you. Because you are such a profitable hire, you will come into demand, and your labor will be bid up. This process can only be prevented by monopoly or government intervention—both of which happened, for example, in South Africa under apartheid. And now in the U.S. under affirmative action.

Talking to FORBES, Becker is very anxious to stress that he is not saying discrimination will be completely competed away. But there is a tendency for it to be competed away. "Competition forces people to face the costs, and therefore reduces the amount of discrimination, when compared with a monopolistic or noncompetitive situation," Becker says.

If you believe that racism stalks America like the Angel of Death and that only federal force can keep it in check, you won't like what Becker is saying. But the evidence clearly supports him.

"Once adjustments are made for factors like age, education and experience, 70% to 85% of the observed differences in income and employment between the various groups in America disappears," says economist Howard R. Bloch of George Mason University. "That's been shown by studies dating back to the mid-1960s. And you can't even be sure that the residual gap is due to discrimination. It could be due to factors we haven't controlled for."

Indeed, Harvard economist Richard Freeman found blacks and whites with the same backgrounds and education had achieved wage parity by 1969, well before quotas had America in their grip.

Even the recent much-touted Federal Reserve Bank of Boston study claiming to prove the existence of racial discrimination in mortgage lending turns out to have made a basic methodological error in its handling of default rates (FORBES, Jan. 4). Perhaps significantly, its co-author, Boston Fed Research Director Alicia H. Munnell, was a featured speaker at the Clinton economic summit.

All of which shows the fallacy of two common arguments for government-imposed quotas: that they are necessary to force corporations to tap new pools of labor, and that corporations need a diverse work force to service an increasingly diverse population. Both simply assume that markets don't operate—that corporations couldn't figure this out themselves.

In fact, it's hard to see any benefits contributed by quotas to the overall economy—as opposed to the benefits they channel to the "protected classes." "Affirmative action is a fairly pure form of rent-seeking," says the University of Arizona's Gordon Tullock, using the concept he developed for special interests' use of political power to extract subsidies for themselves from the economy. "There simply isn't any other economic rationale."

"In 1987, EEOC's local field office wrote me a letter saying they had reason to believe I didn't have enough women 'food servers' and 'busers.' No woman had complained against me. So the EEOC advertised in the local paper to tell women whose job applications we had rejected—or even women who had just thought of applying—that they could be entitled to damages. Twenty-seven women became plaintiffs in a lawsuit against me. The EEOC interviewed me for hours to find out what kind of person I was. I told them in Sicily where I came from I learned to respect women. I supplied them with hundreds of pounds of paper. I had to hire someone full time for a year just to respond to EEOC demands. Six months ago I finally settled. I agreed to pay $150,000 damages, and as jobs open up, to hire the women on the EEOC's list. Even if they don't know what spaghetti looks like! I have to advertise twice a year even if I have no openings, just to add possible female employees to my files. I also had to hire an EEOC-approved person to teach my staff how not to discriminate. I employ 12 food servers in these two restaurants. Gross sales, around $2 million. How much did it all cost me? Cash outlay, about $400,000.

"What the government's done to me—devastating. I wouldn't wish it on my worst enemy."—Thomas Maggiore, Owner of Tomaso's and Chianti restaurants, Phoenix, Ariz.

Economists break the cost of regulation into three parts:

Remember: Thomas Maggiore is precisely the kind of small business person the politicians claim they want so badly to help.

Let's look at some numbers.

Direct Costs. One guess of private sector compliance costs for affirmative action: In 1977 Business Roundtable members spent $217 million complying with equal opportunity regulations. They employed 5% of the non-farm work force; OFCCP regulations cover 42% of the private workforce, implying total costs of $1.8 billion. Adjusted for inflation, that's a current $4.2 billion.

Second guess: In 1981 a study by the Senate Labor & Human Resources Committee suggested compliance costs for the largest 500 companies of about $1 billion. That's $1.8 billion extrapolated over the OFCCP universe. Adjusted for inflation: $2.8 billion.

Neither of these figures includes the EEOC's impact, although it is by far the larger bureaucracy. But the guesstimates are in line with the rule of thumb developed by regulation-watchers from the Center for the Study of American Business at Washington University in St. Louis: Every dollar spent on regulatory enforcement inflicts about $20 in compliance costs. By FORBES' count, the federal government spent some $425 million on civil rights oversight in 1991, of which about $303 million appears to be directed at the private sector. Implied private-sector compliance cost: $6 billion.

To get an estimate of compliance costs in colleges and universities, FORBES turned to John Attarian, a writer and economics Ph.D. who has analyzed the budget of his alma mater, the University of Michigan. Under its "Michigan Mandate," the university is devoting much effort to the recruitment and retention of the "protected classes."

Attarian says about 2.5% of the University of Michigan at Ann Arbor's general budget appears to be devoted to this cause. This does not capture costs buried in department budgets, such as for recruitment. (Minor example: Advertising faculty posts in special minority-oriented publications costs over twice the usual rate.) Still, extrapolated across the estimated $164 billion spent on U.S. higher education in 1992, this suggests total compliance costs of $4.1 billion.

If the same relationship holds true for the $261 billion spent on public and private schools in 1992, their compliance costs would be $6.5 billion. Of course, the problems of schools are different from those of colleges. They may be worse. Busing for racial balance has reportedly caused some school districts to spend over a quarter of their budgets on transportation.

Quotas are just another excuse for the American academic establishment to eschew scholarship for social engineering. Thus, a long survey of "minorities in science" in the Nov. 13 issue of Science magazine reported that the National Science Foundation, which is supposed to be funding research, has spent a staggering $1.5 billion in the last 20 years on fostering black scientists. The magazine describes the results as "dismal."

State and local governments also face compliance costs—and they also inflict them on the private sector. New York State, for example, spent $10.5 million complying with its own and federal laws last year, and $7.5 million on "civil rights" enforcement. In 1990 state and local governments spent some $835 billion. Implied total expended on quota compliance and coercion, given New York's rate: $287 million. Additional private sector compliance costs, given New York's enforcement costs and applying CSAB's 20-to-1 rule of thumb: $2.4 billion.

Note that we include no estimate of what it costs the federal government to comply with its own regulations.

We like to be moderate.

Private-sector compliance costs are apparently much exacerbated by the federal enforcers' arbitrary and erratic behavior. Some rare case studies appeared in the September 1992 issue of the American Academy of Political & Social Science's journal Annals. One victim reported supplying documents nine times because the OFCCP kept losing them. Another, the National Bank of Greenwood, Indiana—$117 million assets, 138 staff, full and part-time—was subject to a grueling and chaotic two-year audit, costing more than $100,000 and 4,000 staff hours, although no complaint had apparently been lodged against it. Later the bank was audited twice more, again apparently without any complaints being lodged. Typically, the Indianapolis-based Merchants National Corp., which has meanwhile taken over the National Bank of Greenwood, refused to allow its officers to talk about the experience.

Total direct costs: $16.5 billion to $19.7 billion. Or about $300 per family of four. Compare it with the $20 billion of "infrastructure spending" Clinton has promised to kick-start the economy. And this is just the tip of the iceberg.

Indirect Costs are the part of the iceberg just under the water—easily seen but involving no direct cash outlay.

"It takes me 50 extra hours to make every faculty hire because of the need to comply with affirmative action rules," says Professor Herbert London, formerly Dean of New York University's Gallatin Division, "even when I end up hiring the person I wanted to hire in the first place."

Naturally, this cost does not appear as a cash item in NYU's operating budget of $627 million, excluding the medical school. (The two-person affirmative action office costs just $172,000—or about $6.50 per fulltime student—although a spokesman tells FORBES that over a hundred people deal with minority recruitment every day.) Nevertheless, the cost is real.

A measure of these indirect costs is provided by the single Ph.D. thesis FORBES found that investigated costs, by Peter Griffin, now assistant professor at California State University at Long Beach. Griffin's rarefied econometric analysis concluded that by 1980, OFCCP regulation had increased federal contractors' labor and capital costs by an average of 6.5%. (As compared with non-contractors—although actually their costs would also have been increased by EEOC requirements.)

The implications of this are substantial. OFCCP regulation covers about 42% of the civilian work force. The contractors' cost of labor alone exceeded $1.4 trillion. The minimum cost of quotas to them, based on Griffin's methodology: about $95 billion—1.7% of GNP.

And the cost to the federal taxpayer is heavy. In 1991, $211 billion was expended on federal contracts with non-government entities. The additional costs inflicted by affirmative action regulation that Griffin's work suggests this sum incorporates: some $13 billion.

Which is on top of the damage inflicted on the taxpayer by "set-asides," the reserving of some portion of federal work entirely for contractors from the "protected classes." About $10 billion of federal contract monies were channeled in this way last year. The premium paid is not supposed to go over 10% (although FORBES has heard of premiums as high as 25%). Additional quota tax: perhaps $1 billion.

Ironic set-aside fact: The law is confused about this type of quota too. In Richmond v. Croson (1989), the Supreme Court ruled that many of the 234 state and local government set-aside programs were unconstitutional, unless actual discrimination could be proved. Local politicians, anxious to continue handing out the pork, instantly created a minor "disparity studies" industry to make the case that discrimination against minorities was widespread. In a detailed account in the January 1993 issue of Public Interest magazine, University of Maryland at Baltimore Professor George La Noue estimates that at least $13 million of taxpayers' money had been fed into this young industry by June 1992, with another $14 million commissioned by the federal Urban Mass Transit Authority alone. Atlanta spent $532,000 for a 1,034-page report coauthored by Ray Marshall, the Carter Administration's Secretary of Labor. [Social Science And Minority "Set-asides"]

Expensive? Well, proving discrimination is hard work. Most localities have long been legally required to accept the lowest bid—a Progressive-era reform aimed precisely at patronage-hungry politicians. And, significantly, cities like Atlanta, which now want to claim they discriminated, have actually been under black political control for years.

Even more ironic set-aside fact: This type of quota has created another industry—corruption. A prime contractor can set up his black electrician, for instance, in "business" as a purchaser. The electrician needn't have credit or contacts with suppliers. He just takes 5% off the top. One "native American" contractor in Tulsa reportedly had blue eyes and an Irish name but had managed to join the Cherokee Nation of Oklahoma on the strength of an alleged great-great-great-great grandparent.

These abuses can only be checked by more supervision. But minority contractors have been quoted complaining the program is too bureaucratic already.

Astoundingly ironic set-aside fact: According to Professor La Noue, over one-half of the Small Business Administration's set-asides go to groups that are composed largely of first or second-generation immigrants. He suspects the same is likely to be true for all set-asides. In Washington, D.C.—where an amazing 90% of the city's road construction contracts have been set aside—one of the largest beneficiaries has been the Fort Myer Construction Corp., owned by a family of Portuguese origin who qualify as Hispanics because they emigrated from Argentina.

Absurdly, all immigrants who fall into the "protected classes" qualify for all U.S. quota programs. Which is a pretty clear indication that quotas are not about righting past wrongs at all, but about asserting political power over the economy.

A further indirect cost of the affirmative action system: litigation. (You thought massive regulation would preclude litigation? This is America!)

The number of discrimination suits in federal courts is rising astronomically—by 2,166% between 1970 and 1989, when some 7,500 were filed, versus an increase of only about 125% in the general federal caseload.

Significantly, suits about discrimination in hiring used to outnumber suits about firing. Today it's the reverse, by a factor of three or more. It's obviously absurd to suppose the same employer discriminates in firing but not in hiring. The civil rights frenzy has simply led to a more litigious, as well as politicized, workplace.

Example: Alabama state law required the Lamar County Board of Education to fire a black teacher after she failed a mandatory competency test five times during the three years allowed. She alleged discrimination because the test failed a disproportionate number of blacks. A judge reinstated her with three years' back salary.

And it's going to get much worse. Preliminary reports are that since the 1991 Civil Rights Act and the 1990 Americans With Disabilities Act (which few people yet realize is also a quota bill) filings have jumped some 30%. Both acts for the first time allow punitive damages, an explicit incentive to contingency-fee trial lawyers.

Opportunity Costs are the base of the quota iceberg, down in the murkiest depths. Unlike the direct and indirect costs of regulations, they don't show up in GNP statistics. They represent what GNP could have been if these more tangible costs had been spent differently—for job-creating investment, say, or for education. But these indirect costs are the most massive of all. For example:

Having the wrong people in the wrong jobs. Corporate America seems to have resigned itself to quotas as yet another tax. But they are a peculiarly debilitating sort of tax, levied not on the bottom line but on every phase of the corporation's activities, increasing inefficiency throughout. Most taxes are a burden to be shouldered. This is an enfeebling drug.

That affirmative action quotas lead to lowered standards is all but guaranteed by the fact that all standards are suspect to Equal Employment enforcers. "Many of these people believe there really is no such thing as job performance or productivity objectively defined, that it's really just a matter of one's cultural definition or cultural orientation," says Frank Schmidt, a University of Iowa industrial psychologist. Increasingly, they have been able to impose this view on American business.

The civil rights revolution has also virtually aborted the use of tests devised by industrial psychologists, which in the 1950s promised to make employee selection a science. Tests came under attack because minorities typically scored lower on them. Today they are only used, if at all, after work-related validation studies that can cost millions of dollars.

Industrial psychologists, however, have gone on believing in their work. Schmidt and John Hunter of Michigan State University have produced numerous studies showing that hiring the able results in enormous productivity increases. Today, Hunter estimates that total U.S. output would be about $150 billion higher if every employer in the country were free to use tests and select on merit. That's about 2.5% of GNP.

Effect on morale. Poor hiring shows up not merely in poor decisions but also in poor morale. Quotas, like income tax (and unlike farm supports), have an immediate and dramatic impact on incentives.

Frank Schmidt put it like this: "When the less competent employees reach a critical mass, their lower performance standards become the standards of the organizations." The longer-established employees who are equipped for the job abandon their old high standards and conform to the new, lower ones. Schmidt and Hunter made no estimate of the impact of this phenomenon. But they have speculated that it lay behind the U.S. productivity stall of the 1970s, as the first effects of the war against testing were being felt.

Misallocation of resources. Monies expended to meet the costs of affirmative action cannot be spent on research and development and plant modernization. The effect of this is cumulative: The growth path of the economy diverges, permanently and increasingly, from its potential. Thus we estimate that an extra $113 billion in direct and indirect costs have been inflicted on the economy annually since 1980. A standard calculation converts this into an estimate of GNP shortfall because of affirmative action: about 1.5 percentage points by 1992.

GNP in 1991 was about $5.7 trillion. The total shortfall quotas may already have caused comes to some 4%. That's well over $225 billion, money that could buy a lot of social programs. Or finance a good deal of job-creating investment.

So quotas cost a lot. But do they do any good at all?

Quotas have obviously failed to prevent continuing catastrophe in much of black America. Prevailing taboos make this subject difficult to discuss. But the distressing facts are powerfully summarized in a remarkable new book, Jared Taylor's Paved With Good Intentions: The Failure of Race Relations in Contemporary America (Carroll & Graf). In 1950 only 9% of black families were headed by a single parent; in 1965, 28%; now, fully half. In 1959 only 15% of black births were illegitimate; in 1992, 66%. One in four black men in their 20s is either in jail, on probation or on parole. Clearly, affirmative action has done nothing to reverse the dismal trends.

Quotas have not decisively improved overall black employment. "Despite all the controversies surrounding affirmative action," says Queens College Professor Andrew Hacker, a supporter of quotas, in his best-selling Two Nations: Black and White, Separate, Hostile, Unequal, "fewer blacks now have steady jobs of any kind and their unemployment rates have been growing progressively worse relative to those recorded for whites."

Quotas' effect on black incomes appears at best mixed. Between 1970 and 1990 black median family income, adjusted for inflation, crept snail-like from $21,151 to $21,423. But the proportion of black families earning above $50,000 jumped sharply, from about 10% to nearly 15%. Dragging down the median: the increase in black families receiving below $15,000, now nearly 40%. So quotas may have helped create a black middle class (although educated blacks might have done well anyway; after all, the proportion of white high income families also rose in this period). But the black poor have not benefited.

Quotas in colleges have not prevented the gap between black and white college participation from widening in the 1980s. By 1976 some 22.6% of black 18-to-24-year-olds enrolled in college, compared with 27.1% of whites. Thereafter black participation declined, then recovered. In 1990, 25.4% blacks enrolled, but meanwhile white participation had grown to 32.5%.

And although crude enrollment numbers are dear to the hearts of college admissions officers, they conceal tragic differences in attrition. For example, only 37.5% of blacks enrolling at Berkeley in 1983 had graduated five years later, compared with 72% of whites. Critics argue that top colleges burn out black students by irresponsibly recruiting them to fill quotas, when they could be successful at less high-pressure schools.

Quotas may have improved the status of women—or they may not. It's easiest to show that women have gained in the last decades—ironic, because their plight was hardly as serious as that of blacks, with whom they are now competing. Women's share of professional degrees grew from 2.7% in 1960 to 36% in 1990, and their average earnings as a percentage of men's has increased from 61% to 72% over the same period.

But quotas may not be responsible. Female participation in the work force has fluctuated widely for generations, correlated with demographic factors like marriage and fertility rates. For example, the Hoover Institution economist (and FORBES columnist) Thomas Sowell has noted that women earned 17% of Ph.D.s in 1921 but only 10% in the early 1960s. Amazingly, as long ago as 1879 women constituted 40% of all college faculty and administrators. Many of these colleges were women-only, but they could still be highly competitive: In 1902 the proportion of women listed in Who's Who was more than double that in 1958.

This problem of apportioning credit bedevils the whole quota debate and, indeed, the entire subject of government-mandated social change. Looking back on the 1964 Civil Rights Act and its controversial enforcement, the American Enterprise Institute's Charles Murray, author of Losing Ground and In Pursuit, offers this startling thought: "There's hardly a single outcome—black voting rights, access to public accommodation, employment, particularly in white-collar jobs—that couldn't have been predicted on the basis of pre-1964 trend-lines." That's pretty devastating. It suggests that we have spent trillions of dollars to create an outcome that would have happened even if the government had done nothing.

From an economic standpoint, quotas work rather like an older form of American neosocialism: price and wage controls. They may seem to produce the desired result. But they could equally well just be simulating it, or even smothering it.

Meanwhile, of course, the economy suffers.

It may be that before America can talk rationally about race, the generation that remembers segregation will have to die off. And we're not talking about liberals. FORBES asked Gary Becker, 62, what he thought would be the ideal public policy in this area.

Becker: I prefer to pass on that one. I have views on it, but I don't want to talk about it at this moment.

Oh. Why not?

Becker: Well, let me just make that judgment. I prefer not to.

Becker's University of Chicago colleague Richard A. Epstein, 49, seems to be less nervous about his popularity in the Faculty Club. His book Forbidden Grounds: The Case Against Employment Discrimination Laws argues that the modern civil rights laws are flawed to their heart because in negating freedom of association they have inexorably led to government coercion that threatens markets and, ultimately, liberty.

"At bottom are only two pure forms of legislation—productive and redistributive," Professor Epstein argues. "Anti-discrimination legislation is always of the second kind. The form of redistribution is covert; it is capricious, it is expensive and it is wasteful."

And Epstein makes the key economic point: If we want to subsidize a "protected class," he writes, it can be done more efficiently by just giving grants.

"I have a dream," Martin Luther King Jr. said 30 years ago, "that my four little children will one day live in a nation where they will not be judged by the color of their skin, but by the content of their character."