Diversity’s Economic Downside

By Edwin Rubenstein
Volume 25, Number 4 (Summer 2015)
Issue theme: "America Unsecured - Pathway to Another 9/11"

America pays a high price for racial and cultural diversity. Most of us have no idea how high the price really is. The economic burden falls hardest on non-Hispanic whites who, as a group, are disadvantaged by policies aimed at increasing diversity. It is increasingly clear, however, that minorities themselves are also hurt by these policies. Indeed, the unintended consequences of race-based preferences may be one of the greatest public policy stories never told.

We analyze here the economic impact of two policies that have changed the rules in favor of racial and ethnic minorities, increased their numbers, or their influence: affirmative action and multiculturalism.

Affirmative Action

Peter Brimelow, writing several years ago in Forbes, calculated cumulative affirmative action costs to the U.S. economy at $225 billion, or 4 percent of Gross Domestic Product, for the year 1991.1 In other words, had the funds spent to enforce and comply with affirmative action programs since their inception been put to other, more productive use—research and development or plant modernization, for example—GDP would have been 4 percent higher in 1991.

If Brimelow’s 4 percent figure is accurate (and to my knowledge no one has ever challenged it), then the cost of affirmative action programs would currently be about $540 billion. But nearly twenty-five years have passed since Brimelow made his calculation. Even if the same affirmative action programs were in place, their impact would be larger today because they have been in existence longer. The economic cost of affirmative action compounds annually, as the growth path of the economy increasingly diverges from its potential.

A recent study published in the Journal of Economic Growth found that between 1949 and 2005 the accumulation of federal regulations slowed U.S. economic growth by an average of 2 percent per year.2 Had the amount of regulation remained at its 1949 level, 2011 gross domestic product (GDP) would have been about $39 trillion—or three and a half times—higher, which translates into a loss of about $129,300 for every person in the United States.

Bottom line: The misallocation that cost 4 percent of that GDP in 1991 could easily cost 8 percent of GDP today (2015). This implies a $1.4 trillion economic loss from affirmative action programs—$4,420 taken from every man, woman, and child in the country.

Affirmative Action in the Workplace

To calculate the cost of affirmative action we start with the Equal Employment Opportunity Commission (EEOC), the federal agency charged with ending employment discrimination through enforcement of the nation’s equal employment opportunity laws. Those laws prohibit workplace discrimination on the basis of race, color, sex, national origin, religion, age, pregnancy, disability, family medical history, and genetic information.

The staffing and budget appropriations for the Equal Employment Opportunity Commission mushroomed during the 1990s, and have continued rising during the Bush and Obama years:


EEOC outlays are quite modest by government standards. But like a pebble dropped in a pool of water, the direct cost of running an agency like EEOC generates far larger costs to a wide swath of the economy.

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. EEOC regulations cover nearly 90 percent of the non-farm private-sector work force.

In recent years (2008 to 2014) an average of 90,000 to 100,000 individuals filed grievances each year with the EEOC—a number substantially more than the 70,000 or so that filed annually during the Clinton Administration’s heyday in the late 1990s.3 Thirty-five percent of EEOC grievances filed in 2014 were for alleged racial discrimination, 30 percent for discrimination based on sex, 23 percent were based on age, and about 11 percent alleged discrimination based on national origin.

Ending, or at least reducing, discrimination in hiring and pay practices is surely a benefit worth paying for. But EEOC defines discrimination in terms of outcomes rather than opportunities. Thus if 50 percent of applicants for a position are black, and only 40 percent of workers actually hired for that position are black, the presumption of discrimination exists.

To protect themselves against EEOC discrimination charges, employers must prepare affirmative action plans laying out, in excruciating detail, the special recruitment, monitoring, notification, review, and record-keeping procedures the company will implement to hire the target percentages of minority employees.

The result is a quota system. In the construction trades, for example, a typical “Affirmative Action/Non-Discrimination/Requirement” document lists “participation goals” as follows: Carpenters are expected to be 42.74 percent minority and 1.58 percent female; iron workers are expected to be 58.53 percent minority and 7.63 percent female.4 There is no clue as to how those numbers were derived—but you can see the value of a minority female ironworker.

To be eligible for federal contracts, private contractors must keep records of each applicant and document reasons for not hiring a woman or minority candidate. Even if there are no openings, help wanted ads must be published—just to add potential female and minority hires to company files.

The federal government is a victim of its own regulations. The Commerce Department, for example, has labeled industries awarded roughly three-quarters of all federal contracts as guilty of discrimination, thereby making female and minority owned firms in these industries eligible for a 10-percent bid shelter.5

Translation: when bidding against non-preferred (i.e., white-owned) businesses, minority contractors can charge 10 percent more and still win a “competitive” bid for a federal project. Unpublished data provided to us by the Federal Procurement Data System show $10.5 billion of federal contracts were awarded to minority-owned firms in fiscal year (FY) 2005. Six years later, in FY 2011, a whopping $36 billion was awarded to minority-owned firms, according to a study by the General Accountability Office.6

These race-based preferences assume that minority firms are discriminated against. Many economists, however, believe racial discrimination is rare or non-existent in a capitalist economy such as ours.

Gary Becker, a 1992 Nobel laureate in economics, wrote the standard analysis, The Economics of Discrimination, in 1957. In a free market, Becker argued, there is an inevitable tendency for everyone to be paid 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 are in an unpopular group—a racial minority, for example—employers may pay you less. But that means they will make more money off you. Because you will be such a profitable hire, your services will be demanded by other employers, and your wages will be bid up.

Employers who pay minority workers less than their “marginal product” will lose those workers to competitors. They will not survive. This process can be forestalled only by monopoly or government intervention—both of which occurred, for example, in South Africa under apartheid. And now in the U.S. under affirmative action.

If you believe that racism is pervasive and that only the federal government 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 percent to 85 percent 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.”7

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. “By the late 1970s,” Freeman writes, “young black male college graduates attained rough income parity with young white graduates.”8

For female black college graduates, the gap more than vanishes. Joseph Conti reports in Profiles of a New Black Vanguard that “black college-educated females currently earn 125 percent of what white college educated females earn.”9

Unintended Consequences

Race-based affirmative action programs, by increasing the threat of lawsuits brought by employees in protected groups, may have actually worsened the plight of minority workers. Case in point: the 1991 Civil Rights Act. The legislation significantly expanded the rights of minority plaintiffs in discrimination complaints to the EEOC and in federal civil court. In particular, the Act made it easier to use statistical evidence, however flawed and misleading, to “prove” discriminatory intent even when no discriminatory intent is evident on the part of the employer.

Research has uncovered three main changes in the hiring and dismissal of protected workers in the wake of the civil rights law. First, a trend toward greater racial and gender integration in labor markets stopped around 1991.10 Industries that historically had low minority and female employment shares were hiring more such workers prior to the legislation, but this pattern did not continue into the post-Act period.

Second, employers changed the way they dismissed protected workers. Because it is more difficult for workers to win employment discrimination lawsuits after losing their job as part a mass layoff, employers delayed dismissing protected workers until the next round of layoffs. While the overall rate of job loss remained constant for both protected and unprotected workers, the fraction of lost jobs coming as part of a mass layoff increased for protected workers, but not for unprotected, after the passage of the Act.11

Third, the wage premium on labor market experience (which economists refer to as returns to experience) increased for some groups of protected workers after the 1991 Act. If employers are concerned about the costs of dismissing potential employees, then increases in those costs will cause firms to reduce their demand for workers who are relatively likely to be dismissed. Because job turnover rates are highest for inexperienced employees, increases in firing costs can increase the returns to experience.

Prior to the 1991 Act, black men were fired about twice as frequently as white men. After the legislation, this gap closed considerably. In fact, in the post-Act period, the firing rates for black and white men were roughly the same—0.5 percent per year.

But this pattern did not signal less frequent job loss for black workers. Economists Paul Oyer and Scott Schaefer found a change in the manner of dismissing minority workers following passage of the civil rights law:

….We discovered that the overall rate of involuntary job loss for black men did not change as a result of the Act. Thus, it appears that firms simply shifted the form of some dismissals for black men from individual firings to layoffs in response to the 1991 Act. We documented that the share of black involuntary job loss coming in the form of firings dropped by around one-third after the passage of the Act, while the share of white involuntary job loss coming in the form of firings was unchanged.12

We think these findings should be troubling for those who supported the 1991 Act in the belief that it would help open labor market opportunities for members of protected groups. As economists and legal scholars have long argued, employment protections that make it costly for firms to fire protected workers are likely to have very different effects from protections that make it costly for firms to fail to hire protected workers. Because the firm feels firing-based costs only if it decides to hire, the costs act as an implicit tax on such hiring. Firing-based protections may therefore lead employers to hire fewer protected workers, not more. This may explain why the Civil Rights Act of 1991…does not appear to have led to gains in wages or employment for protected workers.

To avoid the threat of lawsuits, many U.S. employers have hired illegal immigrants in place of minorities—many of whom may be willing to work for the same wage. That employers take such risks, including hiring people who can barely speak English, suggests that the true cost of hiring protected groups may be huge. Some have put the “implicit tax” of hiring workers protected under the Civil Rights Act at $25,000 per worker.

Affirmative action in college admissions

Elite private universities in the United State routinely bend admission criteria in favor of minority applicants who would not be admitted under a color-blind meritocracy. One result is a shocking differential in acceptance rates between black applicants and those of other races:


At 13 of the 18 upper echelon universities surveyed by the Journal of Blacks in Higher Education, the black student acceptance rate was higher than the acceptance rate for white students. In some cases the difference was substantial. At MIT, for example, black applicants were nearly twice as likely to be accepted as the average applicant. At Harvard blacks are 67 percent more likely to be accepted than the average applicant. At the University of Virginia 62.2 percent of blacks were accepted while 38.2 percent of all applicants received notices of acceptance. http://asianam.org/hall-of-shame/college-admission-officers/

Only six of the eighteen universities surveyed had black acceptance rates lower than the overall acceptance rate. This includes Berkeley and UCLA, which were prohibited from taking race into account during the 2004 admission process. The black acceptance rate was also lower than the white rate at Washington University, Emory, and Wake Forest.

Asian-Americans and whites are the groups most disadvantaged by race-based admissions criteria. Princeton sociologist Thomas Espenshade examined admissions data in No Longer Separate, Not Yet Equal, a book he co-wrote in 2009. Among his conclusions: Asian-Americans need to score 140 points more than whites, 270 points more than Hispanics, and 450 points more than African-Americans out of a maximum 1,600 on the math and reading SAT to have the same chance of admission to a private college.

The pro-minority tilt in college admissions rates is well known to those in the field. Roger Clegg, the president of the Center for Equal Opportunity, a Falls Church nonprofit, that opposes racial preferences in college admissions, is quoted as saying: “Clearly, both whites and Asian-Americans are discriminated against vis a vis African-Americans and Latinos…At some of the more selective schools, Asians are also discriminated against vis a vis whites.” http://www.americanbazaaronline.com/2014/07/21/asian-american-students-battling-discrimination-college-admissions-now-something-startling-race-based-grading/

More troubling still are the race-based biases evident in U.S. professional schools :


Black applicants with low MCAT scores and Grade Point Averages are nine times more likely to be admitted to medical schools than Asians with comparable test and grade levels. Similarly, Hispanics are four to five times more likely to gain admission than Asians of comparable test and grade point achievement levels.

This bias puts many black and Hispanic patients at risk if, as seems likely, black and Hispanic medical school graduates serve disproportionately in communities serving black and Hispanic populations. This “reverse discrimination” reduces competence levels throughout the medical profession.

The newly admitted students may themselves be victims. Students admitted to schools they would otherwise not be capable of attending are more prone to drop out of medical school or, if they graduate, fail in their professional careers. Opponents of race-based affirmative action argue that the programs actually benefit upper- and middle-class blacks and Hispanics at the expense of lower-class white non-Hispanics and Asians. The argument supports the idea that affirmative action programs should be based on class rather than race. A class-based system would channel benefits to the neediest people of color, and reduce the disproportionate benefits received by upper- and middle-class minorities under the present race-based approach.


Multiculturalism is considered the natural order of society today, especially with the racial and ethnic diversity brought on by immigration and globalization. But many economists see a downside to diversity. Cultural differences are often a source of social conflict and often act as a barrier to economic progress as well as personal freedom.

Harvard political scientist Robert Putnam (author of Bowling Alone) finds that the more ethnically diverse a community is, the lower the level of trust in the community. Putnam is not anti-diversity and notes that immigration is beneficial for countries that send and receive immigrants. But his research finds that social connectedness is less likely to be found in areas of the United States most affected by recent waves of immigration. And it isn’t just a lack of trust between different races or cultures, but within them as well.13

Prof. Putnam found trust was lowest in Los Angeles, “the most diverse human habitation in human history,” but his findings also held for rural South Dakota, where “diversity means inviting Swedes to a Norwegians’ picnic.”14

When the data were adjusted for class, income, and other factors, they showed that the more people of different races lived in the same community, the greater the loss of trust. “They don’t trust the local mayor, they don’t trust the local paper, they don’t trust other people, and they don’t trust institutions,” said Prof. Putnam. “The only thing there’s more of is protest marches and TV-watching.”

Putnam is hardly the first learned man to weigh in on the culture-trust relationship. An overwhelming anti-diversity consensus, backed by data quantifying the negative impact of diversity on economic development, is found among serious scholars:

There are scholars who have assessed empirically the influence of cultural diversity on economic development. The primary argument—which can be traced to Aristotle—suggests that diverse states are more susceptible to development-inhibiting internal strife than their homogeneous counterparts are…. Following Tocqueville (1873), Deutsch (1953), and Banks and Textor (1963), Adelman and Morris (1967) gather the data for 74 less developed countries from 1957 to 1962 and rank each country on a 10-point ordinal scale of diversity. Their results, based on factor analysis, support their hypothesis: homogeneous countries typically had higher growth rates. Haug (1967) finds a negative correlation between per capita GNP and cultural diversity based on the data of 114 countries in 1963. Reynolds (1985) compares 37 less developed countries from 1950 to 1980 and, again, indicates that cultural diversity results in lower growth rates. He suggests that this may be due to a sense of alienation among peoples. In other words, reaching a consensus on policies favorable to economic development, especially for the long run, may be difficult when groups have different interpretations of the past and different goals for the future.15

Culture includes learned patterns of behavior, socially acquired traditions, ways of thinking and acting, attitudes, values, and morals. Culture standardizes relationships by allowing people to make reasonably confident assumptions about the reactions of those with whom they interact. There are many dimensions of culture, but race, religion, ethnicity, and language are the principal sources of diversity.

When societies are multicultural, the ethnocentric differences of race, religion, ethnicity, and language often lead to enmity. Even if different groups live together peacefully, the lack of a common language and common norms reduces cooperation and increases the cost of economic transactions.

Economist Gerald W. Scully summarizes the benefits of mono- as opposed to multi-cultural societies in a 1995 paper:

Cultural relativism has made the study of the role of culture in human progress controversial. But there is little disagreement that intergroup enmity is widespread in culturally heterogeneous societies.16

Free markets, private property, rule of law, and eventually representative democracy and universal suffrage arose in culturally homogeneous Western societies where all members of society had equal rights to compete in the marketplace. On the other hand, culturally heterogeneous societies are less likely to adopt the institutions of liberty. Since control of economic resources is essential to political control, dominant cultural groups structure economic institutions to serve their self-interest. And when private property and economic rights are allocated along cultural lines, economic inefficiency is inevitable and societies are less prosperous.17

Language may be the most important component of culture—at least for economic development. A common culture and a common language allow individuals to interact (and trade) more easily: a contract need not be translated if two individuals speak the same language. By contrast, linguistic diversity may lead to cultural shocks and conflicts. In this case, diversity imposes costs to the economy.

Some economists have explicitly factored diversity into economic models. For example, E.P. Lazear assumes that a common culture and a common language facilitate trade between individuals and shows that minorities have incentives to become assimilated and to learn the majority language so that they have a larger pool of potential trading partners.18 Multiculturalism is a bad. In this model, individuals do not properly internalize the social value of assimilation. They ignore the benefits that others receive when they learn the majority language and become assimilated. In the absence of strong offsetting effects, policies that encourage multiculturalism reduce the amount of trade and have adverse welfare consequences.

Europeans have taken these findings to heart. In France, for example, the state is widely viewed as a force for integrating different groups into the same society. Actively multicultural policies are considered dangerous because they divide society. Immigration from Less Developed Countries is restricted far more zealously in Europe than in the U.S. The French are so terrified of threats to their culture that they have restricted American film imports.19

Meanwhile, multiculturalism is celebrated by U.S. policy makers. Public school curricula celebrate black, or Hispanic, or other immigrant cultures—often at the expense of a shared American experience. English as a second language has replaced English immersion. The pluralism ethos treats all cultural values as being equally valid. It should come as no surprise that America has become a Tower of Babel. The economic consequences are profound.


1. Peter Brimelow and Leslie Spencer, “When quotas replace merit, everybody suffers,” Forbes, February 15, 1993. http://www.vdare.com/pb/when_quotas.htm


3. http://www1.eeoc.gov/eeoc/statistics/enforcement/charges-a.cfm

4. Tama Starr, “A small contractor learns affirmative action arithmetic,” Reason, February 1996.

5. Katherine Post, Director of the Center for Enterprise and Opportunity, “White House Makeover for Affirmative Action,” June 30, 1998.

6. http://www.thenewamerican.com/usnews/politics/item/13123-gao-35-of-federal-contract-dollars-went-to-minority-owned-businesses

7. Brimelow and Spencer, op. cit.

8. Ralph R. Reiland, “Affirmative Action or Equal Opportunity?” Regulation, Vol. 18, no. 3, 1995.

9. Reiland, ibid.

10. Paul Oyer and Scott Schaefer, “The Unintended Consequences of the ’91 Civil Rights Act,” Regulation, Summer 2003. http://www.cato.org/pubs/regulation/v26n2-7.pdf

11. Oyer and Schaefer, op. cit.

12. Oyer and Schaefer, op. cit.

13 . John Lloyd, “Harvard study paints bleak picture of ethnic diversity,” Financial Times, October 9, 2006.


14. ibid.

15. Rongxing Guo and Eui-Gak Hwang, “A Theory of Economic Growth in Culturally Diverse Nations,” UNU-WIDER Jubilee Conference, June 2005. http://www.researchgate.net/publication/242274531_A_Theory_of_Economic_Growth_in_Culturally_Diverse_Nations

16. Gerald W. Scully, “Multiculturalism and Economic Growth,” National Center for Policy Analysis,” August 1995. http://www.ncpa.org/studies/s196/s196.html

17. Scully, ibid.

18. E.P. Lazear, “Culture and Language,” National Bureau of Economic Research, Working Paper 5249, 1995. http://www.feem.it/NR/rdonlyres/C5ACFACA-2117-4700-8C94-3ABFA6F9A3D1/690/1203.pdf

19. Scully, op. cit.

About the author

Edwin S. Rubenstein, a regular contributor to The Social Contract, is president of ESR Research, economic consultants. As a journalist, Mr. Rubenstein was a contributing editor at Forbes and economics editor at National Review, where his “Right Data”column was featured for more than a decade. He is the author of The Earned Income Tax Credit and Illegal Immigration: A Study in Fraud, Abuse, and Liberal Activism.