As a general and somewhat loose statement, economic discrimination can be said to occur against members of a group whenever their earnings fall short of the amount “warranted” by their abilities. Public interest has usually been centered on groups that are either numerical minorities (such as Negroes in the United States, Untouchables in India, Jews in the Soviet Union, Laplanders in Sweden, and Chinese in Indonesia) or are political, social, or economic minorities (such as the Blacks in South Africa, immigrants from Africa in Israel, and women in most countries). Until recent years, economic discrimination was generally seen as evidence of exploitation of minorities by majorities. This interpretation is clearest in writings with Marxist sympathies (Aptheker 1946), although it is also found in weaker form in much of the literature (Allport 1954, p. 210).
In the past, economists tended to neglect the study of discrimination against minorities, primarily because they were reluctant to interpret any appreciable economic phenomena in terms of “exploitation,” which stems from what is technically called monopsony power [seeMonopoly]. The growing interest of economists in discrimination during the last decade has been stimulated by an approach that is not based as much on exploitation as on considerations that may involve even a sharper break with traditional economic theory. In this approach, members of a group earn less than their abilities warrant if other persons are willing to “pay” (that is, give up resources) in order to avoid employing, working with, lending to, training, or educating these members.
Discrimination in the market place. Further discussion is facilitated by distinguishing discrimination affecting the talents that are brought to the market place from discrimination in the market place itself. Since the latter is taken to mean that less is earned than is warranted by productivity, such discrimination would not exist if all persons maximized money incomes and if all markets were competitive. For, under these circumstances, the increased money incomes of employers, employees, or consumers, resulting from associating with members of a group receiving less than their productivity, would stimulate a demand for such members that would continue until their earnings rose sufficiently to cover their productivity.
Accordingly, discrimination in competitive markets is said to occur because some participants have tastes for discrimination, more loosely called “prejudice.” Because of these tastes, they are willing to forfeit money income or other resources in order to avoid employing, working with, or buying from members of a particular group. Indeed, the intensity of their prejudice is measured by how much they are willing to forfeit (Becker 1957, chapter 1). The measuring rod of money provides an operational and quantifiable concept of prejudice that is consistent with the economist’s concept of tastes and the statistician’s concept of subjective probabilities (Savage 1954). Objections to such a behavioristic concept of prejudice and hatred might be tempered upon reflecting that a hatred cannot be very strong, in any meaningful sense, if there is a reluctance to satisfy it by parting with some resources. This approach to discrimination not only incorporates common notions about prejudice much more fully than does one based on exploitation, but also is more consistent with the general prevalence of workably competitive markets.
Although the extent of market discrimination depends, of course, on the intensity of the average taste for discrimination, it by no means depends on this alone. Also very relevant are any differences among participants in the desire to discriminate, the relative number and occupational distribution of minority members, the extent of competition in product markets, and the degree of substitution between different groups (Becker 1957). For example, consider discrimination by employers against a group working in a competitive industry. Even if the employers’ average taste for discrimination were large, market discrimination could be negligible if the group were a small fraction of the total labor force in the industry and if some employers’ tastes for discrimination were weak (or even negative). For then, members of the group could be fully employed by these employers. An increase in the size of the group or a decline in the fraction of employers with weak tastes for discrimination would tend to increase market discrimination, perhaps substantially, because members could no longer be fully employed by these employers, and some members would have to find employment among those employers with stronger tastes for discrimination. However, the latter would employ these members only if the cost of discriminating were sufficiently large—that is, the earnings of members sufficiently low—to offset their tastes.
Even the market form in which the discrimination manifests itself can be very important. If the result were simply lower hourly earnings, discriminators would have to balance the gain from discriminating—satisfying their prejudice—against the cost of discriminating—their loss in income from not associating with persons receiving less than their productivity. If, however, because of market imperfections, trade union behavior, or minimum wage or “equal pay” legislation (Alchian & Kessel 1962), hourly earnings of minorities were prevented from falling, the cost of discrimination would be eliminated because no resources would be forfeited by not hiring minorities; and discriminators would be encouraged to discriminate still more. The result would be greater unemployment of minorities, and discrimination would take the market form of, say, lower annual earnings rather than lower hourly earnings.
Quantitative evidence on the extent of market discrimination against different minorities is surprisingly limited; the most extensive evidence relates to Negroes in the United States. Significant market discrimination against them is suggested by the much lower earnings of urban male Negroes than of urban male whites of the same age, years of schooling, and region, and the somewhat greater unemployment of Negro males, even when occupation is held constant. The absolute occupational position of Negroes has risen substantially in both the North and South during the last hundred years. Since, however, the position of whites has risen at about the same rate, the relative position of Negroes has changed surprisingly little, the greatest change (a rise) coming after 1940. One of the most striking findings is that market discrimination is apparently relatively slight against young new entrants into the labor force, but it increases significantly with age. This finding is presumably explained by a combination of inadequate preparation of Negroes for occupational advance and greater market prejudice against more advanced Negroes. Similar findings would probably apply to other minority groups in the world. (For evidence on the economic position of Negroes, see Becker 1957, chapters 7–10; Dewey 1952; Oilman 1963; and Ginzberg et al. 1956.)
Deficiencies brought to the market place. Perhaps more than half the total difference between the earnings and abilities of Negroes in the United States results from deficiencies brought to the market place, rather than from market discrimination itself; again, a similar relation would probably be found for other minorities. These deficiencies include poor health, low morale and motivation, and limited information about opportunities. The most pervasive force, however, is insufficient and inferior education and training. For example, in 1960, Negro males aged 25–64 averaged less than 8 years of schooling, while whites averaged more than 10.5 years; and Negro schooling has clearly been inferior in quality.
An important cause of deficiencies in education and training are the limitations imposed by poverty itself, especially when combined with the usual difficulties in financing large expenditures. Of course, poverty may in part result from either past discrimination, as exemplified by Negro slavery, or current discrimination, thus leading to the so-called vicious circle of discrimination (Myrdal 1944). Discrimination by governments and private institutions in charge of education and training facilities has been common, as illustrated by the government restrictions on the education of Blacks in South Africa and (at least in the past) Jews in eastern Europe, the quotas applied by certain private universities in the United States, or the racial (and even religious) restrictions on admission to apprenticeship and other training programs administered by trade unions in many countries.
Role of government. Undoubtedly, some of the worst economic discrimination, as well as other kinds, is directly traceable to government action. One need only note the restrictions placed on the economic advance of Blacks in South Africa, the harassment of the Chinese in Indonesia and other parts of Asia, or the virtual confiscation of some of the property of Japanese Americans in the United States during World War ii. On the other hand, government action is often a force in widening opportunities and thus breaking down economic discrimination: of greatest influence is the enforcement of equality before the law, although the provision of free or subsidized education has also been important. Legislation of the “fair employment” kind, prohibiting racial, ethnic, religious, and other forms of discrimination in employment, can have some effect too. Very little scholarly analysis has been made of the economic effects resulting from this kind of legislation. In the United States, the evidence provided by the states with such legislation suggests that it somewhat increases both the earnings and unemployment of minorities (Landes 1966).
A few of the more extreme nineteenth-century advocates of a competitive market economy believed that eventually its extension and development would eliminate most economic discrimination and hatred. Unfortunately, this has not yet taken place; discrimination exists, and at times even flourishes, in competitive economies, the position of Negroes in the United States being a clear example. At the same time, one must realize that the pressures of competition, the emphasis on material goods, the impersonality of and the cost accounting in a market economy often help mitigate, and sometimes even eliminate, the effects of prejudice. Indeed, in desperation, market participants in the United States, South Africa, Great Britain, Indonesia, and elsewhere have (successfully) appealed to their governments for protection from the “unfair” competition of various minorities.
In spite of the widespread public interest in economic discrimination throughout the world, relatively little is known about its quantitative importance and socioeconomic determinants. Much greater understanding of the formation of prejudice is necessary, although economists are unlikely to be of great value here. Their comparative advantage lies in analyzing how prejudices combine with various institutional arrangements to produce actual discrimination. It has already been mentioned that the degree of competition, the size of minority groups, and many other factors as yet insufficiently studied also affect the amount of market discrimination.
Likewise, the effect of prejudice, by the electorate, trade union members, or directors of educational institutions, on the extent of nonmarket discrimination is determined by whether different political issues are effectively tied together, whether the costs of discriminating or the gains from not discriminating are hidden, and on many other institutional arrangements (Becker 1957, chapter 5; 1959). Consider, for example, the election by majority rule of representatives to decide on two issues, one dealing with discrimination against a minority. Assume that the minority is more concerned about discrimination while the majority, although more concerned about the other issue, does not agree on how to resolve it. A successful campaign could be conducted by promising to satisfy the minority on discrimination and a segment of the majority on the other issue. The decision on discrimination could, therefore, be more favorable to the minority than if each issue were decided separately (by, say, referendum), or if the majority were less split on the other issue, or if the majority attached greater importance to discrimination. Considerably more study of institutional arrangements is required in order to know more about the factors determining the talents that minorities are permitted to bring to the market place.
Gary S. Becker
Alchian, Armen A.; and Kessel, Reuben A. 1962 Competition, Monopoly, and the Pursuit of Pecuniary Gain. Pages 156–183 in Universities-National Bureau Committee for Economic Research Conference, Princeton, N.J., 1960, Aspects of Labor Economics. Princeton Univ. Press. → Includes a discussion by Gary S. Becker and Martin Bronfenbrenner.
Allport, Gordon W. 1954 The Nature of Prejudice. Reading, Mass.: Addison-Wesley. → An abridged paperback edition was published in 1958 by Doubleday.
Becker, Gary S. 1957 The Economics of Discrimination. Univ. of Chicago Press.
Becker, Gary S. 1959 Union Restrictions on Entry. Pages 209–224 in Philip D. Bradley (editor), The Public Stake in Union Power. Charlottesville: Univ. of Virginia Press.
Dewey, Donald J. 1952 Negro Employment in Southern Industry. Journal of Political Economy 60:279–293.
Gilman, Harry J. 1963 The White/Non-white Unemployment Differential. Pages 75–113 in Mark Perlman (editor), Human Resources in the Urban Economy. Washington: Resources for the Future.
Ginzberg, Eli et al. 1956 The Negro Potential. New York: Columbia Univ. Press. → A paperback edition was published in 1963.
Landes, William 1966 An Economic Analysis of Fair Employment Practice Laws. Ph.D. dissertation, Columbia Univ.
Myrdal, Gunnar (1944) 1962 An American Dilemma: The Negro Problem and Modern Democracy. New York: Harper. → A paperback edition was published in 1964 by McGraw-Hill.
Savage, Leonard J. 1954 The Foundations of Statistics. New York: Wiley.