Economic theory and racial economic inequality

ECONOMIC THEORY AND RACIAL ECONOMIC INEQUALITY

William Darity

Black people have always been troublesome for social scientists studying the United States. Theories that gleam brilliantly in academia's expansive sunlight when focused on the behavior of whites rapidly pale when their explanatory and predictive powers are directed toward the nation's largest racial minority. Political scientists enamored with their particular orthodoxy, pluralist theory, have had to squirm and wiggle in an attempt to preserve their beloved paradigm when faced with an "interest group" that did not comply with the dictates of their model of American political behavior. Generalizations that historians make about developments over time in the United States must include the addendum "except for the Blacks" to retain their accuracy. Sociologists have so consistently found Blacks to be the exception to their theoretical formulations that they now devote more time to the study of "the Blacks" than any other social scientists--so much so that they now have developed numerous theories that apply exclusively to Blacks. Economists also have spent much of their time researching "the black problem." They have looked primarily at the bread and butter issue of why proportionately more Blacks are poor than whites--the issue of blackwhite economic inequality. In a sense, they have been more "fortunate" than other social scientists insofar as black poverty potentially can be explained in the context of the prevailing economic paradigm, neoclassical microeconomic theory, via the concept of human capital. However, as neoclassical theory increasingly comes under fire, a growing school of researchers are contending that orthodox theory is not adequate to explain racial economic inequality. Efforts have begun to explain black poverty in decidedly more unorthodox terms--terms which call into question the

226

The Review of Black Political Economy

continued validity of neoclassical microeconomics altogether. This article will examine and evaluate economists' efforts to explain black poverty. Three major approaches will be analyzed by discussing the major pieces of economic literature that attack the question of why a greater percentage of black people are poorer than whites. These three approaches are applications of (1) conventional trade models, (2) human capital theory, and (3) dual market theory to the phenomenon of racial economic inequality. It is pertinent to add that one behavioral assumption made by the neoclassicists will be retained as valid throughout the discussion--the assumption that individuals in the United States are materialistic, that they will more often than not act in their own self-interest. C O N V E N T I O N A L T R A D E M O D E L S AND B L A C K P O V E R T Y One of the earliest and most ingenious approaches to explain racial economic inequality employed a conventional two-nation trade model with a trade barrier--the taste of the white "nation" for discrimination. When applied to the United States this approach treats the black and white populations as having two separate economies. Given the white taste for discrimination, their greater numerical size, and their advantage in physical capital ownership over Blacks, one could argue that a colonial relationship develops with whites holding a dominant economic position. The Becker Model The seminal work using this type of model was undertaken by Gary Becker 1 in the late 1950' s, preceding the currently popular description of the black community in the United States as an internal colony (popular in radical circles) by more than a decade. Becker proposed that we look at two economies, " W " and " N " in trade relationship. The critical assumption Becker makes that distinguishes his model from later conventional neoclassical explanations of black poverty is t h a t " members of W are perfect substitutes in production for members of N . " Moreover, the two economies share a common production function. Both economies are viewed as being perfectly competitive. Given that capital is more abundant in W and labor more abundant in N, it would be to the mutual benefit of both societies to engage in t r a d e - - W sending capital to N and N sending labor to W--until an optimum allocation of resources is arrived at, an allocation that maximizes the overall income of both societies. Because members of W and N are perfect

RACIAL ECONOMIC INEQUALITY

227

substitutes the equilibrium wage rate of the two societies would be equal in a world where trade could occur freely. In the Becker world a trade barrier prevents such an equilibrium from being attained. The members of W have what Becket calls " a taste for discrimination" toward N. Treating discrimination as an analog to ordinary commodities, Becker argues: " I f an individual has a 'taste for discrimination,' he must act as if he were willing to pay something either directly or in the form of reduced income, to be associated with some persons instead of others." In the Becker world white employers would prefer to hire white employees; white workers would prefer to work with white workers; and white consumers would prefer to purchase white-made products. But such preferences exist only if the wage rates of black and white workers are the same and if the prices of goods produced by Blacks and whites are the same. There is, Becket asserts, a sufficiently low wage and a sufficiently low price where the white employer will take the black employee and the white consumer will choose to buy the black-made product. Presumably, the white employee would be willing to work with Blacks at a sufficiently high wage. Becker therefore differentiates between the m o n e y c o s t s and the net c o s t s that members of W perceive. If a W pays a money wage of q to an N, then W acts as if his net cost is q(1 +a). If a W worker gets a money wage of r when working alongside an N, then W acts as if his net hourly pay is r(1 -b). And, finally, if W buys a commodity produced by N for a price p, then W acts as if the net cost is p(1 +c). The parameters a, b, c are discrimination coefficients (DC). If they are all greater than zero, the DC is associated with "disutility" and becomes a measure of t h e " p s y c h i c costs" of association with the undesirable factor. The DC is not treated as being uniform from person to person in W; its magnitude varies depending upon the tastes of each individual. However, the taste for discrimination occurs with great enough frequency and magnitude among members of W for wage differentials to arise between members of W and members of N. It is "costlier" to hire members of N because of the psychic costs involved in deciding to employ them for members of W. Since N labor is more abundant than N capital in N' s economy while W labor is relatively more scarce in W ' s economy, N labor receives a lower wage than W labor, due to the trade barrier created by W ' s taste for discrimination. Economic inequality arises in the Becker world because of the trade barrier created by W' s taste for discrimination. A further important step in the Becker argument is his conclusion that the

228


white taste for discrimination results in lower aggregate incomes for both W and N, that discrimination is not "optimal" for either society. Less labor and less capital are exported by each society respectively than would be the case in the absence of the trade barrier, thus lowering the equilibrium net incomes of both N and W. Becket's final critical point is that although aggregate net incomes in both societies decline due to discrimination, " . . . all factors are not affected in the same way: the return to W capital and N labor decreases but the return to W labor and N capital actually increases." But the cumulative effect in the Becker model is that the preferences of whites to avoid associations with Blacks lead to an outcome in competitive markets that results in an average income that is lower for Blacks than for whites. Plainly the Becker approach is an imaginative application of neoclassical theory to the problem of racial economic inequality, demonstrating how tastes determined exogenously from the market can lead to black poverty. In the Becker world prejudice is the sore point. There is, however, a fundamental weakness in this model that stems from its static nature. Becker assumes at the outset that N and W economies are competitive. If we distinguish between short-run and long-run competitive equilibria the flaw becomes clear. In the short run perfect competition and the existence of wage differentials among individuals with equivalent marginal productivities are compatible events; in the long run they are not. One of the basic tenets of neoclassical theory is the belief that the presence of competition will eventually lead society to an optimal allocation of resources. Under such circumstances equally productive workers will receive the same wage. Consider a dynamic application of the Becker static model. Assume that in the short run Becket's predicted black-white wage differentials do develop. W receives a hypothetical wage S and N receives a hypothetical wage T. T is lower than S. Next assume, as Becket does, that all firms in W economy are perfect competitors, producing an output at the point where marginal cost is equal to marginal revenue. For simplicity we will assume that all the firms in W produce the same commodity. We also assume that they all hire only W labor at the higher wage S. Because they are all perfect competitors they all earn only " n o r m a l " or " e c o n o m i c " profits, just covering their costs. There are no entry barriers. Suppose an entrepreneur enters the picture whose discrimination coefficient yields a money value that is less than the existing wage differential.


229

He is willing to hire black laborers at the lower wage T. He discovers that he faces lower marginal and average cost curves than all the other firms in W. As a result he can sell his output at a lower price than all the other firms in W. This leaves the other firms with the simple choice to either hire Blacks at the lower wage in order to reduce their own cost curves or to go out of business. Assuming they choose the former course of action the demand for black labor will rise and the demand for white labor will fall until all workers will be paid the same wage, a wage equivalent to their marginal productivity. The existence of competition has thus caused wage differentials to vanish. Racial economic inequality vanishes as well. 2 In short, because Becker assumes that perfect competition exists in both sectors of his model he fails to explain why economic inequality persists between Blacks and whites, as long as the realistic assumption is made that all whites do not share the same hostility toward associations with Blacks (particularly if such associations are profitable). The only potential redemption for Becker's model is to argue that we are still in the short run, that the market has not yet had the chance to "correct" the consequences of white attitudes toward Blacks. But this leads to a rather eerie conception of the short run. After all, there has been close to a century during which entrepreneurs could have exploited the profit-making opportunities afforded by the lower wages they could offer Blacks, and ultimately close the wage (and income) gap in a presumably competitive world. Today wage differentials do exist between Blacks and whites. The average income of Blacks is still little more than 60 percent of the average income of whites. 3 But in long-run equilibrium a perfectly competitive marketplace should not demonstrate such a condition if Blacks and whites are perfect substitutes. A related subsidiary criticism of Becker's model concerns the severing of the neoclassical umbilical cord connecting utility maximization with money income maximization. In the Becker world whites are willing to forego money income for the added satisfaction of staying away from Blacks, but this is not consistent with neoclassical ontology. "Rational" man prefers money above all else--and, as has been argued above, "rational" members of W can associate with Blacks and end wage differentials. At base the observed existence of economic inequality in the United States cannot be explained by Becker's model if the American economy is perfectly competitive and the average productivities of Blacks and whites

230


are the same, two assumptions that Becker makes at the outset of his work. The Krueger Model Anne Krueger's extension of the Becket model is equally ingenious. Although she still uses a conventional trade model, her approach leads to a markedly different explanation of black poverty. She begins by borrowing most of Becker's basic descriptive assumptions--two societies in trade relationship (one white and one black, the latter a numerical and economic minority), each sector perfectly competitive and sharing identical production functions, and black and white factors of production being perfect substitutes. The crucial difference between her model and the Becker model is that she seeks to restore the link between money income maximization and utility maximization for whites. The practice of discrimination is no longer purely a matter of "tastes"; instead it becomes economically rational behavior for whites. The difference between her approach and Becker's becomes clear when the mathematical formulas for white income in each model are compared. For Becket, according to Krueger, the "net income" of whites is equal to the marginal product of capital in the white sector times all white-owned capital (exported and domestically used) plus the white wage bill. The relative scarcity of capital in N causes capital to command a higher price in that sector, due to the white taste for discrimination which precludes exports up to the level where the marginal productivity of capital in both sectors is equal. As a result capital exported to the black sector yields a higher rate of return. But because Becket makes a distinction between money incomes and net incomes (the latter including a discount factor that associates a price with the white taste for discrimination) the white return on exported capital is internalized within the production function. Krueger, on the other hand, makes no such distinction, which enables her to sift out the white return to exported capital and arrive at a slightly different equation for the white sector's net income. It is equal to output in the white sector plus the returns to exported capital. To maximize white income Krueger determines algebraically that "the marginal product of capital in the white sector should be lower than the marginal product of capital in theNegro sector . . . . " The higher price that whites can command for the capital they export by choosing a discriminatory level of exports rather than a mutually optimal level can be viewed as a " t a x " on the black sector. Unlike the Becker model whites gain while Blacks lose, instead of both


231

groups losing as a result of discrimination. Once again, however, the distributional effects of the white gains are not uniform. While white laborers gain from discrimination, white capitalists lose. Economic inequality arises here because whites tax Blacks for the use of capital that they export to the black sector, thereby reducing average black income. It is the lack of uniformity of the distributional effects of the gains reaped by whites through discrimination that makes this explanation of blackwhite income inequality doubtful9 Krueger comments, "White capital owners would not discriminate if they were maximizing their own incomes and had no taste for discrimination 9 In a world where free trade exists the economies of two separate societies can be viewed as functioning as one. Only the introduction of barriers to trade (such as the phenomenon of discrimination) segment the economic environment in a perfectly competitive world. Krueger accepts the Becker assumption that the black and white economies are internally competitive 9 Consequently, within the white sector, white capitalists behave atomistically in their own individual self-interest. For discrimination to occur that maximizes the income of the white population as a whole, white capitalists must act in a collusive manner vis-h-vis the black sector. By doing so they actually suffer a loss. They would maximize their own income if free trade existed. Aware of the paradox in her analysis, Krueger offers the following potential explanation: 9 white capitalists may aim at maximizing the income of the whole white community rather than white capitalist income only. A welfare function of this kind would be quite similar to Becket's, except that discrimination would be directed at maximizing white real income rather than avoiding the distastefulness of working with Negro factors of production. 5 This implies that there is an awareness (or "consciousness' ') on the part of whites that their discriminatory allocation of capital to Blacks will maximize their community's income, and there is a willingness on the part of white capitalists to lose income for the economic gain of the white population as a whole. Krueger herself suggests that conscious efforts to maximize white income relative to black income might be evident "in the allocation of publicly owned capital," but the allocation of publicly owned capital is politically determined, an event external to the marketplace. Therefore, this type of conscious behavior is not the same as discrimination by whites

232


in the marketplace to exploit their capital advantage. This type of conscious behavior ascribes the causes of black poverty to white political control over public resources rather than the collusive actions of white capitalists who stand to lose money by their behavior. The fundamental weakness of the Kmeger model--like the fundamental weakness in the Becket model--is that the long-run equilibrium of a perfectly competitive marketplace does not lead to income differences between individuals who are perfect substitutes as factors of production. In the Becker model we would expect perfect competition to eventually erode wage differentials between Blacks and whites. In the Krueger model the collusive behavior of white capitalists vis-a-vis the black sector is inconsistent with their independent, income-maximizing behavior within the white sector. 6 As long as perfect competition is assumed and Blacks and whites are viewed as perfect substitutes, wage differentials should not persist when the economy is in long-run equilibrium. Competition should lead to the collapse of efforts to collude, particularly if a collusive pact actually leads to the loss of income for a given group of economic actors. To arrive at a sound explanation of black poverty that preserves the behavioral assumption that economic man is materialistic, economic theorists would either have to dispose of the assumption that the American economy is characterized by competition or they would have to dispose of the assumption that Blacks and whites are perfect substitutes as factors of production. The former step was too threatening to the orthodoxy to be pursued immediately. The second step dovetailed neatly with the neoclassical paradigm. It is the step taken by human capital theorists. If any explanation can be considered the conventional wisdom among members of the economics profession, human capital theory is the strongest candidate. H U M A N C A P I T A L T H E O R Y AND B L A C K P O V E R T Y The fundamental shortcoming in the use of conventional trade models to understand why proportionately more black people are poor is their simultaneous assumption that the two societies engaged in trade have perfectly competitive internal economies and that the members of both societies are equally productive workers. Neoclassical theory does not direct us toward a long run equilibrium that sustains a trade barrier yielding economic inequality between the two societies. Market imperfections that block optimal allocations of resources are expected to fade in the face of continued competition in the orthodox view. r A revision in the assumptions that Becker and Krueger make does result


233

in a theoretical approach that can explain the existence of black poverty in a competitive world--human capital theory. The necessary revision is the discarding of the assumption that Blacks and whites are perfect substitutes in production. Human capital theory falls squarely into the neoclassical paradigm. It proposes that low income is the result of low wages. Low wages are the result of workers holding low productivity jobs. Holding low productivity jobs is the result of workers having a lower ability or skill level (or, in economic jargon, " a lower level of human capital," which is the theoretical concept treated as distinct from physical capital, such as machinery, tools, factories, land, or savings accounts). Individual workers can raise their personal level of human capital by investing in increased education or training for themselves that would allow them to fill occupations with higher levels of productivity (and, commensurately, higher wages). Human capital theory preserves the neoclassical link between wages (the price of labor) and the marginal productivity of labor. In this sense, labor is paid what it is " worth " - - n o more, no less. 8 How does this apply to economic inequality between Blacks and whites in the United States? The argument a human capital theorist would make is that Blacks have not accumulated a level of occupational abilities and skill comparable to the level attained by whites. Blacks, in this context, are viewed as having a lower average level of productivity. Race correlates with productivity. Firms are aware of this and their hiring practices reflect this awareness. They expect the white worker to be more productive because whites as a group are better educated. Race then becomes a cheap screening device when a white and a Black apply for the same job. Because human capital theorists would argue that the expectation is correct, race becomes an efficient surrogate variable for allocating jobs in a competitive market. It becomes an accurate means of predicting job performance. After Doeringer and Piore, this is the phenomenon of "statistical discrimination." 9 This perspective is epitomized in the work of Thomas Sowell. 1~ He defines human capital as "the vital accumulation of knowledge, skill and organizational experience," and then proceeds to argue that black people have less of it. For him, this explains why black people are poor: The poverty of the domestic poor generally, or of black people specifically, is not only a poverty in terms of the physical things they own or don't own; their share of the human capital of the country is even more desperately small. 11 Part of these differences is due to educational differences between

234


Blacks and whites, according to Sowell. He presents four sources of differences in the amount of formal education Blacks and whites receive--(1) lower expenditures per student in poor neighborhood schools leading to the hiring of "a large proportion of temporary, substitute, and inexperienced teachers and larger classes"; (2) disproportionate amounts of public funds directed toward middle-class educational needs rather than lower-class educational needs; (3) cost barriers that the poor face in attending four-year colleges and universities; and (4) disproportionate numbers of poor students who must, therefore, attend junior colleges where they receive a lower quality preparation. But Sowell does not limit his analysis of the sources of black-white differences in human capital to differences in formal education--both in quantity and quality--between the two groups. He goes on to observe: Trying to understand these differences in human capital accumulation in terms of years of formal schooling between black and white is barely to scratch the surface of the problem. Human capital does not consist solely or even primarily in the kinds of knowledge acquired in classrooms and certified by embossed pieces of paper. 12 Sowell then makes a culturological argument, obviously borrowing heavily from conventional sociological and anthropological theories about the black community, asserting that the cultural experiences of black people block the development of the kinds of behavioral characteristics and attitudes that are conducive to high productivity performances on the job: The pattern of cumulative inequalities in human capital investment in formal schooling is repeated in the other forms of human capital. The whole way of thinking and behaving appropriate to the more lucrative and responsible occupations is something which comes freely, and even unconsciously, to people reared in families where such occupations have been common for generations, whereas human capital comes to the low-income person only slowly, imperfectly and with great deliberate efforts to break his natural patterns. Such basic traits as punctuality, efficiency and long-run planning are of little use to people who have been limited to menial jobs for generations as with most black Americans. Everyone can understand the economic value of such traits as an abstract intellectual proposition but to understand such qualities abstractly and to have such habits in reality are very different things. Those black people who have such traits have

RACIAL E C O N O M I C I N E Q U A L I T Y

235

typically acquired them through persistent, and sometimes painful, adjustments, which would be difficult to explain to people who grew up with those patterns as a free cultural inheritance. 13 Finally, Sowell argues that a third cause of the lower average level of black human capital accumulation is less access to appropriate contacts: 9 good deal of human capital consists of economically valuable knowledge of particular situations and individuals . . . . Knowledge of specifically who, where, and how to get things done is very unequally distributed and constitutes a tremendous handicap to the poor in a highly complex society. 14

This is a coherent explanation of racial economic inequality, attributing racial economic inequality to causes external to the competitive market--(I) white political control over publicly owned resources (as Krueger suggested) leading to a lower level of investment in the education of Blacks; and (2) white racism and history, isolating Blacks from the kinds of cultural and social experiences that would equalize human capital formation between the races. 15 But when put to empirical test human capital theory virtually explodes as an explanation of racial economic inequality. Indeed, a debate is raging among labor economists over the usefulness of human capital theory as an explanatory tool for wage and income differences among all workers in the American economy. It is a debate which, if the critics of human capital theory prevail, threatens to topple the descriptive power of all of orthodox economics. We need not concern ourselves with that debate here. Suffice it to say that human capital theory does not succeed in explaining why proportionately more black people are poor than whites. The most recent research that presumably lends support to human capital theory as an explanation of black poverty is the work of Richard Freeman. 16 Freeman believes that there has been an observable decline or " 'collapse' of market discrimination." His data focus on the improvements in the economic position of highly-trained and college educated Blacks in recent years relative to the position of their white counterparts. His most provocative observations include the following: (1) economic differences between black women and white women are now negligible (by 1970 he finds that the ratio of incomes had risen to 86 percent compared with 50 percent in 1950); (2) equality of starting salaries has been attained for men with college degrees; and (3) from 1960 to 1970 the proportion of

236


black college men working as managers almost doubled, rising from 7 percent to 13 percent while the white proportion changed only slightly from 19 percent to 22 percent over the same period. In summary Freeman concludes: All told, the d a t a . . , show a remarkable decline in black/white economic differences among women and high level male workers and a smaller decline in over-all male differences. 17 Freeman attributes the income differences to educational attainment treating educational attainment as the indicator of the level of "human capital" the individual has acquired. For Freeman human capital theory is the fundamental explanation for the persistence of black-white economic inequality. But Freeman's work is at best deceptive. Initially, there is the potential for the classic confusion to arise between correlation and causation. Although Freeman might have found a high degree of correlation between differences in educational attainment and incomes for Blacks and whites, differences in income may not be caused by differences in educational attainment. Whether as strong a correlation exists as Freeman suggests is open to doubt. The correlation that exists for an entire sample may not exist for part of the sample. The approaching economic parity for Blacks who have attained educational parity with whites is only occurring for a small percentage of the black population, the group that Freeman himself characterizes as "elite" Blacks. Black women with college degrees earn as much as white women with college degrees ($7200 annually). 10 And indeed, the most recent black male college graduates receive starting salaries equal with those of white males with the same amount of education. Contrast the earnings of the preceding two groups with the experience of other black college graduates. Black male college graduates 35 years and older earn $9300 annually--S300 less than the earnings of white male high school graduates ! And younger black male college graduates do earn more than white high school graduates, but their $8700 median income is much less than the $9200 median for white males who have dropped out of college. 1a These figures do reflect declines in the differentials over time which may support the Freeman view that labor market discrimination is in decline. But they are only declines in income differentials that are significant for "elite" (i.e. college educated) Blacks. Freeman's optimistic reading of his data masks the continuing gap in black-white family income that has remained fairly constant over the past


237

two decades (in the vicinity of 60 percent)9 In fact, a 50,000-family survey undertaken by the Bureau of the Census established that the ratio of black to white family income dropped from 64 percent in 1970 to 62 percent in 1972.2o Clearly the incomes received by Blacks with a high school diploma or less are the major determinants of average black family incomes; otherwise the ratio would more closely reflect the differentials that exist between college-educated Blacks and whites. If Freeman had focused on this portion of his sample his optimism would probably have been severely tempered. The basic question then becomes, can the income differential that exists between "non-elite" Blacks and whites be attributed to differences in educational attainment? Empirical research by Barbara Bergmann, Randall Weiss, and Bennett Harrison suggests that much less can be so attributed than Freeman assumes. Bergmann used 1960 census data to identify "occupations which have historically had low educational requirements. ''21 She identified these occupations by determining which ones had 50 percent or more employees with less than a high school diploma. After identifying the occupations with "low educational requirements" she proceeded to compute " . . . an expected number of nonwhite e m p l o y e e s . . , by assuming that nonwhites would share in employment in each occupation to the degree that they shared in the educational achievement shown by all persons in that occupation." She found: 9 the twenty-nine occupations identified eighteen had significant deficits. Of the eighteen deficit occupations, fifteen had deficits of nonwhites of more than 50 percent. The surpluses of nonwhites were heavily concentrated. Two occupations, service workers and nonfarm laborers, accounted for 82 percent of the surpluses. 22 Bergmann's research uncovers a job allocation process that she describes as occupational crowding, a process whereby Blacks are limited to certain fields by virtue of their blackness, regardless of their educational attainment. This raises the wages of whites in the deficit occupations relative to the wage rate that they would have been able to receive if Blacks could compete freely with them for jobs, and depresses the wages of Blacks in the surplus occupations because their wage rate is a function of an inflated supply curve. Bergmann's findings are damaging to the human capital explanation of economic inequality because (1) they indicate that the presumed slotmachine relationship between years of schooling and income is not the same for Blacks and whites; and (2) the fact that employers have not

238


undercut the crowding phenomenon by hiring Blacks into the deficit occupations at lower wages than they offer whites--eventually leading to a vanishing of the wage differential--casts doubt once again on the neoclassical assumption that the economy is competitive 9 But defenders of human capital theory can still argue that Bergmann's findings are inadequate for disposing of their paradigm's usefulness on this subject because her research did not account for the possibility of interracial differences in educational quality. Or, the argument could be made that one year of education for a white worker endows him with more human capital than one year of education for a black worker because the white worker is more likely to go to a better school. If equal years of schooling do not lead to an equivalent level of educational attainment for Blacks and whites, then Bergmann's findings do not necessarily demonstrate that Blacks are " c r o w d e d " into certain occupations for reasons other than their alleged lower productivity. A major attempt to control for interracial differences in educational quality was undertaken by Randall Weiss in which he compared the earnings of Blacks and whites, who had had similar schooling experiences, by using the Coleman criteria for evaluating school quality. For Blacks and whites with the same number of years of schooling in similar educational settings, Weiss found that whites earned significantly more. 2a A more recent study, also attempting to control for interracial differences in educational quality, was performed by Bennett Harrison. z4 Hardson compared the incomes of Blacks and whites who live in ghetto areas, assuming that black and white neighbors in urban ghettos all are likely to have received low quality schooling. He then compared the earnings of whites and Blacks who have had the same number of years of schooling. His findings have devastating implications for human capital theory: 9 weekly wage of white high school graduates in the pooled set of twelve poverty areas is nearly $25 higher than that of whites who never entered high school. For nonwhites, the difference is only $8.33. High school, therefore, has three times the marginal payoff for ghetto whites as for ghetto nonwhites. On the assumptions of a forty-year working life, a rectangular earnings distribution, and a 6 percent rate of time preference, the present value of the lifetime return to completion of high school is nearly $19,000 for whites but only $6,000 for nonwhites. Clearly, education has a very high opportunity cost for nonwhites living in the urban ghetto. There are any number of (largely illegal) activities out " o n the street" which are capable of returning at least $6,000 in a single year. 2~


239

Differences in education may explain why certain individuals earn more than others (in general, the more education one has, the more one earns), but it certainly does not explain differences in income across racial lines. But, returning to Sowell's formulation of human capital theory, it is only fair to recall that he says that education is only part of the human capital equation--and perhaps not the primary part. The other two factors he cites--sociocultural background and contacts--are more elusive. They have not come under the direct scrutiny of the empiricists and are more difficult to evaluate. Direct tests of their role have not been made probably for two major reasons: (1) Sowell's formulation is somewhat unique; typically, human capital theorists take a narrower view of the concept, limiting human capital solely to the level of education and training an individual has received26; and (2) concepts like "socio-cultural experiences" and "contacts" are harder to test; whereas years of schooling and diplomas held provide a rather direct means of measuring educational attainment, more indirect variables are needed to test for an individual's "socio-cultural experiences" or the number o f " human capital enhancing contacts" they might make. What evidence that can be assembled suggests that the culturologlcal explanation for differences in black/white human capital accumulation resulting in economic inequality is not very powerful. If it was a sound explanation then those Blacks who moved into occupations from which they had previously been excluded could be expected to demonstrate lower productivity characteristics than white workers in the same occupations. But during the mid-1960's, when the labor market was tight, " m a n y disadvantaged workers who had previously been unacceptable to employers because of their alleged lack of skills (widely attributed to the poor quality of their education) were at least temporarily given an opportunity to demonstrate their capabilities," and there was little evidence that their job performances were significantly different from those of their white predecessors. 27 And Peter Doeringer found in examining Boston antipoverty programs that as ghetto workers moved into higher wage occupations they may have demonstrated higher productivity characteristics than their white predecessors; he observed that "for wages higher than the 'prevailing' ghetto wage, disadvantaged workers are more likely to be stable employees than other workers,"~8 indicating that low productivity characteristics such as high rates of turnover are more a function of a worker's occupation (and wage) then his cultural background. The final component of Sowell's human capital vector--special access to economically valuable information--is his argument for the importance

240


of contacts. There is little reason to doubt that black people do have less access to economically valuable information. But one of the major analytical strengths of the application of human capital theory to the problem of black poverty is that observed black/white income differentials are consistent with the assumption of perfect competition (if Blacks, on the average, are less productive than whites). But unequal access to special knowledge is not consistent with a perfectly competitive world--a world where a free and open flow of information is essential. Information monopolies distributed along racial or any other lines means reduced competition. If the informational advantage that whites hold over Blacks is significant enough to cause income differentials, then the assumption of perfect competition must be discarded. Human capital theory itself would no longer be useful as a means of explaining racial wage (and income) differentials. After all, human capital theory posits that the profit maximizing firm must pursue a marginal hiring rule to remain in business, hiring the most productive worker at the going wage rate. Again, in the long run in a truly competitive world, firms would break down informational barriers that prevented them from hiring a particular group of workers who were equally as productive as another group. The analytical validity of human capital theory and perfect markets for labor are tightly interwoven. This leads us finally to the last major approach being pursued by economists to explain black poverty--dual market theory--in which the American marketplace is no longer assumed to be strictly competitive. DUAL MARKET T H E O R Y AND BLACK P O V E R T Y

We have found that black people are a problem for believers in orthodox economic theory after all. In trying to arrive at an " e c o n o m i c " explanation for the lower average income received by Blacks in the United States, theoretical approaches that preserve the neoclassical assumption of perfect competition and treat Blacks and whites as perfect substitutes do not lead to income differentials in long run equilibrium. When they retain the assump: tion of perfect competition and assume that Blacks and whites are imperfect substitutes, empirical research places such theoretical formulations on shaky ground. Nevertheless, many economists cling by nails of faith to the latter explanation despite mounting evidence that says, "it ain't necessarily so." Like the political scientist who still clutches pluralist theory despite what "the Blacks" do to it, many economists still cling to human capital theory and the neoclassical assumption that the economy is competitive despite


241

what "the Blacks" do to it as well. For to do otherwise is to admit that the orthodox economic world that economists have designed for themselves and teach their students may not be approximated in the real world. It is only recently that any economists have begun to look at institutionalized market imperfections as a means of explaining wage differentials between Blacks and whites. One of the earliest attempts to explain black poverty through the use of market imperfections was undertaken by William Tabb. 29 Almost borrowing directly from Becker and Kmeger, he divides the American economy into two sectors, one white and one black. Instead of assuming, as Becker and Krueger did, that both sectors are competitive, Tabb assumes that the white sector is characterized by oligopoly, high wages, and high occupational opportunity while the black sector is characterized by cutthroat competition, low wages, and little occupational opportunity. Tabb says himself that this approach means looking at "the manpower problems of the ghetto" in the context of " a dual labor market." The theory of the dual labor market is rapidly being subsumed under radical economic theory. Although David Gordon takes great pains to emphasize that dual market theory stands in a "neutral corner" between orthodox and radical economic theory, this is a hedge. He admits that the radical theorists have grasped dual market theory for their own, as the most accurate description of the American economy. 30 This is ironic. Although there are hints of a theory of a dual labor market in Galbraith's The New Industrial State, the earliest formalization of such a theory was developed by Robert Averitt--by no means a major critic of the economic status quo. ~1 Averitt did intend his theory of the dual economy to be a challenge to the prevailing orthodoxy offered by neoclassical economics, but when he partitioned the economy into an oligopolistic core and a highly competitive periphery he praised the core firms for being the source of the economy's growth and strength. Radical theorists have readily been able to seize dual market theory, identify the noncompetitive core as the source of villainy, and emerge with a powerful critique of orthodox economics as well as the existing economic order. Unless applications of the theory of the dual labor market to the problem of black poverty are cast in a radical framework, they become mere echoes of human capital theory as an explanation of racial economic inequality. For example, consider Frank Davis's fully developed application of dual market theory to racial inequality. Like Tabb, Davis views the ghetto as having an economy separate from the white majority's. The white sector---external to "central city ghettos"--is characterized by "rapid

242


technological change in manufacturing under conditions of oligopolistic pricing" in low productivity employment and low paying service industries . . . . ,, 32 For Davis, Blacks are poor because they are confined to the ghetto economy in low-wage occupations. The fundamental question then becomes why Blacks are confined to the ghetto economy. Davis describes black workers as being largely unskilled, lacking the productivity characteristics that would enable them to readily fill the occupations in manufacturing industries which dominate the white sector. He criticizes the "war on poverty" for failing to provide a manpower development strategy that would have offered a " m a s s i v e " training and employment program "tied specifically to the higher-paying jobs in the labor market." Actual manpower and training programs were not massive, and they directed participants toward continued low-wage employment. 3a At this level, Davis's explanation for the limited intersectoral mobility of Blacks is not far removed from human capital theory. The only major difference is that the Davis world is not uniformly competitive and as a result the terms of trade between the white and black sectors steadily worsen. But if Blacks had improved education and training their economic condition would also improve markedly. This is precisely the criticism advanced by Robert E. Klitgaard of applications of the theory of the dual labor market to the analysis of poverty. He ascribes the low level of intersectoral mobility experienced by the poor to their own work characteristics--"low learning a b i l i t y . . . tardiness, absenteeism, turnover and [low] motivation." Klitgaard even argues that dual market theory is unnecessary for analyzing poverty because "conventional interpretations bolstered with insights into the motivations and incentives of the disadvantaged do a more elegant and empirically supported job," charging that advocates of the dual labor market explanation "often lapse into conspiratorial accusations which their own evidence makes unnecessary and misleading."a4 Indeed, the Davis model does convey an image of the American economy that is patently antiblack; but he hardly develops a conspiratorial model if the cause of black confinement to the periphery is the lack of skills needed to enter core occupations. If this is the case, core employers behave no differently from perfect competitors, hiring the most productive workers, who happen to be white. If anything is conspiratorial in the Davis world it is the political factors external to the market that cause Blacks to accumulate a lower average level of "human capital." The significant weakness in the Davis model is his failure to truly distinguish his explanation of the causes of black poverty from that offered


243

by human capital theorists (or Klitgaard). A new answer is needed to the " w h y " of black confinement to peripheral occupations, however, because contrary to Klitgaard's view, the "conventional interpretations" are not "empirically-supported." In fact, empirical research is consistently disproving the conventional interpretations' predictions. The application of dual market theory undertaken by Bennett Harrison results in a superior treatment of the " w h y " of confinement because of his reliance on a radical framework. Actually Harrison has two major explanations for black confinement to the periphery--(1) a modified version of the "statistical discrimination" hypothesis, and (2) a neo-Marxian interpretation. Harrison's modified version of the "statistical discrimination" hypothesis involves the argument that firms do indeed assume that the average productivity of Blacks is lower than the average productivity of whites, using educational credentials as an inexpensive proxy variable for arriving at such a conclusion. Where Harrison departs from the human capital theorists is where he argues that employer perceptions are wrong, that the actual productivities of Blacks are equivalent to that of whites for many occupations. Moreover, Harrison argues that education and productivity are poorly correlated, citing evidence from research by sociologist Ivar Berg that demonstrates that in some cases the two variables may be negatively correlated ! Then the screening rule becomes inefficient. He goes on to add, " . . . w h a t all of human capitalists m i s s . . , is that workers adapt to their work environments, so that the job attachment (stability, interest, willingness to take discipline, even health of an underemployed worker will deteriorate over time, making the social decision to reject him a self-fulfilling prophecy!" a5 The problem with this formulation is that it still characterizes noncompetitive employers as behaving like perfect competitors. Although they may be using an incorrect decision rule for hiring workers, they perceive that they are hiring the most productive workers at the going wage. The explanation of black intersectoral immobility is based upon nothing more than the ignorance of white employers. It means that all white employers must be viewed as sharing this ignorance, none of them willing to break new ground and try hiring Blacks. As a result, although they follow marginal hiring rules, they are all failing to minimize their costs because they exclude competent workers from consideration, thus lowering their relevant labor supply curves. White or black entrepreneurs who might doubt the alleged inferiority of Blacks could exploit the profit advantages

244


of hiring them at lower wages--until their wage is driven up to a level equal with whites. In short, it is not terribly persuasive to attribute black poverty solely to white ignorance. The second argument, in which dual market theory is placed in a neo-Marxian framework, is the far more interesting and provocative one. In this view, what underlies the black confinement to the periphery is the desire of the managers of core finns to preserve the existing economic order. If Blacks could enter the primary labor market freely class conflict would be likely to supplant race conflict--and class conflict is potentially far more dangerous to a hierarchical economy. Harrison suggests this when he refers to Barbara Bergmann's crowding study in which she found that "Without any compensating increase in the stock of jobs in the economy, the elimination of occupational discrimination by race (would according to the Bergmann equations) cause a 6 to 9 percent reduction in the incomes of white men lacking an eighth grade diploma--a group which constitutes 14% of all adult white men in the United States and displacement of white women would be even greater." Plainly this significant group of white workers has the most " r e a s o n " for expressing their "oftimes vocal opposition.., to government programs designed to expand job opportunities for blacks . . . . ,,36 The white blue collar worker views the black worker as the greatest threat to what economic hegemony he has achieved, and this keeps him from directing his continued discontentment with "the way things are" at the economic system as a whole. This is, as Klitgaard suggested, a conspiratorial view of the American marketplace. It suggests that corporate managers have a stake in preserving the dual market system--the preservation of their oligopoly profits--and that exclusion of Blacks from the core of the economy enables them to continue. It is a view that also suggests a certain degree of "consciousness" on their part. Historical data being assembled by Reish, Gordon, and Edwards indicate that managers of large corporations actively fostered labor market segmentation, to keep the workers divided politically! 37 In addition, a recent research study by William Comanor led to the finding t h a t " . . . e s t i m a t e d measures of racial discrimination are generally higher in more skilled occupations, more profitable industries, and in industries heavily concentrated in the largest metropolitan areas. ,,38 This particular finding is entirely consistent with the expected hiring pattern for a "typical" core firm. This is an extremely grim vision of the American economy and it demands further research--particularly research that tests dual market theory by making precise distinctions between core and peripheral firms


245

and core and peripheral occupations. What is critical here is that discrimination is viewed as serving the interests of the powerful. It also serves the ;~'interests of white workers who are protected from job competition with Blacks. And it also protects white employers in the periphery who are assured of a continued supply of low-wage labor. In short, all of white society benefits from discrimination against Blacks. The only departure from the Krueger model is the fact that the presence of imperfect competition in the core of the economy enables white capitalists to gain from discrimination rather than lose. That is an all-important departure. CONCLUSIONS The only common implication of all of the models is that what really enables whites to engage in discrimination against Blacks with adverse effects on black income is the black disadvantage in physical capital ownership. If Blacks owned a comparable level of physical capital they could engage in the segregation game without any loss of income. Blacks could ignore the "white sector" altogether if Blacks owned a comparable independent set of resources. This implication is especially evident in the trade and dual market models. The normative implications of each of the three main approaches vary widely. The Becker-Kmeger models would require either (1) a change in white attitudes (improved race relations) or (2) affirmative action or equal employment opportunity efforts to change conditions confronting Blacks. Sowell's theoretical outlook would suggest either (1) increased investments in the education of Blacks until complete parity with whites in educational attainment is achieved (this was fundamentally the "antipoverty" strategy of the mid-1960's--when increased education and training for the black poor were the ticket government officials said would give them passage into the middle class) and (2) increased black exposure to economically valuable cultural experiences (i.e., integration). For the dual market theorist with a radical perspective the situation may well be hopeless. All of white society gains from discrimination. If the power system in the society is not to be successfully attacked or if a major group of whites does not recognize that their "class interests" are best served by opposing the status quo, expectations of any major political strategy being mounted that will change the economic condition of black people must be viewed as slim. Nevertheless, dual market theorists typically propose (1) undertaking community development efforts in urban poverty areas and (2) provision of public employment opportunities at nonpoverty wages. These strategies are viewed as being least likely to

246


threaten the white working class and yet have genuine positive effects on black incomes. Of equal interest is the descriptive difference between the explanation offered by dual market theory (and Anne Krueger's trade model) and the more orthodox models. The Becker model holds the prejudices of whites as the source of racial economic inequality, but there is nothing wrong with the economic system itself which is viewed as competitive. Exogenous tastes are the key. In human capital theory, factors external to the marketplace also are responsible for black poverty--factors which determine the racial distribution of productivities. Again the economic system itself is viewed as sound. But when Kreuger and the dual market theorists identify economic motivations for discrimination--then, and only then--do factors endogenous to the economic system become crucial. The application of dual market theory to this question represents a rejection of the prevailing paradigm in the discipline of economics. It is a theory that scuttles the assumption that the American economy is perfectly competitive, proposing that the causes of economic inequality be explored in a world dominated by firms with market power. The irony of dual market theory is that it also proposes that the neoclassical ideal of perfect competition is only approximated in the peripheral region of the market; only the sector in which Blacks are crowded demonstrates the ideal form of capitalism. Not only is this ironic, it is deeply disturbing--because the probability of substantial social change occurring that will reduce black poverty significantly becomes highly unlikely. The segmented marketplace generates antagonisms among lower-income workers along racial lines that mask "the real problem." The segmented marketplace generates political attitudes that are system preserving, and the perpetuation of those political attitudes depend, in part, upon the system's ability to keep Blacks out of the economy's core. There is no sign of a break in this pattern. In large part this is because the view of the American marketplace that dual market theory provides is widely discounted. After all, economists are as vulnerable to cognitive dissonance as anyone else. NOTES 1. Gary S. Becker, The Economics of Discrimination, (Chicago: The University of Chicago Press, 1957), pp. 6-27. 2. Thiscriticism of the"p ure labormarket" explanationof blackpovertyis also madeby Richard B. Freeman, "Decline of LaborMarketDiscriminationand EconomicAnalysis,"


247

The American Economic Review (May 1973):283-84; and by Joseph Stiglitz, "Approaches to the Economics of Discrimination," The American Economic Review (May 1973):288-89. 3. Herrington J. Bryce, "Putting Black Economic Progress In Perspective," Ebony (August 1973):60. 4. Anne O. Krueger, "The Economics of Discrimination," Journal of Political Economy (October 1973):481-86. 5. Ibid., p. 483. 6. This criticism is also developed by Thomas Sowell, "Race and the Market," The Review of Black Political Economy (Summer 1973): 11. 7. David M. Gordon, Theories of Poverty and Underemployment, (Boston: D.C. Heath and Company, 1972), pp. 90-91. 8. Bennett Harrison, Education, Training and the Urban Ghetto, (Baltimore: The Johns Hopkins Press, 1972), pp. 3-4. 9. Peter B. Doeringer and Michael J. Piore, Internal Labor Markets and Manpower Analysis, (Lexington: Heath Lexington Books, 1971). 10. Thomas Sowell, "Economics and Black People," The Review of Black Political Economy (Winter/Spring 1971):3-21. A large part of this article is Sowell's personal diatribe aimed at other black educators, black colleges, and black "militants." This article will, however, only examine his application of human capital theory to racial economic inequality. l l. Ibid., p. 5. 12. Ibid. 13. Ibid., p. 7. 14. Ibid., pp. 7-8. 15. More invidious extra-market reasons social scientists have offered for the presumed lower level of black human capital accumulation are twofold: (1) Blacks have been given the opportunity to raise their human capital stock but have failed to take advantage of the opportunity and (2) Blacks are genetically inferior to whites in intelligence and cannot be expected to have an equivalent proportion of human capital. These reasons are either explicitly or implicitly advanced in the works of academicians like Daniel Moynihan, Edward Banfietd, Stanley Elkins, Arthur Jensen, William Shockley, and Richard Hernnstein. 16. Freeman, op. cit., pp. 280-86. 17. Ibid., p. 282. 18. "The March to Equality Marks Time," Time (September 8, 1973):74. 19. Ibid., p. 74. 20. Ibid. 21. Barbara R. Bergmann, "The Effect On White Incomes of Discrimination in Employment," Journal of Political Economy (March/April 1971):294-313. 22. Ibid., p. 297. 23. Randall Weiss, "The Effects of Education on the Earnings of Blacks and Whites," Review of Economic Statistics (May 1970): 150-59. 24. Bennett Harrison, "Education and Underemployment in the Urban Ghetto," The American Economic Review (December 1972):796-812. 25. Ibid., pp. 809-10. 26. Human capital theorists are notorious for paradigm stretching because of the flexibility that their conceptual child offers. Like the multiheaded Hydra, when one version of the theory seems to be sliced away another takes its place. Two other versions should be mentioned here. Instead of viewing human capital accumulation as a function of years of schooling, one group of theorists argues that "education" at home during the preschool

248


years is the most important consideration (childhood investments). This may involve something as simple as parents reading to the children each night. However, this would not explain wage differentials that occur between whites and blacks with a high school education or less--whom we would expect to have the least preparation and least time to provide planned cognitive stimulation for their sons and daughters. Another school of theorists has retreated to the point where they view on-the-job training as the only really important type of human capital. But this does not explain why whites have greater access to the entry points for jobs where such training can be received, although their credentials may even be "worse" than those of a black worker. 27. Harrison, op. cit., no. 9, pp. 34-38. 28. As quoted in Bennett Harrison, "Employment, Unemployment, and the Structure of Urban Labor Markets," Wharton Quarterly reprint (Spring 1972):5. 29. William Tabb, The Political Economy of the Black Ghetto, (New York: W.W. Norton & Company, Inc., 1970). 30. Gordon, op. cir., no. 8. 31. Robert T. Averitt, The Dual Economy: The Dynamics of American Industry Structure, (New York: W,W. Norton & Company, Inc., 1968). 32. Frank Davis, The Economics of Black Community Development, (Chicago: Markham Publishing Company, 1972), p. 24. 33. Ibid., pp. 98-110. 34. Robert E. Klitgaard, "The Dual Labor Market and Manpower Policy," The Monthly Labor Review (November 1971):46. 35. Harrison, correspondence. 36. Harrison, op. cit., no. 29, p. 3. 37. Michael Reich, David M. Gordon, and Richard C. Edwards, "A Theory of Labor Market Segmentation," The American Economic Review (May 1973):359-65. 38. William S. Comanor, "Racial Discrimination in American Industry," Economica (November 1973):363-78. In addition a recent case study on St. Louis uncovered employment patterns that demonstrate the existence of racially segmented labor market throughout the entire metropolitan area. See David M. Streifford, "Racial Economic Dualism in St. Louis," The Review of Black Political Economy (Spring, 1974):64-78.