has told us that they do have constitutional rights. So it is wrong to proceed on the premise that their constitutional rights are any less-worthy of being heard than the statutory rights which this bill quite properly protects in its repair of the Lorance case. In this respect paragraph (C) is right off the map. It would allow a third party to be bound by a litigation of which he had no notice, if a court determined that reasonable efforts had been made to give him notice. This violates the constitutional principle that people may not be bound by litigation in which their interests were not represented and in which they did not actually have a chance to participate. It is this provision which suggests that the bill assumes that the rights of these potential intervenors are second rate constitutional rights, and only the constitutional rights of the parties the bill is interested to protect are of the first rate kind. But there are no second rate constitutional rights, just as there are no second class citizens in this country. Paragraph (C) should be abandoned because its premise violates the most fundamental principles of the civil rights movement in this country. The CHAIRMAN. Mr. Loury.

Mr. LOURY. Thank you, Mr. Chairman, Senator Hatch, and Senator Metzenbaum. I appreciate this opportunity to appear before the committee.

As a professor at the Kennedy School of Government at Harvard, I have for some years now been writing and teaching about the problem of racial inequality in our society. In what follows and in my written testimony, I want to offer a brief survey of the dimensions of that problem, drawn from analyses of the social science literature and statistical data published by various Federal agencies, as I interpret them, with the purpose in mind of providing some context for your deliberations on this proposed civil rights legislation.

I believe it important to understand that antidiscrimination laws provide in the larger context a very limited tool with which to address the general problem of group inequality in our society. Not being a lawyer, I am not commenting on the specific legal controversies raised by the bill.

In my summary here I just want to make several points. While substantial differences in income exist between various ethnic groups and while discrimination against women and minority groups in employment has been and continues to be a matter of concern, there is no sound social scientific basis for concluding that the existing economic differences have been caused by or reflect the extent of employment discrimination against various groups.

Indeed, gross statistical disparities are inadequate to identify the presence of discrimination because individuals differ in many subtle ways unlikely to be identified, measured and controlled for when group incomes are compared.

Accordingly, there is no basis for the expectation that antidiscrimination legislation alone will have anything but a marginal effect on these differences.

Second, in the case of blacks, who arguably have experienced the most severely deleterious effects of discrimination in the past, there has been a truly dramatic reduction in the extent of employment discrimination in the period since the passage of the Civil Rights Act of 1964.

Moreover, there has occurred over the last four decades a pronounced improvement in the overall relative economic status of black Americans, an improvement which started before the Civil Rights Act and which is accounted for by beneficial changes in such fundamental economic factors as improved educational attain

ment, long-term economic growth, interregional migration, as well as diminished extent of employment discrimination.

The rate of improvement in the relative earnings of blacks has slowed in the last decade, especially for the younger cohorts-and this should be a matter of some concern-which seems to be due to the slowing in the overall rate of economic growth in our economy as well as to the failure of the quality of education available to black youngsters to continue to improve as it has historically done. Third, despite the long-term upward trend, there nevertheless remain profoundly troubling racial differences in economic advantage which warrant the attention of this committee and of all Americans.

In the case of blacks, the relative labor market gains of individuals have not been matched by comparable gains in the resources available to families. This is because the proportion of families headed by a single parent has risen dramatically among blacks during the same period in which individuals' earnings have improved.

As well, the percentage of black children residing in households in which only one parent is present has risen sharply, indeed continues to rise. Poverty rates among such children are disturbingly high.

More generally, the emergence of what some have called an urban underclass has been noted in many of our cities. Blacks are disproportionately overrepresented in this population, where the problems of drugs, criminal violence, educational failure, homelessness, and family instability are manifest.

It is my conviction that these problems constitute the most important and intractable aspect of racial inequality in our time. Unfortunately, these problems are unlikely to be mitigated by civil rights legislation because they do not derive in any direct way from the practice of employment discrimination.

However, the continued existence of these social problems contributes to the inability of those subject to them to compete effectively in the labor market.

Finally, I am concerned that the politics of civil rights issues have in the last decade taken on a disturbingly symbolic tone and have become divorced from the social reality of racial inequality as it actually exists in our society. This trend is reflected, in my judgment, in the public debates over recent U.S. Supreme Court decisions and the extent to which they do or do not reverse the progress in race relations which we have as a society have made in the last decades.

I am not commenting on the legal correctness of the Court's rulings but rather on the political meaning which has been attached to them by advocates of minority interests. It is my opinion, as a close student of these socioeconomic trends affecting minority groups in our society, that the danger to minority interests of these developments in the law has been significantly overstated.

A far greater threat to the attainment of full social and economic parity for racial minorities is posed by the trends affecting the social, educational, and family life experiences of low-income urban communities. These trends are more difficult to legislate against and do not as readily provide us with villains and heroes. But if

they are not reversed, we will continue to face a serious social and political problem around the issue of race in this country for many years to come.

Thank you.

[The prepared statement of Mr. Loury follows:]

PREPARED STATEMENT OF GLENN C. LOURY

Mr. Chairman, I thank you for this opportunity to appear before the Committee. As Professor of Political Economy at the John F. Kennedy School of Government at Harvard University, I have for some years now been writing and teaching about the problem of group inequality in our society. In what follows I will offer a brief survey of the dimensions of that problem, drawn from analyses in the social science literature of the statistical data collected and published by various federal agencies, as I have interpreted them. I hope with this survey to provide some context for your de liberations on the proposed civil rights legislation. It is important, I believe, to understand that anti-discrimination laws provide a very limited tool with which to redress the problem of racial inequality. I am not qualified to assess the legal issues raised by the legislation, and therefore will not comment on them here. In this testimony I want to stress several points:

(1) While substantial differences in income exist between various ethnic groups, and while discrimination against women and various minority groups in employment has been and continues to be a matter of concern, there is no sound socialscientific basis for concluding that the existing economic differences have been caused by, or reflect the extent of, employment discrimination against various groups. Gross statistical disparities are inadequate to identify the presence of discrimination because individuals differ in many ways likely to affect their earnings capacities which are usually not measured and controlled for when group outcomes are compared. Accordingly, there is no basis for the expectation that anti-discrimination legislation alone will have anything but a marginal effect on these differ

ences.

(2) In the case of blacks, who have arguably experienced the most severely deleterious effects of discrimination in the past, there has been a truly dramatic reduction in the extent of employment discrimination in the period since the passage of the Civil Rights Act of 1964. Moreover, there has occurred over the last four decades a pronounced improvement in the overall relative economic status of black Americans, an improvement which started before the Civil Rights Act, and which is accounted for by beneficial changes in such fundamental economic factors as improved educational attainment, long term economic growth, and interregional migration, as well as by the diminished extent of employment discrimination. The rate of improvement in the relative earnings of blacks has slowed in the last decade, especially for the youngest cohorts. This is a matter of some concern, which seems to be due to a slowing of the overall rate of economic growth in the United States, as well as to the failure for the quality of education available to black youngsters to continue to improve as it has historically done.

(3) Despite the long-term upward trend, there nevertheless remain profoundly troubling racial differences in economic advantage which warrant the attention of this Committee, and of all Americans. In the case of blacks, the relative labor market gains of individuals have not been matched by comparable gains in the resources available to families. this is because the proportion of families headed by a single parent has risen dramatically among blacks during the same period in which individuals' earnings have improved. As well, the percentage of black children residing in households in which only one parent is present has risen sharply. Poverty rates among such children are disturbingly high. More generally, the emergence of what some have called an "urban underclass" has been noted in many of our cities. Blacks are disproportionately overrepresented in this population, where the problems of drugs, criminal violence, educational failure, homelessness and family instability are manifest. It is my conviction that these problems constitute the most important and intractable aspect of racial inequality in our time. unfortunately, these problems are unlikely to be mitigated by civil rights legislation, because they do not derive in any direct way from the practice of employment discrimination. However, the continued existence of these social problems contributes to the inability of those subject to them to compete effectively in the labor market.

(4) I am concerned that the politics of civil rights issues have in the last decade taken on a disturbingly symbolic tone, and have become divorced from the social reality of racial inequality as it actually exists in our society. This trend is reflected,

in my judgment, in the public debates over recent Supreme Court decisions, and the extent to which they do or do not reverse the progress in race relations which we have made as a society in the last decades. I am not commenting on the legal correctness of the Court's rulings, but rather on the political meaning which has been attache dot them by advocates of minority interests. It is my opinion, as a close student of the socio-economic trends affecting minority groups in our society, that the danger to minority interests of these developments in the law has been significantly overstated. A far greater threat to the attainment of full social and economic parity for racial minorities is posed by the trends affecting the social, educational and family life experiences of low-income urban communities. These trends are more difficult to legislate against, and do not as readily provide us with villains and heroes. But if they are not reversed we will continue to face a serious social and political problem of racial inequality in this country for many years to come. Income Differences and Discrimination

While incomes vary considerably between racial and ethnic groups, one cannot draw conclusions about the extent of discrimination from such differences. Some groups which have been discriminated against in the past nevertheless have higher incomes than groups which are presumed to have benefited from discrimination. for example, Asian-Americans have experienced considerable discrimination, yet have incomes often exceeding those of whites. (According to the 1980 Census, the family incomes of Japanese-Americans was 30 percent higher than that of whites in 1979.) Discrimination cannot account for this fact, though other factors may be suggested as a plausible explanation. Family income is affected by factors such as the number of family members in the labor force, and the skills which those persons bring to the labor market. Therefore, family structure (whether families consist of husbandwife couples with children, or mothers raising their children alone, e.g.) will be an important determinant of group differences in family incomes. Also, some groups may exhibit a greater amount of family work effort than others, even when the structure of families is similar. For example, 71 percent of Filipino families have two or more earners compared to 55 percent of white American families. (This consideration is important in accounting for the relative family incomes of Chinese, Japanese, Korean and Cuban families as well.)

Individual skills are also important in accounting for differences between groups in income. Educational attainment varies considerably by race and ethnicity. În 1980, only 58 percent of Hispanic men aged 25-29 had completed high school, compared to 89 percent of Asian men, and 73 percent of black men. If one compares the earnings of minority individuals relative to that of whites, varying the degree of educational attainment, it becomes clear that differences in education contribute in a major way to group income differences, because the ratio of minority to white earnings is consistently higher among persons with the same amount of education than it is among all persons. For example, in the 1980 Census black men aged 25-34 are recorded as having hourly earnings at 84 percent the level of whites; yet among men with 12 years of schooling the relative black hourly earnings was 87 percent of whites, and among men with 16 years of schooling the relative black earnings was 90 percent of whites. Moreover, there are differences in school achievement among persons with the same years of schooling which are likely to explain some earnings differences. These differences, reflected in part by variation across groups in students' performance on achievement tests, may be attributed to differences in the quality of schools attended by group members, as well as differences in the education and income of students' parents.

The problem of inferring discrimination from earnings differentials is further illustrated by reference to the relative earnings of women in various ethnic groups. Among women aged 45-54 with a college education, for example, blacks earned roughly 13 percent more than whites in 1979 despite the fact that they probably experienced more labor market discrimination. Some economists have attributed this fact to the greater work experience and labor market attachment exhibited by black women. Labor force participation rates have traditionally been higher among black women than among whites, and black women have been more likely to work while their children are young, a fact which may derive from the higher probability that a black woman would be the sole support of her children.

It is not possible here to offer a complete discussion of the factors determining individual earnings, nor of the reasons why earnings might differ between groups. The preceding discussion is intended to suggest however that employment discrimination is but one, and by no means the most important of these factors, and that the tendency to conclude that discrimination is a problem from the mere fact that groups have differing incomes should be resisted. This point is underscored by the fact that, even within groups usually treated as a single aggregate for the purpose

of discussing the existence of discrimination, one can observe significant differences in economic experience. Black American families of West Indian descent fare much better than do blacks overall, with incomes comparable to those of whites. And among white men, the group usually offered as a basis of comparison for the purpose of measuring the well-being of others, there are significant differences which presumably cannot be explained by discrimination. for example, even after adjusting for education, geographic region and other factors, Jewish men were found in a recent Civil Rights Commission study to earn about 15 percent more, and Italian men 6 percent more, than white men of British descent.

Economic Progress of Blacks and the Significant Remaining Inequality

There has been significant improvement in the earnings experience of employed black workers over the period 1940-1980, as documented by the decennial census data. This improvement is both absolute, and relative to the earnings of comparable whites. This improvement began well before the onset of the civil rights legislation of the 1960's. It is explained in part by the shift of black workers out of agriculture, where they were concentrated prior to the second world war; by the migration of blacks from countryside to city, and from South to North and West which has occurred in this period; by the improvement in the quality and quantity of education attained by black workers, and by a decline in labor market discrimination against black workers. The pace of improvement historically has been greatest when the overall rate of growth for the economy has been greatest (during the 1940's and 1960's, e.g.), and appears to have slowed during the 1980's. This improvement has been relatively greater for more educated and younger workers. However, since 1960 the gains in relative earnings among employed blacks have been accompanied by a pronounced downward trend in the labor force participation rates of black men, a trend which accelerated during the 1970's. Family incomes have not improved among blacks relative to whites in a manner comparable to the change in the earnings of employed individuals.

These earnings gains are reflected in the following data: (estimated annual earnings of fully employed men in constant 1987 dollars)

Source: Smith and Welch, "Black Economic Progress after Myrdal," Journal of Economic Literature, June 1989

11,441

4,956

These average gains have not, however, been enjoyed equally by all black workers. In the past quarter century earnings inequality within the black population has increased. Since 1959 inequality among black men has been greater than among white men, with the bottom 40 percent of black men earning about 8 percent of total black male earnings in 1959, but only 5 percent in 1984; and with the top 20 percent of black men earning about one-half of total black male earnings in 1959, but roughly 60 percent by 1984. This increasing inequality is due, in large part to the fact that an increasing proportion of black men report no labor market earnings at all, because they are unemployed or not in the workforce. Among black men 24 years old the proportion unemployed, out of the labor force or in jail rose from 13.8 percent in 1950 to 19.8 percent in 1960, 21.1 percent in 1970 and 28.2 percent in 1980; for men 35-36 years old the proportions were 13.5 percent in 1950, 17.1 percent in 1960, 13.7 percent in 1970 and 20.3 percent in 1980. Note the sharp rise during the 1970's.

Family income has also become more unequally among blacks. For example, the proportion of black families with incomes over $35,000 (in 1986 dollars) grew from 15.7 percent to 21.2 percent in the period 1970-1986; the proportion with incomes in excess of $50,000 increased from 4.7 percent to 8.8 percent in the same period; yet over this period the proportion of black families with income of less than $10,000 also grew, from 26.8 percent to 30.2 percent. Among two parent households with children black earnings rose 4 percent between 1973 and 1984, while white earnings fell 4 percent. Earnings for female headed households fell for both blacks and whites (by 9 percent and 8 percent respectively), but there are significantly more black