Context, Please
A new report tries to convince us we haven't gotten very far.


In a recent Washington Times column, Clarence Page expressed surprise that the National Urban League’s 2004 edition of The State of Black America: The Complexity of Black Progress reported that over 40 percent of blacks surveyed for the report believed that there had been little or no improvement in the economic and social mobility of blacks since the passage of the 1964 Civil Rights Act. Such a conclusion seems difficult to square with the facts. However, reading the report provides some basis for understanding why so many black Americans believe so little has changed in 40 years.

As Clarence Page points out in his piece, for the first time the Urban League has attempted to capture the full picture of black economic, social, and political progress in one indicator: The Equality Index. According to the Equality Index, in 2004 African Americans’ economic and social status is only 73 percent of white Americans’, a figure that would certainly seem to support the conclusion among many blacks, and others, that little progress has been achieved. The problem is that the report has a number of flaws.

First, attempting to capture every aspect of black life compared to whites’ in a single indicator is–to say the least–ambitious. Social reformers, researchers, scholars, and policymakers have always been interested in finding indicators that can provide a quantitative measure of success in the area of social change similar to the role that the profit margin plays in the private sector. The difficulty with this approach is that in order to achieve the simplicity and elegance inherent in a single indicator, one has to reduce the complexity and nuance of a multidimensional world into a single number. Even in the business world, profit indicators do not provide a full picture of a business’s operation and long-term success. In fact, because profitability is actually the culmination of a number of factors related to business operation, any good manager spends a considerable amount of time analyzing a number of indicators in addition to a business’s profitability.

Second, the report attempts to talk about black progress, but instead tends to focus on a single snapshot in time. This unfortunately is a very common practice. If you want to show that poverty is deepening, or jobs are declining, or that blacks have made little progress, you focus on a single point in time rather than progress over time. Despite the questionable nature of using a single index to capture the full range of factors related to the progress of blacks compared to whites, the Equality Index might have some limited utility if it were measured over the 40 years since the Civil Rights Act of 1964 was passed. Instead the index simply examines one year. The 73 percent Equality Index score certainly seems to underscore blacks’ lack of progress. However, would we draw the same conclusions if we knew that the Index was 35 percent in 1964? This type of progress would suggest that not only had blacks enjoyed absolute improvements in their status but that over the last 40 years they have progressed faster than whites.

Third, these types of multifactor indicators are very sensitive to how much weight is given to each factor. The actual value of the Equality Index could be easily improved (or made worse) by changing the importance (i.e., the weight) given to each of the factors that compose the index. For example, the Equality Index gives economic factors an importance of 30 percent. That is, the value of economic factors is multiplied by 30 percent and added to the other factors that compose the index. On the other hand, education, health, social justice, and civic engagement are given the weights of 25 percent, 25 percent, 10 percent, and 10 percent respectively. What would happen if we changed the weights assigned to each of these factors? By varying the importance of these factors in the index it is possible to increase the value of the Equality Index to 80 percent or reduce it to 67 percent. This suggests that the index may not tell us the actual gap between blacks and whites; instead, what it may actually reflect is the subjective values of the index’s creators.

Fourth, the greatest potential flaw in the report’s Equality Index, and in the report as a whole, is that it treats race as the only relevant dimension for judging social progress. In this report, differences in education, family status, gender, work history, and residential location have little if any bearing on the analysis. For example, blacks on average tend to have higher unemployment rates and lower earnings than whites. This finding could easily tend to lead one to believe that racial discrimination in the labor market remains an on-going problem. And, while it may be true that there continues to be some level of racial discrimination that affects blacks’ labor-market status, comparing racial differences in average unemployment and earnings is misleading.

For example, it is well documented that workers with less than a high school education experience higher rates of unemployment and lower earnings compared to those with at least a high-school education. By 1999, the percentage of blacks 25 years and over without a high-school education was 28 percent compared to 16 percent for whites, while the proportion of blacks and whites with a college degree was 20 percent and 33 percent respectively. This means that in order to determine the true differences between blacks and whites it is critical to control for the differences in education between the two groups. Using data from the 2000 census, I calculated what the black-white earnings gap would be after controlling for education. It appears that roughly $4,600 in the black-white earnings gap can be explained by the differences in educational attainment. Put another way, if blacks and whites had the same level of educational attainment then $4,600 in the annual earnings gap between the two groups would disappear.

In addition to educational differences, there are other factors that also contribute to the racial-earnings gap. For instance, single female-headed households with children tend to have lower incomes than two-parent households. Because blacks have a disproportionately high number of such households, this also tends to depress racial earnings and labor-market participation vis-à-vis whites. Moreover, these factors not only influence labor-market earnings, but also affect a range of other social and economic indicators, such as employment rates, home ownership, wealth creation, and interaction with the criminal-justice system.

What does all this mean? Does it imply that we can explain away all racial and ethnic differences once we’ve controlled for all the appropriate factors? Not necessarily. What it does suggest, though, is that if one is interested in understanding the actual progress and status of blacks–or for that matter any other group–it is critical to do the following:

First, examine group progress over time. Even if the Urban League’s Equality Index was a perfect measure of racial status, progress can only be determined by assessing the measure over time. As any parent knows, it is hard to appreciate the growth and development of a child by simply looking at the last picture you took at the most recent birthday party. In order to determine if things are changing you need to review a series of pictures over time.

Second, control for differences in education and other factors that tend to have an independent influence on social and economic status. Another way to think of this is, in a democratic society with a market-based economic system, even if everyone were the same race, same ethnicity, and same religion, there would still be certain factors that would determine an individual’s social and economic status. Given this, in order to get a true measure of racial differences these factors must be considered.

Third, use groupings that transcend race and ethnicity. For example, when blacks and whites are grouped according to educational attainment, are there differences in economic and social status between blacks and whites within educational groupings? Given that it is reasonable to expect those with similar levels of educational attainment to have similar social and economic experiences, if black college graduates were to earn less than white college graduates then we may have uncovered a real racial difference. This does not mean that racism is the only explanation, but it does suggest that there is a racial difference that ought to be examined more closely.

Many well-meaning observers will say that it is the historical legacy of racism that is responsible for low levels of educational achievement and problems such as female-headed households. Indeed, some of these observers will even go so far as to assert that it is not just the legacy of historic racism, but on-going racial discrimination that continues to plague African Americans. Even if these observers are correct, they would be hard pressed to demonstrate empirically that there have been no substantive improvements in the economic and social status of blacks in the United States over the last 40 years. Moreover, it would still be better and would lead to more effective policymaking to determine the extent of racial differences after properly controlling for factors such as education and family makeup. For example, given that we know that education is critical to economic success, and if children from low-income families are at a higher risk of not finishing school, or finishing with below average skills, policies aimed at improving school performance such as No Child Left Behind, or school choice, are likely to be the most effective strategies for reducing racial differences.

The challenge for blacks–and for the nation–is to distinguish between the real progress made in black economic and social status over the last 40 years, while also recognizing those arenas where improvements still need to be made. A more balanced report would attempt to do this. Most of all, a balanced report would attempt to treat blacks as a socially and economically diverse group, and move beyond using race as the primary factor in explaining differences in economic and social status between groups. Unfortunately, by using the Equality Index to focus only on the current black-white gap, the National Urban League’s report contributes to the perception that little racial progress has been made and that racial discrimination still remains the primary explanation for the lagging status of blacks.

J. A. Foster-Bey was formerly a senior researcher and director of the Program on Regional Economic Opportunity at the Urban Institute.


Sign up for free NRO e-mails today:

NRO Polls on LockerDome

Subscribe to National Review