Share this:

Despite the progress made in Civil Rights since the 1960s there are still huge wage disparities between black and white workers in America. In new research that examines wage disparities in metropolitan areas in the U.S., Craig Kerr finds that the idea of a national wage gap is misleading. Instead, he argues that wage disparities often differ markedly between different cities and regions due to factors that include black population numbers, workplace discrimination, union membership, segregation, and incarceration rates. With this in mind, he writes that policies aimed at reducing wage inequalities should be targeted at specific areas, rather than using a one size fits all approach. 

In the United States, a worker’s income is significantly tied to their race. Academics and the media often speak on the black-white wage gap, or the difference between the average wage-income of black and white workers.  However, in our recent research, we found that wage gaps differed greatly across metropolitan statistical areas (MSA) and thus the notion of a single overall average wage gap is deceptive. We also found the gaps to be correlated with factors that have historically been thought to influence inequality such as residential segregation or incarceration rates. Our research strongly suggests that policies aimed at reducing racial wage inequality should be geographically targeted.

MSA Level Wage Gaps

Just how much do black-white wage gaps differ across different MSAs?  Figures 1 and 2 display our estimates for the wage gap in each MSA in our sample.  The values of the wage gaps can be interpreted as the percentage difference in the wages of a black worker compared to a white worker when race is the only difference between the two. Wage gaps differ greatly across MSA and take values that imply black workers earn 24.5 percent less than their white counterparts (El Paso, TX in 1990) or up to 12.9 percent more (Muncie, IN in 1990).  As the figures clearly indicate, attributing one overall average wage gap to all locations would be naïve for most MSAs and very misleading for some Midwestern MSAs, where black workers are estimated to earn higher wages than their white counterparts.

Figure 1 – 1990 Wage Gaps

Kerr Fig 1

Figure 2 – 2000 Wage Gaps

Kerr Fig 2

What possible explanations exist for the differences in the estimated wage gaps?  To address this question, we investigated the following factors:

Black Population In the seminal theoretical treatment of discrimination in the labor market, economist Gary Becker showed how the amount of wage discrimination present should increase with the percent of the labor force that is black. The intuition is that black workers will first go to work for the least discriminating firms before obtaining a job with a firm that discriminates. Therefore, the more black workers there are in an MSA, the more likely it is that some must find employment with discriminating firms.  The data strongly agreed with the theory; MSA’s with 10 percent more black workers have, on average, a 3.6 percent wider wage gap in 1990 and between 1.6 percent and 2 percent in 2000. 

Discrimination The “racial tastes” theory (also proposed by Becker) states that employers that do not want to hire black workers will still do so if they can pay them lower wages. The best measure of stated prejudice we could find was only available at the state level and thus we were unable to significantly estimate the relationship between discrimination and wage gaps for the most part.  However, we did find one significant result. When splitting up MSAs into North and South, we saw that the northern MSA’s residing in states that have more stated prejudice actually pay higher wages to black workers. 

Unions Union contracts restrict employers’ ability to pay different wages for the same position and should therefore limit discrimination based on race.  Collective bargaining agreements may then be an important contributing factor to the estimated wage premiums black workers are shown to earn in the Midwest.  This could arise in the following scenario.  Suppose that the typical black worker in any location has less human capital (education, experience, etc.) than the typical white worker and a particular job employs workers whose human capital levels fall within a certain range.   In this scenario, since the average black worker has less human capital than their white counterpart working the same job, the black worker would be paid less absent any prejudice.

However, if the job is covered by a collective bargaining agreement that restricts the employers’ ability to pay different wages for the same job, then we would observe the black worker making the same wage as the white worker.  It would then seem that the black worker is getting paid more because of their race.  The fraction of workers covered by a collective bargaining agreement was found to be greater in MSAs that had smaller wage gaps or wage premiums for black workers.  When we separated the covered workers according to whether they were in the public or private sector, we surprisingly found that it was only the private sector jobs that were positively associated with greater relative black earnings.

Penal System The black population is and has historically been overrepresented in the U.S. penal system. However, the manner in which this should affect local wage gaps is not immediately clear. On the one hand, if black workers are overrepresented in jail, it may be due to prejudice, which could also be reflected by a wider wage gap. However, even if prejudice is not a significant factor in the jail population (the evidence of this is mixed), workers in jail and prison suffer an opportunity cost of lost job experience and the stigma of being incarcerated.   Both effects decrease future wage offers for incarcerated workers.

On the other hand, incarcerated workers are more likely to earn low wages outside of jail and their incarceration prevents them from showing up in our data set.  These “missing” black workers who would presumably earn low wages will then cause our calculated average wages for black workers to increase.  So MSAs with a large incarcerated black population may have artificially high average black wages due to our inability to observe the low wages incarcerated workers would have earned had they not been locked up.

We constructed a measure of black overrepresentation in local jails to test whether this phenomenon was correlated with wage gaps.  For example, if a particular MSA has a population that is 10 percent black and a jail population that is 11 percent black, we would say that the black population was 11 percent – 10 percent = 1 percent overrepresented in jails.  We found that in 1990, the more overrepresented black inmates were in prison, the less black workers made relative to white workers.   In 2000, the relationship turns positive but is not statistically significant.

Combining these results with the fact that 77 percent of the MSAs in our sample increased the relative amount of black males in their jails during this time, it seems that what we may be observing is the removal of black workers that would have earned low wages from the labor market into jails. This suggests that black workers were, on average, earning more in MSAs with high black jail populations in 2000; partially because the low-wage workers were incarcerated and thus not included in the average.

Segregation Theoretically, segregation could generate higher or lower incomes for black workers.  Black incomes could be lower if segregation restricted their employment opportunities by keeping them far from better paying jobs.  This is what is known as the “spatial mismatch” theory.  On the other hand, segregation could improve black earnings through positive peer effects if it forced high-skilled black workers to remain in black neighborhoods.  That is, the effect of having a high-skilled neighbor may influence one’s decisions in a positive manner or allow access to a beneficial network of opportunities.  Although we measured residential segregation in all MSA’s, we only found it to be associated with larger wage gaps in the South.

Segregation likely occurs for different reasons in the North and South. Presumably, the pattern of black neighborhoods in the South are correlated with historic slave populations whereas those in the North are correlated with migration patterns during the “Great Migration” when millions of Southern black workers migrated from the rural South to find employment in urban areas of the North.  If residential segregation is more of a choice in the North and more reflective of historic prejudice in the South, it would make sense that we only observe a correlation between segregation and wage gaps in the South.

Moving Forward

Although we estimate significant relationships between wage gaps and the size of the black working population, the prevalence of collective bargaining, the overrepresented jail population, and segregation in the South, we cannot say at this point that any of these factors are causing wider or narrower wage gaps.

For example, consider the relationship between segregation and the wage gaps in the South.  We cannot say whether the segregated MSAs in the South are causing wage gaps to be larger or whether the segregation is a result of large wage gaps. All we know at this point is that Southern MSAs with larger wage gaps also have significantly more segregation.  What should be taken from this result is that the relationship between segregation and wages in the South appears to differ from the North and is associated with racial inequality at the wage level.

The main implication arising from our results is that policies that aim to diminish racial inequality need to be geographically targeted. The racial wage disparity in New Orleans is very different in magnitude from that in San Francisco and likely exists for different reasons.  A one-size-fits-all approach to reducing racial wage inequality would likely not achieve its desired goals across all MSA’s.

This article is based on the paper ‘Racial Wage Disparity in US Cities’, in Race and Social Problems, coauthored with Dr. Randall Walsh at the University of Pittsburgh. 

Featured image credit: owyzzz (Flickr, CC-BY-NC-SA-2.0)

Please read our comments policy before commenting.           

Note:  This article gives the views of the author, and not the position of U.S.App– American Politics and Policy, nor of the London School of Economics.

Shortened URL for this post:


 About the authors 

Craig Kerr – California State Polytechnic University at Pomona
Craig Kerr is an Assistant Professor in the Department of Economics at California State Polytechnic University at Pomona. 

Print Friendly