Is your name Lakisha or Jamal?  Then you should know that the odds of you getting a callback from a company that you submit your resume to are about 50 percent worse than would be the case if you were named Emily or Greg.  That's according to recently released research from the National Bureau of Economic Research, based upon faxing responses -- with no pictures -- to a sample of 1,300 employment ads.  The depressing conclusion is that even if your name simply sounds African-American you will find the gates of opportunity narrower.

The relevance of race in statistics on economic performance has been a persistent theme in our society.  The erosion in black-white earnings differences, if any, has been glacially slow.  In 1999 the median black household income in Indiana was $29,164, whereas the white household's comparable figure here was $42,744.  Nationally, African-American workers earn about 25 percent less, on average, than white workers, and are twice as likely to be unemployed.

Part of that difference stems from the same factors that explain earnings differentials between individuals. Since blacks tend to have less education, less work experience and less labor force attachment, on average, even in a labor market where there was no racial favoritism, we would expect their incomes to be lower.  But the stubborn truth is that even if we take care to include all of the factors that can legitimately be said to account for earnings differences, race remains a factor.

That's a challenge for our society, certainly, but it's also a challenge to the field of economics.  We struggle to explain how outcomes such as the racial divide in economic prosperity can be consistent with rational, profit maximizing behavior on the part of employers.  It hasn't been easy.

This simple argument reveals why.  If some employers overlook equally qualified black job applicants in favor of whites, they will be at a competitive disadvantage.  That's because the less demanded -- but equally productive -- black workers will have lower wages, and thus lower costs to the non-discriminating employers who hire them.  In a competitive marketplace, those low costs are an advantage that eventually forces the discriminating companies to either reform their hiring practices, or go out of business.

That's not happening, so it’s back to the drawing board.  Truth be told, black-white earnings differences are hardly the only quandary economists face when it comes to the labor market. There is something unique about individuals -- call it "ability" -- that we'll probably never be able to objectively measure, yet it figures hugely in determining our economic fate.  But none of us are prepared to believe that such a thing lines up so closely with skin tone as to explain what we see in the data.

Better education and training, long thought to be the key for all disadvantaged groups, has surprisingly ambiguous outcomes for African-Americans.  The NBER experiment faxed fictitious resumes to employers with different levels of education and experience.  The payback to a thicker resume was positive, in terms of callbacks, for both the Lakisha's and the Emily's, as you would expect.  But the increased response for those with black-sounding names was less than half of what the others experienced.

Thus we enter a destructive cycle, where a lower payback to education and training means less is undertaken, which, in turn, makes pay differences between racial groups even larger. That's a problem our society, and our economists, needs to solve.