Immigration's effect on wages

There has been some controversy about the effect that the large number of immigrants, including illegal immigrants, has had on wages.  To be clear, there is no controversy about what economists expected to happen -- a significant increase in the labor force should, all other things equal, push the supply curve of labor to the right and thereby lower wages.  And a significant increase there has, indeed, been: estimates are that there are up to 16 million foreign-born workers who have recently entered the U.S. labor market, leading to a 15 percent increase in the size of the labor force.  The controversy has arisen over the failure to find any effect of this significant increase on prevailing wage rates.  Of course, it could have been that not all other things were equal -- that, for instance, the demand for labor, particularly for low-skilled labor, had increased more dramatically than had the supply of labor. 

In today's Wall Street Journal George Borjas, a first-rate economist at the Kennedy School of Government at Harvard, summarizes the literature.  (Here's a link to the article.)  His pithy reason for why there has been, so far, very little evidence of a depressive effect of immigration on wages: "the problem has been that economists were looking for the wage effect in all the wrong places."

How so?  Apparently, economists have focused their attention on three cities -- New York, Los Angeles, and Chicago -- where one-third of the immigrants settle and look there for an effect on local wages.  And so far they have found no discernible effect on wages in those three labor markets. 

Borjas points to two effects that would make it difficult to find immigration's effect on wages in local labor markets.  First, immigrants might tend to flock to markets where wages are high (and wages might be high because there is a robust demand in those markets for labor).  Eventually, of course, the influx of immigrants either causes wages to level off or even fall, but initially it might look as if there was a positive correlation between the presence of immigrants and wages (rather than the hypothesized negative correlation).  Second, mobility by businesses and by domestic workers could make it difficult to see a clear effect of immigration on wages.  Suppose, to use Borjas' example, that a business in Michigan observes the influx of immigrants in Los Angeles and, in search of cheaper labor, relocates to southern California.  If enough businesses from other parts of the country move to southern California in search of this expanded land, therefore, cheaper labor pool, then the demand for labor there may increase, offsetting the depressive effect on wages of the increased labor supply.  (Back in Michigan, there may be secondary, adverse effects on the labor market.)  And additionally, domestic workers in southern California may leave that area to avoid competing with the lower-wage-accepting immigrants. 

For all those reasons, Borjas notes, it may be very difficult to find a localized effect of immigration on wages. 

Therefore, he argues in favor of looking for effects on the national level.  By looking at national wage trends for particular skill groups over the past 40 years, Borjas find that a "10 percent increase in the size of a skill group ... reduces the wage of that group by 3 percent to 4 percent."

In looking at the particular effect on native workers' wages of the 1980 - 2000 immigration of Mexican workers, Borjas (and his co-author, Lawrence Katz) found that "immigration lowered the wage of native workers, particularly of those workers with the least education [] by 3 percent for the average worker and by 8 percent for high school dropouts." 

Borjas is careful to conclude that these effects do not "imply that immigration is a net loss for the economy.  After all, the wage losses suffered by workers show up as higher profits to employers and, eventually, as lower prices to consumers.  Immigration policy is just another redistribution program.  In the short run, it transfers wealth from one group (workers) to another (employers)."

Professor Borjas' research on immigration and labor markets is voluminous, as you can see by going to www.ssrn.com and searching by his name.  This paper with Lawrence Katz is the most recent (April, 2005) one on the effects for the U.S. labor market of Mexican immigration.  (This is an NBER Working Paper.  If you are with a nonprofit organization, you can have access to the paper for free.  Otherwise, there will be a fee.) 

TSU