Articles

Yes, Immigration May Lift Wages

The New York Times, "Economic Scene" , November 03, 2005

From 1990 to 2000, the number of people working in the United States grew by more than nine million, or around 8 percent, from immigration alone. What effect did all those new foreign-born workers have on the wages of native-born Americans?

The answer seems obvious at first. An increase in the supply of workers should push down wages, just as a bumper crop of wheat drives down wheat prices.

That is exactly what some influential economic studies, notably by George J. Borjas at Harvard, have found. In a 2003 article, for instance, Professor Borjas calculated that immigrants had depressed the average wage of American-born workers by about 3 percent in the 1990's.

But workers are not as uniform as wheat, and 10 years is a long time -- long enough for employers to invest in new production and take on more workers. The model of surging supply meeting fixed demand is not realistic.

As economists know all too well, changing the assumptions of a model can often change the results. In a new working paper for the National Bureau of Economic Research, two other economists using methodology similar to Professor Borjas's but different assumptions get the opposite result.

In "Rethinking the Gains From Immigration: Theory and Evidence From the U.S.," Gianmarco I. P. Ottaviano of the University of Bologna and Giovanni Peri of the University of California, Davis estimate that immigration in the 1990's increased the average wage of American-born workers by 2.7 percent. (The paper is available here.)

Although it still relies on a highly stylized model of the economy, their paper adds two complexities that bring it closer to reality.

First, the two economists assume that businesses can make additional capital investments to take advantage of the expanded supply of workers. Companies may open new restaurants or stores, add new factory lines or build more houses.

In their model, as in the real world, "investment adjusts not to keep fixed the amount of capital but to keep fixed the return to capital," Professor Peri said. As long as businesses can profitably add new production, they hire more workers, and wages do not necessarily go down. Instead, he said, "more workers means more business."

As businesses expand, hiring foreign-born workers to do one job may also require hiring more native-born workers with complementary skills. Immigrant engineers, for instance, may create demand for native-born patent lawyers and marketing executives.

That is the paper's second refinement. It assumes that immigrants do not always compete for the same jobs as American-born workers. The two groups are not "perfect substitutes," even when they have similar education and the same occupation. A Chinese cook is not the same as a Texas barbecue chef.

Immigrants often bring different skills to the American labor force, and concentrate on different occupations from natives. Among high school dropouts, the paper notes, the "foreign-born are highly overrepresented in professions like tailors (54 percent were foreign-born in 2000) and plaster-stucco masons (44 percent were foreign-born in 2000)." By contrast, American-born workers make up more than 99 percent of all crane operators and sewer-pipe cleaners.

The same is true at the highest educational levels, where foreign-born college graduates make up 44 percent of all medical scientists but only 4 percent of lawyers. (Immigrants tend to be concentrated at the highest and lowest levels of income and education.)

Immigrants do, of course, compete to some extent with native-born workers. The question is how much.

To measure wage competition, the economists looked at how much an increase in the number of foreign-born workers affects the wages of other foreign-born workers versus American-born workers with the same educational background. If the groups were perfect substitutes, the change would be the same.

But there is a difference. When the number of immigrant college graduates goes up by 4 percent, their wages drop by 1 percent more than the wages of native-born college graduates. Immigrants, in other words, compete more with each other than with American-born workers.

Professors Ottaviano and Peri find that recent immigration has had the most negative effects on the least educated.

Immigration in the 1990's, they estimate, raised the wages of native-born high school graduates, college dropouts and college graduates by at least 2.5 percent. By contrast, they estimate that the wages of American-born high school dropouts fell by 2.4 percent because of immigration.

In an interview, however, Professor Peri noted that Americans are increasingly well educated, so that high school dropouts make up a small, rapidly declining portion of today's native-born work force. In 2000, he said, only 9 percent of American-born workers did not have a high school degree.

"If you look at the U.S. labor force," he said, "those people born in the U.S., I am talking about a negative effect for about 9 percent of the population and a positive effect for 91 percent of the population."