Q & A With Giovanni Peri: Noted Expert on Immigration Responds to Three Topical Questions

Peri image

On April 19, 2017, Giovanni Peri gave the final Werner Sichel lecture of the 2016-2017 season at Western Michigan University. Peri is a professor and chair of the economics department at the University of California, Davis, and a research associate at the National Bureau of Economic Research. He is a widely acknowledged research leader in the study of international migration and its consequences on labor markets and productivity. We asked Dr. Peri to address a few questions related to the lecture that he delivered on April 19th.

Question 1: How has immigration to the United States changed over the past 25 years?

Peri: In general, immigrants have become increasingly high-skilled and highly educated over this period. Whereas immigrants composed about 9 percent of the U.S. college graduate population in 1990, today they make up 16 percent of the college graduate population, and this share is higher than the immigrant share of the population as a whole.

Also, the source countries have changed somewhat. Immigrants are more likely to come from China and India and less likely to come from Mexico and Central America. In fact, all immigration from Mexico, documented or not, has fallen by two-thirds since 2000, from about 400,000 per year to about 130,000 per year. And these figures include only Mexicans entering the United States; if one accounts for Mexican immigrants in the United States returning to Mexico, the net flow of Mexican immigration actually turned negative before the Great Recession, a decade ago. More broadly, immigrants to the United States still come on net from poorer countries—the share of immigration from places like Western Europe and Canada is down, and it’s up from places like India, China, and Africa—but the immigrants from those relatively poor places are fairly highly skilled.

However, it’s important to distinguish between the flow of immigrants and the stock of immigrants. Even though net immigration flow to the United States from Mexico and Central America has fallen, the stock of such immigrants in our country is still quite large, and a large number are undocumented. Consequently, immigration policies that attempt to target flows from these countries are likely to be ineffective.

Question 2: What does economic research say about the impact of immigration on wages in the United States?

Peri: First, it’s important to note that wage growth over the past 35 or so years has been highly skewed toward the more-educated, and that wages for those with less than a high school degree have actually fallen (adjusted for inflation). At the same time, as I said earlier, immigration has become more proportionately concentrated among the highly educated. A simple supply-side economic story would say that the faster growth of highly educated immigrants would reduce the wages of highly educated Americans relative to less educated ones, the opposite of what happened. This wrong-signed correlation is even more salient if one looks at changes in relative wages and immigration by decade.

Of course, correlation is not causation, and economics has far more sophisticated tools to tease out the causal impact. In some of my research, for example, we examined changes in the immigrant share of the population and of wages between 1970 and 2010 across labor market areas in the United States. Even when trying to predict the change in the immigrant share using historical data, and controlling for several other factors, we found essentially no correlation between changes in immigration and changes in wages, even when looking at wages of those with no more than a high school degree. A broad reading of the literature suggests that immigration generally has little direct effect on natives’ wages.

Why would that be? First, low-skilled immigrants largely do different jobs than native workers—particularly in agriculture, but also in restaurant kitchens, hotels, and construction sites—and thus are more likely to serve as complements to native workers than as substitutes. Second, firms respond to immigrant workers and become more productive or cost competitive, enabling them to sell more for less.  Lower prices increase demand and in turn put upward pressure on wages. Third, immigrants help grow the population and thus increase local demand, including for goods and services made by natives, and this, too, puts upward pressure on wages.

Economics research suggests other factors, such as automation, shifts in trade, the rising return to skills, unions, and the minimum wage are far more important than immigration in explaining changes in wages and employment opportunities.

Question 3: What would be your key advice to government officials regarding immigration policy?

Peri: Immigration, especially the high-skilled immigration that increasingly represents the case for the United States, is largely beneficial. A growing volume of research suggests that high-skilled immigrants disproportionately contribute toward U.S. innovation, technology, productivity growth, and entrepreneurship. Indeed, there are more Nobel Prize laureates in the sciences who are immigrants to the United States than there are from the next-highest country (the U.K.), period.

Because immigrants tend to hold jobs regardless of their skills, we should promote opportunities for foreign students earning degrees here to stay and work, such as by making it easier to convert student visas to work visas. But we also need to address immigrants with fewer skills who are already here (and largely working). In particular, to increase their chances of social mobility and economic assimilation, we should promote strong schools, especially in immigrant-heavy areas—although this would make sense in economically challenged areas without many immigrants, as well.

Policies that address the current flow of immigrants—mostly high-skilled from India and China—and encourage skill formation among immigrants already here are likely to do more for economic growth, including wages, than policies that are based on a reality that is no longer present.