Discussions of economic inequality are common nowadays thanks to Thomas Piketty’s new book Capital in the Twenty-First Century. There are several good critiques of Piketty’s book and at least one wonderful podcast. I’m not convinced that economic inequality in a (mostly) free-market economy matters one way or the other for economic growth, social stability, or political stability (ceteris paribus), so this blog is a response to those concerned that liberalized immigration could exacerbate wealth inequality.
Papers on how immigrants affect the wages of Americans almost uniformly present the results as relative gains or losses compared to the wages of other workers. While that work is valuable, below I will only discuss papers that focus exclusively on economic inequality caused by immigration.
Borjas et. al. found that immigration (along with trade) only modestly affects earnings inequality – a role not substantial enough to account for more than a small percentage of the change. Instead, he attributes the growth in income inequality to the acceleration of skills-biased technological change (SBTC) and other institutional changes in the labor market.
David Card failed to find a substantially causal relationship between increased immigration and growth in wage inequality. He discovered that immigration explains about 5 percent of the rise in overall wage inequality between 1980 and 2000. An important distinction is between the wage inequality effects of immigration on natives and the effects on wage inequality for immigrants and natives. While 5 percent of the growth in overall wage inequality can be attributed to immigration, immigration’s effect on native wage inequality is negligible. Immigrants tend to have either very high or very low wages compared to natives, meaning that immigrants have a naturally higher residual level of income inequality than natives do. Thus, immigration causes the economy-wide level of wage inequality to increase without changing native wage inequality. Immigration has little, if any, effect on native wage inequality according to Card.
A colloquium of experts held at the Federal Reserve Bank of New York, as reported in the 1997 Economic Report of the President, attributed almost half of the increased income inequality in the 1980s and 1990s to SBTC but only assigned between 5 and 10 percent of the blame to immigration.
Source: Federal Reserve Bank of New York
Economists Brian Hibbs and Gihoon Hong found that immigration is responsible for about 24 percent of the increase in income inequality among U.S. metropolitan areas between 1990 and 2000 – the biggest impact in my reading of the academic literature. They used the Gini index as their measure of inequality. A 1 percent increase in immigrants relative to the population of a metropolitan area increased the Gini coefficient by 0.66 points.
A 1999 paper by Deborah Reed focused on explaining the exceptional rise in household income inequality in California. She found that SBTC and immigration accounted for about half of the increase in income inequality from the late 1960s to 1997. She attributed between 17 percent and 40 percent of the increase in income inequality during that time period to immigration, depending on the assumptions in different models. If immigrants were spread evenly throughout the income distribution, they would have no effect on income inequality in California. However, since many of the immigrants were concentrated in lower-income brackets, the income of the state’s workforce became more unequal.
To introduce another wrinkle, economist Robert I. Lerman attempted to account for rapid immigrant wage gains into any measurement of immigration induced inequality. He undertook this by two means. First, he excluded recent immigrants, who have the lowest wages, from the base and end years of his analysis so he could track the wage inequality of the same group of people over time while ignoring additional workers added to the labor market. Second, he included the immigrant wages in their home countries prior to immigration in his measure of income inequality. This second method takes their massive economic gains into account. Lerman’s first method of excluding the recent immigrants eliminated 20 percent to 25 percent of the standard estimates of the growth in wage inequality. Lerman’s second method eliminated most of the estimated rise in income inequality.
The resulting increase in wage, income, and wealth inequality from immigration appears to be small inside of the United States – if it even exists. Globally, rapid economic growth in Asia has probably reduced income inequality. No doubt immigration affects inequality but it is likely not the primary driver.
Inequality itself does not matter. What matters is how the inequality is produced. If inequality increases because highly skilled workers earn a pay premium due to SBTC, then inequality is a beneficial result of complementary capital investments that will increase economic growth. On the other hand, if inequality increases because the government has decided to expropriate all of the property owned by one group and turn ownership over to another group, then political instability and other perverse economic incentives will emerge that will diminish economic growth. The source of inequality, not the inequality itself, determines whether economic inequality is a problem. In the United States, inequality is only a problem in so far as it is caused by government redistribution. Immigration, as the voluntary movement of people to the United States for economic opportunity, fits into the former positive-sum economic arrangement rather than the latter negative-sum expropriation.
Thanks to Kristina Pepe for her excellent research assistance on this piece.