I worry that the following headline at Ezra Klein’s blog — “Does restricting immigration improve economic performance?” — might give readers the wrong impression. Ezra writes:

There are a lot of studies looking at the economic impacts of immigration, but what about the economic impacts of restricting immigration? Over the past few years, a lot of cities and counties have enacted restrictive laws, which creates a nice natural experiment of sorts: As an area becomes less friendly to immigrants, what happens to its economy?

He then cites a paper by Huyen Pham of Texas Wesleyan University School of Law and Van H. Pham of Baylor University that tracks the impact of local immigration laws.

In the first empirical study of this issue, we analyze the economic impact of these laws. The laws take different forms – some authorize local police to enforce federal immigration laws, some restrict benefits like housing and employment to those with legal immigration status, and some require all government transactions to be conducted in English only. Applying statistical analysis to economic data from the U.S. Census Bureau, we find such laws resulted in a 1 to 2 percent drop in employment, or 337 to 675 lost jobs for the average county. This drop in employment includes both authorized and unauthorized workers. We also find that the laws hurt some industries, such as the restaurant industry, while helping others, such as the grocery and liquor store industry. This suggests that affected workers may be switching jobs, rather than leaving a particular jurisdiction altogether.

I wonder about the limitations of this analysis. I am strongly inclined to be skeptical of local efforts to set immigration policy, and I tend to think English-only laws are counterproductive. But the following passage from “The Evolution of the Mexican-Born Workforce in the United States” by economists George Borjas and Lawrence Katz is in the back of my head:

Recent research … raises two questions about the validity of interpreting near zero cross-city correlations as evidence that immigration has no labor market impact. First, immigrants may not be randomly distributed across labor markets. If immigrants tend to clusterin cities with thriving economies (and high wages), there would be a built-in positive correlationbetween immigration and wages. This positive correlation would certainly attenuate, and perhapseven reverse, whatever negative impact immigration might have had on wages in local labor markets.

Second, natives may respond to the wage impact of immigration by moving their labor or capital to other cities.

Borjas and Katz continue:

Because local labor markets adjust to immigration, a number of recent studies have emphasized that the labor market impact of immigration may be measurable only at the national level. Borjas (2003) used this insight to examine the link between immigration and the evolution of wages for specific skill groups in the past few decades. His study indicates that by analyzing national trends in the labor market and by defining skill groups in terms of both educational attainment and work experience, one can make substantial progress in determininghow immigration alters the employment and earnings opportunities of native workers.