Something that I think isn’t always acknowledged is the extent to which high level of low-skill immigration have a purely statistical impact on the United States. For example, according to the Census Bureau the poverty rate in 2009 was 14.3 percent. For Hispanics it was a much higher 25.3 percent. And of course among the Hispanic population, recent immigrants and their descendants are disproportionately likely to be poor. What this all means is that if we’d had zero immigration from Latin America over the past 20 years the poverty rate would almost certainly be lower. What’s important to see is that doesn’t mean any actual people would have been richer if we’d slammed the doors shut in 1991. It’s just that many people who today are poor and in the United States of America would instead be even poorer and living in Mexico or El Salvador.
That’s an extreme scenario, of course. But it highlights the general fact that we ought to be more interested in the impact of policy on the lives of actual human beings than on its impact on statistical aggregates of various times. If Ward 3 seceded from Washington, DC, it would instantly become an extraordinarily prosperous city. Unfortunately, relative few of the widely available data series about income in the United States let us really look at this issue in the ideal way.