by Conor Friedersdorf

Mickey Kaus has a proposal:

If you're worried about incomes at the bottom... one solution leaps out at you. It's a solution that worked, at least in the late 1990s under Bill Clinton, when wages at the low end of the income ladder rose fairly dramatically. The solution is tight labor markets. Get employers bidding for scarce workers and you'll see incomes rise across the board without the need for government aid programs or tax redistribution. A major enemy of tight labor markets at the bottom is also fairly clear: unchecked immigration by undocumented low-skilled workers. It's hard for a day laborer to command $18 an hour in the market if there are illegals hanging out on the corner willing to work for $7. Even experts who claim illlegal immigration is good for Americans overall admit that it's not good for Americans at the bottom. In other words, it's not good for income equality.

Odd, then that Obama, in his "war on inequality," hasn't made a big effort to prevent illegal immigration--or at least to prevent illegal immigrration from returning with renewed force should the economy recover.

Will Wilkinson objects:

This is badly misleading. A move to United States is an upwardly mobile move for almost all low-skilled immigrant workers, and it tends to reduce inequality on the whole. As a matter of description, Mr Kaus' conclusion follows only if we grant him the premise that the trend in inequality is best measured by looking at the set of people inside a country's borders at one point in time and then comparing it to the set of people inside the country's borders at a later point in time. I propose we reject this premise...

The only reason to make the within-borders population of a nation-state our analytical touchstone is a prior commitment to the idea that the nation-state is the correct unit of normative evaluation. That is to say, an unacknowledged commitment to moral nationalism tends to stand behind the sort of analytical nationalism driving Mr Kaus' inequality accounting trick.

Though I agree with Wilkinson that immigration reduces global inequality – and share his preference for laws that allow more immigrants into the United States – he is flat wrong when he asserts that moral nationalism is the only reason to care about economic inequality within a single nation-state. One coherent reason to fear intra-country inequality: large disparities in wealth might undermine democracy as the poorest citizens wield an ever-shrinking amount of political access, insider savvy, or power (see Bell, California).

People like Kaus argue that economic inequality, while not problematic in itself, excacerbates social inequality, which we ought to care about. (A weakness in his position: wouldn't amnesty for illegal immigrants dramatically increase social equality within the United States? I'm pretty sure he'd respond that amnesty would lead to more illegal immigration and thus ultimately reduce social inequality.)

A third coherent reason: economic inequality eventually results in egalitarian efforts and government programs that critics find undesirable, whether out of principled aversion to redistribution or a prudential concern that these efforts vest the federal government with too much power.

I'd be fascinated to see Kaus and Wilkinson debate immigration as it pertains to social inequality.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.