Bryan Caplan  

Gilens vs. the Political Externalities of Immigration

How Much do 99 Weeks of Unempl... Did the Dot.Com Boom Boost GDP...
Martin Gilens isn't known for his work on immigration.  Yet strangely, his two best-known books - Why Americans Hate Welfare and Affluence and Influence - have major implications about the effect of immigration on American politics.  Everyone worried about the political externalities of immigration ought to read them and reflect.

In Why Americans Hate Welfare, Martin Gilens essentially blames the relatively small size of the American welfare state on racism.  People don't like helping poor people who don't look like them.  As a result, ethnically diverse countries like the U.S. have smaller welfare states.

Although I'm a staunch opponent of the welfare state, Gilens' story makes a lot of sense.  My main reservation: We shouldn't confuse diminished benevolence with malevolence.  White Americans clearly care about non-white Americans; they just don't care about them as much as they care about white Americans.

What would happen, though, if Americans lived side-by-side with people they truly disliked? We should expect an even bigger negative effect on public support for the welfare state.  And if Americans truly dislike anyone, it's "foreigners."  Many can't even pronounce the word "foreigner" without hostility.  The upshot: Even if immigrants vote staunchly in favor of the welfare state, their presence reduces native support for the welfare state.  The net effect of immigration on the size of the welfare state is therefore ambiguous

If you're a conservative or libertarian who fears the "political externalities" of immigration, Why Americans Hate Welfare suggests that your concerns may be displaced.  The key word, though, is "may."  Gilens presents no specific evidence about the effect of immigration on the welfare state.  Two of my former students, Zac Gochenour and Alex Nowrasteh, have been studying this question in depth, but the jury is still out. 

Gilens' latest book, Affluence and Influence, undermines the political externalities argument much more directly.  His main finding is that American democracy is far more responsive to the policy preferences of the rich than the poor.  In fact, the poor and middle class have almost no influence at all.  If Gilens is right, granting citizenship to even staunchly statist immigrants is almost politically harmless... unless they're rich. 

I say "almost" because low-income immigration would slightly alter the identity of the (relatively) rich.  Gilens focuses on the 90th percentile of the income distribution.  So suppose you begin with 100 million native-born Americans.  If you admit 100 million new immigrants with below-median income - a doubling of the population - Americans who used to be at the 80th percentile of the income distribution become the new 90th percentile.  For immigrant inflows of realistic magnitude, though, you'd need a microscope to detect this effect.

Fair disclosure: Gilens' analysis does imply that one kind of immigrant is far more politically dangerous than the staunchest nativists have ever imagined: high-income statists.  While many elites from statist countries come here because they cherish our freer society, this is far from universal.  Plenty come for purely for American money and American lifestyle.  Philosophically, they still embrace their countries' socialist, nationalist, and theocratic traditions.  If these statist elites vote, our politicians will pay attention.  If enough come, we'll suffer under their yoke.

Slogan: The immigrants to fear aren't Mexican laborers, but French professors.

P.S. In case I haven't been blunt enough, I oppose immigration restrictions against people regardless of their political beliefs or income.  My claim is purely descriptive: We probably underestimate the negative political externalities of high-income statists - and overestimate the negative political externalities of everyone else.

COMMENTS (9 to date)
C writes:

Bryan should give us his thoughts on the disparate impact doctrine that is directly tied to changing demographics. That is far more impactful to most "regular" Americans (given its relevance to hiring and education) than the welfare state effects he is describing.

To provide more context, the harmful effects of disparate impact include the destruction of the possibility of a meritocracy, skewed hiring standards, and backlash racism...

Philo writes:

You mention potential immigrants to the U.S. who are “high-income statists,” who “still embrace their [original] countries' socialist, nationalist, and theocratic traditions.” As you say, their socialism will be bad for American politics; but are the nationalism and theocracy worth worrying about? A nationalistic French professor who immigrates to America is likely to remain emotionally attached more to France than to the U.S., so his nationalism will not be effective over here. And the only “theocratic traditions” that have much bite these days (in Catholic countries they are quite feeble) are the Islamic ones. Now, Muslims are such a small minority in the U.S. that an Islamic theocrat would be stymied in the short run; only an implausibly large number of Muslim fellow immigrants, or an unlikely mass conversion of native Americans to Islam, would make Islamic theocracy practicable. Or possibly the theocratic immigrant would convert to Christianity while retaining his theocratic views—but that is unlikely for psychological reasons.

In short: rich socialist immigrants are slightly worrying, but I wouldn’t fret about the nationalists and the theocrats.

Foobarista writes:

This is consistent with the thesis by "Bowling Alone" author Robert D. Putnam: that the more diverse a society, the lower level of trust and comfort you'll have among people in the society, even within the same group.

This appears to be a universal human condition and doesn't just apply to Americans.

Mr. Econotarian writes:

About 10% of French are now Muslims of recent immigration heritage, which has been leading to a lot of racial conflict. Yet they just elected a socialist president.

It may be true that redistributionist political culture may be formed due to racial homogeneity, but it is unclear how long it takes for that political culture to be formed (hundreds of years?) and how long it takes for that culture to be broken down by enhanced racial heterogeneity (hundreds of years?)

There has been a swing to the left in the past decade in areas near Canada. Obviously we must blame Canadian immigrants.

stephen writes:

"American democracy is far more responsive to the policy preferences of the rich than the poor"

South America has some pretty racially diverse populations, and the political spectrum in those countries usually runs from about "extreme commie" to "ultra left social democrat".

Chris H writes:

Two things about South America stephen.

First, it's unfair to generalize about the continent like that. Venezuela and Bolivia are pretty far socialist but other countries like Brazil seem to be following more of a China path (aka, the state is still pretty strong but weaker than it was) and then you have Chile which is down right libertarian in comparison to even the US (they privatized their version of social security years ago).

Second, South America's history is sufficiently different from the US that the historical effects likely trump the effect of distrust of those ethnically different. The big thing I'm thinking here is the degree of rent seeking in South American history. The continent was primarily settled on the basis of feudal holdings in which the owners functioned in the same way as European aristocrats (down to having a strong martial tradition). Wealth was gained by gaining the favor of the crown or his representatives in government. This tradition of rent seeking survived independence in tact (after all the leaders of the revolution were primarily the former elites) and created the tradition of the state as a source of wealth rather than a protector of rights. Thus for South America, the welfare programs function more as helping supporters and hurting adversaries rather than some noble, if very misguided, attempt to improve the lot of the poor. It's just the demagogues currently in charge in many South American states have their supporters as traditionally downtrodden groups of society.

stephen writes:


Thanks for reading and responding. Honestly, your reply is to well thought out for my comment, and my "usually" qualifier.

My main concern (wrt this post) is the idea that popular left wing government and racial conflict are incompatible. From what I can tell, they aren't even negatively correlated. But, again, I admit my ignorance ....

Philo writes:

To quote Gilens: “In Robert Dahl’s formulation, a central characteristic of democracy is ‘the continuing responsiveness of the government to the preferences of its citizens, considered as political equals’.” This is a misunderstanding on Dahl’s part. In a democracy every citizen has the right to cast one vote on certain important public issues (in a representative democracy, largely on the election of people who will make the actual decisions). *Equal right to vote* is the mark of democracy, not *equal influence over governmental decisions*.

For Gilens, differences in influence between citizens constitute *political inequality*, and “the *degree* of political inequality in a society [he means, in a *nation*] . . . tell[s] us much about the quality of the society’s [the nation’s] democracy.” But he offers no support for this normative judgment. On the contrary, I agree with Bryan that the policies of the U.S. government are *better* because low-income citizens do not have equal influence with those with high incomes.

Comments for this entry have been closed
Return to top