People normally assume that immigration will expand the welfare state. The lazy version says (a) immigrants are net beneficiaries of the welfare state, and (b) people vote their self-interest. The better version says that immigrants’ countries of origins favor more redistribution than natives – and immigrants bring their political culture with them.
Both stories, however, ignore the effect of immigration on natives’ support for the welfare state. Researchers – most of whom look kindly upon both immigration and the welfare state – often fear that immigration will sap natives’ support for redistribution by undermining their sense of national cohesion. If they’re right, immigration could easily, on balance, shrink the welfare state rather than expand it.
So what’s the real story? I honestly don’t know, and after reading Soroka et al.’s “Immigration and Redistribution in a Global Era” chapter in Globalization and Egalitarian Redistribution (Princeton University Press, 2006), I’m less sure than ever.
Soroka et al.’s ultimately agree that immigration restrains the welfare state, though the effect is so moderate that it merely slows its rate of growth rather than actually making it smaller:
International migration does seem to matter for the size of the welfare state. Although no welfare state has actually shrunk in the face of the accelerating international movement of people, its rate of growth is smaller the more open a society is to immigration. To the extent that spending growth is inescapable, mandated by the aging of populations in industrial societies, specific parts of the welfare states—especially the parts that redistribute from rich to poor or from the old to the young—may truly have shrunk in the face of migratory pressures. Whatever the details, the typical industrial society might spend 16 or 17% more than it now does on social services had it kept its foreign-born percentage where it was in 1970. [emphasis mine]
But check out these major caveats:
How seriously should we take these propositions? It is an awkward fact that the biggest apparent effect of immigration is in the estimation with the weakest basis: data from two time points in 18 countries. It does seem appropriate that estimated cumulative impact is greater for an implicitly low-frequency estimation than for very high-frequency—annual at the extreme—modelling. But we do feel diffident about the point estimate in our simple cross-sectional estimation. Our anxiety is only increased by that estimate’s vulnerability to inclusion or exclusion of particular cases. Of the cases we include, the USA and the Netherlands carry a heavy burden. If we tell single-country stories, the USA provides an internally consistent one that requires no reference to external migratory pressure. Gilens (1999) argues that Americans have become more resistant to welfare, in particular to programs for the poor, as welfare policy has come to be increasingly racialized in its media presentations. The racial focus is mainly on African Americans. But we also know that immigrants also figure in that country’s discourse. And we can supply no obvious purely domestic story to cover the Netherlands case. An implication of the US and Netherlands role, however, is that modest increments in the foreign-born share carry much less proportionate charge than big ones. The case we exclude, Switzerland, is difficult to discuss because of idiosyncrasies in measurement of the scale and scope of its welfare state. If we are forced to include it, then the immigration-welfare spending relationship disappears.
My main takeaway: If you think you know the effect of immigration on the welfare state, you’re overconfident. Immigration’s effect on the welfare state is too hazy to clearly detect one way or the other. So regardless of your views on the welfare state, you should evaluate the effects of immigration on other grounds.
READER COMMENTS
Weir
Oct 17 2018 at 10:08pm
Suppose your boss has spent the last two weeks ranting about white women. Before that it was prep schools and male entitlement. You’re a white woman yourself, same as your boss, and you don’t take her latest tantrum personally. Your boss is always going off at some group of people she hates. In August it was the NRA.
But you call in sick today. You’re not actually sick. You just don’t feel a lot of responsibility to an employer who is so self-righteous and hectoring. In her own mind, she’s enlightened and virtuous. So you take the day off, and collect a paycheck without having to put up with her lectures for this one shift.
And for the next couple years you’re going to keep trying to get as much free money as you can before the company goes bankrupt. What do you really owe this awful boss? You’re supposed to feel loyalty to an employer who won’t stop ranting about how much she hates everybody outside her little circle?
You’d turn up everyday if your boss was pleasant and treated her employees like human beings. But she has this general hatred she can’t control, so you’ll take the money, and you don’t care what happens to the company now.
Hazel Meade
Oct 22 2018 at 11:10am
This is pretty true of both political parties.
Republicans are always ranting about “liberals”, immigrants, minorities, gay people, Muslims, Europeans, etc.
Democrats are always ranting about capitalism, corporations, “the right wing”, white privilege, the patriarchy, the system, etc.
It gets really tiresome no matter what your political alignment.
Mark
Oct 17 2018 at 10:31pm
The argument that immigrants are more likely to support welfare because of their home countries’ political cultures is odd. The countries that spend the highest % of GDP on government social programs are, by far, white Western European countries. Most of our immigrants are from Asia or Latin America, and Asian countries have much smaller welfare states relative to GDP than European ones, and most Latin American countries are in between. Therefore, if immigrants were to bring the political views of their home countries, one would think they would support lower social spending than European-Americans.
I suggest immigrants’ higher support for the welfare state is a recent phenomenon and is primarily caused by mood affiliation—because liberals are more welcoming to immigrants, immigrants who come in without strong political views of their own will adopt the political views of liberals.
Mark Z
Oct 18 2018 at 8:13pm
Related to your second point (but assuming rationality rather than mood affiliation), Condorcet’s principle (I’ll call it that since it’s not really a paradox) could lead to the same thing. In fact, there could be a positive feedback loop, by which opposition to immigration on the right rooted in immigrants’ tendency to vote Democratic leads more immigrants to vote Democratic whatever their opinions on welfare, which in turn enables Democrats to pursue more welfare programs, bringing us full circle to Republicans seeing this as a reason to further oppose immigration.
I’m inclined to posit a general theorem: whenever you have two mutually exclusive groups that have, on average, differing interests or values, even if the difference is politically irrelevant at the starting point, the state in which each group is proportionally represented in each political party/camp is an unstable equilibrium. Things will always tend, over time, toward a state where one group predominantly votes for one party and the other group for the other party.
Hazel Meade
Oct 22 2018 at 11:15am
That happens in our system because our system is uniquely predisposed to polarization into a binary political dynamic. It probably doesn’t happen the same way in a multi-party parliamentary system. We have this binary dynamic largely because of the electoral college system with winner take all seat allocation at the state level – it creates this large spoiler effect (there are no run-off elections, the plurality winner gets all the seats), and the enhanced spoiler effect punishes people for voting for anyone other then one of the two dominant parties.
nobody.really
Oct 18 2018 at 12:59am
This American Life did a story on immigration (especially illegal immigration), including its financial effects on government. While the show focused on the experiences of Albertville, AL, it also cited research from a 618-page study from the National Academy of Sciences. Among the findings:
Social Security: Immigrants—including illegal immigrants—pay the same FICA taxes as similarly situated native-born US citizens. And legal immigrants derive the same benefits as similarly situated native-born US citizens. But illegal immigrants, typically using false Social Security numbers, don’t; they pay in, but generally don’t draw out. In effect, they’re helping fund our Social Security/Medicare deficit.
SNAP (food stamps): When illegal immigrants move into rural towns (in this case, Albertville, AL), SNAP usage doubled, from 8% of households to 16% over 10 years. But in the rest of Alabama, which had basically no immigration over that time, SNAP usage increased FASTER, from 11% to 26%. Why? In the absence of immigration, rural towns dried up, their schools and businesses closed, and more people went on public assistance.
School spending: Immigrants tend to have 0.75 more kids than native-born couples. The added education spending increased annual local taxes for a typical non-Latino taxpayer in Albertville by $272.
Net effect on government: In the final analysis, there’s an easy rule of thumb. People who get any college education tend, over 75 years, to pay into government more than they cost. People who just graduate from high school break even. And people who never graduate from high school tend to cost more than they pay in.
The bad news is that a disproportionate share of illegal immigrants fall into this latter camp. The good news is the magnitude of the cost. Over 75 years, all levels of government will spend a net $21,000 on an illegal immigrant without a high school diploma—if you also include what is being spent on the immigrant’s kids and grandkids. The number is relatively low because these immigrants generally don’t draw from Social Security or Medicare. In contrast, a native-born US citizen who doesn’t graduate high school will tend to cost government hundreds of thousands of dollars over 75 years.
Alan Goldhammer
Oct 18 2018 at 8:26am
This comment is based on observational data (mine). In our area of suburban Washington, DC almost all those who work for yard service and home cleaning companies are Latino(a)s. Most who do handyman work likewise. I don’t know how recent their arrival to the US is or their immigration status. I only know that they all work hard, earn money, and pay taxes. It would be interesting to hear from Professor Caplan regarding his area on the Virginia side of the Potomac.
Christian K
Oct 18 2018 at 12:20pm
I think that this is an interesting topic to think about. The post says that many believe that more immigration to the US will cause a decrease in support for welfare among existing residents because they will feel less connection to immigrants with a different culture/background. It is difficult for me to see this being an issue because of how diverse and spread out the United States already is. I think this would be a much bigger issue in nations where the cultural and racial makeup is more uniform. I hope that Americans are well adjusted enough to not let race play so much into our feelings about policies such as welfare.
On the flip side, I don’t think that more immigrants would expand the welfare state either. I don’t think most immigrants come from places with more welfare than the US so I doubt that their first thoughts are that US welfare should expand. Just the opposite, it seems to me that most immigrants are usually willing to work the hardest to earn an honest living. In the end I think that the effect on welfare by immigration is probably negligible.
Thaomas
Oct 20 2018 at 8:47am
Immigration by creating wealth should shrink the relative size of the welfare state as fewer people will fall into situations that require food stamps and similar means tested programs. Since SS benefits per capita are fixed in real terms, higher per capita incomes that immigration will help achieve will also lower the share of these transfer payments.
Weir
Oct 22 2018 at 7:25pm
Nima Sanandaji: “Despite the fact that Nordic nations are characterised by good health, only the Netherlands spends more on incapacity-related unemployment than Scandinavian countries. A survey from 2001 showed that 44 per cent believed that it was acceptable to claim sickness benefits if they were dissatisfied with their working environment. Other studies have pointed to increases in sickness absence due to sporting events. For instance, absence among men due to sickness increased by 41 per cent during the 2002 football World Cup.”
So in the first example you feel entitled to free money out of revenge and bitterness. In the second you feel entitled to free money because there’s a match on TV.
Neither example has anything to do with national cohesion. And the story about national cohesion always had some big holes in it. Remember that Bismarck set one part of the nation against another. Being Catholic doesn’t put you outside the nation.
Likewise in Britain. The Labour Party waged its own Kulturkampf against what they called toffs and Tory scum: “The upper classes in every country are selfish, depraved, dissolute and decadent.” When you divide people by religion or class there’s not a lot of national cohesion going on right there.
Class solidarity in Britain. Religious sectarianism in Germany. But not national cohesion.
And that’s before you factor in the biggest explanation for wanting to get the maximum amount of free money for yourself: You’re a sucker if you don’t.
nobody.really
Oct 23 2018 at 1:01am
Did you see how Norway performed during that World Cup? I was sick about it, too.
Mark Z
Oct 23 2018 at 1:06am
Hazel,
“That happens in our system because our system is uniquely predisposed to polarization into a binary political dynamic. It probably doesn’t happen the same way in a multi-party parliamentary system.”
I partly agree, and I think European parliamentary systems do have more prominent third parties for that reason (the Liberal Democrats in the UK, the Free Democrats in Germany), but I think even those systems tend toward a binary model. In democracy in any form, you still have to form a coalition representing at least 50% of the populace. So you still end up with two rival coalitions dominating politics; it’s just that the coalitions are perhaps less stable, with more regularly shifting alliances. Then again, I think the Liberal Democrats and Tories are basically always governing together (and neither with Labour) and the FDP and CDU (Germany’s classical liberal and conservative parties, respectively) basically always coalesce, so even in Europe the party coalitions are fairly consistent.
Comments are closed.