Bryan Caplan  

Solve for This Equilibrium

Response to Professor Stephen ... Scott Alexander IS an Economis...
Tyler Cowen's fond of admonishing us to "solve for the equilibrium."  His maxim came to mind while reading his Bloomberg column last month:
How's this for a simple rule: Open borders for the residents of any democratic country with more generous transfer payments than Uncle Sam's.
What would happen?  Many economists and perhaps laymen would expect democratic countries to ferociously jack up their transfer payments to exploit U.S. immigration policy.  But solving for this equilibrium is a lot harder than it looks. 

1. If voters were rational and selfish, the "transfer payments skyrocket" prediction makes sense.  Open immigration to the U.S. have great value - and if transfer payments induce high emigration, sending countries won't actually have to make good on many of their financial promises.

2. In the real world, however, voters are usually sociotropic but irrational.  A key component of this irrationality is what I call anti-foreign bias - the tendency to underestimate the social benefits of economic interaction with foreigners.  The upshot: Many voters around the world would see the new U.S. immigration policy as nefarious.  While they wouldn't adopt Soviet-type policies to keep their people from leaving, there would be little popular demand to game the U.S. rules.  Some countries might actually curtail transfer payments, especially if they're near the cutoff.

3. Would Tyler's proposal also spur democratization?  If you buy Acemoglu and Robinson, it should: When the benefits of democracy rise, elites who fail to appease the masses risk popular discontent.  But this is also far-fetched.  Even moderately savvy dictators could successful paint Tyler's reform as an attempt to "destroy our country" with dreaded Brain Drain. 

4. While the political effects of Tyler's proposal are messy, the economic effects are fairly clear.  Only rich countries can afford higher transfer payments than the U.S. offers, so there would be no flood of desperate Third World migrants.  Still, hassle-free green cards would still look great to millions of overtaxed, fluent English speakers throughout the First World.  Over the course of ten years, I predict 15 million extra immigrants would come - though perhaps only half would stay permanently.

5. This prediction relies on my stereotypes about which countries would qualify under Tyler's proposal.  I guess Canada, Britain, Australia, Germany, France, and all of Scandinavia fit the bill, but do they?  Any important country I'm missing?

COMMENTS (8 to date)
honeyoak writes:

New Zealand,Netherlands, Belgium, Japan, Austria, Germany, Italy. I don't think Australia fits the bill as their tax rate is quite low.

Robert Kirchoff writes:

I think there might be a few exceptions to 4. depending on how that is calculated. Brazil's absurdly large welfare state maintained despite its middling economic position comes to mind.

As to Cowen's point overall, I'm not sure it would really change much in foreign countries, but it quite possibly could lead to lower political opposition to immigration at home. If this proposal were to break some of the passive opposition to immigration based on vague ideas of immigrant leaches, it would be a win in the short and long term.

Marginal wins are still wins!

Dan Hill writes:

@honeyoak - Bryan refers to transfer payments, not tax rates, though I suppose you need to look at both to understand the net effects. A lot of European welfare is the middle classes taxing themselves at eye watering rates to give themselves lots of "free" government services.

Not withstanding all that, as a dual Australian-US citizen, I can assure you that Australia would meet Bryan's test. The overall tax burden is higher in Australia (government revenues are higher percent of GDP), and the welfare system considerably more generous, especially for those under 65.

My question: How do you arrive at your prediction? Because it would be most certainly false. Unfortunately, we can't bet.

The countries you list have perhaps 300 million inhabitants. So you expect emigration of about 0.5% annually. German emigration peaked at such a level in 1881. And that was at a time when the US was way more attractive relatively than it is today.

The differentials are not that steep: Government spending as a percentage of GDP is 45.4% for Germany and 41.6% for the US. Cf.

There is no flood of desperate Swedes coming to Germany because of a higher differential (51.2% vs. 45.4%). And that's with perfectly open borders. In 2015, there were less than 20K Swedes living in Germany, net immigration from Sweden between 2012 and 2015 was less than 2K (about 0.02% of all Swedes).

Thomas B writes:

I can see a world of trouble in trying to define "more generous transfer payments".

Peter Gerdes writes:

Wait, why would even rational countries want to jack up transfer payments to take advantage of this?

Assume you are a resident of a rich country considering whether to increase your transfer payments to take advantage of this situation. Assuming that you, like a majority of your countrymen, do not anticipate emigration (both social and economic investment in your current location would be forfeit) what are the advantages?

Well, your unproductive countrymen would benefit most by staying where they are (not having to learn new languages, new systems etc..) and taking advantage of the higher transfer payments. In contrast some fraction of your most productive citizens (those graduating college, especially from US college and thus face the least cost of emigration) will be lured by the easy emigration and move to the US costing you tax revenue. That means a higher tax burden on you.

Now transfer payments back home might soften this concern slightly but I doubt it would be sufficient to undermine this analysis.

However, as the other comments point out the rate of emigration to the US is likely to be sufficiently low that other considerations dominate the choice of level of social benefits.


Also, given the various restrictions on immigrants receiving government benefits and the practical difficulty of determining level of transfer payments I suspect an average income floor would be sufficient to achieve the same results.

Maximum Liberty writes:

This is a little similar to something I'd like to see. This is not where I'd want immigration law to end. But the idea is that we could liberalize immigration law in ways that immigration opponents would probably not oppose. It starts with the casual observation that most immigration opponents would not be against fairly open immigration for "real" English, French, and Germans.

So the thought is to give green cards (or some other status that allows work) to anyone who
(a) is a national by birth of a country that
. (1) has a per-person GDP of at least at least 53% of the per-person GDP of the US and
. (2) is a democracy and
(b) was born to parents, both of whom were nationals by birth of that same country.

The GDP requirement means that we would not get a lot of economic migration. But I picked 53% because anything lower seems to start including places that at least many immigration opponents would oppose -- the first questionable one in my mental simulation of an immigration opponent typically being South Korea.

The democracy requirement means we would not get a lot of migration from places that immigration opponents think of as dangerous.

The by-birth requirements mean that those countries do not become waypoints for emigration to the US of people who are not culturally from those countries.

Using the lists on, we'd get potential immigration from such terrible places as Luxembourg, Switzerland, Norway, Singapore, Denmark, Ireland, Australia, Iceland, Sweden, San Marino, United Kingdom, Austria, Netherlands, Canada, Finland, Germany, Belgium, France, New Zealand, Israel, Japan, and Italy.

P. S.

The reason why 0.5% per annum is heavy emigration is that mostly people of young age emigrate. Average age for immigrants to Germany is 25 years. There are some 15% to 20% in an age bracket from 20 to 35 years. So emigration of 0.5% per annum, and 5% per decade means that a quarter to a third of the relevant group actually emigrate over ten years.

That's not some fringe group in a society, but indicates that a lot of people have given up on the country. Annual emigration of 1% to 2% would mean that young people in a society are running for the exits. The percentages might seem low, but they are not. That confuses a lot of people when they see historical or contemporary rates of emigration.


As I also tried to argue with you directly: Life in Germany is broadly on a par with life in the US. There may be some advantages and some disadvantages, but on the whole I'd even think that life in Germany is better than in the US. So there is really not a lot of pressure to move to the US.

For example, I'd rather stay in the 7th most liveable city in the world, have a 20 minute commute, and pay less rent for a large apartment than for a dog kennel in equivalent places in the US, than move to some boring American suburb for a few percent less in taxes and an extra room I don't need. Sorry, if I hurt somebody's national pride here. ;-)

Comments for this entry have been closed
Return to top