Economist Tyler Cowen’s recent post was ostensibly about the labor market effects of immigration and emigration from OECD countries, but the latter half was devoted to a critique of open borders. Cowen:

And no I do not favor open borders even though I do favor a big increase in immigration into the United States, both high- and low-skilled. The simplest argument against open borders is the political one. Try to apply the idea to Cyprus, Taiwan, Israel, Switzerland, and Iceland and see how far you get. Big countries will manage the flow better than the small ones but suddenly the burden of proof is shifted to a new question: can we find any countries big enough (or undesirable enough) where truly open immigration might actually work?

In my view the open borders advocates are doing the pro-immigration cause a disservice. The notion of fully open borders scares people, it should scare people, and it rubs against their risk-averse tendencies the wrong way. I am glad the United States had open borders when it did, but today there is too much global mobility and the institutions and infrastructure and social welfare policies of the United States are, unlike in 1910, already too geared toward higher per capita incomes than what truly free immigration would bring. Plunking 500 million or a billion poor individuals in the United States most likely would destroy the goose laying the golden eggs. (The clever will note that this problem is smaller if all wealthy countries move to free immigration at the same time, but of course that is unlikely.)

The post seems to have generated a lot of buzz in the blogosphere (see here, here, here, here, and here for starters).

First off, although open borders advocates naturally concentrated on the latter half, it’s possible that Cowen actually intended to focus on the earlier half. The confusion about what Cowen intended to highlight is described in this comment by DJ10210:

Tyler’s strategy is interesting here. What’s the proper Straussian reading of this post? (A) Post is intended to be a critique of open borders proponents (e.g. Caplan), but opens with pro-immigration sentiment to signal that he is friendly to the cause he’s critiquing. (B) Post is intended to be a critique of immigration restrictionist, but closes with anti-open borders sentiments to signal that he understands that although he’s pro-immigration he’s not an extremist about it. (C) Both (A) and (B).

I lean toward (A) being the intended message.

I’m a great admirer of Cowen’s quality of thinking about empirical issues. In fact, right now, I’m reading his book, Average is Over, and I’m really liking it (I don’t have enough prior object-level intuition to have a strong view on the accuracy of Cowen’s predictions, but I find it plausible and well-argued). I felt that the post didn’t live up to the standard. So my first reaction to the post was to write something in between a criticism and a point-by-point response. However, after thinking it over, I see that there are a number of reasons why that would be misguided.

• Cowen write about five posts a day, in addition to his teaching, research, administrative duties, and books. His high quantity of reasonably thoughtful output is one reason why he attracts so many readers. But this also means that many individual passages in his blog posts are not subject to the same careful scrutiny and analysis that some other bloggers (such as Bryan Caplan, or, I’d like to think, the Open Borders bloggers) give their own posts. So even though I feel that Cowen wrote these passages somewhat hastily, it’s part of the package one gets with Cowen, and nothing to complain about.
• Cowen is in general skewed toward projecting an image of practicality and moderation, and that is part of what makes him influential as a blogger. This again is the package that his readers and those who choose to benefit from his wisdom sign on to.

With these in mind, I want to take a few minutes to note some possible messages people may take away from Cowen’s post, and why I believe these would be wrong. There is a subtext many people might be reading in Cowen’s text that open borders advocates are anti-empirical and careless and avoid obvious questions that anybody who thinks for a few minutes would come across. While I wouldn’t make generalizations about open borders advocates, I think that this site does not fit the stereotype. We have listed a wide variety of objections from both a restrictionist and a pro-immigration perspective, and attempted to address many of them — perhaps not to many people’s satisfaction, but I think it’d be fair to say that we haven’t ignored the issues. I think the menu options offer a reasonable summary (though doubtless the menu could be improved for better navigation, something that a co-blogger of mine will be working on). We have also discussed — more extensively than Cowen himself appears to have — the objections that Cowen raises in his post. If we haven’t covered a topic in sufficient depth, it is generally because (a) the existing literature and state of knowledge isn’t good enough, or (b) we simply haven’t gotten around it. We are very interested in the empirics of open borders — in understanding what might happen under borders that are open to various degrees. Let’s look at some of Cowen’s most remarkable claims.

Cowen writes:

Try to apply the idea to Cyprus, Taiwan, Israel, Switzerland, and Iceland and see how far you get. Big countries will manage the flow better than the small ones but suddenly the burden of proof is shifted to a new question: can we find any countries big enough (or undesirable enough) where truly open immigration might actually work?

A reader of this passage might believe that advocates of open borders are squarely disconnected from the empirical question of how many people would move under open borders, and that advocates of open borders seem to focus solely on open borders to large countries like the US. Neither assertion is true. Our world map for blog coverage shows how we cover migration-related issues around the world, including cases as diverse as Lebanon and Germany. Nor have we overlooked the significance of some countries being larger or having lower population densities than others. I made some very similar points about the dangers of extrapolating from existing data or historical experience in my blog post back in February 2013 titled open borders is a radical proposal. But for what it’s worth, the value of Cowen’s small country examples is unclear. For one, there does exist a large free movement zone — the Schengen Area, of which Switzerland is a part — and while there has been significant migration (enough to boost the case for the value of free movement) it has hardly been of cataclysmic or existentially threatening proportions. Or at least, that’s the way I interpret it. Does Cowen see things differently?

Cowen has much greater insight into the working of the world than I do, and possibly more than many of the other bloggers on this site. It’s possible that he has sound reasons for his intuition pertaining to Switzerland or Iceland or one of the other countries. It would be nice if he could elaborate more on these reasons.

Cowen also writes:

Plunking 500 million or a billion poor individuals in the United States most likely would destroy the goose laying the golden eggs.

How many would move under open borders? Cowen thinks the number is 500 million or a billion (and his language of “plunking” suggests that they’d all move more or less simultaneously and perhaps not even based on a conscious voluntary decision — but I’ll take that to be artistic license).

Now, I really like the fact that Cowen is providing a concrete estimate. It’s an important question, to be sure, because swamping is a major concern that moderate pro-immigrationers raise when faced with the prospect of open borders. And while there are many approaches (gradually increasing quotas, gradually lowering tariff rates to zero, gradually expanding a free movement zone, etc.) an answer to the abstract question “how many would move under complete open borders?” can be a useful analytical exercise in bounding the problem.

And it’s a question we have looked at repeatedly. We collected a number of links to polling data on migration — the best available data on the stated preferences of potential migrants (for what it’s worth, there are about 135 million people who want to move to the US if given the chance, and about 600-700 million people who want to move to a different country from where they currently are). I raised the “how many would move” question last July, and my co-blogger Chris followed up by asking a more specific question about open borders between Haiti and the US. These are the types of specific, concrete questions where somebody like Cowen can offer specific insight based on his deep understanding of the world — and elaborate on why he thinks open borders may be going too far. Offering the number is a great start. What I’d like from Cowen is an elaboration of how he’s getting at that number, what sort of timeframe he is talking about for the 500 million to 1 billion people, and how he thinks it might be a problem.

Cowen also talks about how open borders may be politically infeasible. We’ve asked this kind of question as well. For instance, this May, I blegged about whether open borders between the US and Canada might pass a referendum. And feasibility is certainly an important consideration when evaluating keyhole solutions.

Finally, the question of killing the goose that lays the golden eggs is an important one to us, and my co-blogger Nathan Smith views it as one of the potentially best arguments against open borders. Nathan wrote a three-part series (here, here, and here) attempting to defend open borders against this line of criticism. It’s one of the arguments we take more seriously on this website. Cowen probably has much to contribute to the discussion again, and I personally would really like to know more about what he sees as the biggest dangers to global innovation and technological progress that arise from moving in the direction of open borders, and how these might be mitigated.

Cowen has a cryptic parenthetical remark:

(The clever will note that this problem is smaller if all wealthy countries move to free immigration at the same time, but of course that is unlikely.)

The “of course that is unlikely” statement is puzzling. Of course, open borders is unlikely for the foreseeable future — whether for one country alone or for many countries together. The relevant question is not so much whether either is likely in absolute terms. The relevant question is about the relative likelihood of the US unilaterally opening its borders versus a number of countries opening borders together. I think history shows that the latter is more likely to happen — countries may form free migration zones, then gradually move to open borders for all. But I’m willing to stand corrected, since I don’t have strong knowledge here.

Perhaps Cowen’s concern is that open borders advocacy itself increases the relative likelihood of unilateral open borders relative to multilateral or universal open borders. I think that’s not the case at all. At least on the Open Borders site, we devote a fair amount of time to the immigration policy of countries around the world, including co-blogger John Lee talking about Malaysia. Does Cowen believe that the United States is uniquely susceptible to a few open borders ideologues promoting global open borders suddenly changing the minds of the powers-that-be? If so, that doesn’t square with what I believe, or what I think he believes, about the US political system. If the data on who favors open borders are any guide, the US is hardly in “danger” of any rapid shift towards open borders. The one rich country that may be in such “danger” is Sweden (also the first country to open its borders to Syrian refugees) but even Sweden has a fair degree of pushback against open borders. Note that, if anything, moderate pro-immigration advocacy tends to be much more rooted in country-specific rhetoric (such as “America is a nation of immigrants”) than the advocacy or discussion of open borders you’ll find on this site, and among other self-proclaimed advocates of open borders. (As a related aside, you might want to check our Carl Shulman’s post titled Open borders in (at least) one (developed) country on his personal blog, arguing that it might be better to attempt open borders in a single country with a relatively smaller population and then expand it to the world).

What I would like from Tyler Cowen is that, when he next discusses open borders, he gives the subject some of the same thought and attention that makes him such a great read on other subjects, and more importantly, that he share his reasoning (thereby avoiding the illusion of transparency and double illusion of transparency traps). Maybe there is a legitimate basis for his figure of 500 million to 1 billion. Perhaps Cowen has some interesting historical understanding that illuminates problems with open borders that we’ve overlooked. But we can only learn from his insight if he shares it.

A plausible response to the above is that it’s sufficient to rely on intuition here, because obviously what Cowen is saying is true. But it would be an inadequate response, given that Cowen himself is pushing back against the restrictionist intuitions expressed in his comment threads about immigrants stealing jobs from natives and turning their destination countries into economic basket cases. Intuition is a starting point, but to communicate and arrive at truths starting from one’s intuitions, it would be helpful to flesh out the rationales more explicitly.