Open Borders and Adverse Selection?
A guest post by Tanmay Khale
Bet On It reader Tanmay Khale sent me a critique of Open Borders that I hadn’t heard before. Reprinted with permission. I’ll post my reply in the coming weeks.
Dear Prof. Caplan,
I have a quick question regarding your arguments in favor of open borders, and particularly the influence of adverse selection.
It seems to me that the main thrust of your argument is "the typical person who comes from a developing country to the US sees their wages multiply by 20 overnight, therefore we should let anyone who wants to enter the country (since they are better off with higher wages and society is better off since the 20x wage multiplier comes from increased productivity).” (To be clear, I know that you have a lot more to say than just this; this is just my one-sentence summary!)
But the argument above seems structurally similar to "the typical person who starts working at Jane Street sees their wages multiply by 5 overnight, so Jane Street should accept anyone who wants to work at Jane Street,” which seems wrong. Both of these arguments seem like “we have made money from lifting offers in market X; therefore, we should lift every offer in market X.” But this seems like begging to be adversely selected against.
It's true that if you lift every offer, you will get more instances of “positive selection”. You will trade with more people for whom the trade is a mutually beneficial transaction for you and your counterparty. But you will also get more instances of adverse selection!
Relatedly, a while ago Byrne Hobart wrote about why backtesting trading strategies is challenging, and while he didn't use "adverse selection" by name, I think it is what he was describing:
There's another element of transaction costs that's even harder to model, because the highest effective transaction cost comes from when you're running a strategy, and someone else is running a smarter version of that strategy. Suppose there's some signal that, as far as you know, leads to above-average returns 55% of the time. And suppose this signal is really made up of two signals, such that if they both fire the hit rate is 60% and if just one of them does, it's a coin toss. If you run the simpler version of this strategy, but a competitor runs the more complex version, what happens is that they routinely outbid you every time both signals say to buy, and you get to make the trade 100% of the time when returns are no better than chance. This is naturally hard to model because, if you knew exactly what you were looking for, you'd be running the smarter strategy.
The reason I bring this up is that in general, it seems like adverse selection makes backtesting challenging, and that's another general reason to be wary of an argument like "people in the past who have come into the US have seen their wages multiply by 20, therefore if we accepted vastly more people, then something like this would continue to hold in the future.”
Finally, I know that you mentioned in a podcast that you visited Poland after a huge number of Ukrainian refugees entered the country (something like 10% of the population of Poland in a few months?), and said "Poland is doing fine / great!" This would seem to be a point in favor of the argument that we can admit vastly more immigrants without fear of adverse selection.
But the Poland example reminds me of the following: Being single at age 40 is in general a negative signal about a person’s desirability as a partner, but that isn't true if you know that their spouse recently died of random causes. In the case of Ukrainian refugees, you have a pretty convincing story for why they are coming (that doesn’t have to do with negative signals about the individuals), and so this seems like exactly the kind of situation where you don't have to worry too much about adverse selection! (The Holocaust is another similar example.) For related reasons, a female acquaintance of mine sorts by reverse height when browsing dating websites for men, since she is indifferent about height, but height and personality are negatively correlated among available dating partners due to selection effects.
But if your position is rephrased to “we should accept as many immigrants as we can while making sure to vet that they will not be a net cost to our society,” then I just don’t think that this sounds very controversial? I do think (from opinion polls) that most Americans would endorse a statement like the above (Americans are overwhelmingly in favor of skilled immigration, for example).
When I say “net cost” above, I want to emphasize that when considering restricting freedoms (such as immigration), the only ways that someone could be a net cost to our society which I think are worth taking seriously are the ways in which they could deteriorate the commons, through crime or milder things (e.g. deterioration of liberal norms like democracy and free speech, if it turns out that it’s more pleasant to be homeless in the US than elsewhere, not-quite-crimes like public disorder / harassment, etc.). Again, I know that you give various arguments for why these costs are relatively minimal among immigrants who currently enter the US, but it is far from obvious to me that this would continue to be true if we accepted vastly more immigrants by lowering the threshold for vetting (because of adverse selection), and especially if we accepted anyone who wanted to come.
I am anticipating the objection that open borders are less like lifting every offer on the immigration market, and more like letting people transact freely across borders. Maybe you think “we aren’t paying to bring anyone who wants to come, we just aren’t preventing them from coming.” My immediate response is that the risks above are a form of cost / payment (paid by the US as a whole). Furthermore, when there is a common interest that people’s individual actions don’t necessarily maximize, I think that there is a role for the government is to make good “trades” to maximize this common interest (and not make the bad trades). I think deciding to let someone live in and transact with the US is such a trade, in the same way that deciding whether or not to let someone be a customer at your exchange based on your assessment of their credit risk is a trade; even if most customers make you money and improve the value of the exchange, adverse selection should make you a little paranoid to lift every available offer, especially since the cost (if the exchange goes under) is paid by all of the customers.
Best,
Tanmay Khale
In general, the larger the deviation one makes from the status quo, the more unpredictable the results. I would like to believe that we could move to something close to open borders. But my Burkean side says to proceed gradually. Radical proposals often make useful thought experiments, but I don’t trust them to work out in practice.
I think a stronger version of Bryan’s argument is something like: Here are a bunch of reasons why open borders could work out really well. Let’s take some incremental steps in that direction and see how it goes. If things go well, we’ll move further. If unexpected problems arise, let’s figure out how to mitigate them before pushing further.
This is reasonable. Emigration is happening in multiple countries and they are all offering different deals, and so it's a dynamic game with different effects on different players. Our current system is let people who don't respect rules in, or patient and people with the ability to and luck to get through the bureaucracy seems like we're not playing the best strategy. I take Brian's moral arguments seriously, and wish to live in a world in which people can move more freely, but you need to acknowledge that some folks have negative value and one goal of our system is to ensuring we have as few of those as possible. One by not letting them in, and two through creating a united culture that makes it makes it unlikely people born here become negative in terms of value.
My synthesis of Bryan and Garret jones is we can let as many people as possible who produce positive value, as long as we can create rules that accomdiate growth easily.
The other issue with Bryans arguments is that it justifies colonialism. Would Mexico be as rich as the US if we just took over and imposed our rules instead of theirs? If yes than immigration isn't needed, if no, then why would you assume immigration of their entire population add value? You can't have it both ways. I can see some ways to reslove this, but not many