In a recent article about why a a guaranteed  income  won’t work in this country, Megan McArdle wrote that:

“There is no way that we are going to admit people to this country in order to hand them, and all of their descendants, a check for a thousand or two every month.”

It seems to be conventional wisdom that a basic income is incompatible with open borders.  Still, I am an advocate of both.  I understand that there is significant tension between them, so let me explain myself.

I may be preaching to the choir, but my primary reasons for supporting open borders are that I think it will result in increased economic activity, it will help many people escape poverty, and it may help avoid some of the tragic circumstances associated with living as an undocumented immigrant.

My reasons for supporting a basic income are probably a bit less familiar, and frankly they may sound a lot like some of the reasons that some people are opposed to open borders.  Namely, we have a duty to look out for our neighbors.

Just as we become vulnerable whenever we are close to someone emotionally, those who live near us gain a certain degree of economic and political power over us.  This is true even if they aren’t citizens.  If people work in our communities, the economy becomes dependent on them.  Thus, everyone who works has some degree of economic power in that they can refuse to continue working.  They also have some ability to actively disrupt economic activity.

Anyone who votes has political power, but even non-voters have some degree of political power because they can become part of a political conversation.  The closer they are, the more visible they are, the more likely it is that people will feel sympathetic to their concerns, and the more likely it is that political powers will take their interests into consideration.

Our duty to our neighbors becomes more pronounced in the face of high levels of inequality.  We cannot expect our neighbors to uphold the rule of law if they are starving.  What argument can I make to one who lacks food for their children that they ought not steal, other than the threat of violence?  Since my neighbors have power over me (for example, the potential to steal from me), I have a strong interest in making sure they respect the rule of law.  Thus, I have two options available to me in the face of high levels of inequality.  I can either increase my threats or I can make sure my neighbors don’t starve.

Let me clarify a bit about the moral responsibilities of starving people.  I personally am not a believer in absolute morality, but you might be.  I am not saying that you are wrong.  I am saying that if a moral relativist is starving and wants to steal from you, you are going to have a very hard time convincing them otherwise based on moral arguments.  The more desperate they are, the more that stealing (or cheating, or engaging in other anti-social behavior) might start to look appealing.

Pretty much every society uses some combination of both violence and welfare support.  But to the extent possible, I think we should always prefer the latter option.  Unless using threats is significantly easier than making sure people don’t starve, we should make sure people don’t starve.

So that is a basic outline of why I support a policy of providing a basic income for anyone living near me.  However, as Megan McArdle points out, giving everyone a basic income can cost a lot of money, and perhaps even worse, it can create a disincentive to work.

I do not take these issues lightly.  I believe that a disincentive to engage in productive work is one of the most serious downsides that a public policy can have.  Thus, my preferred basic income policy would take the form of a work subsidy (e.g., an expansion of the earned income tax credit program).

The simplest example would be to set some wage threshold, say $4,000 per month.  Anyone who accepts a job for less than this amount would be subsidized for half the difference.  Thus, for example, anyone who accepts a full time job that doesn’t pay anything would get a $2,000 check from the government every month.  People who earn more would pay taxes.

Such a program may have some enforcibility issues (people may take fraudulent full time “jobs” that don’t require them to actually do anything).  But people would still prefer to take higher paying jobs, and higher paying jobs would result in lower subsidies, so wage competition should mitigate some of the problems.

OK, so now that you know why I support a basic income, and what sort of basic income policy I prefer, we can get back to the original question.  Is this sort of policy compatible with open borders?

If it were the case that everyone who immigrated to the country just represented another $2,000 check from the government and tax revenues remained constant, the policy would clearly be unsustainable.  However, there is no reason to believe that the marginal immigrant has no impact on tax revenue.

The big question is: for a given level of immigration, are the marginal social externalities greater than or less than the marginal social costs?

I think most advocates of open borders tend to agree that in addition to the benefits that accrue to an immigrant from coming to the US, there are significant social benefits that are not captured by immigrants.  The simplest example is that those who hire immigrants profit from them.  So we should be asking ourselves whether immigrants are zero marginal product workers.

One of the big underlying reasons that I support open borders is that I think some societies are capable of employing workers much more efficiently than others.  That is, the same person working in the US has a higher productivity than they would if they were working in Haiti.

To the extent that workers are (sufficiently) productive, guaranteeing them a basic minimum income won’t really threaten to undermine our economic growth.  As long as our society keeps getting wealthier overall, we can support generous work subsidies.  Even if we don’t capture their productivity in income taxes, we can capture some of it in other ways (i.e., in taxes on the corporations employing them).  The problems arise if we end up guaranteeing the income of a bunch of non-productive people.

There are two parts to this problem.  The first is that people who are inherently non-productive may want to immigrate.  The second is that productive employment may require a certain level of capital, and immigration might outstrip capital growth.

Since I think that having high levels of local inequality is a big problem, I can see why one might be opposed to allowing a bunch of non-productive people into the country.  To mitigate this, we might only open our borders to those who can find productive employment.  But we shouldn’t let people into the country and then let them starve.  As long as people have an incentive to work, keeping people from starving is more efficient than keeping them in line using threats of force.

Limiting immigration to potentially productive people  won’t necessarily resolve the second issue (capital growth).   The main problem arises if there is some ideal level of immigration (based on the relationship between immigration and capital growth) and a basic income would push immigration levels past that limit.  While a basic income might impact actual immigration levels, I don’t think it will have a significant impact on the ideal immigration level.

Many open borders advocates question whether a sovereign nation has the right to control immigration levels.  I do not.  I think that letting people live near us gives them power over us and thus creates strong duties toward them.  It seems possible to me that some immigration scenario would actually overwhelm our society and economy, so we ought to at least think about what the proper level of immigration is.  However, I personally believe that allowing vastly more immigrants than we do now provides some “low hanging fruit” for economic growth and will improve many people’s lives.  If my belief that most immigrants are productive is true, there is no reason to think that allowing them to come would somehow undermine a policy of guaranteed basic income.

Basically, I don’t think that GDP is a zero sum game.  The more people we have, the bigger the pie will get.  As long as we don’t kill the goose that lays the golden eggs (that is, the ability of our economy to productively employ them), providing a basic income will be consistent with much higher levels of immigration.

Note: after reading Paul Crider’s recent post, I would like to note that while I do believe that IQ and culture have some impact on how productive immigrants might be, I am not an advocate of limiting immigration to those from certain countries or with certain job skills.  There are roles in the economy for many different kinds of people, and I don’t think the government should try to decide what kinds of labor we need to import.  I believe that the biggest threat to the “goose” is inequality that might result from having immigration rates higher than capital growth rates.  However, I also think that immigration is a cause of capital growth, so the relationship is complicated.

The suggestion that open borders would (or could) “kill the goose that lays the golden eggs” is in my view one of the strongest arguments against open borders. The argument is that at some level or pace of immigration, open borders could alter a population’s characteristics in so that the very institutions that make the rich world rich could be changed, to everyone’s detriment. An appealing aspect of the Goose argument is that it doesn’t implicitly discount the rights and welfare of foreigners to the absurd degree that most other arguments for restricting immigration do. Indeed if immigration somehow destabilizes the prosperity-generating institutions of the rich world, then the global poor would suffer the loss of aid and technology transfer. The Goose argument has been discussed on this site previously, but mostly in the form of concern about the IQs of immigrants. I find this form of the argument unpersuasive, largely because the universal history of early humanity was one of low IQs and grinding poverty. Differential IQs are unable to explain the sudden onset of both rising economic growth and rising IQs. But you can read more about the IQ Goose from my cobloggers here (including the references therein).

In my view the strongest form of the Goose argument is that the valuable institutions of successful countries rely on certain cultural characteristics that immigrant populations may lack. The cultural traits in question could include general social trust level, religiosity, individualism versus collectivism, the importance of the family in society, beliefs about social mobility and poverty, and so on. Importantly, culture in this context does not refer to specific overall belief systems or ways of life. In other words, in this post I won’t discuss concerns about, e.g., Roman Catholicism, except insofar as such identifiable belief systems are predictive of the more abstract traits mentioned above, like religiosity and family importance.

This doesn’t have to be moralized (and indeed it shouldn’t be): the cultural characteristics of immigrants could be rationally adapted to the institutions of their home countries. An example of this is the oft-cited lower levels of trust exhibited by individuals within some African societies. Low-trust cultural norms among immigrants in developed countries may be mal-adapted, but those norms were optimally adapted to centuries of slave-trading, where there was a constant threat of abduction for enslavement by one’s fellows.

So the concern about mismatched cultural traits is legitimate. Establishing this leaves the question how to proceed with the argument next. The language of “killing the goose that lays the golden eggs” suggests a strong argument, stressing dire, possibly irreversible consequences of permitting an excessive number of culturally mismatched immigrants. But one could also pose a weaker form of the argument, suggesting that, ceteris paribus, permitting too many immigrants from problem cultures will lead to a gradual deterioration of institutions. Appropriate responses differ significantly between the two Geese. I will argue that a realistic treatment of the facts is more consistent with the Weak Goose over the Strong Goose. I’ll begin with the strong version.

If it could be shown with a high degree of confidence that allowing in immigrants from other cultures would indeed destroy the institutions responsible for economic growth, the rule of law, and other desirable characteristics of the rich world, then the argument would succeed in justifying the control of such immigration. There would remain the powerful libertarian and humanitarian cases for free migration, so even the Strong Goose would succeed only in establishing the need to restrict immigration to such limits as are consistent with preserving particular valued institutions. And the argument doesn’t apply at all to immigrants culturally compatible with rich world institutions.

Unfortunately for the Strong Gooser, good evidence for institutional-destruction-by-immigrant-culture doesn’t seem to be in the offing. I found Alberto Alesina’s recent review of the literature on culture and institutions relevant (and fascinating in its own right). First, it should be noted that the literature confirms cultural persistence among immigrants.

By isolating the importance of institutions, the evidence coming from the study of second-generation immigrants implicitly shows that some cultural traits travel with individuals when they move to a society with different institutions and values. Therefore cultural values are persistent, and moving to a place with different institutions does not change them immediately, certainly not within the timeframe of two generations. This finding does not contradict the possibility that the “melting pot” could work; the empirical question is, at what speed do cultural values converge?

The problem is that causality runs in both directions: institutions also affect culture.  Thus there are observable differences in beliefs and preferences between the former West and East Germany, despite cultural uniformity before separation. The market, as an institution, can change culture by shaping incentives and changing what values parents might wish to foster in their children to ensure their success. Longer term, institutional structures from several generations ago correlate to the cultural characteristics we see today. Specifically, the inclusive and democratic polities of yesteryear tend to have greater levels of general trust and universal morality today.

Institutions and culture affect one another, and can  lead to multiple equilibria. Alesina provides the example of the way family importance (culture) and labor regulation (institution) influence one another. “An inherited culture of strong family ties leads to a preference for labor-market rigidities, but the latter in turn makes it optimal to teach and adopt strong family ties.” A weak-family/laissez-faire labor market equilibrium is the other possibility.

Culture and institutions are both subject to shocks. Growing up during a military conflict or during an economic recession results in observable cultural shifts (the former leading to greater in-group egalitarianism and the latter leading to more left-wing political attitudes).  Shocks can come from technological change. There is some evidence that the plethora of new occupations requiring hard work and skill engendered by the Industrial Revolution caused parents to instill middle class values in their children. Shocks can also be purely cultural, as with the feminist and civil rights movements.

Institutions can adapt and transform without shattering. The USA, for example, had open borders for a large stretch of its history, including its earliest years, when its institutions didn’t have the advantage of years of establishment. While institutions changed in that time (a lot has happened in America’s 200+ year history), they still were capable of supporting economic growth and rising living standards. Likewise, there are a variety of societies with different cultural values that are more or less successful.

The point of the above is merely to show that there is no simple, certain, monocausal path from sub-optimal culture to institutional destruction. Culture is just one of many variables determining the fate of societies. Strong Goosers demand that liberal immigration advocates prove that institutions will survive a massive influx due to open borders, but this burden of proof is inappropriately high. The effects of cultural influence are far too vague to support such a deal-breaking requirement. In any case, what would constitute proof?

The Strong Goose resembles the precautionary principle, which posits that some catastrophes are so severe that they must be prevented even at great social cost, and even before the magnitude and probability of the danger is properly understood. When viewed this way, it succumbs to the shortcomings of the precautionary principle. It is exaggerated by the cognitive bias that leads people to suffer (and dread) losses more than they appreciate gains. It’s also double-edged. The same fixation on a vaguely conceived, low-probability catastrophic outcome can be mirrored by vaguely conceived, low-probability positive outcomes. By expanding economic opportunities for individuals everywhere and enabling diaspora dynamics to fuel institutional reform in the poor world, open borders could plausibly end world poverty within two generations. The constant presence of viable exit options to safe and prosperous places already populated by diasporas could plausibly end major conflict in the world; people will leave instead of fight. Not opening borders and thereby ensuring the unnecessary persistence of poverty and conflict could be just as disastrous as the Strong Goose eventuality.

Any deleterious effects of cultural mismatch on institutions are likely to occur over generations. (Incidentally this is another reason why the cultural Goose is more compelling than the IQ Goose–IQs in the second generation will increase with better resources and education, fewer childhood diseases, and more stimulating environments, whereas cultural differences may persist). This is more in line with the Weak Goose, an argument which accepts that the malign effect of some immigrant cultures on institutions is one variable among many. The Weak Goose loses the urgency of the Strong Goose, but it’s far more valuable for its realism.

The best argument that Weak Goosers can make against open borders is that opening wide the gates all at once is unnecessarily risky. Societies with problematic cultural traits could be identified and immigration from those groups could be constrained so that their numbers never exceed some fraction of the native population. The irony with this approach is that individualism, one of the cherished cultural traits of the rich world, would be compromised. Aspiring immigrants would be categorized by their society of birth, regardless of their personal beliefs, histories, and merits. This could be addressed by using some other factor as a proxy for culture, such as a skills-based point system, as is currently done in Canada, or IQ requirements. While this would certainly be better than closed borders, the downsides would be the perpetuation of social class discrimination and the denial of those unskilled workers who could benefit most from immigration. The use of such proxies could also raise uncomfortable questions about how society values its native-born members who fail to live up to the standard.

Restricting immigration has social costs of its own. People will inevitably try to enter the rich world as long as it continues to offer opportunities. Keeping out immigrants who don’t have permission requires abandoning valued institutions like due process and equality before the law, as my co-blogger John Lee has discussed at length. It likely also requires changing the employment institutions to keep out unwelcome immigrants,  which could have deleterious effects on “middle class values” like hard work. In America, a more earnest effort to restrict immigration has turned ordinary law enforcement officers into immigration agents, effectively empowered to demand papers from anyone they suspect of being an immigrant, which often means ethnic profiling. This kind of policy can poison trust in communities with minorities. Deportations rip individuals out of their communities and sometimes even away from their families. This is inconsistent with fostering general trust in society.

One of the cultural traits of the rich world that is considered valuable for sustaining strong institutions and economic growth is “generalized morality”, to be contrasted with “limited morality”. The latter describes morality that applies to family or clan members or otherwise close associates while the former extends moral consideration to strangers. A market order of anonymous buyers and sellers requires this kind of morality, lest transaction costs blow up due to fears of defection. (Just think how easy it would be to shoplift if your scruples didn’t forbid it). Here is another tension with the valued cultural trait and its straightforward application to migration. The bodies of strangers strewn across the American southwest and lost at sea, shipwrecked in the Mediterranean illustrate the paradox of restricting immigration to preserve stranger-regarding morality.

Restricting immigration by appeal to the Weak Goose–warning that too much cultural influence from some societies could gradually weaken institutions–clearly involves some bullet-biting. But perhaps there are more helpful outlets for the Weak Gooser’s laudable caution. A diverse stock of immigrants from multiple source countries would reap the benefits of open borders while reducing the risk of cultural mismatch. Multilateral migration agreements in the style of trade agreements would likewise diffuse risk among several countries. Inclusive policies could more efficiently acculturate immigrants to the values and institutions of successful host societies. Natives of rich countries should also be discouraged from discriminating against immigrants, as such discrimination exacerbates social distrust.

Good institutions don’t necessarily stick around forever. Someone who has never considered the role of culture in the evolution and sustenance of institutions should revise their valuation of rapid border opening marginally downward, and favor somewhat more a selective and/or gradual approach. But the Goose argument isn’t decisive. In the end it must be appraised in the context of improving living standards, diminishing violence, and advancing democratic and market institutions all over the world. In other words, successful institutions do not seem to be on the retreat currently. Culture can and does change, and migration is one way for successful cultures and institutions to spread. It would be a shame if progress in the world were stymied out of exaggerated fears that the world’s best institutions are more fragile than they really are.