Monthly Archives: January 2014

Joseph Carens on the ethics of immigration: part 1

4 Comments

In academic philosophical circles, Joseph Carens is well known as a proponent of open borders. His 1987 article Aliens and Citizens: The Case for Open Borders was included in our pro-open borders reading list since around the time of the site launch, and co-blogger Nathan blogged about the paper back in April 2012. We’ve referenced Carens quite a bit in subsequent blog posts.

I recently learned that Carens has given the philosophical issues surrounding migration the book-length treatment they deserve in the book The Ethics of Immigration (Oxford University Press, 2013). This is the first book-length treatment I’m aware of that deals with migration from a philosophical perspective and is written by a single author (UPDATE: As Paul Crider points out in the comments, Philosophies of Exclusion by Phillip Cole is an earlier book on the subject that I’d forgotten about. I haven’t read it, though). I was quite excited to hear about it, and read it with great eagerness. I found much food for thought in the book. In this blog post series (which may have two, three, or more parts, depending on the amount of material I end up wanting to write up) I will go over the parts of Carens’ book I found most interesting.

#1: Broad strategy followed by Carens

The book is not largely a defense of open borders. In fact, while the author does defend open borders, this is only a couple of chapters near the end of the book, and these chapters operate on somewhat different starting assumptions from the rest of the book. Rather, Carens spends the first ten chapters arguing within the status quo framework, i.e., assuming that it is just that the world is carved into nation-states and that states can exercise significant discretionary control over migration, but he also assumes that these are constrained by what he (inaptly?) terms “democratic principles” — more on that in #3. In the last four chapters, he critiques the status quo itself, and argues for open borders. He also defends himself against the charge of Trojan Horse-ing his way through. Chapters 1-8 come to many mainstream pro-migrant but migration policy-neutral conclusions, while Chapters 9-10 argue for for the right to family reunification and some rights for refugeees. Echoing Nathan’s view that a strong case for freer migration and more migrant rights can be made from communitarian premises, the bulk of Chapters 1-8 argues for migrant rights on communitarian grounds. This isn’t surprising, because communitarian grounds may be the only defensible framework that can simultaneously justify nation-states in the broad sense while still being compatible with moral egalitarian conditions. Roughly, the worldview Carens embraces is that everybody is equal, but many aspects of people’s rights are membership-specific (in relation to their communities) rather than universal moral claims, thereby permitting differential treatment (in some respects) by a state of tourists, temporary migrants, permanent residents, and citizens.

#2: Alleged target audience

Carens claims that his book is targeted at the median resident of the democracies of Europe and North America. This is an improvement over most migration-related books, that are often singularly focused on one specific country. However, I found Carens’ claim disingenuous in two ways:

  • I don’t see a good reason why universal moral arguments should not be applicable to people outside Europe or North America, and Carens’ limited targeting may be viewed as a version of the soft bigotry of low expectations — i.e., that people in India or Malaysia or Australia or Japan or Saudi Arabia or Singapore or Hong Kong or the UAE need to be held to a lower moral bar with respect to migration policy. Carens occasionally cites policies in places like Singapore, Hong Kong, UAE, etc. as policies that no sensible country devoted to “democratic principles” (more on that catchword later) would follow. Contra Carens, I believe not only that the case for open borders is universal, but also that any case that can be made for or against various migrant rights is universal.
  • Carens gives too much credit to the median resident of Europe or North America. The median resident doesn’t buy tracts from a university press that spend 300+ pages pondering over philosophical questions. About 15% of Americans are judged college-ready, and my guess is that the college-readiness benchmark would be a rough minimum to get through Carens’ book (you’d also need to be very interested in the subject). There’d of course be exceptions, but the percentage would overall be less than, not more than, 15%. This per se isn’t worrisome — authors often claim that their works have wider reach than they actually have — but it’s related to other things problematic about Carens’ logic.

#3: The “democratic principles” catchphrase

Carens uses the catchphrase “democratic principles” to describe beliefs that the median resident of Europe or North America might hold, but which seems to me to be (largely) shorthand for the ethical intuitions that people Carens interacts regularly with hold. To be clear, I’m no expert on the median person either, but a lot of the claims that Carens makes about how ordinary people think seem a bit off to me, judging by polling data I’ve seen. I feel like he’s slippery in roughly the same way Michael Huemer is when making claims about reasonable starting points for intuitions that most people hold.

For instance, Joseph Carens argues that it is obvious to any observer today (or at any rate, any observer who is faithful to “democratic principles”) that the Chinese Exclusion Act (CEA) was wrong, because it is obviously wrong to discriminate on the basis of nationality. While I agree that the CEA was wrong (see this lengthy blog post by co-blogger Chris Hendrix), it’s unclear to me that it’s significantly more “obvious” than open borders at large. If you embrace the principle that it’s wrong to discriminate on the basis of nationality to the point that the CEA is obviously wrong, haven’t you more or less embraced open borders (insofar as closed borders discriminate on the basis of nationality in a fairly fundamental way)? Further, to the extent that the CEA is condemnable on the grounds that it discriminated between different foreign nationalities, couldn’t the same be said of free movement within the EU (in that it discriminates between “other EU countries” and “non-EU countries” in its admissions policy)? Empirically, too, it’s unclear that people today have a strong view against the Chinese Exclusion Act. My impression is that the majority of Americans, if polled today, would be largely indifferent and consider it morally acceptable (even if unwise), rather than recoil in horror at the idea that such an act was passed a while back.

The remaining points are all arguments Carens makes presupposing the status quo framework, not necessarily ones he supports in reality, though every argument he makes moves in the “pro-migrant” and/or “open borders” direction once he takes off the hat of presupposing the status quo.

#4: Carens’ argument in favor of local legal equality

In a bow of sorts to territorialism and local inequality aversion, Carens argues that the same legal rules should apply to everybody within the physical territory, as opposed to a multi-tiered legal system. Carens does not propose an actual set of optimal policies, arguing that doing so would be outside the scope of the book. Rather, he uses a meta level argument. He argues that when a government (at a national or provincial level) chooses policies based on a balancing of considerations (e.g., choosing a minimum wage or labor regulation) the optimal policy that applies should be the same for natives as well as non-natives. Therefore, it makes no sense to have different labor regulations or policies for citizens and non-citizen permanent residents and temporary workers (a different policy for tourists is acceptable, because they’re not supposed to work). For instance, if minimum wage requirements are wrong, then they should not be applied to citizens either.

I see two objections to this, the first of which Carens anticipates to some extent, but the second he does not:

  • It can be argued that different subdivisions of the population based on citizenship/residency are statistically different, so the best balancing of interests would suggest different optima for them. This can be analogized to how the optimal labor regulation changes with time — changes with time change the nature of the labor work being done, or the skill level, and therefore change optimal labor regulation. Similarly, different segments of the labor force have different labor needs and different optimal laws.

    Carens addresses this (largely in implicit fashion). He argues that segmenting the force this way is not appropriate, any more than having different labor laws by race is appropriate. If different laws are needed, they should be based on the relevant criterion — occupation or skill level — rather than migration status. To the extent that natives and migrants have different optima, the best overall optimum should be considered.

    This, however, raises an interesting point that Carens does not acknowledge. To the extent that migration policy changes the composition of the labor force, it changes optimal labor policies for the whole labor force. If you’re having a single general minimum wage, and the value of the minimum wage depends on the skill level of the population as a whole, then if large number of people at low skill levels migrate, this could move the optimal minimum wage downward (for instance), for the population as a whole, including natives. Carens’ tone seems to suggest that the optimal policy can be determined just by looking at natives, and once non-natives are added to the mix, they just get subjected to the same policy. But if you’re insisting on one policy for everybody, it needs to take everybody into account. I don’t know if Carens would disagree, but he doesn’t really acknowledge the implications of this (so far) — the idea that changes may need to be made to regulation that move the First World in a potentially “Third World” direction to accommodate the changing composition of the labor force. This seems like the only reasonable alternative to having a two-tiered regulatory system. (As an interesting aside, opponents of expanded migration under the status quo, such as the otherwise pro-migrant Ron Unz, often support increased minimum wages as a way to deter migration).

  • Even if you believe that the optimal policy is independent of the population, the fact that the optimal policy for citizens is the same as the optimal policy for non-citizens doesn’t imply that the current policy for citizens (or for non-citizens for that matter) is close to that optimum. Therefore, moving the current policy for non-citizens in the direction of the current policy for citizens doesn’t make sense unless you already believe that that direction is the same as the direction of optimum. To take an example, suppose you believe that labor regulation X is bad (for everybody), but X applies to citizens currently. You have the opportunity to decide whether to support “not X” for non-citizens. Should you do that? (This also relates to the next point).

#5: Symbolic significance of reasonable measures undertaken in response to anti-immigration sentiment

Carens notes that there may be measures that are not wrong in substance but that have the symbolic significance of being anti-immigrant. He (tentatively) cites the UK’s tightening of birthright citizenship laws (to prevent tourists’ kids from getting such citizenship) as one example of such a measure. He doesn’t see the end result as morally wrong — he doesn’t think tourists’ kids prima facie deserve citizenship, but he believes that the move was in response to anti-immigrant sentiment.

To take another example (not provided by Carens), suppose you’re one of those who believes that “welfare creates a dependency trap that hurts its recipients more than it helps.” Would you vote for a ballot measure that sought to deny such welfare to some subclass of non-citizens? In your view, this denial would be in the non-citizens’ interest, but most likely the symbolic significance of it, and the perceived message, would be that the non-citizens are unwelcome.

#6: Against occupation-specific work visas

Carens offers an interesting argument against having occupation-specific work visas (i.e., work visas where the workers are restricted to a particular occupation). I don’t remember seeing the argument in that precise form before, though on this site we’ve obviously argued for a much more expansive vision of free movement than tying workers to a specific employer or occupation (see here for instance). I’ll take the liberty of paraphrasing Carens’ argument in a manner that will make both the argument and my subsequent critique of it clearer.

Consider these three types of prices of farm work:

  1. The price that farm work commands in the native labor market, without migration.
  2. The price that farm work would command if foreigners were free to migrate for work without being tied to an occupation.
  3. The price that farm work would command if foreigners could be hired to come on a visa restricted to farm work only.

Carens’ point is that (2) would be greater than (3), i.e., if workers had the option of competing on the entire labor market, they could probably command higher wages for farm work. Though Carens doesn’t explicitly say it, his language suggests that he thinks that (1) ~ (2), so that having occupation-restricted work visas distorts prices quite a bit, more than closed borders do. I think the point is theoretically interesting, and regardless of the empirics, is yet another reason to argue against occupation-restricted work visas (though they may still beat out closed borders). Going into the empirics would be too much of a distraction in the context of this post, but it would involve looking at the general issue of the impact that migration has on native wages. To a first approximation, wages are likely to fall in the sectors that experience heavier migration and rise in the other sectors. To the extent that workers are free to move between occupations, both as a matter of law and as a matter of skill level, this would ameliorate the sector-specific wage effects, so Carens’ point does seem to have prima facie merit. However, I still wouldn’t hinge the case for open borders on the general claim that (1) ~ (2), because it is quite possible that even with workers being legally free to move between occupations, wages for some sectors, such as farm work, do fall significantly.

This isn’t the end of my commentary on the book. I’ll be publishing part 2 of the commentary sometime in the next month.

How far are we from open borders?

1 Comment

I’m planning to write a multi-post series on how far the world as of now is from open borders. There are many different angles from which the question can be approached. In this post, I will provide a brief summary of the four major angles I’m considering. In future posts, I’ll elaborate on the individual angles. The four angles are:

  • Legal
  • How many want to move
  • How desperate people are to move
  • How different the world would look (economically, socially, etc.) under open borders

The idea behind the post is quite similar to the idea behind my earlier post titled open borders is a radical proposal. They differ both in rhetorical approach and in the particular points of emphasis. The earlier post focused on how open borders is, in many ways, historically unprecedented whereas this post focuses on how it significantly differs from the current status quo. Rhetorically, while the earlier post viewed open borders as the thing being judged in relation to the status quo, this post judges the status quo as a deviation from open borders.

Legal: Presumption and reviewability

Before looking at the status quo, it might be worth thinking about how an open borders regime might look like. Such a regime is not inherently incompatible with passports and visas. For instance, people need driver’s licenses to drive vehicles on roads, and the test is not completely trivial, but it is generally open and not too difficult for somebody who’s willing to work for it and take the test enough times. Democracies may require voter identification in order to allow people to vote, but they are still considered to have universal (adult) franchise if such identification is easy to obtain.

How closely a passport and visa regime comes to open borders would therefore depend on how procedurally straightforward it is to get a visa or equivalent permission to enter another country. In cases where this is just a matter of paying a small fee to have an application processed, we’d be close to open borders. In cases, however, where visas can be rejected for a variety of reasons, we’d need to start looking more closely at the list of reasons why a visa might be denied.

Legal theory has a useful concept called presumption of innocence, also known as innocent until proven guilty. The principle is generally applied in the context of criminal trials: the legal burden of proof rests on the state (the prosecuting party) that is trying to show that the accused is guilty, rather than on the accused to prove his or her innocence. Part of the justification for this asymmetry is the coercive and destructive nature of the punishment that people suffer once they have been found guilty. There is a strong presumption against forcibly making an innocent person suffer such punishment. The extent to which such a presumption exists, and should exist, is a matter of considerable debate, but the idea is straightforward.

In an open borders world, the analogous doctrine would be a presumption in favor of free movement, and the equivalent slogan would be “unrestricted until proven dangerous” — for approximately the same reasons: denying a person who expresses the desire to move to a new country the ability to do so is a significant infringement of the person’s freedom, and as such, deserves justification. There would be two components to this:

  • When denying a visa, a consulate would need to provide a specific reason for doing so and cite evidence in support of the reason. The evidence would need to be made available to the applicant.
  • The applicant would be in a position to challenge the consulate’s decision in front of a relatively neutral arbiter, who would hear out both sides and come to a decision.

Of course, just having the above doesn’t equate to open borders — the criteria may be very transparently stated but still very stringent. The same principle applies in criminal law: criminalization of a large number of victimless crimes, even if the law is executed fairly, can still be an indicator of an unjust and tyrannical society.

How far is the status quo from this open borders-like scenario? Very far. The United States is perhaps a somewhat extreme example, but not by a huge margin.

According to official estimates (linking HTML page), about 15-20% of applications across all nonimmigrant visa categories to the US in 2012 were rejected initially, and only about a third of the rejected applicants were able to overcome the refusal and get a visa eventually, resulting in a rejection rate of 10-15%. The primary reason for rejection is Clause 214(b): failure to establish entitlement to nonimmigrant status. In other words, the consular officer rejected the visa application on the grounds that the applicant might transition to long-term permanent resident status. Thus, not only does the US lack a direct route for most long-term migrants, it also coercively restricts people who want to visit the US for the short term (for work, study, or tourism) on the grounds that they might stay too long (with no evidence needed that such a long stay would hurt anybody). The US also has a doctrine of consular nonreviewability (see here and here): decisions by consular officers cannot be challenged by law or overturned by anybody, even the US President. Combine consular nonreviewability with Section 214(b), and the paradigm we basically have is the migration analogue of guilty until proven innocent.

How many want to move

I looked at this question in some detail in my earlier posts here and here. But here’s a quick summary: according to polling data on migration (the most recent available poll is here) about 13% of the world’s adults, or 630 million people, say that they are interested in permanently moving to another country. This is a huge number. In a world with open borders, there would still be people who are unable to move to another country due to personal issues, but it wouldn’t approach 13% of the world population. It’s safe to say that this is far from open borders. (The potential distinction between stated and revealed preferences is implicitly handled in the next point, which deals with how desperate people are to migrate). About 138 million people expressed a desire to permanently relocate to the United States. For contrast, the total annual number of people who migrate annually to the United States (through authorized and unauthorized channels) is a little over a million.

In his post titled Some Unpleasant Immigration Arithmetic, Bryan Caplan proposes an Open Borders Index as follows:

Open Borders Index = C/F

where C equals the total number of immigrants who enter the country every year, and F equals the total number of people who would annually enter the country under open borders. Caplan argues that C/F would be 0 under perfect closed borders and 1 under perfect open borders, and therefore it provides a normalized measure of border openness. He estimates that the C/F ratio for the United States is about 0.05 (i.e., about 20-30 million people would migrate to the US annually under open borders), and that the United States is thereby far from open borders. While the specifics of Caplan’s estimate can be disputed, the general idea suggests that the United States in particular is quite far from open borders.

How desperate people are to move

Desperation can be measured by the amount of resources people invest, relative to their current financial situation, to migrate. On the side of migration via legal authorized channels, this includes the fees that people pay as visa fees and lawyer fees. On the side of migration via unauthorized channels, this includes coyote fees as well as fees for document fraud that people who enter in an authorized fashion may pay in order to overstay their authorized stay. The cost measures need to be viewed in conjunction with the number of people who are willing to pay these costs. All these measures point in the direction of the world being quite far from open borders. Coyote fees from Mexico to the US are in the $3000-4000 range, and there are estimated to be millions of illegal immigrants from Mexico to the US, many of whom were smuggled via coyotes (others overstayed legally obtained visas). Coyote fees from China to the US have been estimated at $75,000, and although there are fewer Chinese who use coyotes to get into the US, the number is nontrivial. Note that coyote fees are an underestimate of the costs of moving, because migrants moving illegally often need to take other precautions in order to avoid being caught, and often need to tolerate inhumane conditions during the course of their movement – all costs that would need to be factored in. Finally, these fees should be considered in relation to their home country income. For the profile of people that migrate illegally from Mexico to the United States, coyote fees are generally equivalent to about 1-2 years’ worth of their current income.

How different the world would look if we had open borders

Finally, let’s consider the impact on economic output. Again, the estimation exercise is tricky because of the significant deviation we’re making from reality. A literature review by Clemens (2011) cites estimates suggesting that removing barriers to global labor mobility would yield world GDP gains anywhere between 67% and 147.3%. To rephrase, Clemens estimates that the status quo is shrinking world GDP to somewhere between 40% and 60% of what it might be under open borders. In the median case, open borders would “double world GDP” or equivalently, closed borders are “halving world GDP.” In contrast, ending all trade barriers is estimated to raise world GDP by about 5%. This isn’t surprising. Labor is a large share of the economy, and a lot of the world’s labor is confined to relatively unproductive segments of the world economy. Freeing people to move to places where their labor can be used better would lead to more production. How much more is debatable, but an estimate of doubling world production isn’t completely out of the realm of possibility when viewed in conjunction with the very large number of people who want to move. These same estimates also suggest that much of the gain in production – and consumption – would be experienced by the world’s currently poorest people, leading to a significant reduction in, and perhaps an elimination of, world poverty. If we take utility to grow logarithmically with income, then this distributional aspect argues even more strongly in favor of the idea that open borders would increase global utility tremendously. Open borders would also significantly reduce global inequality. For instance, a paper by Branko Milanovic estimates that under the status quo, country of origin accounts for 2/3 of global inequality (controlling as best as possible for other attributes). While the country of origin would still play a significant role in global inequality under open borders, there’s strong reason to believe that the fraction of global inequality accounted for by country of origin would be far lower than it currently is.

The cultural, social, and political effects of open borders are harder to quantify, but their existence is undisputed. To a large extent, the pushback to open borders is precisely because of the huge perceived cultural, social, and political changes that might be unleashed through open borders. Whether these effects are a net positive or a net negative is a more difficult question that the site at large is devoted to, and is beyond the scope of this post. What’s important is that the effects are significant, indicating that the world is far from open borders in a meaningful manner.

Influencing policy in the direction of open borders: moderate versus radical approaches

6 Comments

I recently responded to Tyler Cowen’s latest critique of open borders. While thinking through my response, my mind wandered over to the question of what the optimal way would be to influence policy in the direction of open borders, and whether one could make a sort of Straussian argument that the best way to move toward open borders is to eschew it overtly, and instead support moderate pro-immigration groups. Since I founded the Open Borders website (even though I’m no longer blogging regularly for the site) this raises the question of whether I did the world a disservice. At any rate, it raises the question of whether I could, instead of spending the several dozens of hours setting up the site, simply have used that time to earn more money at the margin and then donated it to an existing moderate pro-immigration group such as those listed here, or to a lobbying group such as those listed here.

One potential response here is that advocating and/or discussing open borders is virtuous in and of itself, even if it’s very unlikely to have any effect on the status quo. This is the view of open borders advocacy as a part of virtue ethics (cf. Bryan Caplan’s post on his two modes). This response is, however, not the whole story. Advocating and discussing open borders is unlikely to cause any major short-run changes. But, as I attempted to demonstrate in my clumsy imitation of the Drake equation, the expected value of open borders advocacy could still be huge. Roughly, this is because even though the extent to which the world can move toward open borders is very small, the gains are so huge that even multiplying them by a small number still gives a relatively large number that exceeds the cost of trying.

There are two broad strategies that seem to suggest themselves, namely advocacy of the two ends of the moderate versus radical axis:

  • Switching to moderate pro-immigration rhetoric that directly appeals to the masses.
  • Targeting the intelligentsia with our radical ideas, but presented well in a manner that might affect their thinking and writing, even if they don’t explicitly link to us or endorse our proposals wholesale.

Note that the moderate versus radical axis differs conceptually from the direct (try to directly reach out to the public or people in power) versus indirect (try to make general arguments and let things percolate to the masses over time) axis. However, there is a close relationship: direct approaches are statistically more likely to be moderate, simply because it’s difficult to do direct work on a radical stance. That said, some forms of direct action may implicitly carry an extreme pro-open borders message.

In subsequent blog posts, I’ll consider and compare the historical track records of moderate and radical approaches. In the meantime, I’m curious as to what readers think. Of course, one could say “we need both” but I’m looking for an answer to a more specific question: what does your intuition say about which approach is more valuable at the margin? If you think that the approach that’s more valuable at the margin depends on the skill sets of the people involved, what skill sets do you think are more conducive to each approach?

Weekly links roundup 02 2014

Leave a comment

Here’s our weekly installment of links from around the web (see here for all link roundups). As usual, linking does not imply endorsement.