Conversation between Michael Clemens and GiveWell

Post by Vipul Naik (regular blogger and site founder, launched site and started blogging March 2012). See:

Charity evaluator GiveWell often carries out conversations with subject matter experts in a number of areas relevant to the philanthropic directions that GiveWell is considering. Notes from some of these conversations get published on GiveWell’s conversations page.

Recently, GiveWell staffer Alexander Berger had a conversation with Michael Clemens. Clemens is in charge of migration and development at the Center for Global Development, a Washington D.C. think tank. A full PDF of the conversation can be found here.

The part that interested me most in the conversation was the section on migration from Haiti to America:

After the 2010 Haiti Earthquake, the United States strengthened its naval blockade of Haiti to prevent Haitians from moving to the US. Together with others, Clemens worked to loosen the restrictions on a US visa program to allow more Haitians to move to the US legally as a part of the post earthquake relief and recovery effort.

For historical reasons, before 2012 Haitians were not eligible for H-­‐2 visas. After the earthquake, Clemens and his collaborators made a very substantial effort to make Haitians eligible for H-­‐2 visas.

  • They pointed out that if 2,000 Haitians were to get H–2 visas every year,over the course of 10 years, the total amount earned by the migrant workers would be more than $400m, a total exceeding the US relief emergency assistance after the earthquake.
  • They had meetings with the staff from both Florida senators and several Florida Representatives, with the office of John Kerry, with the White House, the State Dept., with several nongovernmental organizations and with the Haitian ambassador.
  • In the end, they were able to get bipartisan support for the proposed policy change. Bill Nelson (a Democratic Senator from Florida) and Marco Rubio (a Republican Senator from Florida) jointly signed a letter to Janet Napolitano (United States Secretary of Homeland Security) requesting that she make Haiti eligible for H–2 visas, and she implemented the change.
  • Last year 58 Haitians obtained H–2 visas. as. The reason that more didn’t obtain visas is the fact that there exists no mechanism to match US employers who need Haitian labor to Haitian workers willing to come to the US on an H–2 visa. No H–2 recruiters are currently active in Haiti, and none has yet been willing to become the first mover.

The next section is about the “Potential for scale-up” of this successful advocacy:

The direct impact of changing migration policy so as to allow 2,000 Haitians to migrate to America each year is very small relative to the scale of poverty in Haiti (which has 10 million people). However:

  1. Before the change, there were no legal migration opportunities for Haitians except for the very small fraction of Haitians who were university educated or who have close family members in the United States.
  2. The success of the effort serves as a proof of concept proof of concept for other advocacy efforts. Before the policy change was secured, many people thought that it would be impossible given that it was an election year and that the US unemployment rate was as high as 10%. The fact that the change occurred despite expectations to the contrary raises the possibility that migration and immigration advocacy in general may be more impactful than has been thought in the past.

I was struck though by the very small number of people who actually made use of the existing system (and in general, by how existing H2 visa quotas go unfilled). Perhaps these numbers should cause one to update one’s priors away from large-scale legal migration under “open borders with keyhole solutions” type policies such as DRITI. In other words, maybe even apparently small frictions can significantly impede migration flows, so a complicated tax/tariff system may dissuade a lot of the migration projected under open borders. If so, this cuts both ways. On the one hand, it allays restrictionist concerns about getting swamped. On the other hand, existing double world GDP estimates simply don’t work if very few people move.

Another part that struck me was how Clemens responded to the idea that open borders is a radical proposal:

About 11% of the American population is foreign born. Those who are opposed to additional migration in the United States often voice concerns that if the migrant population were to rise substantially, the migrants’ lack of common culture and transient presence would cause society to break down. However, there are many examples of well functioning developed countries that have significantly higher migrant populations than the United States does:

  • Australia’s population is 24% foreign born, and Australia is beautiful and has low (5.6%) unemployment.
  • Toronto’s population is 45%-50% foreign born, with the foreign born population consisting primarily of people from developing countries. The city is safe, prosperous and culturally diverse.
  • The UAE has a huge migrant population. About 85% of the population consists of foreign non citizens, and about 98% of the private workforce consists of foreign noncitizens.
  • Singapore, Saudi, Qatar and Kuwait also have heavy international labor forces.

I don’t know enough about the context of the discussion to offer my critical thoughts, but my view on the matter is that these examples don’t do much to allay the concerns of overcrowding and swamping. In my previous post, I pointed out that the extremely low population densities of Australia and Canada undermine their utility as examples. The UAE and other country with significant guest worker programs (which I gather are somewhat similar to what our thoughtful critic BK would endorse) seem to provide stronger support for the idea that open borders with appropriate keyhole solutions would work. Another problem is that the existing differences within the developed world are already more than realistic pessimistic estimates of the effects of open borders. For instance, GDP per capita in Australia is about 10-20% less than in the United States. One might say this is mainly because the US had a head start, but one could also argue that that 10-20% GDP difference may be because Australia has a higher foreign-born population proportion, and that increasing the foreign-born population share of the US may result in a similar perilous 10-20% GDP decline in per capita income. If this decline is mostly due to compositional effects rather than existing natives getting poorer, it is not a bad thing. At any rate, I’d be interested in knowing a lot more about Clemens’ thoughts on the question of what cross-country comparisons say about the viability of open borders.

The conversation concludes with a discussion of why there isn’t more open borders advocacy:

Clemens is not aware of any organizations that have the stated goal of open borders in the United States. Lant Pritchett has asked why people don’t hold demonstrations in Washington DC in favor of labor mobility. Clemens’ best guess as to why there isn’t more interest in the subject is that the issue isn’t salient to people. The people making the decisions about labor mobility are almost entirely immune to the direct consequences of those decisions.

The post had very little discussion of the actual pros and cons of migration and migration policies, possibly because both GiveWell and Michael Clemens already know most of the basic arguments in this regard. Hopefully, future conversations (with Clemens or with others) will delve deeper into some of the specific pros and cons. Currently, in addition to the Clemens conversation, GiveWell has a conversation with Mushfiq Mobarak. Also, part of GiveWell’s conversation with David McKenzie deals with migration.

4 thoughts on “Conversation between Michael Clemens and GiveWell”

  1. The examples given of successful high migration are:

    1) Countries bringing in highly selected migrants who bring up or maintain average human capital in the receiving country, so that integration into democratic welfare states as citizens does not hurt natives (via transfer programs, politics, and crime) .

    2) Countries that adopt strongly citizenist (rather than territorialist) stances and strongly restrict the legal, welfare, and political rights of temporary migrants, using measures that open borders advocates would call draconian to make sure temporary really is temporary.

    What’s missing is:

    3) Countries that have received mass immigration of poor-quality migrants, who would be screened out by the selective permanent migration systems of Canada or Australia or Singapore, with the migrants being made into politically powerful citizens.

    One can make analogies to demographically diverse countries, the end of colonial or European rule in some countries, but this is what has not been tested.

    “Australia’s population is 24% foreign born, and Australia is beautiful and has low (5.6%) unemployment…Toronto’s population is 45%-50% foreign born, with the foreign born population consisting primarily of people from developing countries. ”

    http://en.wikipedia.org/wiki/Demographics_of_Toronto#Immigration_patterns

    Immigration to both Canada and Australia is highly selected: immigrants have high levels of education and skills relative to the sender countries, and are disproportionately drawn from high-IQ populations.

    Again, take a look at these numbers on education levels for migrants to the OECD countries vs education levels in the sender countries:

    http://perso.uclouvain.be/frederic.docquier/filePDF/DM_ozdenschiff.pdf

    42.6% of emigrants from Sub-Saharan Africa have tertiary education, versus 2.8% of residents of SSA.

    For South-central Asia (India and surrounds), 52.5% of migrants fit the bill, versus 5.0% of residents.

    For the Caribbean, 38.6% vs 9.3%.

    For East Asia 55% vs 6.3%.

    Because of the immigrant mixes in Australia and Canada, migrants and their children are correctly expected to on average be net taxpayers and support rather than drain the public fisc, to provide good peer effects in the schools, not to commit much crime, and generally to be good neighbors. There is less worry about less selected illegal immigrants in Canada, which is far from poor countries, and the island nation of Australia.

    “The UAE has a huge migrant population. About 85% of the population consists of foreign non citizens, and about 98% of the private workforce consists of foreign noncitizens.”

    The UAE is extreme in the citizenism/monarchism of its policies. It freely deports migrants, throws them in prison if they can’t pay their debts, uses brutal policing tactics, rejects birthright citizenship and the logic of the DREAM act, gives migrants no political power, and is a monarchy. The natives don’t get private sector jobs, but are bribed with huge numbers of make-work public sector jobs reserved for natives to make them accept the system.

    The bloggers here are opposed to almost every element of the UAE system, except the high number of migrants and perhaps compensation to natives.

    “even apparently small frictions can significantly impede migration flows”

    The failure of many people to take advantage of “permanent residency for cash” schemes when bureaucratic hassles are high supports this too. Current immigrants are unusually high in patience and other desirable traits that let them overcome such frictions.

    “complicated tax/tariff system may dissuade a lot of the migration projected under open borders”

    Looking at the different programs, it seems that delays, uncertainty, requirements for paperwork that is hard to assemble, and especially the involvement of host-country 3rd parties are the big problems. The U.S. could probably increase immigration rather heavily just by increasing its staffing levels to process applications immediately, in the style of Singapore.

    Paying a cash entry fee or posting a bond, biometric identification, and issuance of an identity card can all take place very quickly indeed. Also, with really open borders, there would be economies of scale that would favor the emergence of large organizations selling assistance with the migration process, loans for fees and bonds, and so on. The problem described for Haiti is a diseconomy of small scale.

    1. To avoid missing the tree for the forest: more H-1 and H-2 workers getting recruited and then going home with their wages is good for all concerned, and if an immigrant-recruiting organization such as CITA does that, it would be a positive use of charitable funds, even if it is behind saving lives from infectious disease.

Leave a Reply