Tag Archives: effective altruism

Overview of the Open Philanthropy Project’s work on migration liberalisation

[A draft of this post was reviewed by Alexander Berger, Program Officer for US Policy at GiveWell, and a number of changes were made to it based on his comments and corrections.]

Charity evaluator GiveWell seeks to identify underfunded charities that can provide clear evidence of positive impact. Making their list of top charities therefore requires that one do good in sufficiently uncomplicated ways, ideally through a straightforward chain of cause and effect. Open borders activism does not fit this description. However, in early 2013, GiveWell (GW) broadened their focus to include less tractable causes through the Open Philanthropy Project, a joint project of GW and the philanthropic foundation Good Ventures (GV). Among a few dozen general causes including criminal justice reform and geoengineering research, “international labor mobility” was put on the agenda no later than in May 2013. This post will give an overview of the work the Open Philanthropy Project (OPP) has done in investigating and funding migration related efforts in the last two years.

A shallow overview of “labor mobility” was posted on GW’s homepage in May 2013. The page credits two specific sources with raising GW and GV researchers’ interest in this cause: Michael Clemens’s article “Economics and Emigration” (the origin of the “double world GDP” estimate), and the conversation that GW and GV staff held with Lant Pritchett in June 2012.

The Open Philanthropy Project’s assessment of free migration as a philanthropic cause

Since the inception of the OPP, the researchers’ stated position has been that labour mobility holds potential for very large gains, mainly in the form of large wage increases for workers who migrate from low-income-countries to high-income-countries. This is in line with Michael Clemens’s argumentation, although the OPP’s position is more guarded in its assessment of the magnitude of the gains, stating little confidence in the output of the relevant models. (Note that Open Borders bloggers have also argued for a lower estimate than Clemens’s.) A back-of-the envelope calculation provided on the GW website nonetheless states that it may be appropriate to consider the “importance” of labor mobility to be in the low trillions of $/year, based on the assumption of 10% as much migration as expected under full liberalisation in the models used by Michael Clemens. Efforts to facilitate legal migration through information sharing and coordination are estimated to hold potential corresponding to hundreds of millions of additional $/year, and the Senate Comprehensive Immigration Reform bill is estimated to represent a plausible US policy path that would carry benefits in the low hundreds of billions of $/year for future migrants (these gains would be realised in 2033 had the bill passed in 2013).

Characteristically concerned with room for more funding, the OPP’s assessment considers the extent to which the cause is already crowded by other philanthropic organisations. Policy work concerned with the treatment of undocumented immigrants in the US and with high-skilled labour for US businesses is seen as very crowded. Crucially, however, the OPP’s globalist humanitarian perspective sets it apart from the vast majority of active philanthropists working on US immigration policy, whose focus seems strongly influenced by citizenism and territorialism: The OPP’s focus is primarily on the interest of the immigrants, not on the interests of US employers in search of labour. And their priority lies with low-skilled immigrants, who have the most to gain from labour mobility. And here, the cause is everything but crowded.

The shallow assessment of labour mobility from May 2013 raises the possibility of important downsides of migration liberalisation as requiring research, and takes no position on this side of the issue. A post published in July 2014 states (citing a conversation with Michael Clemens and announcing a forthcoming writeup of the evidence)

our current understanding is that best evidence suggests that both lower- and higher-skill immigration are net beneficial for current residents, though they have somewhat different distributional effects.

On the 3rd of September 2014, Holden Karnofsky posted a draft writeup on the likely impact of increased immigration on current US residents’ wages, which the OPP had commissioned David Roodman to write, stating

We haven’t yet fully vetted this writeup (something we are planning to do), but we believe it gives a thorough and convincing picture of the literature, and provides some reason to believe that immigration is unlikely to result in substantially lower wages (particularly over the long run) for current residents.

(See also Open Borders’ reference page on the potential suppression of wages of natives.)

As for how highly they have prioritised this cause compared with the other philanthropic causes on their list:

An update on the Open Philanthropy Project posted on the GW blog on 26 September 2013 described “deep investigations” of 7 philanthropic causes as a crucial next step, involving proactive grantmaking. Labour mobility is on top of the list.

A much later post from 29 May 2014 on “Potential U.S. policy focus areas” groups labour mobility together with “macroeconomic policy” under the heading “Ambitious longshots: outstanding importance”, and places “deep investigation” of these two causes on top of the agenda, as investigation into the more time-sensitive “criminal justice reform” was being paused at that point in time.

A new Open Philanthropy Project update on US Policy related causes was posted on 10 March 2015. It states:

Our highest priority is to make a full-time hire for criminal justice reform, factory farming (pending a last bit of cause investigation, focused on the prospects for research on meat alternatives), or macroeconomic policy. Our second-highest priority is to further explore international labor mobility and land use reform, areas that we find conceptually very promising but in which we aren’t currently aware of (multiple promising-seeming) potential grant opportunities, and accordingly aren’t ready to make full-time hires in. These priorities are followed by several issues on which we have a relatively specific idea of what we could fund, and the next steps would be to investigate in much greater depth to decide whether the specific potential grants were worth making.

A spreadsheet linked to from last week’s OPP update explicitly gives “labor mobility” the highest importance out of all OPP causes. (See the  “Importance” column.) Unfortunately, this importance is not reflected by a corresponding number of funding opportunities.

Taking action

Since many of the causes taken on in the Open Philanthropy Project call for policy changes, GW’s and GV’s researchers have investigated expected costs and benefits of policy reform strategies. Vipul has written an Open Borders post about the conversation they’ve held on the topic with Steve Teles, and they have also held two conversations with Mark Schmitt. A series of  blog posts from October and November 2013 outline some general conclusions on policy oriented philanthropy.

As previously mentioned, the “deep investigation” of the causes was to involve proactive grantmaking. A blog post from May 2014 describes how GW’s and GV’s researchers came to adopt this approach:

from observing the behavior of potential grantees and other funders, we came to believe that a funder must be highly prepared (and likely) to make grants in an area in order to find giving opportunities in that area. Many people will only make the relevant referrals, propose relevant ideas, etc. once they are convinced of a philanthropist’s serious interest in providing funding.

The term “Earning to give” is often used in the Effective Altruism community, and I imagine the parallel terminology here is intentional:

“Giving to learn” can mean multiple things. It can mean (a) funding research in order to gain specific knowledge; it can also mean (b) funding a project in order to learn from following the project’s progress. The dynamic laid out in the above bullet points represents perhaps the most counterintuitive meaning: “giving to learn” can mean (c) offering funding in order to learn from the process of finding grantees.

[Update: Alexander Berger tells me the parallelism is not intentional.]

Three grants and one potential top charity

 The Center for Global Development (CGD) was awarded a grant for $1,184,720 over 3 years in March 2014.

This is the nonprofit think tank that employs Michael Clemens. As mentioned above, his publications were important in bringing the issue of labour mobility to GW and GV researchers’ attention.

In the aftermath of the 2010 earthquake in Haiti, Michael Clemens had advocated for making Haiti eligible for access to the H-2 temporary work visa program, as an outstandingly effective form of disaster relief. GW and GV researchers estimate that his efforts contributed significantly to the U.S. government’s decision to accept this proposal.

The grant will fund further research by Michael Clemens on “both marginal and more ambitious” changes to migration policy and its possible role in disaster relief. CGD will further use the grant money to launch a Working Group on Designing and Evaluating Bilateral Low-Skill Labor Mobility Agreements between high and low income countries. A Working Group on Creating a Migration-for Development Unit within the US Government will possibly also be launched.

While GW and GV are unsure of the marginal contribution the grant money will make to the CGD’s productivity in this area, they note that Michael Clemens’s work had very few sources of funding.

Follow-up is a crucial part of the  Open Philanthropy Project’s process. The writeup states that they “expect to have a conversation with Dr. Clemens every 3-6 months for the duration of the grant to learn about the status of his research and advocacy efforts, with public notes if the conversation warrants it.”

Notes on a conversation with Michael Clemens held on 21 January 2015 were published last week. Highlights:

Recently, most of Dr. Clemens’ time has been dedicated to three working groups and one study group:

  • A working group on a bilateral labor agreement between the U.S. and Mexico. This project has been funded by Good Ventures’ grant.

  • A working group on creating a migration and development bureau within the U.S. government. This project has been funded by Good Ventures’ grant.

  • A working group on implementing global skill partnerships. This project is currently stalled, and it is unlikely that CGD will become involved in any global skill partnerships within the next year.

  • The Beyond the Fence study group, focused on the indirect effects of the drug war in the U.S., Mexico and Central America. This group’s work has been fairly light so far.

Some details on the first of those working groups:

The exact output that the working group will produce is itself a subject of discussion. It may decide to produce a document outlining particular features that a practical agreement would require and suggesting research needed. This could build upon current bilateral, interministerial cooperation happening between the U.S. and Mexico.

(…)

A primary goal of this group is to design a better system for pairing migrant workers with employers than the current H-2A temporary agricultural worker program. Employers perceive the H-2A program as an obstacle. The U.S. Department of Labor could potentially create a pilot of a program that is instead a useful service for employers, similar to New Zealand’s Recognized Seasonal Employer Work Policy or the work of CITA Independent Agricultural Workers Center.

The second half of the conversation notes provides a lot of detail on Michael Clemens’s numerous migration related research projects. The last section of the document states:

Dr. Clemens does not have a good metric for determining the influence of his work. His papers are frequently included in course syllabi, and two of his papers in particular, “Economics and Emigration: Trillion-Dollar Bills on the Sidewalk?” and “The Place Premium,” seem to receive significant attention.

The U.S. Association for International Migration and the International Organization for Migration were awarded a grant for $1,490,500 over approximately 14 months in July 2014, for a jointly submitted proposal that will involve three further organisations. Among these is the Center for Global Development, which will conduct an evaluation of the program to assess its impact.

The grant will fund a pilot program to familiarise U.S. employers with Haitian lower skill workers, and ensure the legitimate uptake of available temporary H-2 working visas.

Potential upsides of the project include the continuation of the program after the pilot study, and policy changes in response to the results of the evaluation.

A December update reports that the first phase of this program has gone satisfactorily (one of the initial criteria for disbursing a second tranche of funding was waived, as it was recognised in hindsight as unrealistic), and announces the launch of its second phase.

The same document on the conversation with Michael Clemens on 21 January 2015 as cited above also includes two paragraphs giving further updates on this program:

Sarah Williamson (Protect the People) and her team have not yet finalized the employers who will participate in IOM’s program to bring Haitian workers into the U.S. via the H-2A program. IOM plans to take leaders of Haitian agricultural associations on a “study tour” of American farms, with the hope that farmers will put in orders for Haitian workers after meeting these leaders in person.

CGD is preparing to run a survey to measure the effects of the program. (…)

ImmigrationWorks was awarded a grant for $285,000 in July 2014.

Quotes from the writeup to ponder:

We were not able to find any advocacy organizations dedicated to making the case that more lower-skill workers should be allowed to migrate on humanitarian grounds, and experts generally told us that they felt that there was not a major constituency for such a message. The only groups we were able to find advocating for more lower-skill migrants represent business in some capacity, and they are relatively small or do not focus primarily on lower-skill immigrants (…)

Said groups numbered 3, counting ImmigrationWorks. Further:

our understanding is that ImmigrationWorks is the only one for which lower-skill immigration is the top priority, and that it is much smaller than the others.

ImmigrationWorks’ stated mission is to organise small employers of lower-skill immigrants, and mobilise them to advocate in Washington D.C. and across the U.S.

Their stated principles involve: bringing annual legal intake of foreign workers in line with “the country’s labor needs”, ensuring better enforcement of immigration laws, finding “a way to deal realistically with” existing illegal immigrants (which can be neither amnesty nor deportation, as those are both deemed “unacceptable”), and making sure that immigration policy is handled at the federal level.

The writeup acknowledges a (low) risk that ImmigrationWorks will use the grant to move policy in a direction that GW and GV would consider actively harmful.

Beyond closing the organisations projected funding gap for 2014, the proposed uses for the grant are:

  • Advocacy for immigration reform (…) that includes an ample less-skilled worker visa program, by mobilizing business to advocate to “business-minded Democrats and pro-immigration Republicans.”

  • Public opinion research (…) to try to determine which messages work to persuade people of the need for lower-skill immigrant workers

  • Building consensus around policy (…) with the business community

Conversations with IW founder Tamar Jacoby are expected “every 2-3 months over the course of the year-long grant.” No update has yet been published (which does not mean that no conversations were held, as notes are published only for a minority of conversations).

Migration within national borders

Domestic migration may not be of obvious concern to bloggers devoted to Open Borders, insofar as the obstacles faced by the migrants do not include any political borders. But the work on seasonal migration within low income countries that GW and GV researchers have been following and funding is quite relevant to Open Borders advocacy as well.

Bryan, Chowdhury, and Mobarak have run randomised controlled trials in  Rangpur, “a region of rural Bangladesh that persistently suffers from pre-harvest famines.” The trials were conducted over three years and involved 100 villages. This research finds that providing subsidies for seasonal migration can effectively increase migration and household consumption.

Evidence Action, the organisation that manages the GW Top Charity Deworm the World Initiative, is currently funding a 4,000 household study in northern Bangladesh “to explore further the potential of scaling up a migration subsidy program”.  The OPP has made a $250,000 grant to support this work in March 2014, with the stated aim of supporting the creation of future Top Charities.

A more specific goal of this research is to empirically investigate a number of questions on unintended consequences of migration – some of which are frequently discussed here on Open Borders:

  • Does sending many unskilled laborers to a single city change wages?

  • Does migration influence housing prices at destination cities?

  • What kinds of housing opportunities are migrants finding?

  • Does migration affect food prices in villages of origin?

  • Does migration change gender dynamics (e.g., what changes occur when women are left at home to manage home finances when men migrate)?

  • Are there are any unintended consequences for households who do not send a migrant?

Provided that the results of this research are encouraging with respect to scalability, Evidence Action intend to significantly scale up their seasonal migration support program. We can hope to see a funding proposal later this year.

Conclusion

I am very impressed with the Open Philanthropy Project’s work on labour mobility. It is exciting to read about the specific action undertaken, and I can imagine their sheer demonstration of initiative having considerable power to shift people’s thinking on migration.

The researchers’ careful evaluation both of the importance of the cause of migration liberalisation, and of the amount of effort currently invested in the cause, seem to me to strongly confirm the views generally held on these issues by Open Borders bloggers. To recap some relevant highlights:

  • The OPP come out prioritising the cause of free migration very highly. If other causes are currently prioritised more highly, the stated reason for this is always that they are able to identify more funding opportunities in these other domains. Thus, when it comes to launching additional efforts to further a cause, increasing freedom of migration between low-income and high-income countries seems to be a plausible candidate for “most high impact cause to take on”.
  • The OPP have found no political advocacy group in the U.S. that promotes immigration of low-skilled workers on humanitarian grounds.
  • The OPP have found only three political advocacy groups in the U.S. that promote immigration of low-skilled workers at all, and they all do so with the aim of “advancing the interests of U.S. businesses”.
  • In contrast, there is plenty of philanthropic engagement in immigration-related causes that are consistent with extreme citizenism (bringing in more high-skilled labour to advance U.S. economic interests) and territorialism (defending rights of existing immigrants, but not the right to immigrate).

Related reading

Some related reading from Open Borders: The Case and others:

Open borders advocacy: a Drake equation

The Drake equation is a probabilistic argument used to estimate the number of active, communicative extraterrestrial civilizations in the Milky Way galaxy. The idea is to express the number of such civilizations as a product of quantities in a manner that’s true by definition, but also such that one can talk somewhat more intelligently about estimating the individual factors than one can talk about directly estimating the number of active, communicative extraterrestrial civilizations. XKCD has poked fun at the Drake equation in at least two comics. Viewed as an exercise aimed at obtaining precise actionable estimates, the Drake equation is probably futile. But viewed as a way to start thinking about the problem, it is arguably useful. The main reason it’s bad for estimation is that the multiplicative nature of the model means that the huge uncertainty in measurement for each of the factors is also multiplicative, leading to a gigantic uncertainty in the overall estimation.

Here’s my Drake-like attempt:

$latex \text{Utility of a particular form of open borders advocacy} = Wxyz$

Here:

  • $latex W$ is the naive estimate of the gains from complete open borders (using, for instance, the double world GDP ballpark).
  • $latex x$ is a fudge factor to represent the idea that “things rarely turn out as well as we expect them to.” If we set $latex x = 0.1$, for instance, that’s tantamount to saying that, due to all the numerous problems that our naive models fail to account for, the actual gains from open borders would be only 10% of the advertised gains. The product so far, namely $latex Wx$, describes what we really expect the gains from open borders to be.
  • $latex y$ is the fraction to which the world can realistically move in the direction of open borders. The product $latex Wxy$ is total expected gain from however far one can realistically move in the open borders direction.
  • $latex z$ is the extent to which a particular effort at advocacy or discussions moves the world toward open borders, as a fraction of what is realistically possible. For instance, setting $latex z = 10^{-4}$ for Open Borders the website would mean that the creation of the website, and work on the website, has moved the world 1/10,000 of the way it feasibly could in the direction of open borders.

The restrictionist or pessimist might well view $latex x$ as a negative number, making open borders advocacy a great disservice to humanity. For our purposes, however, we’ll consider estimates where the values are positive, yet sufficiently small as to account for considerable uncertainty. Let’s say that, for the Open Borders website, the numbers look as follows, with the numbers in US dollars (note that of the four numbers, $latex z$ is the only one that requires particular knowledge of the Open Borders website):

$latex W = \$ 50 \text{ trillion}, x = 0.01, y = 0.001, z = 0.0001$

The 50 trillion figure can be calculated as just one year’s gain based on the double world GDP estimates. Note that there are some complications when considering potential delays in opening borders, as well as discount rates for the future and economic growth in the future. But since the starting numbers are anyway very rough guesses, there’s not much point in trying to do a very elaborate estimation exercise to calculate $latex W$ (for what it’s worth, I did some estimates based on assumptions about discount rates and economic growth, and I got a figure of about twice that much in expected value even if open borders are delayed by several years and the gains are slow to arrive and temporary). Note also that the fudge factor $latex x$ of 0.01 is essentially taking a very pessimistic view of the estimation exercise, by claiming that 99% of the claimed gains will not in fact materialize, or will be canceled by other losses.

With these numbers, the value of the website comes out to 50,000 US dollars. That’s not huge, but it’s about the same order of magnitude as the cost of time spent on the website (about 1500-2000 hours). With these numbers, therefore, the site just about breaks even in terms of social value generated versus time spent.

Here’s an optimistic version of the numbers:

$latex W = \$ 50 \text{ trillion}, x = 0.1, y = 0.1, z = 0.0001$

With this view, the naive estimate overstates the gains, but only ten-fold, it’s also possible for the world to realistically move 10% of the way toward open borders, and Open Borders the website has moved the world 1/10,000 of the way toward the theoretically possible limit. With these numbers, the expected value of Open Borders comes to about $50 million.

Obviously, the above estimation exercises are very naive, and there’s a sense in which this might feel like Pascal’s mugging. The key point that emerges here, though, is that the position yes, open borders would have gains, but the gains from what’s realistically possible in that direction are too small to be worthwhile isn’t a very tenable position. Open borders is a radical proposal — for better or worse. To arrive at such a position, you’d need to have $latex x,y,z$ all very small — but still positive. If you’re coming that close to zero, then you might as well offer some good reason why you don’t go all the way to zero — or beyond, to the negative territory. If the restrictionist position were right, then, it would entail showing at least one of these (or more precisely, an odd number of these, but never mind that):

  • $latex x$ is zero or negative: Economists have badly estimated not just the magnitude, but rather, the sign of the effect of open borders. The best attempts in the direction of demonstrating that the expected sign is negative is the killing the goose that lays the golden eggs argument. And while I think there’s considerable plausibility to that argument, and it may well point toward certain keyhole solutions being desirable, I am not convinced that these come anywhere near toggling the expected sign of the gains from open borders.
  • $latex y$ is zero or negative: It’s impossible to move in the direction of open borders at all.
  • $latex z$ is zero or negative: Open borders advocacy (or at any rate, the specific advocacy effort being considered) hurts the move toward open borders more than it helps. Tyler Cowen took this sort of approach in his recent blog post that generated considerable response (including from Nathan and John).

An alternative position is that we just don’t know enough to even estimate the signs of the quantities, and that more research is needed. I certainly agree about the need for more research, and I think a strong case could be made for an agenda that focuses extensively on research before clearly coming down on one side or another, while favoring continued experimentation with liberalization and keyhole solutions at the margin. But what’s not justifiable is the absolute certainty that many people seem to have that the status quo is approximately optimal, or that radical liberalization of movement simply isn’t a paradigm worth investigating because the gains are too small.