Tag Archives: Federation of American Scientists

Understanding Open Philanthropy’s work on migration policy

1 Comment

NOTE: I sent a draft of this post to Open Philanthropy for comment. They wrote:

We may share an update on our plans for future work on this cause at some point, but don’t have anything we’re ready to discuss publicly just yet.

My co-blogger Sebastian wrote posts in 2015 and 2016 reviewing Open Philanthropy’s thinking and grantmaking work in the migration space. It’s now five years since the last post so I thought it’d be worth revisiting the body of work.

This post is structured a bit differently than Sebastian’s posts, which went into detail on the individual grants. Since we now have a larger set of grants (22 grants directly identified as being related to migration policy) it’s not that valuable to discuss them individually, or even the ones made since Sebastian’s last post. Instead, the post focuses on Open Philanthropy’s overall strategy changes, how these changes are rooted in their broader worldview and strategy, and the implications of these changes.

The post is structured as follows:

  • Evidence that Open Philanthropy is reducing its involvement in and commitments to migration policy
  • Factors influencing the reduced involvement
  • Discussion of the implications, both in terms of what we can learn from Open Philanthropy and what it means for the crowdedness or lack thereof of the space

General links for reference:

Evidence that Open Philanthropy is reducing its involvement in and commitments to migration policy

Implicit evidence from the pattern of grantmaking

A few lines of evidence are suggestive:

  • The grants database includes only one grant in 2021 (of $600,000 to the Federation of American Scientists, which is a different flavor than past grants) and otherwise no other grants since November 2020.

  • Open Philanthropy has been using exit grants as well as reducing levels of commitment even for grantees that they are continuing to support:

    • The March 2020 grant to the Center for Global Development for the migration program (led by Michael Clemens) appears to reduce ongoing support for the program from the $600,000/year for the previous grant to a funding level of $200,000/year (after one year at the old funding level) that “may be more sustainable for the long run.”

    • They exit-granted Labor Mobility Partnerships (LaMP) in November 2020 (see note below on the LaMP exit grant being a little different and possibly not evidence of general exit from the space) and Niskanen Center in January 2020.

    • They had previously exit-granted Protect the People back in 2016, though that exit grant was tied more to providing initial philanthropic funding to a project whose goal was to be sustainable. More details on Open Philanthropy’s learnings from the program are here.

The exit grant to LaMP is a little different from the other exit grants: I have it on good authority that Open Philanthropy did not intend, even at the time of incubation, to support LaMP in the long term — just to get it off the ground. The “exit grant” should therefore not be thought of as a reduction in a previous planned commitment, and therefore shouldn’t be thought of as a sign that Open Philanthropy is trying to exit the migration policy space. It is, however, tentative evidence suggesting that the level of future grantmaking may be less than the level of grantmaking in 2019 and 2020 (when LaMP was receiving grants).

A counterpoint to the narrative of exit is a look at the raw numbers which actually shows 2020 as the biggest year in terms of money donated for migration policy (at $3.7 million donated). That number alone is misleading because grants often have multi-year timeframes, and depending on when the first grant was made and what the periodicity is, the grants can sometimes bunch up in a year. Indeed, this was the case for 2020:

  • The March 2020 grant to the Center for Global Development happened at the end of the three-year timeframe of the previous March 2017 grant. The amount of the March 2020 grant was less than that of the previous March 2017 grant, as discussed above.

  • The January 2020 exit grant to Niskanen Center was at the end of the two-year timeframe of the previous January 2018 grant. The amount was also half — reflecting a shorter period of support, but the same rate per year.

  • The August 2020 exit grant to Labor Mobility Partnerships was a little before the end of the 1.5-year timeframe of the previous 1.5-year March 2019 incubation grant. The amount was also corresponding lower but at roughly the same rate per year.

Based on giving in 2020, the two main programs that Open Philanthropy seems interested in continuing to support on an ongoing basis are the Center for Global Development migration program (at $200,000/year) and the International Refugee Assistance Project general support (at $500,000/year).

One positive sign of increased giving was a grant to Mercy Corps for seasonal migration in Nigeria. Open Philanthropy has previously funded studies of migration around the world, and this seems in line with that past funding — but the magnitude of the Mercy Corps funding is significant at $1,000,000. It is unclear, however, if this was a one-off, and moreover, this is a bit outside what we normally think of as migration policy (as the barriers being addressed when promoting seasonal migration are more financial and educational than policy restrictions).

Explicit statements from Open Philanthropy

I could not find a lot of public discussion by Open Philanthropy of strategy changes tied to funding specifically in the migration policy space. However, grantmaking in migration policy is mentioned in their annual progress reports. These reports make clear that Open Philanthropy is interested in continuing to maintain support at relatively low levels in this cause area for now. What they don’t make clear is whether there are any changes to the relative importance over the years.

The 2017 progress / 2018 plans post says:

By default, we plan to continue with our relatively low level of effort and resources in other focus areas (e.g., macroeconomic stabilization policy, or “other” global catastrophic risks).

Although migration policy is not listed here, it is implicit, as becomes clear from the evaluation in the 2018 progress / 2019 plans post post:

We also wrote:

By default, we plan to continue with our relatively low level of effort and resources in other focus areas (e.g., macroeconomic stabilization policy, or “other” global catastrophic risks).

Other grants included the Center for Popular Democracy and the Economic Policy Institute (macroeconomic stabilization policy), the International Refugee Assistance Project (immigration reform), and California YIMBY (land use reform).

The 2018 progress / 2019 plans post sets a similar goal for 2019:

By default, we plan to continue with our relatively low level of focus and resource deployment in other areas (e.g., macroeconomic stabilization policy).

The 2019 progress / 2020 plans post confirms this:

We also wrote:

By default, we plan to continue with our relatively low level of focus and resource deployment in other areas (e.g., macroeconomic stabilization policy).

Other grants included the Center for Global Development (Global Health and Development), California YIMBY (Land Use Reform), the International Refugee Assistance (Immigration Policy), Employ America (Macroeconomic Stabilization Policy), and the Center for Election Science (other).

The 2019 progress / 2020 plans post sets a similar goal for 2020:

By default, we plan to continue with our relatively low level of focus and resource deployment in other areas (e.g., macroeconomic stabilization policy).

The 2020 progress / 2021 plans confirms this:

We also wrote:

By default, we plan to continue with our relatively low level of focus and resource deployment in other areas (e.g., macroeconomic stabilization policy).

Our giving in causes beyond those listed above remained at comparatively low levels. Grants in these areas included the Center for Global Development (Immigration Policy), Employ America (Macroeconomic Stabilization Policy), Mercy Corps (Immigration Policy), the International Refugee Assistance Project (Immigration Policy), and YIMBY Law (Land Use Reform).

And it says the same for 2021:

By default, we plan to continue with our relatively low level of grantmaking in other areas (e.g., macroeconomic stabilization policy).

How the evolution of grantmaking in migration policy compares with other cause areas

Certain cause areas have been prioritized by Open Philanthropy; this prioritization can be seen in the annual progress reports and is also reflected when we look at the pattern of grantmaking. Example areas include AI safety, animal welfare, and biosecurity and pandemic preparedness. Some indicators of continued focus in these areas include:

  • At least a few grants in 2021.
  • Large amount of annual grantmaking (at least several million a year).
  • Diversity of grantees (large numbers of distinct grantees).
  • Identification as a high-priority area in blog posts and public materials.

Migration policy clearly isn’t in league with these. But perhaps it’s better to compare migration policy with other areas that are more structurally similar: land use reform and macroeconomic stabilization policy. Here is what the comparison indicates:

  • All three areas get a similar treatment in Open Philanthropy’s annual progress / plans report.

  • Migration policy has slightly higher overall spend across the years, though with the dubious distinction of a lot of exit grants, which is more evidence of Open Philanthropy scaling back in the area.

  • Land use reform, like migration policy, has had a relatively quiet last couple of years; although the amount donated hasn’t fallen much, the diversity of organizations receiving grants in recent years has reduced. However, it doesn’t seem to have had explicitly announced exit grants, and one of the factors causing infrequent granting is the long timeframes for grants (meaning less frequent renewal). So there isn’t that much change to the number of organizations being supported if you consider whether they are stilll within the timeframe of a previous grant. The graphic here helps visualize this.

  • Macroeconomic stabilization policy, relatively speaking, is teeming with activity in terms of diversity of grantees.

My overall assessment is that migration policy is pretty comparable with land use reform, though with slight pattern differences. Macroeconomic stabilization policy is a little ahead right now.

Factors influencing the reduced involvement

When Open Philanthropy published its Labor Mobility cause report back in May 2013, it considered the area to have high potential. Similar cause reports were published for other areas. Over time, some of these areas “took off” with Open Phil deciding to spend much more on them, and others, such as migration policy and land use reform, continued at maintenance levels. What happened in the intervening years that affected this? Let’s understand a bit about how Open Phil evaluates causes, and then look both at how their general thinking has evolved and their specific learnings in the migration policy space.

Importance, neglectedness, and tractability and BOTECs

Open Phil follows a framework similar to the framework of importance, neglectedness, and tractability: they want to pick important causes that have been neglected by other funders and are tractable.

The initial cause report established migration policy as important but was ambivalent about the questions of neglectedness and tractability. Over the years, Open Phil has acquired more evidence on neglectedness and tractability — particularly tractability of the kinds of areas where money can move things forward.

One of the practices followed commonly in Open Philanthropy’s internal grant writeups is that of BOTECs (back-of-the-envelope-calculations) that roughly estimate the cost-effectiveness of a grant. BOTECs incorporate the ideas of importance, neglectedness, and tractability all into a single formula.

In addition, beyond individual grants, Open Philanthropy uses similar cost-effectiveness calculations to inform how much to focus on particular cause areas.

Evolution of the “bar” for grants

In thinking about different buckets of money allocation, Open Philanthropy separates long-termist grantmaking from near-termist grantmaking (recently renamed global health and wellbeing), and further separates human-centric grantmaking from animal-centric grantmaking. The “near-termist, human-centric” group of grants includes several areas such as global health and development, scientific research, immigration policy, land use reform, and macroeconomic stabilization policy.

One key evolution in Open Philanthropy’s thinking, particularly as it applies to near-termist, human-centric grantmaking, is in the “bar” that grants must satisfy in terms of the cost-effectiveness multiplier expected. This evolution is described in detail in the 2019 blog post GiveWell’s Top Charities Are (Increasingly) Hard to Beat. As the blog post describes, Open Philanthropy’s original bar for cost-effectiveness was that it should beat out the 100x multiplier expected for unconditional cash transfers. However, by the time of the 2019 blog post, Open Philanthropy is considering a bar of 500-1,500x, comparable with GiveWell top charities.

The raising of the bar is a reflection of greater confidence in the cost-effectiveness estimates underlying GiveWell’s top charity selection, as well as more confidence that there is enough room for more funding for these charities to use up. In a review of past BOTECs, this led to the conclusion that out of 33 grants whose internal writeups including ex ante BOTECs, all passed the original 100x bar but only 8 of the 33 passed the 1000x bar.

A predicted consequence of this raising of the bar would be a reduction in grantmaking in near-termist, human-centric areas including criminal justice reform, migration policy, land use reform, and macroeconomic stabilization policy. However, the blog post disavows any immediate major changes to grantmaking patterns:

We are still in the process of thinking through the implications of these claims, and we are not planning any rapid changes to our grantmaking at this time. We currently plan to continue making grants in our current focus areas at approximately the same level as we have for the last few years while we try to come to more confident conclusions about the balance of considerations above. As Holden outlined in a recent blog post, a major priority for the next couple years is building out our impact evaluation function. We expect that will help us develop a more confident read on our impact in our most mature portfolio areas, and accordingly will place us in a better position to approach big programmatic decisions. We will hopefully improve the overall quality of our BOTECs in other ways as well.

If, after building out this impact evaluation function and applying it to our work to date, we decided to substantially reduce or wind down our giving in any of our current focus areas, we’d do so gradually and responsibly, with ample warning and at least a year or more of additional funding (as much as we feel is necessary for a responsible transition) to our key partner organizations. We have no current plans to do this, and we know funders communicating openly about this kind of uncertainty is unusual and can be unnerving, but our hope is that sharing our latest thinking will be useful for others.

More recent developments, however, suggest that Open Philanthropy is following through on its plans to scale back in areas of grantmaking where this bar is not met. The most salient example is the spinout of the criminal justice reform grantmaking to its own organization, Just Impact. While the scale of giving in migration policy is probably too small to justify a spinout to a separate organization, the general thinking around scaling back grantmaking may also apply to migration policy and other areas.

[Added after the original post]: Open Philanthropy reaffirms in a new post that it has been using the “roughly 1,000x” bar for funding new programs since the 2019 post:

As of 2019, we switched to tentatively thinking of “roughly 1,000x” as our bar for new programs, because that was roughly our estimate of the unfunded margin of the GiveWell top charities, and we thought we would be able to find enough other opportunities at that cost-effectiveness to hit our overall spending targets.

Specific learning by Open Philanthropy about migration

Combining information gleaned from various grant write-ups, the following are some learnings Open Philanthropy appears to have acquired about the migration space:

  • They have become a bit more bullish on refugee migration. Here’s a blurb found on a few recent refugee migration grants such as the 2019 IRAP grant:

    In the past, our immigration policy work has not focused much on refugee resettlement, which we had assumed would be more crowded than other aspects of immigration policy with funders aimed at supporting increased opportunities for people to move to the U.S. for humanitarian reasons. While we continue to believe that is directionally correct, our increased interest in supporting advocacy around refugee resettlement is partially based on learning more about the fairly limited foundation funding for advocacy around refugee resettlement.

    The largest ongoing commitment seems to be support for IRAP at a rate of $500,000/year.

  • They have struggled to find success in the space of influencing policy around labor mobility or migration more generally. They’ve exited the space in various ways, including exiting Niskanen Center, exiting Labor Mobility Partnerships after helping start it, and reducing ongoing commitment to the Center for Global Development’s migration program to $200,000/year.

  • Their efforts to fund programs that directly support worker visas (such as grants to the U.S. Association for International Migration and Protect the People for Haiti H-2A visas) didn’t give the desired magnitude of results. Though not mentioned in their write-ups, their efforts to create sustainable flows from Haiti got a big blow when Haiti was removed from the Eligible Countries List in January 2018. This was after they published their October 2017 retrospective.

Discussion of the implications, both in terms of what we can learn from Open Philanthropy and what it means for the crowdedness or lack thereof of the space

Open Philanthropy’s reasons for not focusing on migration policy are related to tractability

As best as I can make out, Open Philanthropy has not significantly changed its thinking on the importance of the issue or its neglectedness (with the exception of refugee migration where it has updated toward that area being more neglected). Rather, the biggest updates have been around tractability (where there was initially a lot of uncertainty). To break that down further, this is talking about the tractability of what Open Philanthropy and its grantees can accomplish, and not the tractability of what they could accomplish if they had the ability to directly write migration policies.

One important multiplier that significantly diminished tractability is the uncertainty of pushing desired changes through political systems. This applies particularly to political advocacy, think tank work, etc.

Another factor affecting tractability is that individual programs to navigate complex systems, such as the H-2A use by Haiti, can suffer from program design challenges and limitations. New patterns of migration tend to start slowly and grow over time, which also makes the evaluation of programs to push for these more complicated.

Neither of these reasons has much effect on our estimate of the value of changing migration policies themselves. It’s more about Open Philanthropy’s relative lack of power in affecting those changes.

This is an update of sorts against migration philanthropy for donors with a similar quantitative focus

Open Philanthropy is well-funded and has devoted a fair amount of time and energy to exploring possibilities in migration philanthropy. If they have not found strong opportunities for migration philanthropy, that suggests that such options are probably either not available or hard to find. So while this shouldn’t change our view on the moral justness or practical consequences of immigration restrictions, it should make us more pessimistic about the ability of philanthropy to meaningfully move this. For donors who are moderately cause-agnostic, it does look like, relative to other causes, migration policy may not be the right thing to focus on.

There is space for other donors passionate about migration policy to make donations

Even if we update against migration policy from a purely near-termist cost-effectiveness angle, there’s still plenty of room to donate in the space for donors who are interested both in helping people now and in shaping the long-term discourse.

Open Philanthropy has transformed spaces like animal welfare and AI safety by pumping huge amounts of money into these spaces. In animal welfare, they were a major source of funding for cage-free campaigns. With AI safety, they’re a major source of funding for several AI safety organizations (along with Jann Tallinn via the Survival and Flourishing Fund).

In contrast, because of the various reasons discussed in the post, they have not flooded the migration policy space with money. For donors who are not looking at immediate cost-effectiveness, and are passionate about this cause, the many opportunities — including ones that Open Phil has funded and ones it passed up on — would be worth looking into.

A comparison of different migration policy philanthropy options would be a separate post; I did one years ago collating comments on an EconLog post. I do have more updated thoughts on the matter, but it would take some time to write them up.

Conclusion

Although funding migration policy did not “take off” as a cause for Open Philanthropy, they don’t appear to have updated away from the importance or neglectedness of the issue. Rather, they have struggled to find philanthropic opportunities that are cost-effective. Donors with a similar mindset as Open Philanthropy should use this information to also update away from migration policy philanthropy; those who are passionate about migration policy per se should consider this evidence that the space is still uncrowded. We probably shouldn’t update much on the subject of free migration itself.