# Potential U.S. policy focus areas

Throughout the post, “we” refers to GiveWell and Good Ventures, who work as partners on GiveWell Labs. [Added August 27, 2014: GiveWell Labs is now known as the Open Philanthropy Project.]

Previously, we laid out our basic framework and reasoning for selecting U.S. policy causes to focus on for GiveWell Labs. This post goes through the specific causes that we’re most likely to commit to (and are accordingly performing in-depth investigations of, with some preliminary grantmaking, at the moment).

A few preliminary notes:

• This post does not offer the same sort of thoroughness and comprehensiveness that people might be accustomed to from our research on top charities. Part of this is because this post is more preliminary than our charity recommendations are, and much of the purpose is to elicit feedback and determine which questions most merit further investigation. (We are not yet committing to causes, and are aiming to do so near the end of the calendar year.) However, we also feel that as we move forward on GiveWell Labs, “accomplishing as much good as possible” and “thoroughly examining every question one might ask” will come into conflict, and the former is more important to us. We are aiming for transparency in the sense of making clear what sort of support we have for all of our major beliefs and statements; we are not aiming for comprehensiveness of investigation.
• Some of our beliefs at this point come from intuitions that we’re unable to trace back to a particular source – intuitions that have come from the aggregate of many conversations as well as generally following and discussing policy-related topics. We try to make clear what our beliefs are based on, to the extent we’re able, and hope the ensuing discussion will highlight the areas where we have the most work to do in re-examining the bases for our statements.

We are trying to evaluate causes to “commit” to (as discussed previously), and “committing” could end up meaning many different sorts of things in terms of what sort of work we support. In a given cause, we could end up focusing on (a) supporting better research to determine optimal policy; (b) supporting information, education, and advocacy to push for particular policies; (c) working within an already-changing policy landscape and trying to affect the details of how policies change; (d) something else. We’ve tried to assess the importance, tractability, and crowdedness of causes with this broad potential mandate in mind, and to focus on causes that seem to be quite broadly important/tractable/uncrowded rather than simply presenting an opportunity for a specific narrow intervention.

As discussed previously, the causes we find most promising generally stand out on at least one of our three key criteria: tractability, importance, and crowdedness. As such, our discussion of causes is organized by criterion – we discuss which causes stand out on each dimension, followed by discussion of other causes that we find worth discussing for other reasons.

Contents of this post:

A few key resources that provide partial support for much of the reasoning in this post:

Windows of opportunity: outstanding tractability
As discussed previously, it can be very difficult to predict whether and when a policy area might become tractable (i.e., when it might become possible for advocacy infrastructure to play a major role in how policy develops in that area) in the long run. Paying too much attention to very short-run tractability (for example, what issue is in the news or being debated in Congress at the moment) seems inappropriate given the nature of what we’re trying to do: pick areas to commit to and build infrastructure in for several years.

With that said, we’ve come across a few causes that seem to present unusual “windows of opportunity,” in which something highly relevant in the political landscape seems to be changing in a way that could make the issue unusually prone to change for the next several years (and possibly beyond), and we could imagine our involvement helping to shape the specific way in which changes play out.

Perhaps the best example we’ve seen is the criminal justice policy space, which we’ve done a medium-depth writeup on. This space came up as promising early in our conversations with generalists, and was particularly emphasized by Steven Teles. There has long been a humanitarian argument (generally emphasized by people on the political left) for the importance of reducing unnecessary incarceration and the suffering associated with it; what seems to have changed relatively recently, however, is a combination of historically high incarceration rates, declining crime rates, and state budget difficulties – accompanied by a growing interest among the political right in reducing incarceration rates if it can be done without reducing public safety (e.g., Right on Crime.) Between the excitement we saw about tractability and the concrete opportunities we saw to support promising-seeming approaches, we saw this cause as a good one for our first medium-depth investigation in the policy arena; having investigated further and made some grants, we believe there are many promising underfunded approaches, real opportunities to influence policy, and reasonably high humanitarian stakes. More at our writeup on this topic.

Other “window of opportunity” causes that have come up:

• Public opinion on marijuana over the last ~15 years has shifted dramatically, leading to state-level changes in drug policy and seeming potential for more change. Good Ventures has done some funding and a fair number of conversations in this space, with GiveWell providing support; in addition, there is some overlap between criminal justice and drug policy, particularly the research by Mark Kleiman that we are supporting. We don’t see this cause as particularly crowded, and we see work on designing good regulation (such as Prof. Kleiman’s research) as having a great deal of room for more funding. However, GiveWell’s perception at this moment (though open to revision) is that the net humanitarian benefits of marijuana legalization are unlikely to be as high as for potential reforms in the criminal justice space, and that the aspects of this space that don’t touch on either marijuana legalization or criminal justice policy do not necessarily have much “window of opportunity.”
• Public opinion has also been shifting on the topic of same-sex marriage, though the people we’ve spoken with have often expressed the sentiment that we’re “in the endgame” at this point, and that the entrance of an additional major funder wouldn’t be likely to have much impact.
• A couple of people have raised the possibility that surveillance – e.g., policy around what sorts of information U.S. security agencies collect – is becoming a more dynamic area, due to recent revelations (e.g., Edward Snowden’s leaks) and changes in what’s technologically possible. Our impression, based on our general perceptions of what is at stake and common attitudes toward this issue, is that this area is likely to be or become relatively crowded, and that we are unlikely to see opportunities with comparable humanitarian significance to what we’ve seen in the criminal justice space.

We have largely relied on impressions from our conversations with generalists in order to identify the most promising “window of opportunity” causes. Of these, our view is that criminal justice reform is the most promising, having equal or greater humanitarian significance and equal or lesser crowdedness compared to the others.

Ambitious longshots: outstanding importance
There are a couple of obstacles to identifying policy areas with outstanding importance:

• In order to assess potential humanitarian gains, one must have some idea of what sort of policy change is possible. This seems very hard to assess, and we hope that gaining more intimate familiarity with specific policy areas will give us a better idea of how to do so.
• The calculations we’ve done attempting to compare the significance of different policy areas are extremely non-robust, and arguably have little value-added on top of intuition. With that said, it does seem to us that one can reasonably distinguish between “contenders for the most important policy area,” “policy areas that have fairly small implications” and “somewhere in between.”

Another challenge here is that from what we’ve seen, the causes that seem like strongest contenders for “most important” tend to have relatively poor, or at least highly ambiguous, scores on the other two criteria (more details below). We haven’t seen any such cause where we are (at this early stage) convinced of a clear opening for a philanthropist and an opportunity to make tangible progress.

With that said, we see some compelling reasons to get deeply involved with at least one “ambitious longshot” cause, even if the prospects for change seem doubtful and/or the space seems relatively crowded:

• It’s possible that our assessments of the other two criteria are highly unreliable. It’s possible that the right attitude is: “Any cause has potential for change over the long run, and any cause has plenty of space for a new philanthropist to add value, if one is sufficiently committed. What’s most knowable is which policy areas have the highest stakes.” In particular, if we get deeply involved in a space that initially seems “crowded,” we may discover that there is more value to add than we would have guessed.
• As discussed in a later section, we are thinking about the long-term goal of promoting a broader political platform. More broadly, we hope to see dramatically higher money moved and influence over time. With this in mind, it is probably worth attacking the question, “What causes would we encourage massively more people to support – with their giving and with their talents – if we could?” For that question, focusing on a relatively overlooked cause of extreme importance – rather than simply on causes that currently seem to present opportunities for major gains with relatively small investments – seems valuable.

What follows is the set of causes that we believe to have overwhelming humanitarian importance (in the sense that an imaginable policy change would create large amounts of economic value and/or affect large numbers of people significantly). They are listed in order of how promising we find them, taking into account other criteria (tractability, crowdedness). Note that the way we’re using “importance” here attempts, when feasible, to incorporate not just the size of the problem, but the likely impact of an improvement in policy if the improvement could be implemented. (In other words, a major problem may still fall short on “importance” if it seems unlikely that one could identify a change in legislation with large expected impact on the problem.) With that said, there are many cases in which we know very little about the details of possible policy fixes, and try to approximate “importance” based primarily on the size of the problem and very rough intuition about how policy change might affect it.

We have created a collection of back-of-the-envelope estimates on the likely impact of policy reform in different areas, which informs the comments below in general, though we do not place much confidence in the particular estimates.

Labor mobility

It appears to us that moving from a lower-income country to a higher-income country can bring about enormous increases in a person’s income (e.g., multiplying it several-fold), dwarfing the effect of any direct-aid intervention we’re aware of. As such, labor mobility seems to us to be an enormously high-stakes issue, whether based on our own back-of-the-envelope calculations for possible legislative changes, academic estimates that sufficient increases in immigration could create value on the order of 50% of world GDP, or just the observation that changes on a per-person-affected basis are impressive.

Additionally, it appears to us that there is relatively little attention paid to this cause in some sense: the humanitarian benefits of migration seem to receive little discussion and emphasis generally, we have not identified any other philanthropic funding focused on labor mobility as an anti-poverty issue, and we note that immediately prior to our involvement, Michael Clemens’s work on this issue at Center for Global Development was in the relatively unusual position of not having specific private support (though it had been supported previously).

With that said, there is another sense in which this cause is quite “crowded”: U.S. immigration policy more broadly is an extremely salient and heavily contested issue, with significant philanthropic involvement as well as interest in allowing more migration from the business community. The debates taking place at the moment seem to center mostly around the treatment of undocumented immigrants, with labor mobility as a secondary issue. Thus, the question of how “crowded” this space is – and what a new funder might be able to contribute – remains very much an open question for us, and one that we are trying to address with deeper investigation and declared interest in funding.

There are a couple of other challenges with this area:

• There is a high degree of controversy over this issue. We hope to conduct a thorough review of arguments and counterarguments, which we have done some work on but have not yet completed.
• While we generally feel that “political tractability” is difficult to predict past the short term, immigration seems to be a particularly charged issue where the fundamental obstacles to change may be extremely strong.

Macroeconomic policy

Macroeconomic policy appears to be an area with extraordinarily high stakes, in that a small number of decisions can arguably have substantial effects on national (and global) growth and unemployment.

Our aim in this space would likely be focused on generating better information and new ideas, rather than coming down on one side or the other of a partisan battle.

The question of whether we ought to consider this space “crowded” is a difficult one.

• Arguments to consider this space “crowded”:
• Argument to consider this space “uncrowded”:
• We’re not aware of any major philanthropic funder that has made this area a top priority.
• The early conversations we’ve had have given us some reason to think that there are certain types of research that are not supported by existing infrastructure.
• A number of people we spoke to noted the outsize influence of the Fed in monetary policy and macroeconomic research, and argued that support for more independent research and thinking could be useful.
• We’re not aware of work in this area that focuses on formulating workable legislation to improve on the status quo (e.g., working out the details of “automatic stabilizers” that could be budget-neutral over time, which could be an important criterion for winning support or acceptance from both the right and left). The new Hutchins Center may end up working in this area.
• It generally seems to us that the importance of macroeconomic policy is underappreciated and rarely discussed outside of a few narrow circles (in particular, the academic field of macroeconomics and a particular set of bloggers and journalists).

We are currently conducting deeper investigation accompanied by readiness to provide funding. We expect to learn more about what gaps and opportunities exist.

Foreign aid and global poverty

We have had a number of conversations about the policy landscape around issues directly affecting the global poor, such as the U.S. foreign aid budget and allocation, trade policy related to the developing world, etc. (We have unfortunately not been able to publish notes from a number of these conversations; others are linked to from our shallow writeup on this topic.)

We see this general cluster of issues as having potentially overwhelming importance, because of the direct relevance to the global poor (whose numbers and degree of poverty both exceed those of the U.S. population).

Our impression is that there is a substantial amount of philanthropic involvement in this area, and a relatively strong infrastructure that analyzes and advocates for policies that benefit the global poor. This infrastructure includes The Center for Global Development (CGD), a think tank we perceive as highly intelligent and effective in developing new ideas; we have supported CGD and may increase our level of support over time, but we also note that CGD has expressed a lack of desire to expand much further. It also includes the ONE Campaign (with a budget of roughly $30 million/year, supported by the Gates Foundation and others) and a network of large aid organizations. It has been argued to us that this infrastructure has been highly successful in preventing cuts to foreign aid despite recent concern over budget balance, and we find this a strong argument. The Bill and Melinda Gates Foundation has made influencing policy in this space a clear priority. As such, we aren’t sure how much can be accomplished by a new funder in this space, at least at the level we’re currently contemplating (in the$5-25 million per year range). A few possibilities we’ve considered:

• One issue we perceive relatively little attention to within the anti-global-poverty community is labor mobility, discussed above.
• We also have the impression that there is relatively little advocacy – in the US policy arena (as opposed to the community around multilateral funders, etc.) – around allocating foreign aid for maximum humanitarian “bang for the buck.” This could mean, e.g., advocating for relatively more to be spent on proven cost-effective global health programs relative to more expensive and/or less proven programs; it could also mean advocating for more structural reforms, of the sort promoted by CGD (e.g., Cash on Delivery Aid). We see some reason to believe work along these lines could do more harm than good (by undermining the attempt to preserve/expand the total level of aid), and we believe that other funders in this space (particularly the Gates Foundation) recognize the importance of these issues, so we are not highly optimistic about pursuing this line of reasoning, but we may do so if capacity permits.

Because we perceive the infrastructure in this space as relatively strong and successful, we’ve considered providing funding and spending time in this area as a way of learning more about what a strong advocacy infrastructure looks like.

Improving democracy

We’ve been following the Hewlett Foundation’s evolving initiative on aiming to improve the general functioning of the U.S. democratic system, particularly with regard to the highly polarized current environment. We have reviewed an early report on this initiative (not public) and spoken with Daniel Stid and Hewlett Foundation President Larry Kramer about it.

It seems to us that federal politics are currently deeply dysfunctional, and we could imagine enormous gains (though it is hard to lay out the likely specifics of such gains) if we could help ameliorate this issue. However, “size of the problem” is only one part of our definition of “importance.” The other part – “likely impact of hoped-for legislative reforms” – is much less clear for us. It seems to us that past attempts at reforming the political system as a whole haven’t clearly done more good than harm (see, for example, points 1 and 2 at Wonkblog’s discussion of U.S. political dysfunction, which I see as a good concise summary of the major potential factors overall). Reviewing the fairly broad list of potential interventions laid out by Hewlett (in its not-yet-released document, and summarized to some degree in our conversation notes), we are ambivalent regarding what the likely impact of legislative reforms would be, assuming political victory.

“Crowdedness” is somewhat difficult to assess for this cause. The Hewlett Foundation seems likely to make it a real priority, and to try to interest other foundations in it too, which could dramatically increase the amount of philanthropic investment. It’s hard to say, at this point, to what degree this will happen and how much space (and what sort of space) will remain for us to potentially fill.

Overall, we are glad to see that the Hewlett Foundation is taking on what we believe is one of the world’s most pressing issues, and we plan to follow its work with interest. At this time, we see greater likelihood of getting heavily involved (in the sense of “committing” to) the causes listed above, though that may change as we continue to follow Hewlett’s work.

Climate change

We have done a shallow-depth investigation of climate change, an area that gets a great deal of philanthropic attention compared to all of the above causes. The potential impact of climate change mitigation is enormous, but not (by our estimates, based on mainstream projections) clearly larger than that of other causes we’ve classified as “ambitious longshots.” We do see a case that climate change deserves special attention because of its potential as a global catastrophic risk: there is a risk that mainstream projections are badly off and that the consequences will be much worse than currently projected. We will discuss this aspect of climate change (and the interventions we feel are most appropriate to deal with this relatively low-probability, high-impact scenario) in an upcoming discussion of global catastrophic risks.

Tax policy

Tax policy, like macroeconomic policy, has theoretically huge economic stakes and a good deal of attention from intellectuals. We see it as having substantially more attention from funders and nonprofits, and (likely as a consequence) fewer gaps in the work done by intellectuals (particularly with regard to developing workable policy proposals). We also see less room for impact from new academic research on related matters, as the main bottleneck to improved policy seems to be politics (in particular, resistance from groups like Americans for Tax Reform to changes that would involve new taxes or reduced tax expenditures) rather than knowledge. We have done a shallow investigation of this area and will be writing it up in the future.

Green fields: outstanding “room for more philanthropy”
We’ve identified a small number of causes that seem to have at least moderate importance and potential tractability, while being extremely “empty” – very little infrastructure in place pushing for what we would see as positive policy change.

One such area is what Steven Teles calls “rent seeking.” The broad idea is that there are some industries in which government regulation has been captured in a manner that makes it unnecessarily difficult and expensive to provide a service, so the existing providers of this service benefit from inefficiently low levels of competition. Consequently, existing providers tend to push for preserving and expanding such regulation. A classic example would be that of taxis: an artificially restricted supply of taxi medallions makes it artificially difficult and expensive to become a taxi driver, and the existing medallion owners have an interest in continuing to artificially restrict the supply. This dynamic results in unnecessarily high taxi costs, low taxi supply, and fewer job options for people who would consider being taxi drivers. It’s been claimed that similar dynamics apply, to varying degrees, to a broad range of occupations, both lower-skilled and higher-skilled (such as doctors, dentists, lawyers, and accountants).

Prof. Teles believes that there is little in the way of concentrated advocacy groups to counteract “rent seeking” in occupational licensing (by arguing for less protective regulation and more permissiveness in who can e.g. drive taxis), and that even creating a small advocacy infrastructure could make a big difference in combating artificial supply restrictions. Most importantly, a small number of victories at the local and/or state level could (he argues) raise the general profile of these issues, create a model for people in other areas, and lead to “compounding” policy change at the state and local level. We expect that efforts focusing on higher-skilled occupations would have quite a different profile than efforts focusing on lower-skilled occupations, and we do not have a strong sense of which is likely to be more promising.

Other causes in this category:

• Zoning reform to enable more construction and urban density. It seems possible that there is a currently excessive level of regulation held up by those (property owners) who benefit from a restriction in supply of housing, business space, etc. While specific developers may advocate heavily on behalf of specific projects, it seems to us (from initial conversations) that there is very little advocacy infrastructure making the public-interest case for general increases in how much development is allowed.
• Incentives for organ donation. GiveWell Senior Research Analyst Alexander Berger has an unusual degree of familiarity with this area. It appears to us that there is practically no work being done on finding, and promoting, ethical and safe ways to provide incentives for organ donation, something that could have large health benefits and save a significant amount of money for the health system.

Other causes of interest
We are interested in a few other causes that don’t fit into any of the above categories.

• We’ve had some conversations about the idea of improving the general quality of policy analysis available to state-level governments, where there may be a type of void that doesn’t exist at the federal level. This is a very preliminary idea at the moment and we will likely be writing more about it.
• We are investigating the treatment of animals in industrial agriculture at a medium-depth level, due largely to a particular interest on the part of one of our employees. This is a cause that preliminarily appears relatively “uncrowded,” and according to some moral frameworks could be seen as having enormous importance as well. We’re also intrigued by the possibility that Steve Teles raised of working more generally toward accountability of industrial agriculture companies on a broad array of issues; this could have implications for animal welfare, climate change, antibiotic resistance, farm subsidies, and potentially nutrition as well.
• Intellectual property reform could present an unusual combination of unusually high tractability (see notes from our conversation with the Electronic Frontier Foundation), unusually high uncrowdedness (see our writeup on software patent reform, though intellectual property reform need not confine itself to software), and reasonably high importance (though we’ve had a good deal of trouble estimating this last piece). There could also be connections with trade policy, as mentioned in our conversations with Steven Teles.

Some major issue areas that we are less likely to prioritize
There are a other issue areas that we may investigate at some point, though we consider them less promising than the issues listed above.

• U.S. education generally is a popular area among philanthropists, and the education policy space generally seems to be heavily influenced by the agendas of three major foundations: Gates, Broad and Walton (references to this in conversation notes here and here). At the levels of funding we’re currently contemplating, we have difficulty imagining that we could substantially contribute to or alter this agenda.
• Health care policy is highly important, and there is arguably some degree of “window of opportunity” to affect the specifics of how the system changes in reaction to the recent passage of the Affordable Care Act. However, our loose impression is that (a) this is the major priority of one of the major U.S. foundations (Robert Wood Johnson) and that (b) more generally, this area seems highly crowded and we haven’t become aware of any likely promising angles that we could take on.
• A very broad area of policy, with potentially very far-reaching repercussions, is the issue of inequality and the question of the extent to which (and manner in which) U.S. governments, at the federal, state and local level, should redistribute wealth. Our impression is that this cause gets far more attention from philanthropists, nonprofits and intellectuals than questions about helping the global poor, who we feel are more numerous and benefit more from redistribution relative to the U.S. poor. We have not come across any aspects of this broad space that seem appealing by the criteria we’ve laid out.
• Our impression is that environmental issues (aside from climate change, discussed in an earlier section) also receive a great deal of attention from philanthropy, and are not particularly likely to be of comparable humanitarian significance.
• Trade policy is another major policy area. The main potential benefits we see to working on trade policy pertain to the impact on the developing world, so we’re inclined to classify it with the set of developing-world-oriented policy areas discussed above. Generally, it seems to us (based on loose impressions and no particular source) that there is reasonably strong infrastructure in place representing most relevant perspectives in trade policy, though we have not done a shallow investigation and may in the future.
• Defense policy seems clearly important, and we aren’t aware of much in the way of advocacy infrastructure pushing to reduce unnecessary military expenditures and unnecessary military engagements. We plan to investigate this area at some point, but intuitively feel that philanthropy is unlikely to have much impact on this front and that the causes discussed earlier are more promising.

Other causes we may focus on in the future, but are not including in the categories above
Helping to strengthen a broad political platform. It can be argued that the strongest impact of philanthropic engagement with policy has been long-term promotion and development of a movement. (For instance, Steve Teles has notably made this argument with respect to the conservative legal movement). Rather than picking individual policy issues in which to invest, a philanthropist with interests in a number of a causes and clear set of values might achieve more by promoting their general values, along with the people and organizations that share them (since much of the long term benefits of investment in a given area may be in the form of empowering the particular individuals who receive support, who may go on to other things). However, we do not feel that our values are broadly shared by any existing, easily located major political movements.

In particular, we generally favor policy focused on benefiting low-income and otherwise disadvantaged people, even when it involves active government – an attitude often associated with the U.S. political “left” – but we place particularly high value on the developing world. Additionally, we place high emphasis on the value of economic growth and innovation (which we feel are likely to benefit future people). In the long term, we could imagine exploring the possibility of helping to promote a political platform consistent with these values and trying to find, connect, and support people and organizations supporting this platform. We’re aware that people who share these values will have many disagreements over policy, but feel that there could nonetheless be major benefits to laying out, and promoting, a platform that emphasizes both global humanitarianism and economic development.

We think of this as a long-term possibility with highly uncertain value. We are doing some very preliminary work now to explore the idea, but feel that more direct engagement with specific issues will make us better-informed, better-connected, and overall better-positioned to explore such a possibility further down the line.

Policy related to global catastrophic risks. We are treating “global catastrophic risks” as a separate category of work at the moment, and we will be writing more later this year about our likely priorities in that category. So far, we haven’t identified clear cases in which a particular policy change seems highly important for one of what we consider the most important global catastrophic risks (other than climate change, discussed above), though this may change. We’re looking to build our general capacity for policy-oriented philanthropy by working on other causes, and will hopefully be well-positioned to do relevant policy-oriented work if and when it becomes important to do so.

Policy related to scientific research. We see policy around scientific research (for example, the budget, mandate and policies of the National Institutes of Health) as potentially extremely important, but at this time we don’t feel that we have strong enough scientific advisory capacity to have a good grasp on the relevant issues. We are building our scientific advisory capacity via separate projects, and will be writing about this more in the future. Again, we will hopefully be well-positioned to do relevant policy work if and when it becomes important to do so.

Other categories. This post has focused exclusively on our medium-term plans for U.S. policy. We continue to explore a broad variety of other sorts of philanthropic work, which we will be writing about in the future.

Bottom line and our plans from here
We’ve spent a good deal of time investigating potential focus areas in U.S. policy, and we have a very large number of questions remaining. There are many causes that we have much to learn about on many dimensions, including both questions like “How should policy change and why?” and questions like “How can a philanthropist increase the odds of a particular policy change?” One of the aims of this post is to stimulate discussion and help determine which questions are most important to focus on. Our hope is to finalize “commitments” to causes by the end of this calendar year.

Our current working agenda is as follows:

Deep investigations of cause areas: looking actively for funding opportunities and being highly open to funding them.

• We are exploring both labor mobility and macroeconomic policy at this level.
• We have done a fair amount of work on criminal justice reform, and are pausing our investigation of it for the moment.
• In addition to finding funding opportunities, we are also interested in (a) doing thorough reviews of academic literature to assess the best arguments on each side of the relevant policy debates; (b) trying to substantially refine our “importance” estimates after gaining more context. Both (a) and (b) could be substantial projects, and we are likely to do them only for causes that we do deep investigations of and seriously consider committing to.
• Depending on our capacity and on the results of lower-depth investigations, we may do this sort of “deep investigation” of other causes as well.

Medium-depth investigations of cause areas: having 5+ conversations per cause area to get a good sense of the overall landscape.

• We are hoping to explore the “rent seeking” and “zoning” causes discussed above (under “Green fields”) at this level.
• We are also conducting a number of conversations on factory farming, currently with a focus on animal welfare implications.
• There are several other cases in which we have done a medium level of investigation, including foreign aid and organ donation (in the latter case, we feel we have a strong understanding of the issue largely through Alexander Berger’s personal background, as mentioned above).
• We are likely to do a future investigation on improving the general quality of policy analysis available to state-level governments, which we will be writing about more in the future.
• Other causes we may investigate at this level include tax policy and intellectual property reform.

Shallow-level investigations of cause areas: having a few conversations to get a basic picture of an area. We hope to look into some of the causes we have done little investigation of, such as health care policy. However, this area is a lower priority than the above, and we aren’t sure whether we’ll get to it this year (whereas we do expect to make significant progress on all of the above points).

Hiring. Having a decent sense of our likely interests, we are working on hiring U.S.-policy-specific staff, so that when we do make commitments, we’ll have the staff available to execute on them. We have a major hire starting in June whom we will be writing more about in the future.

Limited time and capacity. At the moment, we are executing on the above agenda; if and when we complete currently-in-progress items and have more capacity, we may promote some causes from the “medium” to the “deep” level of investigation or (less likely) from “shallow” to “medium.” However, around the end of the calendar year, we expect to use whatever information and staff we have at that time to make commitments.

• Andrew on May 29, 2014 at 3:48 pm said:

As a doctor and public health professional I’d argue your survey of the healthcare and education spaces is making you overlook key areas for major impact.

As you have moved from beyond the non-profit sector to include other niches of social impact within politics and policy, I’d encourage you to think beyond the typical healthcare policy debates (payer reform, insurance access, payment reform, HIT, biomedical research) to include other areas that are neglected: higher levels of evidence in medicine (e.g. systematic review), implementation science, deimplementation science, medical education science, the trend of CAM licensing and payment, reimbursement for public health and wellness interventions, prizes/patent pools for medical innovation reform, and many, many more areas.

Medicine is a $2.7 trillion sector in the US alone, there are certainly many niches that RWJ, IHI, etc. are missing. Similarly, while the education debates focus narrowly on charters, teacher training, unions, standards, testing, etc. there are many areas ripe for advocacy/policy reform: early childhood literacy, preschool, adult learning, individually paced learning, HS externships, and more. In a$500+ billion sector, again, it’s easy to miss key niches.

• chaos on May 29, 2014 at 6:23 pm said:

Perhaps you’d benefit from adding a fourth criterion that dealt with meta-level epistemic uncertainty or outside-view uncertainty. Certain issues in certain fields, like economics, might appear to have good evidence supporting certain politics, but your overall confidence in those policies should still be low. It can sometimes be difficult to find these type of problems unless you explicitly search for them, which is an argument in favor of adding this as a separate criterion.

• chaos on May 29, 2014 at 6:26 pm said:

More specifically, I think that some of the fields in which there are highly politicized recommendations on both sides can either be considered more trustworthy or less trustworthy, and that this is a problem.

In one sense, if there is biased argumentation there are likely biased conclusions. But in another sense, motivated thinking does a fantastic job of rooting out difficult to recognize problems in policies. Areas in which there are more arguments or researchers are in some sense worse off and in other senses better off in terms of accuracy. I’m unsure how anyone could resolve this tradeoff.

• Rafael Prado Proenca on May 30, 2014 at 2:33 pm said:

Great, fantastic post! Still, here are two areas that I find missing. Maybe you have already given good reasons for not looking into them and I have overlooked them since I do not follow the blog regularly. If so, can you please point them out?

1- Scientific publication system reform. I think it is very important because science might have a very big positive impact (through, eg, AI, drugs, agricultural productivity), thus making science progress faster might have an even bigger impact. The current system slows down progress by doing a bad job in fostering large scale collaboration and data sharing, and by increasing errors by not giving the right incentives for reproducing results, fact checking or sharing detailed procedures for replicating experiments. Peer review seems to be broken (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3005733/ , please take a look since it presents the problems much better than I). The problem is big and it could be improved in many ways. There are efforts ongoing (http://blogs.bmj.com/bmj/2010/08/05/richard-smith-enter-the-%E2%80%9Cliquid-journal%E2%80%9D/) and lots of discussion on the topic, but little money as far as I know (Nature just launched a database publication journal http://www.nature.com/news/announcement-launch-of-an-online-data-journal-1.13906, but they are not looking into fixing the peer review system). I am of the (not so informed) opinion that efforts like these could benefit a lot from some high profile attention and (not so big) resources.

2- International security and peace. US military budget was around 640 Bn in 2013. There is an arms race starting in asia (source: http://www.ft.com/intl/cms/s/0/9d83bf62-b9b9-11e3-a3ef-00144feabdc0.html#axzz33DHzVYch). Finding clever ways of (even marginally) increasing collaboration and decreasing mistrust among nations can free up a huge amount of money for higher impact causes. Not to mention the benefits of killing less people and generating less suffering by decreasing actual conflicts.

• Rafael Prado Proenca on May 30, 2014 at 2:49 pm said:

edit: I just got very happy to find this:

https://blog.givewell.org/2013/07/11/our-landscape-of-the-open-science-community/

you are indeed wired, smart guys. I know international security is very polemic, but please take a deeper look. There seems to be clever ways to make a difference (http://www.foreignpolicy.com/articles/2014/04/25/can_big_data_stop_wars_preemptive_peace_technology_conflict).

• Holden on May 30, 2014 at 6:54 pm said:

Andrew, thanks for the thoughts. I agree with you that when it comes to very broad areas such as health care and education, we could be failing to divide up the area into sub-issues and find specifics ones that perform well on our criteria. Ideally we would put concerted effort into exploring areas like this (climate change is a possibility as well) to find such opportunities. I can’t guarantee that we will do so; we don’t think it’s exceedingly likely that we’ll find more promising areas than our current priority areas, and we don’t think it’s realistic to be fully comprehensive in our investigations before making commitments. However, health care is one of the areas we’re most likely to explore in this way. We appreciate your list of potential niches and may follow up privately at a later date for more information/thoughts on them.

chaos, can you clarify a bit? Are you arguing for or against entering heavily-studied areas?

• Vasee on May 31, 2014 at 4:42 am said:

How about funding for advocacy on correcting cinical trial non-reporting in the public heath and heath science space by mandating that all cinical trials are reported through legislation. The current % of non-reported cinical trials undermines the entire evidence-based medicine architecture, and this in turn biases decisions about: R&D funding allocation as these are based on a biased understanding of the science state-of-the-art, regulatory decision-making on new interventions, financing allocations, policy recommendations across the entire drugs, vaccines and diagnostics field. There is also a major ethical consideration. Athough the Altrials campaign is an indication of some movement in the area, there is an ongoing pushback from those with vested interests. A GiveWell analysis could be very hepful. Please consider it, if you haven’t addressed this already. The fact that there are reasonaby simple legislative fixes to structural problems in this space makes it an attractive area for action.

• Mary on June 2, 2014 at 8:10 pm said:

You have probably already thought of this, but just in case . . . You discuss international aid and global poverty. I have heard a number of anecdotes suggesting that emergency response in particular is poorly coordinated, with too many aid agencies providing what is not needed, failing to provide what is, and generally running over themselves and each other. I’m not aware of anything being done to work towards more coordination. I’d be interested in knowing more about this as a potential area of involvement.

• Holden on June 3, 2014 at 5:50 pm said:

Mary, we agree that this is an area worth thinking about. We have classified it under the “foreign aid” heading rather than under the “U.S. policy” heading, since the most likely players are likely to be international bodies/aid agencies.