Summary

There is a lack of advice in EA for individuals on where to donate if they value global development and are okay with high risk, high reward interventions - hit based giving.

This gap could be closed by using the EA fund for global development. This could also lead to increased funding for more experimental, neglected organisations and interventions that are less likely to receive institutional money.

Problem

There may be a missing gap when it comes to effective altruism and systemic change in global development. This post from 80,000 Hours in 2015 - “Effective altruists love systemic change” - highlights many ways in which effective altruism can be used to pursue legal, cultural and political changes. But it seems that when it comes to individuals donating, who value global development, most advice is still to give via charities that have evidence mainly based on randomised control trials rather than higher risk, higher reward options.

When looking at other cause areas traditionally supported within effective altruism, these kind of interventions are often seen as worth funding. ACE recommends many system changing and research organisations. This AI alignment charity comparison post highlights organisations working on research and lobbying. Open Philanthropy has expanded their giving - but focused on U.S. policy, science funding and history of philanthropy research.

For individuals who want to support potentially more impactful organisations there isn’t much dedicated research out there.

GiveWell

Givewell does look at qualitative info, but reputationally are looking for harder evidence, so may not want to expand their criteria. These two quotes below highlight their position.

“We have strict criteria about the sorts of charities we recommend. These criteria are partly about achieving maximum impact, but partly about having recommendations that others can fairly easily be confident in...Thus, we think there may be many giving opportunities that are better than our top charities but don't meet our criteria and/or are not known to us.”

“Root-causes-based approaches are, in our view, the kind of speculative and long-term undertakings that are best suited to highly engaged donors”

New Approach

There is starting to be a shift in this area though. Founders Pledge have a new report on evidence based policy. In January 2019, the EA fund for global development gave $1,000,000 to J-PAL's Innovation in Government Initiative.

“[IGI] plans to make grants to partnerships between governments, J-PAL offices, and affiliated researchers to help pilot and scale evidence-informed programs in education, health, and social assistance.”

There is also a chance that donations from people interested in effective altruism can be given to more experimental and/or newer organisations and interventions.

This might be the main advantage of EA donations in this space, willingness to fund hard or impossible to measure interventions that might be neglected by larger funders. This could be donating to meta organisations like GiveWell or AidGrade. It could be donations for organisations working directly on lobbying, research, accountability checking, advocacy, improving journalism, data collection, etc. It could also fund think tanks, such as CGD, JPAL.

Possible Next Steps

If there is positive feedback about the recent donation choice by the EA fund for global development it may make more sense to use that fund to donate to organisations in a more portfolio based approach similar to the EA fund for animal welfare. - “The fund [animal welfare] could support newer but still promising organizations with less evidence to support them or organizations with smaller funding gaps”.

If there is negative feedback about the direction the EA fund for global development has taken with it’s latest donation there could be a new fund, managed by a team of people who have a variety of backgrounds in global development. This allows donors to give to the fund that matches their risk appetite.

A third but less likely option is setting up an organisation similar to GiveWell or ACE to research these kinds of interventions from the point of view of individual donors.