Summary of my academic paper “Effective Altruism and Systemic Change”
I’ve recently published a paper entitled “Effective Altruism and Systemic Change” in the academic journal Utilitas (a philosophy journal). In this post I would like to introduce its main ideas for a non-academic EA audience and highlight what I think is most interesting to EAs.
There is already a significant literature dedicated to EA ideas in academic philosophy (53 results in the category “Effective Altruism” on PhilPapers), but there is little interaction between this research and the actual EA movement. There are good reasons for this: most of this research takes a fairly “external” perspective on EA, which might be considered by many EAs as having little relevance to what they do, and academic writing is often intricate and convoluted. By writing this post, I hope to contribute to bridge this gap, and encourage others to do so. I’m also looking forward to feedback on these ideas!
Summary of the paper in plain English
The paper can be found here. It deals with the systemic (or institutional) change objection against EA, according to which EA neglects changes that affect large-scale institutions or social systems. Most of the intuitive interpretations of this objection seem to fail (see Brian Berkey’s paper for a review), so the challenge has been to come up with a more compelling interpretation.
The paper has two distinct and fairly independent parts. In the first one (Sections I and II), I criticize a recent version of the systemic change objection put forward by Alexander Dietz. I think his objection has some theoretical appeal, but ultimately does not work very well (in particular, I don’t see how it can be applied to real-life situations).
In Section III, I propose a new way of understanding the systemic change objection. The ideas it is based on are pretty intuitive, so I would guess many in the EA movement have already thought about them. Though in the paper it is introduced informally, all of this could be more rigorously put in mathematical terms. Here is the bottom line:
I take systemic change to be a result from actions that affect large-scale political, social and economic institutions. Institutions can be difficult to disrupt, but when they are, after a critical mass of effort is exerted on them, the disruption can potentially bring about very positive change and enable us to reach an improved equilibrium. Intuitively, there are increasing marginal returns (IMR) at the threshold point where the critical mass of efforts bears its fruit and starts to bring about systemic change. Here is a toy example: resources put into lobbying MPs to vote for a law have little impact until the majority of MPs have been talked into voting for the law, at which point the impact made is huge. This suggests that cause areas that deal with systemic change are likely to have IMR when relatively large amounts of resources are put into them.
The methodology used in the EA movement does not seem to be suited for dealing with IMR, and thus with systemic change. There are two potential problems:
The assumption of diminishing marginal returns (DMR) is regularly made (see for example William MacAskill’s talk and 80,000 Hours’ prioritization framework) and arguably underlies the Neglectedness factor in the Scale-Neglectedness-Tractability framework (see here and here for previous discussions of this in the EA literature), which directly contradicts the possibility of IMR.
The recommendation of going for the low-hanging fruits seems to be based on the assumption of DMR. Low-hanging fruits are interventions which are initially very effective and with rapidly DMR. In terms of Cotton-Barratt’s metaphor, they are nuggets of gold lying on the ground that we can easily pick up. Interventions with IMR, in contrast, can be seen as seams of gold buried deep in the ground which require a lot of efforts to reach. If they are large enough, the recommendation of picking up the nuggets of gold first (i.e. going for the low-hanging fruits) is suboptimal: it would be better not to spend any time picking up the easy gold, and go instead for the larger seams.
This could explain in which sense EA may be considered as overly individualistic (this is a common criticism made against EA, see for example here). The existence of cause areas with IMR implies that we may have some reason to coordinate our efforts and pool our resources to spend them on these cause areas to achieve systemic change (i.e. reach one of the larger seams deep in the ground). If IMR are ignored, this kind of collective cooperation is unnecessary: EAs may instead focus on filling the needs of the most effective cause areas, up to the point where their marginal impact decreases so much that it equals that of the next most effective cause areas, then fill the needs of the latter, and so on.
It is sometimes thought that low-hanging fruits and systemic change are antagonistic: going for the low-hanging fruits may undermine the impact of interventions which bring about systemic change (think about the opposition between reform and revolution in Marxist thought, or between welfarist and abolitionist strategies in the animalist movement). If this is the case, it is even more important to pay attention to those interventions that bring about systemic change, since by picking up some adjacent low-hanging fruits we may undermine their cost-effectiveness.
To be clear, I do not think this objection is convincing as it stands. The main takeaway of the paper seems to me that it provides some common ground for constructive discussions between EAs and some of their critics, and within the EA movement. Two directions for future research strike me as particularly interesting:
To what extent is this interpretation of the objection a steelman of what some EA critics have in mind? I’m rather skeptical that it is close to what EA critics have in mind, but it might still help make sense of some of their intuitions. For example, the focus on undertaking collective actions and contributing to existing social movements (like anti-capitalist movements) could suggest that they think there are collective actions in which the more participants there are, the more effective the contribution of the next participant will be (which amounts to positing IMR). In any case, I would be interested in hearing more from EA critics.
Are IMR commonly found in cause areas? This is obviously a difficult question. The objection is associated with the idea that there are IMR in some cause areas in such a way that it could be very cost-effective to invest large amounts of resources in these cause areas. If this were true, there could be a blind spot in current EA methodology, and it would be worth thinking about potential solutions. One example could be to consider different ranges of resources and rank cause areas by their expected marginal impact per [x] dollars, instead of ranking cause areas by their expected marginal impact per dollar. But if it turns out that this kind of IMR is relatively rare, the objection would be likely to fail. General considerations against IMR in cause areas have been put forward (see Cotton-Barratt’s arguments; see also the discussion on the EA forum here), and it seems plausible to assume DMR in specific domains like research. I would be interested in seeing more specific considerations bearing on other cause areas (including the cause areas EA critics are most interested in).
I would like to thank Florent Berthet, Guillaume Corlouer, Romain Destenay and Laura Green for very useful feedback on earlier versions of this post.
- Formalizing the cause prioritization framework by (5 Nov 2019 18:09 UTC; 50 points)
- Does EA need better explanations of why Climate Change isn’t so popular a cause area? by (4 Oct 2022 13:28 UTC; 19 points)
- 's comment on Hilary Greaves: The collectivist critique of the EA movement by (20 Jan 2021 18:55 UTC; 7 points)
My sense of what is happening regarding discussions of EA and systemic change is:
Actual EA is able to do assessments of systemic change interventions including electoral politics and policy change, and has done so a number of times
Empirical data on the impact of votes, the effectiveness of lobbying and campaign spending work out without any problems of fancy decision theory or increasing marginal returns
E.g. Andrew Gelman’s data on US Presidential elections shows that given polling and forecasting uncertainty a marginal vote in a swing state average something like a 1 in 10 million chance of swinging an election over multiple elections (and one can save to make campaign contributions
80,000 Hours has a page (there have been a number of other such posts and discussion, note that ‘worth voting’ and ‘worth buying a vote through campaign spending or GOTV’ are two quite different thresholds) discussing this data and approaches to valuing differences in political outcomes between candidates; these suggest that a swing state vote might be worth tens of thousands of dollars of income to rich country citizens
But if one thinks that charities like AMF do 100x or more good per dollar by saving the lives of the global poor so cheaply, then these are compatible with a vote being worth only a few hundred dollars
If one thinks that some other interventions, such as gene drives for malaria eradication, animal advocacy, or existential risk interventions are much more cost-effective than AMF, that would lower the value further except insofar as one could identify strong variation in more highly-valued effects
Experimental data on the effects of campaign contributions suggest a cost of a few hundred dollars per marginal vote (see, e.g. Gerber’s work on GOTV experiments)
Prediction markets and polling models give a good basis for assessing the chance of billions of dollars of campaign funds swinging an election
If there are increasing returns to scale from large-scale spending, small donors can convert their funds into a small chance of huge funds, e.g. using a lottery, or more efficiently (more than 1⁄1,000 chance of more than 1000x funds) through making longshot bets in financial markets, so IMR are never a bar to action (also see the donor lottery)
The main thing needed to improve precision for such estimation of electoral politics spending is carefully cataloging and valuing different channels of impact (cost per vote and electoral impact per vote are well-understood)
More broadly there are also likely higher returns than campaign spending in some areas such as think tanks, lobbying, and grassroots movement-building; ballot initiative campaign spending is one example that seems like it may have better returns than spending on candidates (and EAs have supported several ballot initiatives financially, such as restoration of voting rights to convicts in Florida, cage bans, and increased foreign spending)
A recent blog post by the Open Philanthropy Project describes their cost-effectiveness estimates from policy search in human-oriented US domestic policy, including criminal justice reform, housing reform, and others
It states that thus far even ex ante estimates of effect there seem to have only rarely outperformed GiveWell style charities
However it says: “One hypothesis we’re interested in exploring is the idea of combining multiple sources of leverage for philanthropic impact (e.g., advocacy, scientific research, helping the global poor) to get more humanitarian impact per dollar (for instance via advocacy around scientific research funding or policies, or scientific research around global health interventions, or policy around global health and development). Additionally, on the advocacy side, we’re interested in exploring opportunities outside the U.S.; we initially focused on U.S. policy for epistemic rather than moral reasons, and expect most of the most promising opportunities to be elsewhere. ”
Let’s Fund’s fundraising for climate policy work similarly made an attempt to estimate the impacts of their proposed intervention in this sort of fashion; without endorsing all the details of their analysis, I think it is an example of EA methodologies being quite capable of modeling systemic interventions
Animal advocates in EA have obviously pursued corporate campaigns and ballot initiatives which look like systemic change to me, including quantitative estimates of the impact of the changes and the effects of the campaigns
The great majority of critics of EA invoking systemic change fail to present the simple sort of quantitative analysis given above for the interventions they claim, and frequently when such analysis is done the intervention does not look competitive by EA lights
A common reason for this is EAs taking into account the welfare of foreigners, nonhuman animals and future generations; critics may propose to get leverage by working through the political system but give up on leverage from concern for neglected beneficiaries, and in other cases the competition is interventions that get leverage from advocacy or science combined with a focus on neglected beneficiaries
Sometimes systemic change critiques come from a Marxist perspective that assumes Marxist revolution will produce a utopia, whereas empirically such revolution has been responsible for impoverishing billions of people, mass killing, the Cold War, (with risk of nuclear war) and increased tensions between China and democracies, creating large object-level disagreements with many EAs who want to accurately forecast the results of political action
Nonetheless, my view is that historical data do show that the most efficient political/advocacy spending, particularly aiming at candidates and issues selected with an eye to global poverty or the long term, does have higher returns than GiveWell top charities (even ignoring nonhumans and future generations or future technologies); one can connect the systemic change critique as a position in intramural debates among EAs about the degree to which one should focus on highly linear, giving as consumption, type interventions
E.g. I would rather see $1000 go to something like the Center for Global Development, Target Malaria’s gene drive effort, or the Swiss effective foreign aid ballot initiative than the Against Malaria Foundation
I do think it is true that well-targeted electoral politics spending has higher returns than AMF, because of the impacts of elections on things such as science, foreign aid, great power war, AI policy, etc, provided that one actually directs one’s efforts based on the neglected considerations
EAs who are willing to consider riskier and less linear interventions are mostly already pursuing fairly dramatic systemic change, in areas with budgets that are small relative to political spending (unlike foreign aid),
Global catastrophic risks work is focused on research and advocacy to shift the direction of society as a whole on critical issues, and the collapse of human civilization or its replacement by an undesirable successor would certainly be a systemic change
As mentioned previously, short-term animal EA work is overwhelmingly focused on systemic changes, through changing norms and laws, or producing technologies that would replace and eliminate the factory farming system
A number of EA global poverty focused donors do give to organizations like CGD, meta interventions to grow the EA movement (which can eventually be cashed in for larger systemic change), and groups like GiveWell or the Poverty Action Lab
Although there is a relative gap in longtermist and high-risk global poverty work compared to other cause areas, that does make sense in terms of ceiling effects, arguments for the importance of trajectory changes from a longtermist perspective, and the role of GiveWell as a respected charity evaluator providing a service lacking for other areas
Issue-specific focus in advocacy makes sense for these areas given the view that they are much more important than the average issue and with very low spending
As funding expands in focused EA priority issues, eventually diminishing returns there will equalize with returns for broader political spending, and activity in the latter area could increase enormously: since broad political impact per dollar is flatter over a large range political spending should either be a very small or very large portion of EA activity
Essentially, the cost per vote achieved through things like campaign spending is currently set by the broader political culture and has the capacity to absorb billions of dollars at similar cost-effectiveness to the current level, so it should either be the case that EA funds very little of it or enormous amounts of it
There is a complication in that close elections or other opportunities can vary the effectiveness of political spending over time, which would suggest saving most funds for those
The considerations are similar to GiveDirectly: since cash transfers could absorb all EA funds many times over at similar cost-effectiveness (with continued rapid scaling), it should take in either very little or almost all EA funding; in a forced choice it should either be the case that most funding goes to cash transfers, whereas for other interventions with diminishing returns on the relevant scale as mixed portfolio will yield more impact
For now areas like animal advocacy and AI safety with budgets of only tens of millions of dollars are very small relative to political spending, and the impact of the focused work (including relevant movement building) makes more of a difference to those areas than a typical difference between political candidates; but if billions of dollars were being spent in those areas it would seem that political activity could be a competitive use (e.g. supporting pro-animal candidates for office)
I think this is a really good comment and probably should be it’s own post.
A minor point, I would say that giving to the Against Malaria Foundation may not on it’s own be systemic change, but if enough people keep on giving to charities that are evaluated on their impact, it could change individual giving as a system, as charities respond to these new incentives.
Agreed (see this post for an argument along these lines), but it would require much higher adoption and so merits the critique relative to alternatives where the donations can be used more effectively.
I have reposted the comment as a top-level post.
Regarding increasing marginal returns (IMR), which seems to be the primary contribution of this paper and not obviously addressed by replies to other types of systemic change objections:
Perhaps rather than ‘Are IMR commonly found in cause areas?‘, I would ask ‘where are IMR found?’ and, for the purposes of testing the critique, ‘in which cases are relevant actors not already aware of those IMR?’ This is because I expect the prevalence of IMR to vary substantially between areas. (I see that you also call for concrete examples in your paper.) Some examples, sorted from strategic to individual decision-making:
DMR at a high level within causes: you may have seen this list of reasons to expect DMR when looking at a cause area at a sufficiently high level. I would argue that since
the methodology you point to is largely about high-level cause prioritisation, and
the neglectedness criterion works if there are roughly† logarithmic diminishing marginal returns within causes,
the IMR objection does not apply to them.
IMR when starting new projects: in their 2011 post on why they are not using donation matching, GiveWell discusses ‘coordination matching’, where the project only goes ahead if sufficient funds are committed (like Kickstarter), as a legitimate form of donation matching. [Minor: this may be IMR only with respect to money ultimately committed. A large donor making a partial, initial commitment may in fact have outsized impact relative to later funders due to the earlier donor causing later donors to act.]
IMR when coordinating individual careers: this section of William MacAskill’s 2017 EAGx Berlin talk encourages more attention to coordinated action. This includes things closely analogous to the coordination games you discuss in the paper, in which one person switching to their comparative advantage is a cost, but two people switching is a gain. I would guess that you would also like 80,000 Hours’ writing on comparative advantage and other coordination concepts.
Of course, my examples of IMR are drawn from people who helped create the effective altruism community, so they help with ‘where are IMR found?’ but not with identifying points where IMR are underappreciated by the community. It certainly seems possible that there are issues I am unaware of. I see that in your paper you mention ‘the problem of competition between NGOs and states’ as a problem for which the solution which might have IMR. To rehash an old counterargument, I would guess that this is not really a conflict, in that substantially reducing disease prevalence (the relevant comparable on the scale of the funding required for other mass empowerment efforts) is hugely empowering. However, this is an off-the-cuff comment. I also know little about the example of whether corporate campaigns accelerate or decelerate abolition, but I believe that this was and probably still is an active debate among effectiveness-focused animal advocates.
† In fact, I am a little worried that the logarithmic assumption is sufficiently wrong to cause problems, while I am confident that there is DMR in general at this level.
I don’t see that IMR poses any challenge to the standard EA cause prioritization method. IMR can be easily modeled as a tractability function that is increasing for some part of its domain. Depending on funding levels, causes with IMR can have the highest marginal utility per dollar, and hence would be prioritized according to the standard framework.