RSS

Distri­bu­tion of cost-effectiveness

TagLast edit: 22 Apr 2023 16:36 UTC by Pablo

Not all causes are equally cost-effective. In fact, the distribution of cost-effectiveness is heavy-tailed: some causes are thousands of times more cost-effective than others.

Such variation in cost-effectiveness in part follows from the existence of immense global inequality, which implies that many people in the world suffer from problems that could be solved with only small amounts of money. Consider, for instance, a developing world charity that spends $20 per person on a surgery to prevent blindness, compared to a charity that spends $40,000 per person to provide guide dogs to blind people in the United States. But even charities that think globally are far from equal in their abilities to turn the donations they receive into real improvements in people’s lives.

Within a given focus area, we can understand variation in cost-effectiveness as arising from both underlying variation in the impact of the interventions that charities carry out and variation in how much charities spend to carry them out. For instance, the cost-effectiveness of a charity that combats malaria will depend both on whether it distributes medication or bednets and on how much wasteful spending it engages in. People in the effective altruism community tend to believe that the former source of variation is more significant than the latter.

One general finding is that, in the long run, the cost-effectiveness of additional donations to many charities will diminish as the amount they have already received grows. Because of this phenomenon, we would expect the cost-effectiveness of charities to become more and more equal in a world where donations were based on cost-effectiveness.

Further reading

Caviola, Lucius et al. (2020) Donors vastly underestimate differences in charities’ effectiveness, Judgment and Decision Making , vol. 15, pp. 509–516.
Shows that lay people estimate that the variation in charity cost-effectiveness is limited, whereas experts estimate them to be much larger.

Daniel, Max (2020) Collection of existing material on ‘impact being heavy-tailed’, Effective Altruism Forum, June 26.
An annotated bibliography of writings related to the distribution of cost-effectiveness.

Ord, Toby (2019) The moral imperative toward cost-effectiveness in global health, in Hilary Greaves & Theron Pummer (eds.) Effective Altruism: Philosophical Issues, Oxford: Oxford University Press, pp. 29–36.
Discusses the moral relevance of variations in cost-effectiveness.

Tomasik, Brian (2014) Why charities usually don’t differ astronomically in expected cost-effectiveness, Essays on Reducing Suffering, January 5 (updated 16 September 2017).
Argues that the variation in cost-effectiveness of the charity sector isn’t as significant as some people believe.

Related entries

cause prioritization | cost-effectiveness | impact assessment | intervention evaluation | ITN framework

Data on how much solu­tions differ in effectiveness

Benjamin_Todd17 Feb 2023 20:22 UTC
100 points
11 comments3 min readEA link
(80000hours.org)

Effec­tive­ness is a Con­junc­tion of Multipliers

Thomas Kwa25 Mar 2022 18:44 UTC
244 points
34 comments3 min readEA link

Un­rav­el­ling the Mys­tery of Distri­bu­tions of Im­pact: Power law or Log­nor­mal?

Michael22 Apr 2023 9:40 UTC
30 points
3 comments5 min readEA link

Sav­ing lives in nor­mal times is bet­ter to im­prove the longterm fu­ture than do­ing so in catas­tro­phes?

Vasco Grilo20 Apr 2024 8:37 UTC
11 points
23 comments9 min readEA link

Cri­tique of the no­tion that im­pact fol­lows a power-law distribution

Sarah Weiler14 Mar 2024 10:28 UTC
87 points
89 comments18 min readEA link

Coun­ter­pro­duc­tive Altru­ism: The Other Heavy Tail

Vasco Grilo1 Mar 2023 9:58 UTC
186 points
8 comments7 min readEA link
(onlinelibrary.wiley.com)

Pareto-Distributed Op­por­tu­ni­ties Im­ply Isoe­las­tic Utility

ABlank30 Mar 2024 20:23 UTC
33 points
5 comments3 min readEA link

Pur­chase fuzzies and utilons separately

EliezerYudkowsky27 Dec 2019 2:21 UTC
110 points
3 comments5 min readEA link
(www.lesswrong.com)

Char­ity En­trepreneur­ship’s re­search into large-scale global health interventions

CE16 May 2023 11:51 UTC
77 points
0 comments4 min readEA link

Quan­tify­ing Uncer­tainty in GiveWell Cost-Effec­tive­ness Analyses

SamNolan31 Oct 2022 14:31 UTC
116 points
7 comments20 min readEA link
(observablehq.com)

My notes on: Search­ing for outliers

Vasco Grilo3 Jun 2022 16:19 UTC
9 points
0 comments2 min readEA link

List of ways in which cost-effec­tive­ness es­ti­mates can be misleading

saulius20 Aug 2019 18:05 UTC
228 points
33 comments13 min readEA link

The value of money go­ing to differ­ent groups

Toby_Ord16 May 2017 13:11 UTC
22 points
14 comments1 min readEA link

Adquira sen­ti­men­tos calorosos e útilons separadamente

AE Brasil / EA Brazil20 Jul 2023 18:48 UTC
4 points
0 comments5 min readEA link

【練習問題】インパクトの差

EA Japan25 Jul 2023 6:29 UTC
2 points
0 comments1 min readEA link

Com­par­ing Health In­ter­ven­tions in Colom­bia and Nige­ria: Which are More Effec­tive and by How Much?

Alejandro Acelas24 Mar 2023 13:48 UTC
49 points
1 comment12 min readEA link

[Question] On GiveWell’s es­ti­mates of the cost of sav­ing a life

aaron_mai1 Oct 2020 10:22 UTC
16 points
4 comments1 min readEA link

Cor­po­rate cam­paigns for chicken welfare are 10,000 times as effec­tive as GiveWell’s Max­i­mum Im­pact Fund?

Vasco Grilo28 Jul 2022 8:22 UTC
77 points
19 comments11 min readEA link

Longterm cost-effec­tive­ness of Founders Pledge’s Cli­mate Change Fund

Vasco Grilo14 Sep 2022 15:11 UTC
35 points
9 comments6 min readEA link

Five steps for quan­tify­ing spec­u­la­tive interventions

NunoSempere18 Feb 2022 20:39 UTC
93 points
8 comments11 min readEA link

Cost-effec­tive­ness dis­tri­bu­tions, power laws and scale invariance

Stijn18 Mar 2021 19:27 UTC
30 points
4 comments10 min readEA link

[Question] When should the in­verse-var­i­ance method be ap­plied to dis­tri­bu­tions?

Vasco Grilo14 Jun 2022 14:33 UTC
5 points
6 comments1 min readEA link

Why Char­i­ties Usu­ally Don’t Differ Astro­nom­i­cally in Ex­pected Cost-Effectiveness

Brian_Tomasik16 Sep 2017 10:40 UTC
94 points
6 comments28 min readEA link
(reducing-suffering.org)

How to de­ter­mine dis­tri­bu­tion pa­ram­e­ters from quantiles

Vasco Grilo30 May 2022 15:20 UTC
23 points
6 comments1 min readEA link
No comments.