I am a generalist quantitative researcher. I am open to volunteering and paid work. I welcome suggestions for posts. You can give me feedback here (anonymously or not).
Vasco Grilođ¸
Karma: 11,019
I worry just 4 human-anchored pain intensities are not enough for reliable comparisons, even for an early stage. For shrimp-anchored annoying pain 10^-6 times as intense as human-anchored annoying pain (the ratio between the individual number of neurons of shrimps and humans), and this 10^-6 times as intense as human-anchored excruciating pain, shrimp-anchored annoying pain would be 10^-12 (= (10^-6)^2) times as intense as human-anchored excruciating pain. It seems super hard to cover such a wide range of pain intensities with any significant reliability using just 4 values?
Thanks for sharing, Nicoll. Could you share how much demand you have found for cost-effectiveness analyses (CEAs)? I am open to reviewing these for free.
Hi Jamie. I agree.
The Happier Lives Instituteâs investigations into intervention cost-effectiveness find that some mental health interventions are over 5x more effective than GiveDirectlyâs cash transfers
Nitpick. I do not think the above is exactly supported by the graph with the cost-effectiveness comparison. According to this, only Friendship Bench (48.5 WELLBY/âk$) is over 5 times more cost-effective than GiveDirectly, which corresponds to having a cost-effectiveness over 45.6 WELLBY/âk$ (= (1 + 5)*7.6/â10^3). There are 2 charities, Friendship Bench and StrongMinds (40.4 WELLBY/âk$), which are more than 5 times as cost-effective as GiveDirectly, which corresponds to having a cost-effectiveness over 38.0 WELLBY/âk$ (= 5*7.6/â10^3), but they both deliver the same type of mental health intervention (âTask-shifted group
psychotherapyâ).
Thanks for the update. I like your BHAG of reaching 1 million pledgers.
This year, we shift from laying foundations to strengthening them while starting to push more seriously on growth. Weâre aiming to grow 40% on most of our key pledge and donations metric (e.g. aiming for roughly 1500 new đ¸10% Pledges). This puts us on a trajectory of achieving our BHAG by 2040.
Below is the growth of the number of 10 % Pledges from 2009 to 2022. An annual growth of 40 %/âyear from 2025 to 2040 seems very ambitious relative to 2018-2022. On the other hand, you now have â16 core staffâ, whereas you only had a few during 2018-2022?
Thanks for the post, Dilan.
Likewise, if weâre optimising for the reduction of near-term animal suffering, then we can compare welfare improvement projects using a standardised measure of suffering like Ambitious Impactâs âSuffering-Adjusted Daysâ (SADs) and invest in the ones which reduce the most SADs per dollar.
1 SAD corresponds to 1 human-day of disabling pain. So it is not time-bound. It can be used to measure reductions in suffering even if they happen far into the future (although I guess more than 90 % of the overall effect materialises in the 1st 100 years for the vast majority of interventions).
Many advocates are working towards a broader project of ending industrial food animal production (IFAP).
There is a sense in which reducing animal suffering (as measured by SADs) is a broader project than ending large-scale farming? Farmed animals account for a tiny fraction of all animals and neurons.
Each of these is attacking IFAP from a different angle. Therefore itâs extremely hardâif not outright impossibleâto compare their cost-effectiveness, because each projectâs âunit of impactâ is completely different.
Since you mention âif not outright impossibleâ, do you think ending industrial farming is intrinsically good? I believe what matters is just decreasing suffering, and increasing happiness. So I would ultimately assess interventions decreasing industrial farming in these terms. I worry about using non-welfare final outcomes because animals in large-scale farming can have positive lives, with more happiness than suffering, under some conditions. In this case, they would prefer existing over not existing. I also think positive lives are very much possible in practice. I estimate that the welfare per chicken-year of slower growth broilers is 92.9 % higher than that of fast growth broilers, and that of layers in barns 80.4 % higher than that of layers in battery cages, which suggests lives in the improved conditions that are negative, but close to neutral (estimates of 100 % higher welfare would suggest neutral lives). I guess cows, broilers, and layers in organic production have positive lives.
I was not clear, but I meant the image suggests âorganismâs resolution from welfare intensity A to Bâ = ânumber of different welfare intensities the organism can experience between A and Bâ/â(BâA), which depends on the organism, A, and B. Is this what you have in mind?
Thanks for the good points, Matt.
So we canât assume that CG is filling EG organizationsâ budgets until their multiplier is about 1.
What if funding caps are part of CGâs strategy to maximise impact? They may result in greater diversification of funding sources, and therefore greater resilience against shortfalls in CGâs funding, and potentially more funding longterm. The benefits will not be observed nearterm. So the expected marginal multiplier may be closer to 1 than the observed nearterm marginal multiplier.
The ideal funding cap would vary by grantee neglecting CGâs assessment costs. However, having a single or a few funding limits could save time, and therefore be closer to optimal.
@Melanie Basnakđ¸, do you have any thoughts?
I thought ii) in my past comment was the resolution over the whole range of welfare intensities. The image below of the post suggests âresolution from welfare intensity A to Bâ = ânumber of different welfare intensities between A and Bâ/â(BâA)? More precisely, it looks like resolution is the derivative of the number of different welfare intensities with respect to welfare intensity. This still leaves open how ii) relates to the whole range of welfare intensities. A system can have high resolution, and a narrow whole range of welfare intensities in the same way that a car can move fast over a short distance (even though the average speed over a distance can be calculated from the ratio between the distance covered, and time spent covering it).
Is it an RCT? If not, do you have strong reason to believe that all else but the intervention is approximately equal between the treatment and control groups?
Our World in Data (OWID) has a great post describing the procedures that strengthen RCTs against bias. âIn medicine, this commonly includes the use of controls, placebos, experimentation, randomization, concealment, blinding, intention-to-treat analysis, and pre-registrationâ. Relatedly, I think it is useful to assess studies based on the hierarchy of evidence.
You are welcome, Seth. I liked the approach, and broadly agree with your takes. Here is how to get notifications of comments on the post.
How I read studies
Thanks for the relevant discussion, Jim and Wladimir. Wladimir, in your framework, is resolution i) the total number of distinct welfare intensities, ii) the ratio between i) and the difference between the maximum and minimum welfare intensities, or iii) something else?
Hi Ozymandias. Thanks for the interesting and relevant survey.
Thanks for sharing. Have you considered having a shorter version of the survey (with questions from the longer one such that you can still compare results)? A survey which takes 5 instead of 15 min to fill may result in significantly more responses.
Thank you for the post, Oscar. I found it quite funny.
Thanks for the post. The top 2 interventions are more in the area of global health and development (GHD) than in that of mental health, considering they mostly increase income (Pure Earth) and improve physical health (Taimaka)?
How much of a post are you comfortable for AI to write?
I think what matters is the content. So I am in principle comfortable with AI writing all of the words of posts. However, in practice, AI has written basically no words of my posts.
I strongly endorse expectationaltotal hedonistic utilitarianism. So I would say the pain of an individual has 3 relevant dimensions. Probability, duration, and hedonistic intensity. I personally only care about the product between these. However, I think my argument works even for more dimensions. I believe any pains are quantitatively comparable if the pains of any 2 infinitesimally different states are quantitatively comparable.
Hi Anthony. Do you think the expected welfare of 2 states of the world which only differ infinitesimally can be incomparable? I do not see how this could be possible. For example, it feels super counterintuitive to me that, given 2 identical states, moving an electron by 10^-100 m in one of the states would make their expected welfare incomparable. I guess one can get from any state of the universe to another in an astronomical number of infinitesimal steps, and I believe any 2 states which only differ infinitesimally are comparable. So I conclude any 2 states are comparable too, even if it is very hard to compare them, to the point that I do not know if electrically stunning shrimps increases welfare in expectation.
Thanks, David. Did you have the chance to look into this?