Thank you, this is exactly the kind of list of examples I was looking for.

Martin Gould's Five insights from farm animal economics over at Open Phil's FAW newsletter points out that (quote) "blocking local factory farms can mean animals are farmed in worse conditions elsewhere": 

Consider the UK: Local groups celebrate blocking new chicken farms. But because UK chicken demand keeps growing — it rose 24% from 2012-2022 — the result of fewer new UK chicken farms is just that the UK imports more chicken: it almost doubled its chicken imports over the same time period. While most chicken imported into the UK comes from the EU, where conditions for chickens are similar, a growing share comes from Brazil and Thailand, where regulations are nonexistent. Blocking local farms may slightly reduce demand via higher prices, but it also risks sentencing animals to worse conditions abroad.

The same problem haunts government welfare reforms — stronger standards in one country can just shift production to places with worse standards.

This reminded me of what Will MacAskill wrote in Doing Good Better on anti-sweatshop protests being potentially misguided because the alternative for sweatshop workers is worse (long quote):

...  those who protest sweatshops by refusing to buy goods produced in them are making the mistake of failing to consider what would happen otherwise. In developing countries, sweatshop jobs are the good jobs. The alternatives are typically worse, such as backbreaking, low-paid farm labor, scavenging, or unemployment. 

A clear indicator that sweatshops provide comparatively good jobs is the great demand for them among people in developing countries. Almost all workers in sweatshops chose to work there, and some go to great lengths to do so. In the early 2000s, nearly four million people from Laos, Cambodia, & Burma immigrated to Thailand to take sweatshop jobs, and many Bolivians risk deportation by illegally entering Brazil in order to work in the sweatshops there. The average earnings of a sweatshop worker in Brazil are $2,000/year — not very much, but $600/year more than the average earnings in Bolivia, where people generally work in agriculture or mining. Similarly, the average earnings among sweatshop workers are: $2/day in Bangladesh, $5.50/day in Cambodia, $7/day in Haiti, and $8/day in India. These wages are tiny, but when compared to the $1.25 a day many citizens of these countries live in, the demand for these jobs seem more understandable. 

It’s difficult for us to imagine that people would risk deportation just to work in sweatshops. But that’s because the extremity of global poverty is almost unimaginable. 

Among economists, there’s no question that sweatshops benefit those in poor countries and that they are ‘tremendous good news for the world’s poor.’ One said, ‘My concern is not that there are too many sweatshops but that there are too few.’ Low-wage, labor-intensive manufacturing is a stepping-stone that helps an economy based around cash crops develop into an industrialized, rich country. During the Industrial Revolution, for example, Europe and America spent more than 100 years using sweatshop labor, emerging with much higher living standards as a result. It took many decades to pass through this stage because the tech to industralize was new, and the 20th century has seen countries pass through this stage of development much more rapidly because the tech is already in place. The four East Asian ‘Tiger economies’ — Hong Kong, Singapore, South Korea, and Taiwan — exemplify speedy development, having evolved from very poor, agrarian societies in the early 20th century to manufacturing-oriented sweatshop countries mid-century, and finally emerging as industrialized economic powerhouses in recent decades. Because sweatshops are good for poor countries, if we boycott them, we make people in poor countries worse off. 

We should certainly feel outrage and horror at the conditions sweatshop laborers toll under. The correct response, however, is not to give sweatshop-produced goods in favor of domestically produced goods. The correct response is to try to end the extreme poverty that makes sweatshops desirable places to work in the first place. What about buying products from companies that employ people in poor countries but claim to have higher labor standards, like People Tree, Indigenous, and Kuyichi? By doing this, we would avoid the use of sweatshops, while at the same time providing even better job opportunities for the extreme poor.

This made me wonder about 2 things: 

1. Zooming out: if you buy that the two examples above form a natural category (of "noble intentions misguided by poor reasoning about counterfactuals / second-order effects", say), what other examples are there of such altruistic mistakes that we might be making?
2. Zooming in: what kind of intervention is analogous to "buy from People Tree" in the FAW context? Is this a promising avenue at all?

I know very little about FAW, but I'd guess the answer to #2 is "not promising" mainly because it isn't what advocates do. Instead, and again quoting from Gould's writeup, they do this:

... advocates are getting smarter about this. They're pushing for laws that tackle both production and imports at once. US states like California have done this — when it banned battery cages, it also banned selling eggs from hens caged anywhere. The EU is considering the same approach. It's a crucial shift: without these import restrictions, both farm bans and welfare reforms risk exporting animal suffering to places with even worse conditions. And advocates have prioritized corporate policies, which avoid this problem, as companies pledge to stop selling products associated with the worst animal suffering (like caged eggs), regardless of where they are produced.

Wow, that's nuts. Thanks for the pointer.

"Quantification isn't meant to replace our empathy—it's meant to extend it, direct it" is beautifully put. In the same vein, Brian Tomasik wrote of triage as being "warm and calculating", a reframing (and phrasing) which stuck with me.

(Tangent: "big tent EA" originally referred to encouraging a broad set of views among EAs while ensuring EA is presented as a question, but semantic drift I suppose...)

I again want to say that I resonate a lot with what you're trying to do, since I've tried (and mostly failed) to do something similar myself before. 

I worry a bit that you're prematurely optimising approach-wise when you conclude the thing to focus on is figuring out the cost-effectiveness upper limit at which you can tell people honestly and confidently that their donation is doing that much good, instead of asking donors how they think about giving (which you did). For instance, Sawyer's comment reiterates the sentiment I mentioned earlier that 

Most potential donors are not really risk neutral, and would rather spend $5,001 to definitely save one life than $5,000 to have a 10% chance of saving 10 lives. Risk neutrality is a totally defensible position, but so is non-neutrality. It's good to have the option of paying a "premium" for a higher confidence (but lower risk-neutral EV).

and Jason's comment seems relevant as well: 

Orthogonally, I think most people are willing to pay more for a more legible/direct theory of impact. 

"I give $2800, this kid has lifesaving heart surgery" is certainly more legible and direct than a GiveWell-type charity. In the former case, the donor doesn't have to trust GiveWell's methodologies, data gathering abilities, and freedom from bias.  

and these aren't necessarily obvious in advance, so if you start from a (simplistic) model of giving advisory effectiveness as being [number of prospective donors in your circle] x [fraction of donors who find argument X persuasive] x [$ per donor] x [expected good per $], then starting with an argument that works on yourself (and on me too — I don't think we're that representative of the donor pool) doesn't let you get empirical input on the 2nd term, while asking donors does.

Thanks for digging up that plot, I'd been looking for annual data instead of 3-year rolling averages.

Here's what WHR say about their methodology which seems relevant. 

What is your sample size?

The number of people and countries surveyed varies year to year but, in general, more than 100,000 people in 140 countries and territories participate in the Gallup World Poll each year.

In most countries, approximately 1,000 people are contacted by telephone or face-to-face each year. Tables 1-5 in the Statistical Appendix show the sample size for each country since 2005. Gallup’s website provides more details on their data collection methods. ...

What time of year is the data collected?

The Gallup World Poll collects data throughout the year, taking into account religious observances, weather patterns, pandemics, war, and other local factors. Variation in collection timing is not a serious obstacle to analysis as there are established techniques to test for seasonal effects and adjust for them (see this paper for an example).

That Gallup website doesn't say if they've changed their methodology over time; that said, they seem to try their best to maintain a similar sample over time, e.g. 

With some exceptions, all samples are probability based and nationally representative of the resident population aged 15 and older. The coverage area is the entire country including rural areas, and the sampling frame represents the entire civilian, non-institutionalized adult population of the country.

I remain as baffled as you are. 

Tangentially re: protest, I think things are slowly shifting, due to the work of folks like James Özden founding Social Change Lab to understand how social change movements can be more evidence-based and effective. For instance, James changed my mind on the effectiveness of radical protest tactics in What’s everyone got against throwing soup?, which drew upon this literature review to conclude that 

A nonviolent radical flank is likely to help, not hinder, a social movement. Specifically, we think there’s good evidence it can increase support for more moderate groups and increase the salience of an issue without harming support for the overall movement’s policy goals.

I'd also signal-boost James' article Protest Movements Could Be More Effective Than the Best Charities published in the Stanford Social Innovation Review. You should always take charts like the one below claiming superlative cost-eff with a metric ton of salt, but I mostly trust the general quality of his analysis and think his bottomline holds up.

That said, James seems to be the only person banging this drum, so I suppose your observation still broadly holds true.

The first thing I wondered about when this report came out was: how did India do? 

In that quick take I asked how India's self-reported life satisfaction dropped an astounding -1.20 points (4.97 to 3.78) from 2011 to 2021, even as its GDP per capita rose +51% in the same period; China in contrast gained about as much self-reported life satisfaction as you'd expect given its GDP per capita rise. This "happiness catastrophe" should be alarming to folks who consider happiness and life satisfaction what ultimately matters (like HLI), since given India's population such a drop over time adds up to roughly ~5 billion LS-years lost since 2011, very roughly ballparking (for context, and keeping in mind that LS-years and DALYs aren't the same thing, the entire world's DALY burden is ~2.5 billion DALYs p.a.). Even on a personal level -1.20 points is huge: 10x(!) larger than the effect of doubling income at +0.12 LS points (Clarke et al 2018 p199, via HLI's report), and comparable to major negative life events like widowhood and extended unemployment. So it mystified me that nobody seems to be talking about it.

Last year's WHR reported a 4.05 rating averaged over the 3-year window 2021-23, improving +0.27 points. This year's WHR (linking to the country rankings dashboard) reports a 4.39 rating over 2022-24 i.e. +0.34 points, continuing the improvement trend. So I'm guessing this is some sort of mean reversion effect, and that the 3.78 LS averaged over 2018-20 was just anomalously low somehow...? Some commenters pointed to rising inequality and falling social support as potential factors. I still find myself confused.

Just so people know what you're referring to, this is Figure 4: 

Ben West noted in the blog post that 

We think these results help resolve the apparent contradiction between superhuman performance on many benchmarks and the common empirical observations that models do not seem to be robustly helpful in automating parts of people’s day-to-day work: the best current models—such as Claude 3.7 Sonnet—are capable of some tasks that take even expert humans hours, but can only reliably complete tasks of up to a few minutes long.