Yep. A significant portion of the relevant health economics literature Givewell researchers will be familiar with uses measures which do treat lives as non-equal, typically the “value of a statistical life” which represents how much society is willing to pay to save that life which is broadly proportional to GDP per capita. The rationale is basically that survivors in richer societies are capable of generating enough wealth to cover the costs of their treatment, but if you’re valuing lives from an altruistic perspective then you really, really don’t want to weight it based on future ability to pay...
That “value of a statistical life” obviously factors in differences in opportunities and values positive externalities generated from surviving, but vastly overweights differences in actual quality of life—and even on value-of-a-statistical-life grounds malaria nets and vitamin supplementation in Sub Saharan Africa is generally still seen as cost effective.[1] From a pure hedonic utilitarian perspective you might want to use some sort of subjective wellbeing factor instead. Multiply that by the expected future life of the person saved and you get the WELLBY as an alternative metric [2]
But the difference in average self-reported subjective wellbeing on a linear scale is… really not very big compared with the differences in costs between countries, and probably isn’t going to change their recommendations very much. Taking the example of the Democratic Republic of Congo, and Congoese people polled do indeed value their happiness at lower than many other countries on the World Happiness Poll’s nominally linear scale at only 3.3 out of 10. But India and Bangladesh, highlighted in the post as countries which don’t have ongoing conflict and plausibly have better economic opportunities, score only 4.1 and 3.8 respectively so factoring in the weightings of subjective wellbeing—if you believe them to be accurate—would change very little. (The main reason why comparatively few nets are dispensed in India and Bangladesh is that the local malaria variety is a lot less prevalent and a lot less lethal. The life expectancy difference to Congo shrinks if you factor out malaria too...). And if children survive infancy, their lives are typically lived over spans of 60-70 years. It’s unlikely the global distribution of happiness will be identical 30 years from now, and entirely possible that the countries with the lowest happiness will see the biggest improvement
So whilst GiveWell may have made the judgement to weight lives equally on ideological grounds, the actual data you’d need to create a robust argument for doing things differently tends to not be there or broadly inclined to what they’re already doing...
people in richer countries not only face proportionally higher healthcare costs in general, but also diminishing returns since the treatments they’re at risk of missing out on tend to be expensive and complex surgery and new experimental drugs, rather than vitamins and nets...
using national life expectancy figures which are significantly affected by malaria prevalence in infants as weights which discourage supplying malaria nets is questionable, but in theory life expectancy measures could be adjusted to factor malaria out....
Yep. A significant portion of the relevant health economics literature Givewell researchers will be familiar with uses measures which do treat lives as non-equal, typically the “value of a statistical life” which represents how much society is willing to pay to save that life which is broadly proportional to GDP per capita. The rationale is basically that survivors in richer societies are capable of generating enough wealth to cover the costs of their treatment, but if you’re valuing lives from an altruistic perspective then you really, really don’t want to weight it based on future ability to pay...
That “value of a statistical life” obviously factors in differences in opportunities and values positive externalities generated from surviving, but vastly overweights differences in actual quality of life—and even on value-of-a-statistical-life grounds malaria nets and vitamin supplementation in Sub Saharan Africa is generally still seen as cost effective.[1] From a pure hedonic utilitarian perspective you might want to use some sort of subjective wellbeing factor instead. Multiply that by the expected future life of the person saved and you get the WELLBY as an alternative metric [2]
But the difference in average self-reported subjective wellbeing on a linear scale is… really not very big compared with the differences in costs between countries, and probably isn’t going to change their recommendations very much. Taking the example of the Democratic Republic of Congo, and Congoese people polled do indeed value their happiness at lower than many other countries on the World Happiness Poll’s nominally linear scale at only 3.3 out of 10. But India and Bangladesh, highlighted in the post as countries which don’t have ongoing conflict and plausibly have better economic opportunities, score only 4.1 and 3.8 respectively so factoring in the weightings of subjective wellbeing—if you believe them to be accurate—would change very little. (The main reason why comparatively few nets are dispensed in India and Bangladesh is that the local malaria variety is a lot less prevalent and a lot less lethal. The life expectancy difference to Congo shrinks if you factor out malaria too...). And if children survive infancy, their lives are typically lived over spans of 60-70 years. It’s unlikely the global distribution of happiness will be identical 30 years from now, and entirely possible that the countries with the lowest happiness will see the biggest improvement
So whilst GiveWell may have made the judgement to weight lives equally on ideological grounds, the actual data you’d need to create a robust argument for doing things differently tends to not be there or broadly inclined to what they’re already doing...
people in richer countries not only face proportionally higher healthcare costs in general, but also diminishing returns since the treatments they’re at risk of missing out on tend to be expensive and complex surgery and new experimental drugs, rather than vitamins and nets...
using national life expectancy figures which are significantly affected by malaria prevalence in infants as weights which discourage supplying malaria nets is questionable, but in theory life expectancy measures could be adjusted to factor malaria out....