One point I’m curious about, however, is whether more economic growth really will contribute to higher wellbeing.
So far, the relationship seems slim, as argued here. In many rich countries, subjective well-being levels have stagnated since the 1970s, even though GDP has more than doubled. At a given point in time, richer people are more satisfied than poorer people, and richer countries are more satisfied than poorer countries, but over the course of time, countries which grow faster don’t seem to get happier faster. The effect of more money is not based on how objectively wealthy we are, but how much we have compared to others.
For India specifically, happiness levels seem to have been declining between 2008 and 2018, despite strong economic growth.
This answer to the post argues that although the effect on individuals is slim, there can be an aggregate effect. Michael Plant answers that even though there can be an effect, we don’t know if it’s more than chance.
Growth is correlated with all the good stuff as argued here with good back-and-forth in the comments. One thing it’s not (robustly) correlated with is subjective wellbeing. I personally see that as one of the many issues with SWB measures and I place almost no stock in them.
Interesting Karthik. Why does economic growth not being robustly associated with SWB make you less confident in the measure particularly?
Putting “almost no stock” in a self-reported measurement which is fairly well correlated with measures like good health, income etc. seems like a strong response, but responding to that might be too long for this thread!
I’m also interested in why you think orgs like One Acre fund and Givedirectly are more “Growth” orientated than GiveWell interventions like deworming, which has been shown to improve education outcomes (a big deal for growth) or even mosquito nets which through avoiding sickness help improve productivity and energy levels a huge amount. On this note it would be interesting to see some kind of specific comparative analysis of potential pro-growth spillover effects of different interventions that are already highly rated. Obviously uncertainty of second-order effects will be through the roof but the analysis might be worth doing.
Putting “almost no stock” in a self-reported measurement which is fairly well correlated with measures like good health, income etc. seems like a strong response, but responding to that might be too long for this thread!
GDP per capita is also strongly correlated with all of those things… but yes it’s a bit long of a discussion for this, see some comments here for a good discussion.
There are a couple of papers showing that disease eradication has real but quantitatively small effects on income. (Acemoglu and Johnson 2007, Bleakley 2007, Bleakley 2010) They are severely problematic in many ways but they are the best evidence we have and they don’t point to large effects. So health interventions are just not that promising in that regard. I plan to elaborate on this argument in the final post of my growth series… some day...
Edit: I should note that Bleakley is more positive than me in his interpretation, but I think the effect sizes are just not large and certainly wouldn’t survive any skeptical adjustments downwards (of which many are warranted)
Thanks so much that’s great. And yeah I commented on that thread a bunch :D :D :D.
I’m interested in what attracts you to the impact of eradication on economic growth as the best evidence we have. Intuitively it seems to me like not a great case study, as moving from low malaria prevalence to eradication may only improve productivity for a small percentage of people who were getting malaria. Anywhere where malaria has been “eradicated”, seems unlikely to have had malaria as a massive economic issue in the 30 yeas before eradication.
Wheras Here in a Ugandan town though with high prevalence where most people get sick with malaria every year I would say, even with nets and prompt treatment, it really seems to affect productivity and motivation. Also malaria causes anemia and iron deficiency which obviously can reduce productivity in the long term.
There’s a bit of research on malaria and economic burden as well, but obviously. The systemic review below is interesting, it seems like it has potential be a fairly big deal on a number of measures including catastrophic expenditure for families, absenteeism, even GDP loss. Obviously those are very much proxys (and you could argue some aren’t even that) for economic growth but at least they can be robustly measured.
The systemic review below is interesting, it seems like it has potential be a fairly big deal on a number of measures including catastrophic expenditure for families, absenteeism, even GDP loss.
Most studies in this space are just correlational, and having high burden of malaria is obviously correlated with lots of bad things—that doesn’t tell us anything meaningful. For example, being poor could cause countries to be unable to deal with malaria, and also cause all those bad things. It looks like the systematic review is also correlational. The studies I linked are the only ones I know of that have a quasi-experimental approach, which is why I lean on them.
as moving from low malaria prevalence to eradication may only improve productivity for a small percentage of people who were getting malaria. Anywhere where malaria has been “eradicated”, seems unlikely to have had malaria as a massive economic issue in the 30 yeas before eradication.
Broadly I don’t think this is true. DDT was an incredibly powerful anti-malarial tool and caused near-eradication even in places with quite high burdens. Malaria has always been endemic to the Americas and my impression is that DDT is the reason it’s mostly gone, though it’s still a real public health problem in e.g. Brazil.
even with nets and prompt treatment, it really seems to affect productivity and motivation.
I’m sure this is true, but the productivity and motivation of individual workers isn’t the big constraint on growth. A lack of jobs, mobility frictions, inability to invest in growing firms, etc—so even if you made every worker able to 2x their working hours because of better health, that would have <<2x impact on GDP.
I generally think that people with less wealth would trivially swap to having more wealth if they could, which suggests to me that they prefer the latter to the former. I get that wellbeing may not scale linearly with wealth but I would be surprised if it wasn’t positively correlated.
Let’s formulate this another way. Suppose I propose to give you a Ferrari. Would you accept?
Of course you would. However, will that really make you happier in the long term? Or will you just take the Ferrari for granted in 3 months?
Think of the last 3 items you bought on Amazon, or you received as a gift. Did they make you significantly happier compared to a universe where you didn’t receive them?
One aspect of having more money is that we get used to it. Fast! And we underestimate how fast we adapt to these kind of changes.
Sure, but I think we should be wary of thinking that things people choose aren’t on some deep level what they want. It’s a small but useful amount of information.
Well, I’m personally a bit sceptical of that, due to the fact that many things influence what we choose or not. We’re not that good at predicting what will make us happier, and a large amount of what people choose derives from several influences like marketing.
To take the Ferrari example—the marketing team at Ferrari has been very good at making this kind of car desirable, by associating it with status. But status is relative—in a world with no Ferrari we’d use other symbols (and we did). Marketing is pretty strong when it comes manipulation.
The fact that chasing one’s desires (or, more exactly, craving), does not contribute to happiness has been documented in many spiritualities, and I think this still applies today. I recommend this text : https://www.dhamma.org/en/about/art
(of course, this does not apply to basic stuff like eating, sleeping, etc.)
I don’t want to detail, but I have one example in mind : one survey looked at lottery winners (who presumably wanted to win it), and it turns out that one year later, they were less happy than before, on average. This is because they got used to their wealth, but the quality of their relationships with others (friends, family) declined, as people around them started seeing them differently.
Some people in this comment thread are repeatedly making the mistake of talking about the increase in satisfaction due to increased wealth of already-wealthy/developed-country-people. Increases in satisfaction related to increases in income only start to plateau around the $80k per year mark.
https://images.app.goo.gl/7jDJFqbj3Eq1ePeY7
The per capita global income is around $10,000. Most people in the world would be significantly more satisfied with 2x, or even 7x, their current income.
The lack of improvement in developed countries is relatively easy to explain: they had a high starting base and many of the material gains accrued to those at the top—wide distribution of gains is really important when you’re measuring average happiness. We do see rich countries having fairly consistently higher reported happiness levels than poor ones, it’s just that average differences of subjective wellbeing are only a couple of points despite a very large difference in average incomes.
How much of that is genuine hedonic adaptation and how much of it is a culture of not reporting being very unhappy because your neighbours are even poorer is open for debate (it’s probably both, and personally I’d hate people to make too many decisions about my welfare based on a semi arbitrary self assessment on a ten point scale). But it might be a moot point anyway, because if GDP growth mean the country can afford a range of interventions to halve infant mortality, that’s a lot of extra WELLBYS even if average happiness only changes from 5.2 to 5.4 (Granted, “promoting economic growth” may not be the most efficient way to do this, but ultimately the history of economic growth in much of the world is why there are people with enough disposable income to consider donating to other ways of improving happiness...).
I also think that it’s worth drawing a distinction between the direct interventions which aim to boost individual businesses like the One Acre Fund and the indirect interventions aimed at policy reform like supporting think tanks. The former are much more tractable, they have relatively fast positive effects on individuals (relative to others so hedonic adaptation and Easterlin Paradox arguments don’t really apply) and there’s definitely room for more spending
On the other hand, promoting economic reform is an all-or-nothing intervention; either the government acts or it doesn’t. And whilst advocacy has tractable results in other fields: few are as politicised and competitive and thoroughly studied as development economics, so it’s very hard to see where an EA institute or funding would have any impact at the margin. Many plausible economic policy interventions also can be expected have negative impacts on welfare or even growth if actually implemented [in the wrong circumstances]. So I think “many strategies to promote economic growth probably don’t work” is a better critique than macro data on subjective wellbeing
Thanks for the post.
One point I’m curious about, however, is whether more economic growth really will contribute to higher wellbeing.
So far, the relationship seems slim, as argued here. In many rich countries, subjective well-being levels have stagnated since the 1970s, even though GDP has more than doubled. At a given point in time, richer people are more satisfied than poorer people, and richer countries are more satisfied than poorer countries, but over the course of time, countries which grow faster don’t seem to get happier faster. The effect of more money is not based on how objectively wealthy we are, but how much we have compared to others.
For India specifically, happiness levels seem to have been declining between 2008 and 2018, despite strong economic growth.
https://forum.effectivealtruism.org/posts/gCDsAj3K5gcZvGgbg/will-faster-economic-growth-make-us-happier-the-relevance-of#1__Introduction
This answer to the post argues that although the effect on individuals is slim, there can be an aggregate effect. Michael Plant answers that even though there can be an effect, we don’t know if it’s more than chance.
Growth is correlated with all the good stuff as argued here with good back-and-forth in the comments. One thing it’s not (robustly) correlated with is subjective wellbeing. I personally see that as one of the many issues with SWB measures and I place almost no stock in them.
Interesting Karthik. Why does economic growth not being robustly associated with SWB make you less confident in the measure particularly?
Putting “almost no stock” in a self-reported measurement which is fairly well correlated with measures like good health, income etc. seems like a strong response, but responding to that might be too long for this thread!
I’m also interested in why you think orgs like One Acre fund and Givedirectly are more “Growth” orientated than GiveWell interventions like deworming, which has been shown to improve education outcomes (a big deal for growth) or even mosquito nets which through avoiding sickness help improve productivity and energy levels a huge amount. On this note it would be interesting to see some kind of specific comparative analysis of potential pro-growth spillover effects of different interventions that are already highly rated. Obviously uncertainty of second-order effects will be through the roof but the analysis might be worth doing.
GDP per capita is also strongly correlated with all of those things… but yes it’s a bit long of a discussion for this, see some comments here for a good discussion.
There are a couple of papers showing that disease eradication has real but quantitatively small effects on income. (Acemoglu and Johnson 2007, Bleakley 2007, Bleakley 2010) They are severely problematic in many ways but they are the best evidence we have and they don’t point to large effects. So health interventions are just not that promising in that regard. I plan to elaborate on this argument in the final post of my growth series… some day...
Edit: I should note that Bleakley is more positive than me in his interpretation, but I think the effect sizes are just not large and certainly wouldn’t survive any skeptical adjustments downwards (of which many are warranted)
Thanks so much that’s great. And yeah I commented on that thread a bunch :D :D :D.
I’m interested in what attracts you to the impact of eradication on economic growth as the best evidence we have. Intuitively it seems to me like not a great case study, as moving from low malaria prevalence to eradication may only improve productivity for a small percentage of people who were getting malaria. Anywhere where malaria has been “eradicated”, seems unlikely to have had malaria as a massive economic issue in the 30 yeas before eradication.
Wheras Here in a Ugandan town though with high prevalence where most people get sick with malaria every year I would say, even with nets and prompt treatment, it really seems to affect productivity and motivation. Also malaria causes anemia and iron deficiency which obviously can reduce productivity in the long term.
There’s a bit of research on malaria and economic burden as well, but obviously. The systemic review below is interesting, it seems like it has potential be a fairly big deal on a number of measures including catastrophic expenditure for families, absenteeism, even GDP loss. Obviously those are very much proxys (and you could argue some aren’t even that) for economic growth but at least they can be robustly measured.
https://malariajournal.biomedcentral.com/articles/10.1186/s12936-022-04303-6
Most studies in this space are just correlational, and having high burden of malaria is obviously correlated with lots of bad things—that doesn’t tell us anything meaningful. For example, being poor could cause countries to be unable to deal with malaria, and also cause all those bad things. It looks like the systematic review is also correlational. The studies I linked are the only ones I know of that have a quasi-experimental approach, which is why I lean on them.
Broadly I don’t think this is true. DDT was an incredibly powerful anti-malarial tool and caused near-eradication even in places with quite high burdens. Malaria has always been endemic to the Americas and my impression is that DDT is the reason it’s mostly gone, though it’s still a real public health problem in e.g. Brazil.
I’m sure this is true, but the productivity and motivation of individual workers isn’t the big constraint on growth. A lack of jobs, mobility frictions, inability to invest in growing firms, etc—so even if you made every worker able to 2x their working hours because of better health, that would have <<2x impact on GDP.
I generally think that people with less wealth would trivially swap to having more wealth if they could, which suggests to me that they prefer the latter to the former. I get that wellbeing may not scale linearly with wealth but I would be surprised if it wasn’t positively correlated.
Let’s formulate this another way. Suppose I propose to give you a Ferrari. Would you accept?
Of course you would. However, will that really make you happier in the long term? Or will you just take the Ferrari for granted in 3 months?
Think of the last 3 items you bought on Amazon, or you received as a gift. Did they make you significantly happier compared to a universe where you didn’t receive them?
One aspect of having more money is that we get used to it. Fast! And we underestimate how fast we adapt to these kind of changes.
Sure, but I think we should be wary of thinking that things people choose aren’t on some deep level what they want. It’s a small but useful amount of information.
Well, I’m personally a bit sceptical of that, due to the fact that many things influence what we choose or not. We’re not that good at predicting what will make us happier, and a large amount of what people choose derives from several influences like marketing.
To take the Ferrari example—the marketing team at Ferrari has been very good at making this kind of car desirable, by associating it with status. But status is relative—in a world with no Ferrari we’d use other symbols (and we did). Marketing is pretty strong when it comes manipulation.
The fact that chasing one’s desires (or, more exactly, craving), does not contribute to happiness has been documented in many spiritualities, and I think this still applies today. I recommend this text : https://www.dhamma.org/en/about/art
(of course, this does not apply to basic stuff like eating, sleeping, etc.)
I don’t want to detail, but I have one example in mind : one survey looked at lottery winners (who presumably wanted to win it), and it turns out that one year later, they were less happy than before, on average. This is because they got used to their wealth, but the quality of their relationships with others (friends, family) declined, as people around them started seeing them differently.
Some people in this comment thread are repeatedly making the mistake of talking about the increase in satisfaction due to increased wealth of already-wealthy/developed-country-people. Increases in satisfaction related to increases in income only start to plateau around the $80k per year mark. https://images.app.goo.gl/7jDJFqbj3Eq1ePeY7
The per capita global income is around $10,000. Most people in the world would be significantly more satisfied with 2x, or even 7x, their current income.
The lack of improvement in developed countries is relatively easy to explain: they had a high starting base and many of the material gains accrued to those at the top—wide distribution of gains is really important when you’re measuring average happiness. We do see rich countries having fairly consistently higher reported happiness levels than poor ones, it’s just that average differences of subjective wellbeing are only a couple of points despite a very large difference in average incomes.
How much of that is genuine hedonic adaptation and how much of it is a culture of not reporting being very unhappy because your neighbours are even poorer is open for debate (it’s probably both, and personally I’d hate people to make too many decisions about my welfare based on a semi arbitrary self assessment on a ten point scale). But it might be a moot point anyway, because if GDP growth mean the country can afford a range of interventions to halve infant mortality, that’s a lot of extra WELLBYS even if average happiness only changes from 5.2 to 5.4 (Granted, “promoting economic growth” may not be the most efficient way to do this, but ultimately the history of economic growth in much of the world is why there are people with enough disposable income to consider donating to other ways of improving happiness...).
I also think that it’s worth drawing a distinction between the direct interventions which aim to boost individual businesses like the One Acre Fund and the indirect interventions aimed at policy reform like supporting think tanks. The former are much more tractable, they have relatively fast positive effects on individuals (relative to others so hedonic adaptation and Easterlin Paradox arguments don’t really apply) and there’s definitely room for more spending
On the other hand, promoting economic reform is an all-or-nothing intervention; either the government acts or it doesn’t. And whilst advocacy has tractable results in other fields: few are as politicised and competitive and thoroughly studied as development economics, so it’s very hard to see where an EA institute or funding would have any impact at the margin. Many plausible economic policy interventions also can be expected have negative impacts on welfare or even growth if actually implemented [in the wrong circumstances]. So I think “many strategies to promote economic growth probably don’t work” is a better critique than macro data on subjective wellbeing