Chris Blattman actually carried out some randomised controlled experiments with sweatshops in Ethiopia and published a paper in 2016 on the results. (The paper refers to these as “industrial jobs” but Blattmanelsewhere has described them as “sweatshops”.) They found that, contrary to traditional economic thinking, workers who got jobs in sweatshops were not necessarily better off than those who did not.
Most who got sweatshop jobs quit within just a few months, leaving the sector entirely. Those who stayed experience higher rates of serious health problems and disability compared to the control group. Income effects after a year did seem to be slightly higher for those in sweatshops compared to the control group, but not statistically significant.
The paper concludes:
Overall, the results suggest that that industrial firms in Ethiopia paid no better than worker’s informal alternatives, so that most workers were at best indifferent between these forms of work. This suggests a formal and informal labor market that was more fluid and competitive than expected, at least for the young, unskilled, and capital-poor.
Usual caveats apply about this being a limited study in Ethiopia, and more studies would be needed to see if the results generalise. But it is nevertheless an important study that challenges the traditional economic narrative of “sweatshop jobs are better than no jobs”.
It seems like even if workers are indifferent between the sweatshop jobs and the informal jobs, it could still be the case that the introduction of sweatshop jobs into the labor marketplace results in improvements for workers’ conditions. For example, it could drive up the wages of informal labor.
most workers were at best indifferent between these forms of work
Wouldn’t traditional economic thinking basically predict that the equilibrium for wages in industrial jobs be at the point where workers are indifferent to them? As in, there’s no incentive for the wages in industrial jobs to be any higher than the level which is just high enough to compensate for the downsides (e.g. health risks) compared to the informal labour market? (I’m not an economist, and maybe this depends on what’s considered “traditional”.)
Good point. The study undermines the traditional economic narrative that sweatshop jobs are clearly better than workers’ alternatives—e.g. it would challenge this statement in the original post:
Economists across the political spectrum have pointed out that for many sweatshop workers the alternatives are much, much worse.
But it doesn’t go so far as to show that sweatshops are clearly worse than workers’ alternatives, either, which is what sweatshop opponents often claim. So it seems to be a middle-of-the-road finding that supports your idea that the Ethiopian labour market was actually more efficient than people had previously assumed. It’s still notable because it challenges conventional wisdom on both sides of the debate.
The study had a third “entrepreneurship” condition that involved giving people $300 plus some business training. Unsurprisingly (imo), participants under that condition did better than both the sweatshop and control conditions. So while I don’t think boycotting companies that use sweatshops is likely to be a very effective way of doing good, I don’t think going out of your way to support sweatshops is likely to be worthwhile, either—there are likely to be better ways to promote economic development.
So while I don’t think boycotting companies that use sweatshops is likely to be a very effective way of doing good, I don’t think going out of your way to support sweatshops is likely to be worthwhile, either—there are likely to be better ways to promote economic development.
Agreed! I recommend donating to the High Impact Philanthropy Fund (HIPF) from the Centre for Exploratory Altruism Research (CEARCH) to increase human and animal welfare as cost-effectively as possible. Joel Tan, CEARCH’s founder and managing director, guessed on 28 May 2025 donating to HIPF was 9 to 15 times as cost-effective as donating to GiveWell’s top charities, for which the mean cost of saving a life from 2021 to 2023 was4.13 k$/life. Belows is some context about CEARCH’s cost-effectiveness estimates.
[...] CEARCH estimated the cost-effectiveness accounting only for effects on humans, as a fraction of that of GiveWell’s top charities, of donating to Giving What We Can (GWWC) in 2025 to be 13, that of “advocacy for top sodium control policies to control hypertension” to be31, that of advocating for “increasing the degree to which governments respond with effective food distribution measures, continued trade, and adaptations to the agricultural sector” in “global agricultural crises [such as nuclear and volcanic winters]” to be33 (although I estimated this should be 12.4 % as high), and that of “advocacy for sugar-sweetened beverages [SSBs] taxes to control diabetes mellitus type 2” to be55. Joel disclaimed he thinks the cost-effectiveness estimates from CEARCH’s deep reports, such as the ones I just mentioned, as well as (similarly elaborate) estimates from other impact-focussed evaluators, are 3 times as high as they should be. This largely explains why the mean between Joel’s lower and upper bound for the marginal cost-effectiveness of HIPF is only 21.8 % (= 12⁄55) of CEARCH’s highest cost-effectiveness estimate respecting advocacy for taxing SSBs.
Thanks for sharing! Strongly upvoted. I like the study randomised people to receiving job offers instead of being placed into jobs. The opportunity to reject the offer makes it more like the real world.
Here is how the authors describe the effects on income.
On balance the factory job offer seems to have no significant effect on income by any of the three measures, and a family index of the three increases only 0.014 standard deviations.37 These income measures tend to be skewed and highly variable, however, and so our estimates are imprecise. Thus while the average effect is close to zero, the confidence interval on income includes moderate increases and decreases in income from the industrial job offer.
I guess the effect was slightly positive, but that the study was underpowered to detect it. The authors also note the workers think the increased health risks are worth it.
Naturally, workers were probably not perfectly informed about job risks and quality of these jobs, and there is some evidence that they underestimated the risk. Nonetheless, our data suggest that workers understood the health risks, at least in part, did not update their assessment of the risks as a result of spending more time in industry, and that they were willing to bear these risks to cope with temporary unemployment spells.
workers who got jobs in sweatshops were not necessarily better off than those who did
Chris Blattman actually carried out some randomised controlled experiments with sweatshops in Ethiopia and published a paper in 2016 on the results. (The paper refers to these as “industrial jobs” but Blattman elsewhere has described them as “sweatshops”.) They found that, contrary to traditional economic thinking, workers who got jobs in sweatshops were not necessarily better off than those who did not.
Most who got sweatshop jobs quit within just a few months, leaving the sector entirely. Those who stayed experience higher rates of serious health problems and disability compared to the control group. Income effects after a year did seem to be slightly higher for those in sweatshops compared to the control group, but not statistically significant.
The paper concludes:
Usual caveats apply about this being a limited study in Ethiopia, and more studies would be needed to see if the results generalise. But it is nevertheless an important study that challenges the traditional economic narrative of “sweatshop jobs are better than no jobs”.
It seems like even if workers are indifferent between the sweatshop jobs and the informal jobs, it could still be the case that the introduction of sweatshop jobs into the labor marketplace results in improvements for workers’ conditions. For example, it could drive up the wages of informal labor.
Thanks for the comment, Ian! Strongly upvoted, and agreed.
Wouldn’t traditional economic thinking basically predict that the equilibrium for wages in industrial jobs be at the point where workers are indifferent to them? As in, there’s no incentive for the wages in industrial jobs to be any higher than the level which is just high enough to compensate for the downsides (e.g. health risks) compared to the informal labour market? (I’m not an economist, and maybe this depends on what’s considered “traditional”.)
Good point. The study undermines the traditional economic narrative that sweatshop jobs are clearly better than workers’ alternatives—e.g. it would challenge this statement in the original post:
But it doesn’t go so far as to show that sweatshops are clearly worse than workers’ alternatives, either, which is what sweatshop opponents often claim. So it seems to be a middle-of-the-road finding that supports your idea that the Ethiopian labour market was actually more efficient than people had previously assumed. It’s still notable because it challenges conventional wisdom on both sides of the debate.
The study had a third “entrepreneurship” condition that involved giving people $300 plus some business training. Unsurprisingly (imo), participants under that condition did better than both the sweatshop and control conditions. So while I don’t think boycotting companies that use sweatshops is likely to be a very effective way of doing good, I don’t think going out of your way to support sweatshops is likely to be worthwhile, either—there are likely to be better ways to promote economic development.
Agreed! I recommend donating to the High Impact Philanthropy Fund (HIPF) from the Centre for Exploratory Altruism Research (CEARCH) to increase human and animal welfare as cost-effectively as possible. Joel Tan, CEARCH’s founder and managing director, guessed on 28 May 2025 donating to HIPF was 9 to 15 times as cost-effective as donating to GiveWell’s top charities, for which the mean cost of saving a life from 2021 to 2023 was 4.13 k$/life. Belows is some context about CEARCH’s cost-effectiveness estimates.
Thanks for the comment, Hugh! Strongly upvoted, and agreed.
Blattman discusses this on the Rationally Speaking podcast
Thanks for sharing! Strongly upvoted. I like the study randomised people to receiving job offers instead of being placed into jobs. The opportunity to reject the offer makes it more like the real world.
Here is how the authors describe the effects on income.
I guess the effect was slightly positive, but that the study was underpowered to detect it. The authors also note the workers think the increased health risks are worth it.
Nitpick. There is a “not” missing at the end.
Thanks for the nitpick, I’ve fixed it in my post. Agree that the randomisation element is very useful.