One of the cool things about blogging is it documents how one is thinking (and how these thoughts have changed) over time. Back in 2015 I wrote a post entitled, “So the garment industry makes you uncomfortable…” It was my reaction to a wave of guilt being tossed around about participating in “fast fashion” and buying clothing made in sweatshops. At the time, my thinking was basically, “Well if you think sweatshops are bad, you should see the alternatives – the situations those who willingly accept sweatshop jobs are coming from.”
New research suggests I may have been too bold.
Chris Blattman and Stefan Dercon have a new working paper entitled, “Occupational Choice in Early Industrializing Societies: Experimental Evidence on the Income and Health Effects of Industrial and Entrepreneurial Work” in which they find that the majority of people who got a “sweatshop” job quit within the first few months. Here are some key details of the study:
- The authors randomized entry-level applicants to one of three treatment arms of a randomized control trial (RCT): (i) an industrial job offer, (ii) a control group, and (iii) an “entrepreneurship program with a $300 grant plus business training.
- These individuals were followed up with after a year.
- Individuals who worked for the industrial job worked more hours compared to the control group, but there was no meaningful difference in incomes between these two groups. This is likely due to the relatively low wages of the industrial job.
- Those working in industrial jobs reported more health problems and many ended up quitting their job within just a few months.
- Meanwhile, the entrepreneurship program increased self-employment and raised income by 33% and decreased the likelihood of taking an industrial job by half.
- On average, it seems people preferred entrepreneurial work to industrial labor.
The authors also wrote an op-ed that appeared in The New York Times entitled, “Everything We Knew About Sweatshops Was Wrong“. This title is a head-scratcher, because the study was performed with 5 manufacturing firms in Addis Ababa, Ethiopia. This is not a study on the impact of industrial labor worldwide. Obviously, the authors know this and I suspect this title was influenced by editors at the NYT. I think this is an important and worthwhile study that just so happens to have found results that were contrary to my priors. This study, however, still presents preliminary results and should encourage more research into this topic. A better title might have been “Some of the things some economists think about sweatshops in Ethiopia may be incomplete and we need to do more research”, but who would click on that article…
In fact the NYT piece was followed this line of reasoning much more closely. Here is Blattman and Dercon offering a much more nuanced conclusion:
We have to be careful about generalizing from five businesses in one country, but this study has still shaped our views of factory work. Industrialization is not a quick fix. The first defense of industry probably still holds: Over time, a booming sector tends to improve labor conditions and bid up wages as more businesses compete for workers. But the path there isn’t smooth. In the short run workers seem to share few of the benefits but a heavy burden of the risks — a burden borne by the desperate and the uninformed.
For poor countries to develop, we simply do not know of any alternative to industrialization. The sooner that happens, the sooner the world will end extreme poverty. As we look at our results, we are conflicted: We do not want to see workers exposed to hazardous risks, but we also worry that regulating or improving the jobs too much too quickly will keep that industrial boom from happening.
It is a difficult path to walk. But supporting insurance systems and encouraging companies to adopt modern management strategies and worker protections could be a way to travel that path faster and more safely.