What fraction of output was produced by enslaved people?

 

Paul Rhode has an important new paper in the January issue of Explorations in Economic History ("What fraction of antebellum US national product did the enslaved produce?." 2024. Explorations in Economic History 91). Rhode frames the argument against Ed Baptist’s claim that “almost half of the economic activity in the United States in 1836, derived directly or indirectly from cotton produced by the million-odd slaves…”, which has been not just repeated but exaggerated by others. It was easy to show that Baptist’s claim had no foundation in either theory or evidence and was purely a creation of Baptist’s imagination (see here), but the question of how large a fraction of output was produced by enslaved labor remained unanswered.

 

I have for years suggested that the place to begin an answer to this question is the labor supply. Begin with the percentage of the labor supply accounted for by enslaved people and then ask why the percentage of output would be either higher or lower than the percentage of labor (see for instance here in my thoughts on Stelzner and Beckert’s attempt to answer the question). I was too lazy to do the work, but fortunately for us Paul Rhode was not.

He estimates that the percentage of output was probably about the same as the percentage of the population, around 12 percent. He also does a series of robustness checks using alternative assumptions that raise or lower the estimate a little bit. As with all such estimates people will be able to quibble, but I think he makes a pretty strong case that it is difficult to produce an estimate that is much larger than the percentage of the population.

Rhode’s conclusion is not just important because he debunks Baptist. The flaws in Baptist’s work were so obvious that only people so enamored with his conclusions that they were willing to completely disregard all evidence continued to support his work. Rhode’s estimate is important because, like recent work by economists Hornbeck and Logan and the economic historian Joe Francis, it lays waste to a tradition rooted in the work of Fogel and Engerman. In Fogel and Engerman, slavery, although morally repugnant, was not just profitable it was efficient and highly productive. Later economists, including Engerman and Sokolof, would argue that despite its productivity slavery had negative long run consequences (see here for instance). But this recent work says that slavery did not just have negative long -term consequences, it was a massively inefficient misallocation of resources while it was taking place. Rhode’s conclusion that the fraction of output produced by enslaved labor was about the same as the fraction of the population accounted for by enslaved people means that the fraction was much less than the percentage of the labor force accounted for by enslaved people, about 22 percent in 1860.