The great divergence: history or path dependence? Results from the Americas*

This paper synthesizes our results from work on growth in the Americas and especially on the former slave regions. Firstly, we found that significant differences in GDPpp emerged only after slavery was abandoned. At abolition, there was a fall in almost all of them but most started to grow again, some quite rapidly. Secondly, we formalized the notion of history dependence in annual time series of GDP per person as a homogenous Poisson stochastic process and tested the series for Brazil, 1822-2000 and the USA, 1869-1996. The US passed, but Brazil did not, due to its stagnation in the late 19th century. Thirdly the principal-agent theory of the firm is then used to explain the stagnation. By at least the early 19th century, New World slave plantations had evolved some of the mechanisms firms use to supervise the effort of their workers at multiple tasks, which increase productivity. Since abolition meant they had to pay their workers more, then without technical progress or changes in product prices, they may become unviable and the economy can collapse into lower-productivity family farms or worse.