The first experimental paper in economic history

... hits the working paper stage. I meant to do this for... ? maybe a decade? Anyway, I finally found the right co-authors, time, and money to do it. I always loved the elegance of experimental economics; no problem with internal validity here. What do we learn? A lot, in general, but external validity can be an issue... So why and how to apply this to history? The idea is simple, actually - instead of using market conditions in the lab that are at best loose metaphors for financial markets, we use concrete, detailed features of a historical asset market where things really went wrong: the market for South Sea shares in 1720. We then switch of these features one by one, in a bid to pin down what was really responsible for the "mother of all bubbles". The paper is over at SSRN and the abstract is

Major bubble episodes are rare events. In this paper, we examine what factors might cause some asset price bubbles to become very large. We recreate, in a laboratory setting, some of the specific institutional features investors in the South Sea Company faced in 1720. Several factors have been proposed as potentially contributing to one of the greatest periods of asset overvaluation in history: an intricate debt-for-equity swap, deferred payment for these shares, and the possibility of default on the deferred payments. We consider which aspect might have had the most impact in creating the South Sea bubble. The results of the experiment suggest that the company’s attempt to exchange its shares for government debt was the single biggest contributor to the stock price explosion, because of the manner in which the swap affected fundamental value. Issuing new shares with only partial payments required, in conjunction with the debt-equity swap, also had a significant effect on the size of the bubble. Limited contract enforcement, on the other hand, does not appear to have contributed significantly.