The Paris financial market in the 19th century: an efficient multi-polar organization?

The literature in financial history usually considers London as the only centre of the late 19th century's financial globalization, and explains it at least in part by the efficient microstructure (organization) of the London Stock Exchange (LSE). The LSE is characterized as having been a little regulated market, where entry was easy both for traders and issuers [Michie (1998), Neal (2004), White (2006)]. The LSE microstructure is also considered as the natural and optimal one by much of the theoretical literature on stock markets, which argues that free entry decreases transaction costs and increases both liquidity and diversification, resulting in economies of scale attracting traders, issuers and buyers. Our paper tries to explain why the Paris Bourse was able to be so successful in spite of the supposedly inefficient monopoly and regulations that the State imposed it. We focus on the fact that the Paris market actually included several different market places: the Parquet (the official Bourse, organized by the agents de change), the Coulisse, the Marché libre, and inter-bank direct operations. We argue that this multi-polar organization, was efficient, relying on the specialization it allowed, and the complementarities it helped develop among markets. We incorporate in the discussion the recent theoretical literature that shows that no single market can satisfy the heterogeneous preferences of all issuers and investors, so that a multi-polar organization can be a superior solution. We demonstrate our claim by looking not only at the rules but also at the actual functioning of the Parquet thanks to its archives which we recently classified. These archives also allow us to build new statistical series which permit evaluating the performances of the Parquet during the 19th century: volumes traded, seat prices, transaction costs, and operational risks. If one supposes that the Parquet was the least efficient segment of the Parisian market, this will provide us with a lower bound for the global efficiency of that market, which should be compared with other markets on similar concrete grounds.