Peter Koudijs

Job market candidate

Contact information

C. Ramon Trias Fargas, 25-27
Barcelona
Tel. +31 623 752 321
koudijs_peter@gsb.stanford.edu

Research interests

Economic History, Finance, International Finance

Placement officer

Joachim Voth
jvoth@crei.cat

References

Research

"The Boats that did not sail. News, Trading and Asset Price Volatility in a Natural Experiment" (Job Market Paper)

Abstract:
How much of the short run volatility of asset prices is due to the arrival of news? How much can be accounted for by other factors like behavioral biases or frictions in the market microstructure? In today's markets these questions are difficult to answer because of the complexity of information flows. I use a natural experiment to answer them. During the 18th century a number of British stocks were traded on the Amsterdam exchange and all relevant information from England reached Amsterdam through mail boats. I reconstruct the arrival dates of these boats. This allows me to identify the flow of information directly. I then measure the effect of information on the volatility of the British stocks traded in Amsterdam. Stock prices moved significantly more after the arrival of news. Nevertheless in the absence of new, public information, price changes were still considerable. Volatility in "quiet" periods amounted to between 50 and 70% of that observed in periods with news. I construct a model of share trading with frictions that explains why asset prices move in periods when no new public information reaches the market. The model analyzes how an insider trades on his private information. An informed agent unveils his information only slowly and information asymmetry, although decreasing, remains. Prices will be inherently volatile even if no public news reaches the market. Private information also has an effect on uninformed trading. As information asymmetries gradually decrease, it becomes less costly for uninformed agents to trade. In consequence, uninformed trading increases over time. When the impact of private information on volatility becomes less dominant, a larger fraction of asset price movements is explained by uninformed trading. Empirical results for share trading in 18th century Amsterdam are consistent with the model's predictions.

Tim Hartford (Undercover Economist), Financial Times, April 26, 2008

Winner New Researcher prize, Economic History Society Conference 2008

Research in progress:

"'Common Thieves, Cheaters and Swindlers.' Predatory Trading in the 18th century." (joint with Joachim Voth)
Abstract:
We document two episodes of predatory trading (Brunnermeier and Pedersen JF 2005) in the Amsterdam market for British East India Company Stock (EIC) during June 1772 and January 1773. We first of all provide qualitative evidence that during both episodes certain market participants traded aggressively to benefit from other traders' need to reduce their positions. We then use the fact that EIC stock prices were quoted in both London and Amsterdam to identify the impact of this trading behavior on the stock price. The econometric evidence points to severe mispricing in Amsterdam that lasted up to two or three weeks.

"Secondary markets and sovereign debt. Evidence from the Dutch capital market during the Revolutionary and Napoleonic Wars."
Abstract: What sustains sovereign debt? Broner, Martin and Ventura (AER 2010) argue that the existence of secondary markets ensures that debt ends up in the hands of investors that are most likely to be repaid. Ex ante this enables sovereigns to borrow. I use unique transaction data from the Amsterdam capital market during the Revolutionary and Napoleonic wars (1792 - 1815) to test this hypothesis. During the second half of the 18th century a large number of foreign countries issued bonds in Amsterdam. The risk on these bonds increased after the start of international conflict in 1792. I document the national and religious background of brokers active in this market for foreign bonds and I argue that some brokers were more likely to have international connections than others (e.g. German Lutherans or French Huguenots). I show that well connected brokers were more likely to buy bonds with higher default risk. This supports the secondary market explanation of sovereign debt.