Using a new daily dataset for all stocks traded on the New York Stock Exchange from 1905 to 1910, we provide the first in-depth, microstructure analysis of the Panic of 1907 — one of the most severe financial crises of the 20th century — and quantify the critical role of asymmetric information in the generation of panic conditions. We first show that quoted equity bid-ask spreads rose six-fold, from 0.5% to 3%, during the peak weeks of the crisis. We then implement a spread decomposition procedure and pinpoint the source of the illiquidity spike in the adverse selection component, that is, the fear of informed trading. Moreover, we show that information costs rose most steeply in the mining sector — the origin of the panic rumors — and in other sectors with opaque corporate reporting standards. In addition to wider spreads and tight money markets, we find other hallmarks of information-based liquidity freezes: trading volume dropped and price impact rose. Importantly, despite short term cash infusions into the market, and abatement of the run, we find that the market remained relatively illiquid for several months following the panic. We go on to show that rising illiquidity enters positively in the cross section of stock returns. Thus, our findings demonstrate how opaque markets can easily transmit an idiosyncratic rumor into a long-lasting, market-wide crisis. Our results also demonstrate the usefulness of illiquidity measures to alert market participants to impending market runs. Joint work with Caroline Fohlin and Thomas Gehrig.

Link to paper: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2591343

Discussant: Marius Zoican (VU University Amsterdam)