Rumors and Runs in Opaque Markets: Evidence from the Panic of 1907

Working Paper: CEPR ID: DP10497

Authors: Caroline Fohlin; Thomas Gehrig; Marlene Haas

Abstract: Using a new daily dataset for all stocks traded on the New York Stock Exchange between 1905 and 1910, we study the impact of information asymmetry during the liquidity freeze and market run of October 1907 - one of the most severe financial crises of the 20th century. We estimate that the market run drove up spreads from 0.5% to 3% during the peak of the crisis and, using a spread decomposition, we identify information risk as the largest component of illiquidity. Information costs rose most in the mining sector - the origin of the stock corner and a sector with among the worst track records of corporate governance and accounting. We find other hallmarks of information-based illiquidity: trading volume dropped and price impact rose. Despite short-term cash infusions into the market, the market remained relatively illiquid for several months following the peak of the panic. Notably, market illiquidity risk is priced in the cross section of stock returns. Thus, our findings demonstrate how opaque systems allow idiosyncratic rumors to spread and amplify into a long-lasting, market-wide crisis.

Keywords: Information Risk; Liquidity Risk; Price Discovery; Rumour-based Panic; Microstructure

JEL Codes: G00; G14; N00; N2


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Panic of 1907 (N22)increase in bid-ask spreads (G19)
information risk (D80)illiquidity (G33)
worse corporate governance and accounting practices (G38)increased illiquidity (G19)
decreased trading volume (G14)increased price impact (F69)
market illiquidity risk (G10)priced in stock returns (G12)

Back to index