Stock Price Fragility

by Robin Greenwood & David Thesmar

Overview — Does the composition of ownership of a financial asset influence future returns and risk? Previous economic research has documented significant price effects of investor demand in numerous settings, including retail demand for options, investor demand for bonds, and mutual funds' flow-driven demand for stocks. This paper provides a methodology to identify assets that are vulnerable to such investor demand shocks. The central idea is that assets are risky if the current owners of the asset face correlated liquidity shocks—i.e., they buy and sell at the same time. We call assets with a high concentration of owners who trade in the same direction "fragile." A related concept is "co-fragility." Two assets are "co-fragile" if their owners have correlated trading needs, even if the holdings of these owners do not directly overlap. The authors build measures of fragility for U.S. stocks between 1990 and 2007. Consistent with their predictions, more fragile stocks are more volatile, and two co-fragile stocks exhibit high correlations among their stock returns. Key concepts include:

The link between ownership structure and non-fundamental risk.
The link between common ownership structure and commonality in returns.
Relating the liquidity needs of an asset's owners to the risks of the asset.
The concept of fragility expresses the three reasons why a stock may be volatile: ownership concentration, volatility of liquidity needs, and correlation of liquidity needs across owners.

Author Abstract

We investigate the relationship between ownership structure of financial assets and non-fundamental risk. An asset is fragile if its owners collectively have to buy or sell. Such assets are susceptible to non-fundamental price movements. An asset can be fragile because of concentrated ownership, or because its owners face correlated liquidity shocks, i.e., they must buy or sell at the same time. Two assets are "co-fragile" if their owners have correlated trading needs, even if the holdings of these owners do not directly overlap. We formalize this idea and apply it to the ownership of US stocks between 1990 and 2007. Consistent with our predictions, fragility strongly predicts future price volatility, and co-fragility predicts cross-stock return comovement. 48 pages.