Nanyang Business School Forum on Risk Management and Insurance
Catastrophic Risk and Institutional Investors: Evidence from Institutional Trading around 9/11
Over the last several decades, institutional investors (e.g., mutual funds and pension funds) have come to dominate global financial markets. Collectively, institutional investors are the majority shareholders of most publicly-traded companies. As a result, institutional investors have been playing an increasingly significant role in almost all aspects of financial markets. An interesting question that naturally arises is: what role do institutional investors play during financial market crises caused by catastrophic events? Are they sophisticated investors who provide a “steady hand” in stabilizing financial markets, or do they “panic” like many retail investors and thereby exacerbate such crises? This is an important research question with relevant practical and policy implications.
We attempt to address this question by analysing detailed institutional trading records around the series of terrorist attacks on September 11, 2001 (hereafter 9/11). We choose to focus on 9/11 because it is an unexpected, unprecedented, sudden exogenous shock which caused large market-wide losses. It is therefore an ideal natural experiment and “stress test” to examine the role played by institutional investors during market crises caused by catastrophic events.
We are not alone in choosing to focus on 9/11. For example, Chen, Doerpinghaus, Lin, and Yu (2008) study the short-run claim effect and long-run growth effect of 9/11 on insurance companies. Although terrorist attacks happened before, they do not have such tremendous and wide-spread impact as 9/11. It changed the way we think of terrorism (Karolyi and Martell 2010). 9/11 also differs from other catastrophic events such as earthquakes (Shelor, Anderson, and Cross 1990; Li, Tang, and Liao 2015) and hurricanes (Lamb 2005). Most natural disasters only have local effects, while 9/11 has market-wide impact on the economy and financial markets. Moreover, natural disasters are typically anticipated at least to some extent, while 9/11 was completely sudden and unexpected. U.S. stock markets were forced to close before opening on September 11, and they remained closed until September 17, making it the longest market closure since the Great Depression. Focusing on this important event also allows us to conduct in-depth and thorough analysis of its impact across hundreds of institutional investors and thousands of stocks.
Examining the role of institutional investors during extreme market movements, Dennis and Strickland (2002) find that quarterly institutional ownership is positively related to the magnitude of stock returns on the days when the absolute value of CRSP market index return is greater than 2%. Their findings are consistent with institutional investors contributing to market volatility when there are large market movements. This implies a destabilizing role played by institutional investors, which is somewhat puzzling and surprising given that institutional investors are typically considered to be sophisticated investors. On the other hand, their inferences are based on public institutional holdings data which are only available quarterly. We make use of proprietary transaction-level institutional trading data provided by Abel Noser, which contain detailed information on transactions executed by hundreds of institutions, including shares traded, trading prices, as well as trading direction (buy or sell) of each transaction. This enables us to provide direct evidences on the role of institutional investors during market crises such as 9/11.
Our main findings can be summarized as follows. Our sample institutions are net buyers for both pre- and post-9/11 periods. Both buying and selling activities of our sample institutions increase after 9/11. Daily average principle traded increases by 46% immediately after 9/11. Institutions’ total trading activities significantly increase while their buying relative to selling does not change much. Average daily buy proportion remains almost unchanged at around 52% after 9/11. Hence, we provide direct evidences that institutions act as liquidity providers and play a stabilizing role rather than engaging in panic selling as the stock market plummeted after 9/11. We also study whether institutions are able to profit while they provide liquidity during this market crisis period. We divide stocks into quintiles by institutional net buying (shares bought minus shares sold normalized by shares outstanding) and find that stocks most heavily bought by institutional investors earn higher abnormal returns than stocks most heavily sold by institutional investors, indicating their trading is profitable while providing liquidity to the overall market.
We further compare trading patterns of two types of institutional investors: investment managers versus plan sponsors. We find that the buy proportion after 9/11 increases for plan sponsors while decreases for investment managers. This is consistent with the notion that investment managers conduct more forced selling because of redemption by fund investors, while plan sponsors may face less redemption pressure from plan participants. We also study institutional trading for various industry sectors. We find that institutional investors are net buyers (sellers) for most sectors with negative (positive) contemporaneous market-adjusted returns. We also find that institutional investors earn positive abnormal returns for the sectors that they are providing liquidity while earn negative returns for the sectors that they are not providing liquidity. This implies that institutions’ liquidity-provision trading can positively predict future returns during 9/11. Finally, in a regression framework, we find that institutional trading can only positively predict returns for stocks that institutional investors are providing liquidity. A one standard deviation increase in institutional net buying of liquidity provision stocks contributes to a 10.8% increase in returns for the following 3-month period. This indicates that the profitability of institutional investors mainly comes from stocks for which they are providing liquidity.
The complete paper is available at: