Decoding Inside Information

The following post comes to us from Lauren Cohen and Christopher Malloy, both of Harvard Business School, and Lukasz Pomorski of the Department of Finance at the University of Toronto.

In our paper, Decoding Inside Information, forthcoming in the Journal of Finance, we employ a simple empirical strategy to decode the information in insider trading. Our analysis rests on the basic premise that insiders, while possessing private information, trade for many reasons, and that by identifying ex-ante those insiders whose trades are “routine” (and hence uninformative), one can better isolate the true information that insiders contain about the future of firms. Using simple definitions of routine traders, we are able to systematically and predictably identify insiders as either opportunistic or routine throughout our sample. We show that stripping away the uninformative signals of routine traders leaves a set of information-rich opportunistic trades that are powerful predictors of future firm returns, news, and events.

We show that while the abnormal returns associated with routine traders are essentially zero, a portfolio strategy that instead focuses solely on opportunistic insider trades yields value-weighted (equal-weighted) abnormal returns of 82 basis points per month (180 basis points per month). Similarly, in a regression context the combined differences in the coefficients between opportunistic trades and routine trades translate into an increase of 158 basis points per month in the predictive ability of opportunistic trades relative to routine trades. Further, this effect increases with the strength of the opportunistic signal (as measured by the number of trades or trade-size intensity), but is unrelated to the strength of the routine signal.

In exploring the mechanism at work behind our identification and return results, we demonstrate that trades by opportunistic insiders predict future firm news and future firm events, while trades by routine insiders do not. Opportunistic traders have particular predictive ability for the information in future firm-level announcements such as earnings forecasts and managerial forecasts. The most informed opportunistic traders tend to be local non-senior opportunistic insiders, and these traders are likely to come from geographically concentrated firms and poorly governed firms.

Further, we find that opportunistic trading increases the likelihood of SEC enforcement against insiders (while routine trading has no impact). Also, opportunistic traders decrease their trading activity in the wake of increases in the number of news releases by the SEC regarding illegal insider trading cases, consistent with the idea that opportunistic traders are sensitive to the potential costs of illegal insider trading. Finally, we find suggestive evidence that institutional investors mimic the trading of opportunistic (but not routine) traders, and yet provide contemporaneous liquidity to (less informative) routine traders.

Collectively our results suggest that it is possible to identify, out of the hundreds of thousands of investors who trade in the market each year, which traders are truly informed. More generally, decoding the true information in other activities in the market, such as the trades made by banks or institutional investors, may also be a promising avenue. These insights could allow us to understand new characteristics of other informed financial agents and the exact type of information these informed agents possess, which would help investors, market regulators, and all active participants in securities markets develop a clearer picture of the information environment that helps form asset prices.

The full paper is available for download here.

Both comments and trackbacks are currently closed.