Insider Trading – Chasing Private Information

Marcin T. Kacperczyk

Imperial College London – Accounting, Finance, and Macroeconomics; National Bureau of Economic Research (NBER)

Emiliano Pagnotta

Imperial College Business School

November 24, 2015


Abstract Using a sample of 3,586 illegal insider trades documented by the SEC, we examine whether asset prices and volume reveal firm-specific private information to markets. We find that information embedded in equity markets is a weak signal of private information. In turn, information embedded in option markets offers a strong signal of informed trading. We show that volume is generally more informative about insider trading than are prices, and that the most robust metrics combine both option and stock volume. Further, we show that the impact of insider trades holds irrespective of whether insiders trade strategically or not. Finally, we document significant information spillovers from equity to option markets, but not vice versa. Overall, our results provide strong and novel guidance in the search for private information.

Insider Trading – Chasing Private Information – Introduction

Asymmetric information is ubiquitous in economics and finance (e.g., Akerlof, 1970; Grossman and Stiglitz, 1980). A large number of theories in asset pricing and corporate finance, ranging from intermediation to asset management and market efficiency, rely on the presence of privately informed investors. To test these theories one faces the challenge that information sets are almost never observable. In an attempt to identify private information, the literature has proposed a wide array of empirical proxies that are based on publicly available data.1 Such measures, however, may not reflect private information but a number of spurious factors. For example, changing levels of illiquidity may be due to a systematic liquidity component or uninformed demand pressure.

Ultimately, the decision maker is never sure whether the signals she observes are informative or not. In this paper, we consider a novel setting–insider trading cases–in which we can directly evaluate the quality of market signals. Specifically, we hand-collect a comprehensive sample of insider trading investigations by the SEC documenting how certain individuals trade on secret material information. Our sample of legal cases involves a large number of trades in several hundred companies over the period 1995-2012. The advantage of using insider trading data is that we can observe the dynamics of market signals at times when private information is used and, therefore, we can assess their ability to identify private information.

Our choice of market signals is guided by prior theoretical and empirical research. In particular, models of adverse selection imply that both the amount of trading and asset prices are affected by the presence of informed traders. This premise motivates our use of two types of potential information measures: (i) those that rely on volume and (ii) those that rely on prices. Similarly, some of our measures are based on stock-level and some are based on option-level data. Our results carry two main messages: (1) information measures which are based on options data perform better than those that are based on equity data; (2) private information is better reflected in volume than in prices.

A typical approach to evaluate the quality of any information measure has been to test whether such measure predicts stock returns. We begin our analysis by assessing the predictive power of our information measures for next-day and next-month stock returns. To this end, we analyze the data on individual daily stock returns of U.S. companies for the period 1995-2012 corresponding to our insider trading data. We consider two different groups of measures to capture private information: the first group includes nine measures derived from the stock market data, such as various spread measures, price impact, volatility, Kyle’s lambda, absolute order imbalance, and illiquidity; the second group includes nine measures derived from the options market data, such as quoted spread, implied volatility (its spread and skewness), levered option volume, and option/stock volume. Our regression results show that several of the measures from both markets correlate strongly with future returns. Using this evidence one could conclude that the measures are informative about the extent of private information revealed in financial markets. However, this interpretation is only one among many others simply because the measures could be correlated with returns for reasons other than information. This is a spirit with which we believe our empirical context of insider trading is a superior way to evaluate the informational content of the signals.

Our experiment is based on a comprehensive sample of 370 insider trading cases filed by the SEC over the period 2001-2012. Each case includes a detailed description of situations in which individuals execute their trades using material and non-public information. An example of such trade would be buying stocks of a company by a spouse of a CEO based on the private information about the exceptional quality of the company’s earnings. We collect detailed information about insider traders, the companies they trade, the exact dates of the trade and information acquisition, the types of instruments that are being used to trade, and the corporate events to which the trades correspond. We, additionally, collect information about the dates such information is released to the public. Overall, our final sample covers 3586 trades that involve 547 companies. The trades span a long period of 1995-2012 and represent the vast majority of industry sectors.

insider trading

insider trading

See full PDF below.