Are there certain kinds of insider trades that are more exploitable than others? In their August 2012 paper entitled “Insider Trading Patterns”, David Cicero and Babajide Wintoki define and examine two kinds of insider trading: (1) isolated trades (no trades in prior or subsequent months; and, (2) sequenced trades (occurring in successive months). They hypothesize that when insiders have short-maturity (long-maturity) information, they tend toward isolated (sequenced) trading. They measure insider trading informativeness via post-trade abnormal returns to associated stocks, calculated relative to returns for matched stocks with no insider trading or via adjustment based on a four-factor (market, size, book-to-market, momentum) model of stock returns. Using a broad sample of insider trades in U.S. stocks aggregated monthly by insider, along with associated stock prices and firm characteristics, during January 1986 through December 2011, they find that:
- Isolated (sequenced) trades constitute about 40% of insider trades and 25% of active insider trading months over the sample period. Typical sequenced trading spans roughly three months. Sequenced trades are relatively more frequent for large, value stocks.
- Depending on the method of calculating abnormal returns, the average next-month gross abnormal return for stocks with isolated insider sales (purchases) is 0.6% to 1.0% lower (0.6% to 1.5% higher) than that for stocks with sequenced sales (purchases).
- Over the three months following the end of a sequence of insider sales (purchases), the average gross abnormal return is 1.5% to 2.0% (2.0% to 3.0%).
- These patterns are more pronounced when limited to the trades of top executives.
- A trading strategy focused on the ends of trade sequences (by waiting for a month with no trades) generates significant four-factor gross alphas. Specifically, a trading strategy that each month buys stocks after sequences of purchases and sells stocks after sequences of sales by all insiders (top executives only) and holds for one month generates a gross monthly alpha of 1.7% (2.4%), or 22.6% (32.5%) annualized. Most of the alpha (about three quarters for top executives) comes from the long side. After accounting for trading frictions and shorting costs, the short side may not be profitable.
In summary, evidence indicates that investors may be able to isolate the most exploitable informed trading of insiders by focusing on the ends of sequences of purchases by top executives.
Cautions regarding findings include:
- As noted in the paper, reported returns are gross, not net. Including trading frictions would reduce these returns. Monthly turnover for the specified trading strategy is apparently, by definition, 100%.
- There may be more informative ways to define trading sequences than calendar month aggregation.
- The informativeness of insider trades may change with regulatory shocks, as indicated in “CFOs Still the Best Inside Traders?”.
