Discussion of To Trade or Not To Trade: The Strategic Trading of Insiders around News Announcements

Abstract

Korczak, Korczak and Lasfer (2010) (hereafter KKL) examine the likelihood and amount of corporate insider (company director) trading before a news announcement about a corporate event. The novel feature of the paper is that it examines insider trading after distinguishing between good and bad subsequent news announcements. The authors construct a unique dataset of 119,179 news announcements and 8,086 insider trades for 797 UK companies over the period 1999-2002. After separating out the non-confounding announcements, they divide the sample into 39,617 good news announcements and 38,634 bad news announcements, where the definition of good and bad news depends on the two-day stock market return around the day of the announcement. This good-bad classification is determined by how the stock market responds to the announcement relative to the value of the stock on the day prior to the announcement. They match these announcements with the shares traded by corporate insiders in the 30 days before the news announcement. This yields 4,083 good news events and 1,262 bad news events preceded by insider trading. KKL’s approach is then to use a logit analysis to examine the determinants of whether or not insiders trade before a news announcement (the occurrence of insider trading) and an OLS regression of the net quantity of insider trades in the month before the announcement to identify the determinants of the magnitude of insider trades. KKL record two major findings. First, corporate insider trading is more prevalent before good news than bad. Second, the quantity of insider trading before good news is a concave function of the information content (revealed at announcement of the corporate event). Insiders’ trades increase as the information content of the news increases, but it is argued due to regulatory scrutiny, for news announcements with large price impacts, the quantity of insider trading levels out. In