Banks

The crisis test

Banks are one of the few sectors with their own evidence. Cziraki (2018, Journal of Financial Intermediation) shows that from mid-2006, when US house prices began to fall, insiders at banks with high housing exposure sold more than insiders at low-exposure banks, with a wider gap for CEOs, and that insider trading in 2006 predicted stock returns during the crisis.

Akin, Marín and Peydró (2020, Economic Policy) find that share sales by the top five executives before the real estate reversal of 2006 predict bank performance during the crisis of 2007 and 2008.

Opacity in normal times

Outside a crisis the picture is weaker. In a Bank for International Settlements working paper (No 697, 2018, not a journal article), Spargoli and Upper find that bank insider sales earn no abnormal return and do not predict returns, while purchases do predict returns, though less than at other firms. They find no evidence that banks are more opaque than other companies.

Reading bank filings

The crisis studies drew their results from patterns across many banks over months, not from single trades. Insurers and asset managers sit in the same bucket on this site but were not part of these bank samples.