Evidence on gains trading by financial institutions: Its incidence, motivations, and the security market's reaction

Much attention has been paid by the business press, regulators, and standard setters to the practice of gains trading, selling appreciated securities to recognize gains while securities with unrealized losses are held to avoid recognizing those losses. Two standards were enacted to help curtail the practice, or at least make its incidence more apparent to financial statement users. At present, it is unclear what impact, if any, the new standards have had on the perceived problem.; My first research question concerns the extent of gains trading by banks. I examine the proportion of banks engaging in gains trading, as well as the magnitude of gains trading, and consider whether the standards enacted since 1990 helped to curtail the practice. The second research question concerns economic incentives for management to engage in gains trading. Have these incentives changed as new standards were enacted? My final research question concerns the market's perception of gains trading. How have investors valued gains and losses arising from gains trading, and how has the valuation changed as disclosures changed?; For the first research question, I find that while the proportion of banks engaging in gains trading does not decrease after enactment of either standard, the magnitude of gains trading does decline after enactment of SFAS No. 115.; In examining possible incentives for managers to engage in gains trading, I focus on an earnings management effect. However, I find only weak evidence that gains trading is related to a decrease in current earnings, and this relation does not moderate with enactment of the new accounting standards. Alternatively, results suggest that the practice of gains trading may be motivated by bank regulatory requirements. Finally, I find no support for either tax minimization or liquidity motivations for the practice of gains trading.; With respect to the market's perception of gains trading, there is no indication that the standards affected the pricing of realized gains. However, the standards appear to be associated with increases in the valuation weight on unrealized losses, conditional on gains trading.