Catherine D'HONDT, Patrick ROGER
Sentiment measures, based on the trading activity of retail investors, carry some predictive power of future market returns. In this paper, we use such a sentiment measure on two samples of approximately 25,000 individual investors, who differ in their choices when answering MiFID questionnaires, especially in terms of their appetite for information and professional recommendations. Our data covers 51 months from January 2008 to March 2012. We show that the sentiment of investors who disregard free information and professional advice is the best predictor of future returns on a long-short portfolio based on size. Our findings remain valid when controlling for investor characteristics like spoken language (French or Dutch), portfolio value and financial literacy. Our results bring evidence that sentiment is essentially driven by underdiversification and narrow framing by retail investors. When shared by many investors, sentiment can generate long-lived mispricing, which is, therefore, difficult to arbitrage.
Sami ATTAOUI, Moez BENNOURI, Imen MEJRI
We propose a new mechanism of finite-maturity performance-sensitive debt (PSD). Unlike typical PSD, our mechanism relates positively the firm’s performance, captured by the level of its assets, to the coupon rate. That is, when its performance improves, the firm pays a high coupon rate, and when its performance deteriorates, the firm pays a low coupon rate. In a fairly general setting, we provide the necessary and sufficient conditions for our PSD to be efficient, and show that this mechanism can be immune against risk-shifting when augmented with a covenant deterring managers from changing risk following debt issuance. We analyze the characteristics of an example of such a mechanism where the coupon rate is a stepwise function of assets, and find that intermediate credit quality firms benefit most from it. This suggests that this new PSD should be considered as a complement to typical ones.
Christophe MOUSSU, Arthur PETIT-ROMEC
Return on equity (ROE) is a central measure of performance in the banking industry. The reliance on ROE emerged with the risk management approach that inspired bank capital regulation. In this paper, focusing on the 2007-2008 crisis, we empirically assess the validity of ROE as a performance measure in the banking industry. We document that pre-crisis ROE is a strong predictor of both bank standalone and systemic risk during the crisis. These results are unchanged for the 1998 crisis. Banks appear to be special as the same association between pre-crisis performance measures and the materialization of risk in crisis periods is not observed for firms outside the banking industry. Complementary tests confirm the existence of monetary incentives associated to ROE. Overall, our findings challenge the use of ROE as a main performance measure in banks and its incorporation in bank executives’ compensation contracts.