|
|
Authors: Bruno M Parigi, Loriana Pelizzon and Ernst-Ludwig von Thadden
The paper shows that corporate governance affects firms’ stock returns and also how the quality of corporate governance is chosen endogenously. The model predicts that in equilibrium the quality of corporate governance correlates positively with β and idiosyncratic volatility and negatively with returns on assets. Various tests with U.S. firm data using the corporate governance index of Gompers, Ishii, and Metrick (2003) confirm these predictions.
This paper is motivated by the attempt to investigate why corporate governance choices matter for stock returns if the stock market understands and discounts the managerial agency problem correctly. To address this question researchers have constructed a model that incorporates two key elements of the managerial agency problem into the CAPM, with counterveiling effects of strict corporate governance. On the one hand, corporate governance affects the amount of private benefits a manager can enjoy, on the other it affects the disutility of effort in providing flexibility and initiative in managerial decision making. The trade off between these two motives yields an optimal degree of strictness of corporate governance.
The governance choice affects cash flows and through cash flows affect the firm’s β and idiosyncratic risk. Interestingly, the partial equilibrium effect, excluding the optimal ownership adjustment on the capital market, yields a negative relation between governance strictness and β, very much in the spirit of classic theories of levered beta: the stricter is governance, the less the manager can divert, the larger is the pie to distribute to shareholders, and thus the smaller β (which is a per unit measure). But in full equilibrium, taking the manager’s choice of ownership stake into account, this effect is reversed: researchers predict the quality of corporate governance to correlate positively both with β and with idiosyncratic risk. The reason is that as long as effort is not just a scale factor but is able to improve the trade off between the variability of cash flows and their average, the value of the firm is larger but the risk does not change. Thus as governance worsens the unit of risk (systematic and idiosyncratic) for each dollar invested is lower.
These predictions are strongly confirmed in researchers´ empirical analysis. Researchers´model also makes predictions about the relationship between the variables just discussed, and book value measures of returns, such as ROA, and managerial ownership. While researchers can confirm the prediction about returns on assets with our data, too, the test of their predictions about ownership is quite di¢cult because they do not have sufficiently good ownership data. A comprehensive answer is left for further research.