With the exception of Ross’ APT, classical theories of asset pricing are based on the assumption that the market is populated entirely by rational expected utility maximizing individuals. This assumption was not unreasonable in the mid-1960’s when the CAPM was formulated, since around 85% of US equities were then held by domestic households. However it has become increasingly hard to maintain in subsequent years since, as shown in Figure 1, the share of US common stocks held by households has now dropped to only about 35%.
The share held by domestic institutions is currently 48.8%, while the share held by foreigners is 16%. If we combine the foreign holdings, which are mainly institutional, with the holdings of domestic institutions, then the share of US equities held by institutions is of the order of 65%. Institutional investor demands differ from those of individual investors because of the agency problem that arises from delegated portfolio management: while direct investors are typically concerned only with the return characteristics of their portfolios, investment managers, like corporate managers, have other concerns. Therefore it seems likely that the institutionalization of the equity markets will have a significant effect on the pricing of securities calling for a model of equilibrium that takes account of agency effect.
In this paper we are concerned with the effects on stock prices of the agency problems of institutional investment managers whose performance is evaluated relative to a stock market index or benchmark. For such managers, the benchmark plays the role of the riskless asset in conventional portfolio theory and, as a result, in equilibrium the expected returns on securities which covary with the benchmark returns are depressed relative to the predictions of a classical asset pricing model.
The use of a benchmark return to evaluate investment performance can be justified at several levels. First, since the return on a managed portfolio depends on both the random state of nature and the action of the manager, the analysis of Holmstrom (1979) suggests that it will in general be optimal for the portfolio manager’s reward to depend, not only on the portfolio return, but also on any signal that is informative about the manager’s action. One such signal is the return on an unmanaged portfolio such as a market index, and Carpenter et al. (2006) provide conditions under which it is optimal to reward the manager with a bonus which is proportional to the difference between the returns on the managed portfolio and the returns on the bench mark portfolio. On the other hand, Admati and Pfleiderer (1997) question whether benchmark-adjusted compensation schemes are useful.
More pragmatically, Cornell and Roll (2005, p61) argue that it makes sense to use a benchmark because ‘the client’s alternative to hiring an active manager is to use a passive manager who matches a widely followed benchmark’. Chevalier and Ellison (1999) provide empirical evidence that the probability that a mutual fund manager leaves his or her position is strongly dependent on the realized Jensen’s alpha, which is closely related to the excess return relative to the market index. Finally, flows into mutual funds depend strongly on relative performance, which will be well approximated by the excess return relative to a benchmark portfolio that may represent the market return or the return on a segment of the market such as growth or value stocks.
Since asset prices are determined by the supply and demand for securities, we should expect the concerns that affect the demands of institutional as well as of individual investors to be reflected in prices, unless institutions are sufficiently transparent that the principals or individual investors on whose behalf the portfolios are managed are able to offset the distortions in institutional portfolios that arise from agency considerations, or the resulting price effects are arbitraged away. It seems unlikely that individual investors have the knowledge or the expertise to make transactions to offset those of the managers, since a primary reason for committing their funds to managers in the first place is a lack of investment expertise. Therefore the ‘limits to arbitrage’ arguments of Shleifer and Vishny (1997) suggest that the concerns of institutional investors are likely to leave at least residual traces in the data.
Download
PDF Ebook Agency and Asset Pricing
