RDP 9802: Systematic Risk Characteristics of Corporate Equity 1. Empirical Objectives

What can be inferred about the behaviour of publicly listed corporations from the behaviour of their security returns? This paper develops a body of evidence characterising the degree of co-movement between equity returns of individual firms and the return on the entire equity market. The focus is on how the degree of co-movement changes through time and on how it is related to observed characteristics of the firms. The evidence strongly suggests that the degree of co-movement associated with a given firm's equity converges, through time, to the market average degree of co-movement. Previously, this convergence phenomena was thought to have been a statistical artefact of the estimation techniques. Robust evidence in this paper refutes this hypothesis, suggesting instead that the driving force behind convergence is the preferences of the managers or owners who control firms.

The degree of co-movement between a firm's equity return and the return on the equity market is commonly referred to as systematic risk. In finance parlance, systematic risk is that component of risk which cannot be diversified away by investing across a wide variety of assets. In this sense, the systematic risk of an individual asset return is that part of equity return volatility driven by economy-wide shocks rather than idiosyncratic or asset-specific shocks. Generally, the systematic component of equity risk is estimated using some normalisation of the covariance between the return on a firm's equity and the return on the market, however, broadly the market is defined. A firm's idiosyncratic equity risk is then defined as the residual variation in the firm's equity return.

In this paper, systematic risk is estimated using monthly equity returns on all stocks listed on the New York Stock Exchange (NYSE) from January 1926 to December 1992. The data are obtained from the Center for Research in Security Prices (CRSP) database. Systematic risk is estimated for slightly fewer than 4,000 firms. In estimating systematic risk, the ‘market’ return is measured as the NYSE weighted-average equity return. This corresponds to the interpretation of systematic risk as a measure of the degree of co-movement between the rate of return on the individual firm and the rate of return on the entire equity market.

The evidence is presented in two stages. First, the law of motion governing the systematic risk is estimated for various sub-samples. Second, the relationship between systematic risk and both the age and size of firms is characterised. The methodology for estimating both the law of motion and the relationships between systematic risk and observed features of the firms is similar to that used by Quah (1996) to examine macroeconomic convergence across nations. The methodology characterises the entire distribution of observations rather than focusing on the representative firm. This type of analysis is shown to deliver greater insight, especially in the analysis of the link between firm size and systematic risk.

Aside from providing robust evidence that the convergence of systematic risk is driven by the behaviour of firms rather than by flaws in the estimation techniques, this paper develops several interesting relationships between the systematic risk of equity returns and characteristics of firms. For example, firms with extremely high systematic risk and firms with extremely low systematic risk both have a relatively high probability of being restructured compared to firms with systematic risk that is closer to the market average. This paper also shows that larger firms and older firms tend to have systematic risk exposure that is closer to the market average than do smaller and more recently listed firms. Together, the wealth of empirical regularities are strongly suggestive that systematic risk is manipulated within firms. The apparent relationships between systematic risk and firm characteristics reinforces this message because they would not arise if the observed behaviour of systematic risk was driven by estimation techniques alone.

The remainder of this paper is structured as follows. Section 2 examines the older literature associated with tests of market efficiency and asset-pricing models. This examination is necessary to understand many of the issues addressed by an alternative methodology, presented in Section 4, for analysing systematic risk convergence. Section 3 describes the techniques used to estimate systematic risk for each month and firm in the dataset. It describes the data used and documents some of the more elementary features of the systematic risk estimates. Section 4 presents the key findings, beginning with a thorough characterisation of the time-series behaviour of systematic risk. The systematic risk estimates are then related to other properties of equity returns and, importantly, to the size and the age of firms. Finally, Section 5 summarises the findings and ties them into the conclusion that systematic risk is being manipulated by firms.