|
Traditional asset pricing frameworks distinguish the determinants of the risk of assets as systematic components, i.e., those driven by common, explicit risk factors, and ‘residual’, ‘non-systematic’ or ‘idiosyncratic’ components. The idiosyncratic component can be thought of as the component of returns that is specific to each single security—and also as the residual component from a model that is by nature incomplete.
Within a broader stream of literature that looks to identify empirical pricing ‘anomalies’, several researchers have studied the behavior and role of the idiosyncratic component of stock returns volatility, often documenting aggregate or cross-sectional effects partly incompatible with the assumptions of the traditional models. For instance, the CAPM assumption that idiosyncratic risk should not be priced has been challenged both theoretically and empirically, although with mixed results. But from a modeling standpoint, the behavior of idiosyncratic risk and its interaction with other risk sources is important for a number of reasons. Firm-level volatility accounts for a large share—over 60% according to some estimates—of total volatility and volatility variation. In addition, asset holders and portfolio managers are exposed to idiosyncratic risk because of portfolio or wealth constraints, transaction costs, regulation, informational differences and explicit choice. As a result, improving the knowledge of the determinants of such risk will bear consequence on asset allocation decisions and from a risk-management standpoint improve the assessment of portfolio risk exposures, in terms of breaking down the overall risk into factor exposures and evaluating scenarios for the movement and correlation between them.
|
