Risk and return revisited

Traditional asset pricing models, such as the celebrated capital asset pricing model, argue that only systematic risk should be priced in the market; specific or idiosyncratic risk does not deserve a risk premium. Recent empirical findings have raised serious challenges to this belief. It appears that beta - a measure of systematic risk - has little power in explaining cross-sectional risk and return relationships over long periods of times, while other variables, such as firm size, appear to be more useful risk proxies. The authors find that idiosyncratic volatility is highly correlated with firm size, and that it plays a powerful role in explaining expected returns. Some possible explanations for this phenomenon are offered.