دانلود رایگان مقاله نوسانات ویژه در بازارهای آتی کالای قیمت گذاری شده

Abstract

This article investigates the relationship between expected returns and past idiosyncratic volatility in commodity futures markets. Measuring the idiosyncratic volatility of 27 commodity futures contracts with traditional pricing models that fail to account for backwardation and contango leads to the puzzling finding that idiosyncratic volatility is significantly negatively priced cross-sectionally. However, idiosyncratic volatility is not priced when the phases of backwardation and contango are suitably factored in the pricing model. A time-series portfolio analysis similarly suggests that failing to recognize the fundamental risk associated with the inexorable phases of backwardation and contango leads to overstated profitability of the idiosyncratic volatility mimicking portfolios.

5. Conclusions

The commodity pricing literature documents that commodity futures risk premia depend on considerations relating to inventory levels, roll-yields, hedging pressure, past performance, total volatility, skewness or liquidity. This article considers idiosyncratic volatility as another potential signal of expected commodity futures returns. Theoretically, the presence of trading costs and the non-marketability of producers' claims suggest that idiosyncratic volatility should be priced (Hirshleifer, 1988). Empirically, however, the evidence to date is not clear-cut (Bessembinder, 1992; Rouwenhorst & Tang, 2012). A pitfall of the extant empirical studies is that they employ pricing models that fail to recognize the fundamentals of backwardation and contango. This calls for a reassessment of the evidence on the pricing of idiosyncratic volatility using both traditional and long-short commodity pricing models. Using a similar methodology as that employed by Ang et al. (2006, 2009) to analyze the pricing of idiosyncratic volatility in equity markets, we establish that inferences on the relation between idiosyncratic volatility and expected commodity futures returns crucially hinge on the asset pricing model used to measure the idiosyncratic volatility signal. When the asset pricing model fails to recognize the backwardation and contango dynamics, idiosyncratic volatility seemingly commands a puzzling negative risk premium and, relatedly, mimicking portfolios that systematically buy low idiosyncratic volatility commodities and short high idiosyncratic volatility commodities offer sizeable abnormal returns. Prima facie these results extend those of Ang et al. (2009) from international equity markets to commodity futures markets. However, if the methodology of Ang et al. (2009) is instead deployed using suitable benchmarks that incorporate the commodity risk premia related to the backwardation versus contango fundamentals, then idiosyncratic volatility is not priced cross-sectionally and the abnormal performance of long-short idiosyncratic volatility mimicking portfolios vanishes. These conclusions are unchallenged when illiquidity is considered as a risk factor. They are also robust to the inclusion in the various pricing models of “missing” factors such as expected idiosyncratic skewness or past returns that have been shown to rationalize the puzzling negative pricing of idiosyncratic volatility in equity markets.