Measuring The Oil Risk Effect On Industry Volatility Shocks

Measuring The Oil Risk Effect On Industry Volatility Shocks

Bernard Ben Sita

Lebanese American University

January 19, 2016


I examine the information sequential hypothesis in complementary oil markets. Unlike the underreaction hypothesis suggested as an explanation to the lagged negative oil effect of financial return, a sequential information schedule through crude oil and gasoline provides a differential dynamic in the way oil risk is channeled to financial markets. Not only do I find that the market response to oil volatility risk is contemporaneous, but that crude oil triggers financial risk at the time of information, whereas gasoline effects of financial risk are subsequent to crude oil effects.

Measuring The Oil Risk Effect On Industry Volatility Shocks – Introduction

There is an abundant literature on the transmission of oil risk to stock markets. In this literature, crude oil risk is perceived as the only global oil source of systematic variation of returns of locally-traded financial securities. However, evidence shows that not only gasoline price incorporates crude oil price information, but also market beliefs at large. In line with for instance Borenstein, Cameron, and Gilbert (1997), the observed lags in gasoline price adjustment to crude oil price decreases may be partly due to the existence of several gasoline markets. Since many gasoline prices are generated for a single crude price, the gasoline market aggregates consumers-beliefs about future crude prices, monopolistic gains, consumer search and learning costs, inventories, refinery capacity and taxation policies1. Despite the systematic nature of the determinants of gasoline prices, the current financial economics literature is silent about the complementarity of the gasoline and crude oil markets in pricing financial assets.

Kaufmann and Ullman (2009) ask where do innovations in world prices enter the market? They find that innovations move sequentially from one crude oil spot market (Dubai-Fateh) to other crude oil spot and futures markets, due partly to the trading behavior of speculators. The sequential nature of price movements across crude oil spot and futures markets is much similar to stock price discovery in financial markets, that is a result of the trading behavior of informed and uninformed traders. If crude oil information is segmented, the question to be asked then is how this information is transmitted to financial markets?

Driesprong, Jacobsen, and Maat (2008) use monthly returns data to examine the relationship between crude oil price change and stock market return, and find that stock market return response to crude oil price changes is negative and first-order lagged. Expecting an immediate response of stock market return to public oil information, the first-order 23 lagged negative coe¢ cient is viewed as puzzling. However, the existence of complementary oil markets, each with a bit of oil information, may ease the empirical puzzle. In fact, Kilian (2009) argues for a differentiation of sources of oil shocks since these impact macroeconomic outcomes differently.

Since a lead-lag relationship exists between crude oil and gasoline (Ben Sita and Abosedra 2013), oil-based innovations to financial markets can originate from either oil market. The purpose of this paper is to investigate the effect of both crude oil and gasoline volatility risk on industry volatility shocks. However, since crude oil and gasoline share the same long-run volatility trend and are strongly correlated in their returns in the short-run, a correlation-weighted idiosyncratic volatility factor is developed. Discriminating this factor for possible crude oil or gasoline information leadership in volatility, I examine whether the sequential information hypothesis better explains the transmission of oil risk to stock markets than the underreaction hypothesis.

The contribution of this paper is twofold. First, the paper sheds light on the nature of oil risk that is sequentially transmitted to financial markets by breaking up oil risk into a crude oil and gasoline risk component. The main objective with the decomposition of oil risk is to include the two strongly correlated oil factors into a typical econometric model and to quantify the contribution of each source of oil risk to the formation of financial volatility risk. This study is by no means the first to examine the relationship between oil return (volatility) and financial return (volatility)3. However, this study differs in that it recognizes that oil risk may have a crude oil information tag or another oil product information tag. Second, in line with Hansen, Lunde, and Voev (2014), the paper provides a richer econometric framework, where not only the exponential Generalized AutoRegressive Conditional Heteroskedasticity (EGARCH)4 process comes into play, but also a model for realized volatility is included to generate uncorrelated volatility shocks that are related to the two oil risk components. Unlike, for instance, Elyasiani, Mansur, and Odusami (2011), including the crude oil effect into a GARCH model that is a forecasting volatility model, the oil risk effect is identified as it impacts the covariance between return and measurement innovations.

Volatility Shocks

See full PDF below.