تنوع بخشی پرتفوی در بازارهای انرژی

This paper's results indicate that futures for crude oil, natural gas and unleaded gasoline fail to enhance the performance of representative energy stocks in terms of return to risk, but do decrease the overall level of risk exposure borne by passive equity investors. Our findings suggest that futures contracts on energy commodities are valuable to market participants with an interest in hedging against price fluctuations in energy markets by buy-and-hold strategies. However, this conclusion is reversed when one takes the perspective of traders whose core interests can be better approximated through the return to risk-bearing. In fact, this paper documents that return-to-risk maximizing agents are unlikely to profit from trading energy futures in addition to energy stocks. Moreover, futures for energy commodities fail to offer significant diversification gains with respect to energy stocks once investors adopt simple dynamic trading strategies that rely on readily available pricing information.

What could investors in North American energy stocks potentially gain by diversifying their portfolios with investments on energy commodities markets? This is the key question addressed by this paper. Textbook portfolio theory teaches us that investors can improve portfolio performance by diversifying across different classes of assets. The argument is that allocating funds to imperfectly correlated securities decreases portfolio risk without necessarily affecting the level of expected return. However, utility-based portfolio selection models do not provide a consensus regarding the advantages of expanding the collection of tradable assets.1 This is especially true if the additional investment opportunities are dominated by the securities that are already in investors' portfolios.2 The systematic evaluation of the benefits from exposure to alternative markets has attracted enormous attention from individuals and institutional investors. The main line of study in this field concerns the assessment of the advantages that are associated with expanding the set of tradable assets from the domestic to the international equity market (among others, Bekaert and Urias, 1996, Errunza et al., 1999 and De Roon et al., 2001). Other works have investigated the gains from portfolio diversification across different classes or sets of assets (e.g., Eun et al., 2008). Ours is the first paper that applies these methodologies to assess the value of portfolio diversification within energy asset and commodity markets. This study addresses the question of whether energy futures improve the investment possibilities offered by a representative selection of crude oil and natural gas producers, refiners, and integrated firms. Our empirical findings indicate that futures for crude oil, natural gas, and unleaded gasoline are effective in decreasing the overall level of risk borne by passive investors. That is, trading energy futures is a valuable strategy for investors that have an interest in hedging against energy commodities' price fluctuations by means of buy-and-hold portfolios. This might be the perspective of an agent who regards futures contracts as insurance policies rather than as profitable investments. The opposite conclusion can be reached when one takes the perspective of market participants whose core interest lays in maximizing their compensation for risk-bearing. Our analysis clearly indicates that no additional reward to risk-taking is entailed by including energy futures in an efficient portfolio of energy stocks. This paper also documents that the returns to energy futures contracts can be replicated–up to some zero-mean error term–by simple dynamic strategies of stock holdings. In fact, a dynamic portfolio of energy stocks that mimics the net positions of commercial traders (i.e., large hedgers) is shown to exhaust all the diversification gains offered by the futures contracts considered. Hence, this paper's findings document the existence of simple trading strategies, that rely on readily accessible information and require infrequent portfolio rebalancing, for which the selected energy futures fail to offer significant diversification benefits over a representative pool of energy stocks. Our empirical results are in line with some known stylized facts on commodities markets. Erb and Harvey (2006) documented that for most commodities the average return on futures contracts is indistinguishable from zero, a fact that finds support in our sample. Gorton and Rouwenhorst (2006) noted that the returns on commodity futures are negatively correlated with those on equities. Standard portfolio management principles therefore suggest that futures might decrease the level of risk-bearing entailed by stocks. In contrast to Gorton and Rouwenhorst's study, however, our analysis relies on utility-based measures of diversification benefits which disentangle gains that come in terms of improved return to investment from those entailed by a lowered risk exposure. The remainder of the paper is structured as follows. The next section provides an analytical framework and the econometric approach that we adopt. The data library utilized for our empirical analysis is described in Section 3, which is followed in Section 4 by our first set of results. Section 5 refines the analysis of the diversification benefits associated with energy futures by introducing dynamic trading. A short statement of conclusions follows in Section 6. Finally, three appendices provide additional description of the data used, some robustness checks of our empirical analysis, and the mathematical derivation of some of the tests for spanning.

This paper began with a simple question: what could investors in North American energy stocks potentially gain by diversifying their portfolios with investments on energy commodity markets? To address this question, we used mean-variance tests of diversification benefits to assess whether energy futures (on WTI and Brent crudes, natural gas, and gasoline) improve the investment opportunities offered by a sample of fifteen U.S.-based oil and gas producers, refiners, and integrated companies. These tests gauge the changes of the efficient frontier, i.e. the collection of variance-minimizing portfolios, that are associated with broadening the investment opportunities set. Daily data were obtained for the period extending from the beginning of January 1990 to the end of February 2008. Our analysis suggests that futures on energy commodities fail to increase the return to risk-bearing associated with efficient energy stock portfolios held by market participants who adopt a buy-and-hold strategy. However, energy futures contracts do allow such passive investors to decrease the overall level of risk exposure that can be obtained by holding positions in the selected energy stocks. The rejection of the spanning hypothesis here indicates that broadening energy portfolios to include futures yields additional benefits to market participants who value hedging against risk. These results hold whether all equities are grouped together or each industry sub-group is treated separately. Similarly, the same results are obtained whether the energy commodities included in our sample are grouped together or considered individually. More generally, however, if market participants are thought of as adopting simple dynamic trading strategies that rely on available pricing information (such as data on long and short hedging positions maintained by the U.S. Commodity Futures Trading Commission), then a different picture emerges. In such cases, the null hypothesis of spanning cannot be rejected and futures on energy commodities fail to offer any significant diversification gains to market participants. Here again, these results hold whether all equities are treated as a whole or each sub-group is considered separately, and whether the energy commodities included are grouped together or considered individually. Overall, our results suggest that the only source of potential benefits available through broadening energy portfolios to include commodities will take the form of reducing the level of unavoidable risk offered by energy stocks to market participants who adopt buy-and-hold strategies. When dynamic trading strategies are considered our results indicate that the introduction of energy futures ceases to be a source of significant hedging benefits. In neither case, however, does our analysis yield any evidence that energy futures provide a way to enhance the return to risk associated with equities traded on North American markets.