A Risk-Hedging View to Refinery Capacity Investment in OPEC Countries.

• Author: Ghoddusi, Hamed
• Position: Organization of the Petroleum Exporting Countries

1. INTRODUCTION

   Investing in the downstream sector to export refined products, as opposed to the export of crude oil, is an appealing and popular policy slogan in many oil-producing countries, including OPEC members. In 2010, OPEC's secretary-general predicted that "over the next decade, members were expected to invest around $40 billion in refining capacity expansion." (1) The ambitious idea to invest in downstream is, however, not new. An Iranian government stamp published in 1973 promotes a national dream for a full-vertical integration in the country's oil industry from well to the wheel. Also, a recent Bloomberg news item discusses UAE's ambitious plans to heavily invest in the downstream. (2)

   The incentive for downstream investment is strong because it is tempting to export final products instead of the raw material. Anecdotal evidences (3) suggest that the policy-makers in developing countries occasionally consider the value-added in the downstream sector (i.e., processed primary commodities) to be significantly larger than the upstream. (4) The idea that a possible negative correlation between upstream and downstream profits can motivate vertical integration in the oil industry is first introduced by (McLean and Haigh, 1954). Opponents of the vertical integration policy believe that the value-added in the oil refinery sector is limited, and not much can be gained while exposing the country to substantial capital investment commitments, taking financial risks of the downstream business, and in some cases hiring expensive foreign labor.

   To shed some light on the policy debate, we offer an analysis of the optimal downstream investing from a risk-hedging perspective. We build a static (single-period) model and a dynamic, forward-looking one. The first model highlights the trade-off between return and risk-reduction features of upstream/downstream sectors. The dynamic model characterizes the volatility of the total budgetary revenue of each sector. We take both models to the data to provide some quantitative insights in the case of crude oil refinery investment decisions. Our analysis is normative in natures; thus, we are not aiming at providing an explanation for the observed refinery capacities OPEC countries. Instead, our goal is to provide a framework to analyze such decisions critically.

   To better motivate the relevance of the downstream investment opportunities in OPEC countries, Figure 1 shows the ratio of the domestic refining capacity to oil production capacity of countries. (5) One observes a significant degree of heterogeneity among the major oil-producing nations. One also notes that due to a higher ratio of domestic consumption to production, non-OPEC oil producers (e.g., U.S. and China) tend to have a much larger ratio of downstream to upstream, compared to OPEC members. (6)

   Volatile crude oil prices expose oil-exporting countries to major foreign exchange and government revenue risks, resulting in macroeconomic instabilities (especially in the presence of rigid exchange rate regimes) and causing the so-called resource curse effect (Van der Ploeg and Poelhekke, 2009). Commodity stabilization funds (Arrau and Claessens, 1992) and/or hedging through financial instruments (e.g. futures and options) are two commonly proposed methods to manage volatile oil prices (Devlin and Titman, 2004). Vertical integration along the supply chain is the third strategy, which we will discuss in more details. The low correlation of refinery markups and crude oil prices, as well as their different time-series dynamics, can potentially provide some degree of hedging to the current account of the oil-exporting country. Our theoretical models provide several empirical hypotheses for the relative value of the investment in the upstream and downstream sectors.

   The optimal degree of vertical integration is a key input for the high-level energy and development discourse of oil-producing economies. Despite the obvious policy relevance and the potentially large resource commitment to investing in such industries, there is very little academic research on this topic (especially in recent years). To the best of our knowledge, our paper is one of the very few academic papers in the past two decades, specifically focusing on formal models and empirical results to analyze downstream investment in oil-rich countries. There are older papers (e.g. Al-Monsef, 1998; Al-Obaidan and Scully, 1993) which consider the problem of vertical integration for national oil companies. Also, Mabro (2006) provides a non-technical overview of the issue in a chapter. Finally, a small body of literature focuses on the energy policy choices of individual countries. For example, Krane (2015) discusses the incentives of Saudi Arabia for investing in downstream industries. We contribute to the literature by first extending the existing optimal portfolio models of the energy sector, and then offering an up-to-date empirical analysis of the problem.

   In short, our contribution has two major dimensions. First, we extend current theoretical models to formally characterize hedging incentives for the downstream investment. Second, we show that the time-series dynamics of profits in the upstream and downstream sectors have different properties. More precisely, due to the mean-reverting nature of cash-flows in the refinery sector, the present value of the downstream revenues is significantly less volatile than of the upstream.

2. LITERATURE REVIEW

   Our work is built on insights from research in the natural resource and energy economics as well as the industrial organization (IO) literature. In a broad sense, our work is related to the large and mature literature on the resource curse (Frankel, 2010) and the political economy of oil-producing countries (Beland and Tiagi, 2009; Ross, 1999). The resource curse literature not only highlights the role of institutions (e.g., Cabrales and Hauk, 2011) but also emphasizes that the way natural resource revenues are spent plays a critical role. Commodity price volatility has also been identified as a major source of the resource curse in resource-rich countries (Van der Ploeg and Poelhekke, 2009). Volatile terms of trade can suppress productivity growth, even in the presence of large capital accumulations.

   Investment in downstream could be a potential remedy for the resource curse if it helps oil-rich countries alleviate some of the negative features of exporting crude oil, such as the volatility in the export revenues. Merener and Steglich (2018) consider the role of price correlation to gauge the price performance of diversified economies and conclude that diversified commodity-producing countries face a significantly lower risk than specialized producers. Borensztein et al. (2013) quantify the welfare gains of hedging against the commodity price risk for commodity-exporting countries and highlight the first-order effect of reducing precautionary saving. Van der Ploeg and Venables (2011) discuss policy options for spending resource revenues. Export diversification is a key suggestion to reduce the magnitude of the resource curse. Herzer and Nowak-Lehnmann D (2006) and Bertinelli et al. (2009), among others, empirically examine the export diversification structure of resource-rich countries and conclude that considerable welfare can be gained if these countries move toward an optimal export portfolio. Alwang and Siegel (1994) evaluate the usefulness of portfolio models in advising export diversification policies for resource-rich countries. Labys and Lord (1990) use portfolio optimization techniques to determine the optimal export diversification strategy for Latin American countries. Massol and Banal-Estanol (2014) apply an optimization model to identify the optimal downstream investment for gas-rich countries. Cherif and Hasanov (2013) model the consumption, saving, and investment decisions of oil-exporting countries and show that a sizable precautionary saving is optimal for such economies. If the downstream investment can reduce the volatility of export revenues, a lower level of precautionary saving might be optimal. Thus, precautionary saving and downstream investment can be considered as potentially substitute policies.

   A variety of reasons (e.g., transaction costs, property rights, agency issues) are offered for vertical integration and have been extensively discussed in the industrial organization (IO) literature (see Carlton, 1979; Lieberman, 1991; and Joskow, 2012). Suzuki et al. (2011) provide an interesting rationale for partial vertical integration when small suppliers have a superior ability to absorb demand shocks. Also, Aid et al. (2011) and Leautier and Rochet (2014) discuss the risk-reduction incentives of vertical integration. However, some insights from that literature are not directly applicable to the oil industry. From the perspective of the refinery industry, crude oil has no economic substitute as the major input. Thus, the downstream of the oil industry is not making a strategic choice of input, and there is little room for the upstream monopolist to influence the downstream decisions. This eliminates strategic considerations that are typical in the IO literature.

   Levin (1981) and Barrera-Rey (1995) study the effect of vertical integration on the performance of oil companies and find no impact on the profitability but a small effect on risk reduction. Norton (1993) shows that vertical integration reduces systematic risks for refinery companies. The optimal hedging strategy for refiners has been studied by several papers (e.g. Sykuta, 1996; Sukcharoen and Leatham, 2017). Alexander et al. (2013) criticize the merits of the standard mean-variance optimization methods for determining the optimal hedging policy. Our paper differs from this literature by focusing on the profitability of the upstream rather than the profitability of the refinery. Moreover, we address the problem from a...