استراتژی ذخایر برای ذخیره ارزی چین: یک روش برنامه ریزی پویا

China is currently accelerating construction of its strategic petroleum reserves. How should China fill the SPR in a cost-effective manner in the short-run? How might this affect world oil prices? Using a dynamic programming model to answer these questions, the objective of this paper is to minimize the stockpiling costs, including consumer surplus as well as crude acquisition and holding costs. The crude oil acquisition price in the model is determined by global equilibrium between supply and demand. Demand, in turn, depends on world market conditions including China׳s stockpile filling rate. Our empirical study under different market conditions shows that China׳s optimal stockpile acquisition rate varies from 9 to 19 million barrels per month, and the optimal stockpiling drives up the world oil price by 3–7%. The endogenous price increase accounts for 52% of total stockpiling costs in the base case. When the market is tighter or the demand function is more inelastic, the stockpiling affects the market more significantly and pushes prices even higher. Alternatively, in a disruption, drawdown from the stockpile can effectively dampen soaring prices, though the shortage is likely to leave the price higher than before the disruption.

China׳s oil dilemma stems from the expanding supply–demand gap along with rapid economic growth. As the second largest energy consumer in the world, China consumed 483.7 million tonnes of oil in 2012 which accounted for 11.7% of global consumption. However, domestic oil supply was only 207.5 million tonnes. China׳s reliance on international oil supplies reached over 40% (British Petroleum, 2013) and has been steadily increasing since China became a net importer in 1993. China׳s sense of energy vulnerability is related to external threats as well as internal challenges. Now China imports more than 70% of its oil, mainly from the Middle East and Africa. Rising tensions in these regions strengthen concerns over China’s future oil supply security (Bai et al., 2012a). The rapidly increasing imports of oil have catalysed the Chinese government into establishing strategic petroleum reserves (SPRs). It has been reported that the government aims to comply with the IEA standard on stock requirements for its member countries, i.e. holding crude oil stockpiles equivalent to 90 days of net petroleum imports. Construction of China’s SPRs began in 2004 and they are expected to be fully completed in three phases over 15 years. Phase I (2004–2009) is completed, with 102 million barrels of storage capacity in four separate sites (e.g. Zhenhai, Zhoushan, Dalian and Huangdao). Between the second half of 2008 and the first half of 2009, it is estimated that 102 million barrels of strategic reserves were filled into Phase I SPR facilities, implying an average stockpiling cost of $58/bbl and a total crude acquisition cost of RMB 6 billion (Goldman Sachs, 2010). The construction of Phase II, with a total of 169 million barrels, began in late 2008. Phase III of the programme is still in the planning stage. Fig. 1 shows the locations of China’s SPRs at different phases. At the same time, China has been establishing an SPR management system. The National Oil Reserve Centre (NORC), established in 2007, manages the SPR construction and filling. It, in turn, is overseen by the National Energy Administration (NEA). Full-size image (58 K) Fig. 1. Layout of China’s SPR sites for three phases.

In recent years, China has been accelerating its construction of SPRs. We make three important contributions towards minimizing the short-run cost of this expensive project. First, we take the size of the reserve as given and minimize the short term fill costs, including changes in consumer surplus, oil acquisition and carrying costs. Second and more importantly, we examine the short run oil price trajectory from this crude acquisition. These price changes, which account for 52% of reserve costs in the base case, are not mentioned in any of the other studies we have found. Third, not only have we considered the optimal short-term drawdown rate and its effects on the short run oil price trajectory, but we also show that the cost of an optimal drawdown strategy in a disruption is comparable with that of fully drawing down the reserve. In the case study, we first examine results for a base case of the model together with sensitivities of the results to variations in the parameter assumptions. The results suggest an optimal stockpile acquisition rate of 9–19 million barrels and the optimal stockpiling drives up the price by 3–7%. Second, special cases for disruption are studied. When a 2% shortfall occurs, the best drawdown rate is 9 million barrels per month. If the disruption lasts longer, i.e. 3 months or 6 months, it is recommended that the government reduce the drawdown rate month by month. Based on our results, we see the following policy implications. First, China could consider acquiring oil at the rate of 9–19 million barrels per month depending on oil price. When the price is low, the government could absorb oil more quickly. Otherwise, the acquisition rate should slow down. Second, the influence of oil acquisition is more remarkable when the market is tight. Regarding the recent global economic recession and demand decline, we suggest that the Chinese government seize the opportunity and expand its SPR size. Third, although the drawdown of the SPR can effectively dampen soaring prices, it seems the shortage leaves the price high. Therefore we suggest that the government face potential risks with more solutions, i.e. energy diversification and oil substitution, rather than solely rely on SPRs. It is worth noting that the price is assumed to be affected only by market supply and demand. However, in reality the price is easily influenced by many factors such as market speculation, politics and so on. By making supply constant and only changing demand with China׳s stockpile changes, we have implicitly assumed there are no changes in other government stockpiling programs or commercial inventories. Nor have we considered non-competitive behaviour on the part of suppliers since we consider the stockpiling and drawdown strategies in the short-term and exporters cannot adjust their production capacity immediately. Future research could provide a richer and more detailed depiction of the other market factors mentioned above.