Thursday, June 02, 2011

Hedging the Risks of Selling Oil from the Strategic Petroleum Reserve

I see that the administration has asked Congress to approve a non-emergency sale of oil from the US Strategic Petroleum Reserve (SPR), in order to allow a storage cavern to be repaired before it starts to leak. That's fine, as far as it goes, though the article I read suggested this would be done as a net sale into the market, rather than an exchange for future oil, as has been done for many previous SPR releases. The distinction means that the government will either be exposed to buying the oil back at higher prices later, or would simply forgo refilling that portion of the reserve. The current shape of the oil futures market provides another alternative, though without the presumed political benefits of being seen to sell SPR oil when gasoline prices are high.

The sale in question was included in the administration's annual budget request and identified 6 million barrels to be sold "for operational purposes." That amounts to less than 1% of the 727 million barrels of oil currently in the SPR, equating to a little more than one day of import disruption insurance at the SPR's maximum output of 4.4 million barrels per day. Of course at current oil prices it would be worth over a half-billion bucks, so I can understand the appeal of doing this when federal finances are tight. However, the purpose of the reserve was never to speculate on the price of oil and harvest those gains when we came up short elsewhere; the oil is there to mitigate a serious disruption in the roughly 9 million barrels per day of oil imports on which our economy depends. Unless the administration now wants to undertake a comprehensive review of our SPR strategy--something I've advocated for several years--it is more or less obligated to replace the oil once the cavern has been fixed.

In that case, selling the oil, rather than offering it to refiners on a time-trade, will expose the government to a substantial amount of price risk while repairs are completed. For example, if they had sold this oil last fall and needed to buy it back now, the Department of Energy would have incurred a loss of up to $180 million, based on the increase in oil prices in general and the divergence of physical markets, which tend to track UK Brent Crude, from the futures market in West Texas Intermediate. Prices might fall in the meantime, but it is not the role of the DOE to bet on that prospect. The futures market offers a uniquely better alternative today.

Most of the time, the oil futures price curve is bent either up or down, in "contango" or "backwardation" in trader's terms, with oil for delivery several months or more from now selling for considerably more or less than for prompt delivery. That's usually an indication of expectations that the balance between supply and demand will be either tighter or looser in the months ahead, compared to today. The contango that prevailed until recently has flattened dramatically, so that if it acted quickly, the DOE could sell the oil from the caverns to refiners and lock in its future repurchase price on the futures market at only a dollar or two per barrel more than the sales price. Of course this would involve having the government participate in the dreaded futures market, even though it wouldn't be for the purpose of manipulation or stabilization, but for simple hedging of the kind that producers and refiners do every day of the week. (Backwardation would offer an even better deal, and the Brent market is currently mildly backwardated, but I can only imagine the hullabaloo if the US government hedged SPR oil on a European exchange.)

We would argue all day about which approach is riskier: hedging the oil sold from the SPR with futures contracts or waiting to buy back at whatever price prevailed later. In the larger scheme of things, neither looks as risky as emptying the cavern and not refilling it at all. Based on my experience and at least in this special case, hedging seems like a good way to ensure that the SPR cavern repair doesn't end up costing a lot more than the DOE expects, if its managers ignored oil-price risk.

No comments: