How Should We Think About the Strategic Petroleum Reserve?

Lessons about strategic reserves from price theory

There has been a lot of discussion recently about the drawing down of the U.S.’s Strategic Petroleum Reserve. As a result, I thought that today would be a good day to discuss the price dynamics of exhaustible resources like oil and what happens when a big player, like the U.S. government, sits on a large stock of that resource.

The theory of exhaustible resources gives us an idea of how the real price of such resources should behave over time. In particular, the real price of an exhaustible resource should grow at a rate equal to the real interest rate. Why? Suppose that you are sitting on land that contains an exhaustible resource or that you already have a stock of that resource and you are waiting to sell it. If you extract that resource and sell it (or simply sell your existing stock), a reasonable expectation is that you take the proceeds and save them and earn the real rate of interest on those proceeds. However, if the real price of the exhaustible resource is growing at a rate higher than the real interest rate, it makes no sense to extract the resource and sell them. Since the real price is rising at a rate faster than the real interest rate, you would be better off holding that resource and selling later. You’d earn a higher real rate of return that way.

That makes sense for you as a producer. Nonetheless, some consumers of that resource might want it now. In the event that is true, the price of the resource will tend to “jump” to a higher price such that the expected rate of return on the resource declines and is equal to the real interest rate. This jump is necessary to get the producer to actually extract the resource or sell it from their existing stock.

This price behavior is not only important for the producer, but the path of the price is also important for managing an exhaustible resource over time. Since the resource is exhaustible, there could come a point in time when the resource is no longer available. Given that the real price of the resource will rise over time, the quantity demanded of that resource will decline over time. (Note that this could mean, and models suggest, that people can consume a constant fraction of the known stock of the resource per period — and could do so indefinitely as the existing stock becomes infinitesimal.) In other words, higher prices lead people naturally to reduce their consumption of the exhaustible resource over time.

All of this sounds well and good. It makes perfect sense. However, in the late 1960s and early 1970s, the real price of gold rose at a rate persistently higher than the real rate of interest. This is hard to explain, given the previous discussion. If the real rate of return on gold is above the real rate of interest, one would expect people to “hoard” gold to sell it at a higher price later. That should induce the price to “jump” to a higher level and then proceed to grow at the real rate of interest. However, this didn’t happen.

An astute reader might point out that this was the end of the Bretton Woods era in which the price of the dollar was fixed in terms of gold and other countries had fixed exchange rates with the dollar. While this is technically correct, central banks had by this point stopping managing their pegs.

So what explains this persistent higher real rate of return?

Stephen Salant and Dale Henderson provided an answer to that. The reason that the market reacted the way that it did was due to anticipation that the U.S. and the IMF would auction off some of their existing gold stock. To understand why this would matter, suppose that one day the U.S. unexpectedly sold a significant amount of gold reserves at auction. The auction price of those reserves would necessarily be lower than the existing market price because of the unexpected increase in the available supply. Thus, even the expectation that this is possible would lead the real rate of return on gold to be higher than the real interest rate. The reason is quite simple. The fact that the U.S. is sitting on a pile of gold that could be sold at any moment presents a risk to others holding gold since an auction could generate a capital loss. In order for other gold owners to be willing to continue to hold gold, they must be compensated for that risk. Thus, even the prospect that this occurs will tend to lower the price today and subsequently have a higher real rate of return than the real interest rate (and what would otherwise be the case).

Although their paper focuses on the market for gold, the lessons of their approach apply to any exhaustible resource in which a big player, like the U.S. government is sitting on a large stock of reserves. In other words, there are also lessons here for the Strategic Petroleum Reserve.

As Salant and Henderson discuss, a government sitting on a large stock of reserves of an exhaustible resource has a couple of policy tools. The first is to sell off some of the reserves when the state finds it strategically advantageous to do so. The second option is that the state could use the reserves to enforce a price ceiling or to try to maintain a particular real price.

The latter option is unlikely to be successful. To understand why, consider a scenario in which the U.S. decides to impose a price ceiling on oil. It could so by promising to sell oil whenever the price of oil rose above the price ceiling. This would attract speculation. What speculators should do is wait until the strategic reserves have dwindled and then conduct a speculative attack, buying as much oil from the U.S. as possible at the price ceiling until the reserves are drained and then profit from the subsequent dramatic increase in the real price of gold when the U.S. is no longer able to defend the price ceiling. The same would be true of attempts to fix the real price oil.

The other option is strategic deployment. This could be done if the U.S. was shut off from oil supplied by its adversaries. Alternatively, this could be done to influence the real price of oil. Like in the gold example, strategic deployment will cause the real price to fall today. However, what happens next depends on whether this is a once-and-for-all deployment. If the U.S. did this once and could perfectly commit to this never happening again, the real price of oil would continue to appreciate at the real interest rate. Of course, it is unlikely that the U.S. could commit to do this, especially if it didn’t sell all of its reserves. Thus, the more likely scenario is that the real rate of return on oil would rise above the real interest in anticipation of unexpected capital losses caused by the U.S.’s strategic deployments.

Although the focus of these strategic deployments tends to be on the current, real price of oil, in reality such deployments affect the price path of the real price of oil. As a result, it also changes the intertemporal consumption path of oil. A reduction in the real price of oil today will result in higher oil consumption today (all else equal). In fact, that is often the point of the strategic deployment. In the face of supply shocks, strategic deployment acts as a counterbalance to what would otherwise be a reduction in consumption. Nonetheless, this comes at a cost. The high real rate of return on oil implies that future prices will be higher than they otherwise would have been thereby reducing future consumption.

It is also important to consider how this affects the management of the exhaustible resource over time. For any exhaustible resource, we typically think of there being a sufficiently high real price at which consumption would be so low that it no longer makes economic sense to extract what is left of the remaining resource. Strategic deployment can affect how quickly one gets there. Given that the threat of strategic deployment tends to raise the real rate of return on the exhaustible resource, this means that the price will reach its “choke price” sooner than would be the case otherwise. Thus, too frequent reliance on strategic deployment might actually contribute to more rapid depletion of the resource than would otherwise be the case.

Also, and as something that might seem paradoxical, the recent drawdown of the Strategic Petroleum Reserve might have actually reduced the real rate of return on oil since further strategic deployment is not possible. Of course, this would mean that replenishing the strategic reserves requires some degree of care. Large purchases of oil by the U.S. could cause current prices to rise. At the same time, the replenishment of those reserves raise the prospect of strategic deployment in the future. Thus, it might not just raise the current real price of oil but also increase the real rate of return on oil. As a result, there might be strategic reasons for the U.S. not to replenish those reserves if their desire is to limit the effect of the current geopolitical conflicts on the current real price of oil.

These are just some things to think about when people discuss the deployment of the U.S.’s Strategic Petroleum Reserves.