I received some thoughtful (offline) responses to my post yesterday on the impact of speculators on oil prices – or, to be more precise, the relationship between financial and physical oil markets. One group insisted that financial markets can’t push up the price of physical oil without also leading to substantial inventory buildups. (Higher prices mean more supply and less demand, and thus, everything else being equal, growing inventories.) Since in many of the cases where people blame speculators for high prices, we haven’t seen big inventory buildups, financial markets can’t be responsible. The other group insisted that huge volumes of paper oil trade mean that financial markets can overwhelm physical ones, and basically dictate prices, at least for some time. In particular, these people noted that most financial traders sell out their futures contracts before they expire, rather than taking delivery of physical oil. Thus, they argued, it is possible for financial markets and physical markets to become largely decoupled from each other.
I don’t buy either argument. Let’s start with the first point. Futures markets tend to serve a price discovery function: if people bid up near-term futures, producers will price their physical oil at that higher futures price. That should boost supply and curb demand relative to what it otherwise would have been, which in turn should lead to inventory buildups. Since we don’t see big inventory buildups, people argue, speculation can’t be having a big impact.
There are two big problems with this. First, short-term elasticities of supply and demand are very low, and possibly negative in the case of supply, which means inventory buildups can be small or nonexistent. Most producers can’t respond to higher prices by quickly pumping out more oil; indeed many believe that some OPEC producers respond to higher prices by cutting back production (keeping their revenue constant). On the demand side, motorists are pretty unresponsive to high gas prices (up until a point). Even worse, since refiners are currently minting money, they can to some extent suck up higher oil prices and be slow to pass them along to their customers. To the extent that higher oil prices don’t pass through, final demand remains unchanged. (Basically, in this case, refiners’ elasticity of demand is even smaller than motorists’.) All this means that prices can get jacked up considerably while only introducing small imbalances in the physical market, and hence small inventory buildups.
Which leads us to the second big problem: There are many other things that influence inventories. (If you don’t believe me, ask yourself whether inventories were constant before the growth of financial markets. They weren’t.) In the presence of this noise, it can become all but impossible to separate out relatively small speculation-driven inventory buildups. At a minimum, anyone who has sophisticated enough techniques to see that sort of thing is probably trading oil, not writing blog posts.
On the flipside, I also don’t buy the “financial and physical markets are decoupled” argument. There are two problems with it. First, (financial) futures prices ultimately converge with (physical) spot prices. Those physical prices need to be consistent with what’s going on in the physical market. By implication, so do the financial prices (as the relevant futures contracts get close to maturity). If they don’t, we’ll see futures prices failing to converge with spot prices – but we don’t. (All of this is true regardless of whether financial traders are buy and selling a gazillion paper barrels each day.) Second, if for some weird reason futures prices didn’t converge with spot prices, why would physical consumers care if those futures prices were high? I care about what I pay for gasoline, not what it’s trading for on the NYMEX. And if someone wants to respond by saying “but if people push up prices on the NYMEX, you’ll end up paying more”, they’ve just made my point – financial prices translate into physical ones.
To be sure, futures that don’t mature for a while can get way out of whack with fundamental values. This is interesting, and can have an impact on production and consumption decisions that have long lead times, but they don’t have a direct impact on what you pay at the pump today. What matters for the “high gas prices” discussion is physical prices and near-term futures.
One last note: Don’t read my discussion about whether speculators can influence oil prices as a discussion of whether speculation is good or bad. Alas, ninety-nine percent of commentary about speculation seems to conflate “speculators are influential” with “speculation is bad”, and “speculators aren’t influential” with “speculation is fine as it is”. But there is no reason why one’s understanding of how reality works should dictate one’s value judgment as to whether a particular activity is good or bad. Unfortunately, when we tie the two together, our value judgments tend to dictate (and thus warp) our analytical findings. That’s not good for anyone. It’s certainly not good for informed debate.