Rabah Arezki and Olivier Blanchard write,
Futures markets suggest that oil prices will rebound but will remain below the level of recent years.
Pointer from Mark Thoma.
Futures markets are bound to tell you that oil prices will remain near current levels, with a tendency to rise. If not, there is an arbitrage opportunity. If futures prices were far below spot prices, then producers would pump lots of oil and holders of inventories would try to sell every drop as soon as possible, until current prices fell. If futures prices were far above spot prices, then producers would cap wells and holders of inventories would fill every storage tank to the brim, until current prices rose.
I find the Hotelling model of resource pricing persuasive. In that model, the futures price and the spot price do not contain independent information. The relationship between the two is governed by storage cost and the rate of interest.
Then there is this:
central banks’ forward guidance is crucial to anchor medium-term inflation expectations in the face of falling oil prices.
This statement confused me on many levels.
1. The drop in oil prices that is already behind us would not seem to cause a downward shift in medium-term inflation expectations. The quoted phrase seems to equate the future with the past and levels with rates of change.
2. Elsewhere in the article, the authors point out that for oil importing countries, a drop in oil prices will boost aggregate demand. Well, from a conventional macroeconomic standpoint, that is the end of the story. There is no reason at all for monetary authorities to think that all of a sudden they need to counter the deflationary impact of an increase in aggregate demand. That is an oxymoron.
The arbitrage is throughout financial markets too. People use currency forward rates as predictors of where currencies will go, when the value of the forward is driven entirely by an arbitrage of the spot value and different interest rates.
The authors seemed to limit their views on stimulative lower oil prices to a ceteris paribus, supply driven decline. The last section then includes the context of weak aggregate demand indicated by a zero lower bound. The collapse in nominal expectations, supported by monetary ineffectiveness, makes temporary the real income and profit boosts.
1. You are correct, and the Hotelling view is actually the mainstream cost-of-carry model: F(t,T) = S(t) exp( (r+w) (T-t) ). The link between futures and spot is entirely determined by interest rate, r, and cost of storage, w (throw in convenience yield if you want). Convenience yield explains (but does it really?) why futures prices can be lower than spot prices.
2. Futures prices are not today’s expectation of prices in the future. They are price expectations under the risk-neutral probability measure, so they reflect both true expectations and risk premia. Disentangling the two is difficult.