Commodity bubbles or successful monetary policy?
The blogosphere has been discussing a Federal Reserve Board of St Louis paper by Richard Anderson arguing that what appears to be an asset price bubble may actually be the normal result of expansionary monetary policy.
Rapid increases in commodity and financial market prices by themselves, however, are not reliable indicators of potential bubbles because such increases also occur as part of normal monetary policy. ... Disappointingly low returns on short-term, low-risk investments prompt investors to move to longer-term, higher-risk investments in financial instruments, commodities, and durable goods.
...
One factor is the potential success of expansionary monetary policy: If economic activity expands, demand for commodities likely will increase, pushing futures prices upward, which, in turn, tends to increase current-period prices. ... A second factor is the decreased foreign exchange value of the dollar as a result of aggressive monetary policy.
...
As long as the FOMC’s pursuit of highly expansionary policy continues, households and businesses remain pessimistic, and demand is sluggish, the potential exists for asset prices to deviate from their long-run levels by large amounts and for long periods. Such increases per se are not bubbles but a commonplace reaction of the monetary transmission mechanism. ... Whether bubbles have been generated remains to be seen.
It is simplest to evaluate this argument for commodity prices because to a first approximation their “long-run levels” can be regarded as constant in real terms (real commodity prices are stationary to a first approximation absent secular supply or demand shocks).
A large increase above this long-run level implies a large negative expected real return on commodities. This is difficult to reconcile with positive yields on long-term inflation indexed bonds. Even at shorter maturities, inflation indexed yields are only mildly negative, and even a modest risk premium would lead to a positive required rate of return on commodities.
Moreover, if the price pressure on the spot prices is coming from elevated futures prices as Anderson argues, the implication is that nominal prices are expected to rise. That too does not appear consistent with falling real prices given modest inflation rates in the developed world.
It appears to me that one would have to assume severe supply constraints (of the peak oil variety) to provide a non-bubble explanation for a sharp rise in real commodity prices above their “long-run levels”.
Similar problems would arise in the case of other assets as well. If one assumes that prices are above their long-run levels, then expected risk adjusted returns would be negative which would be inconsistent with positive or even near zero yields on inflation indexed bonds.
Depressed expected rates of return can be attributed to “a commonplace reaction of the monetary transmission mechanism,” but it is difficult to see how highly negative expected rates of return (implied by large deviations from long-run levels) can be so explained.
Of course, one can argue that inflation indexed yields are only reflecting large risk premiums for sovereign default and that the true real risk-free rate is hugely negative. But that would be an even more scary story than an asset price bubble.
Posted at 5:47 pm IST on Fri, 21 Jan 2011 permanent link
Comments