Covariances reveal differences between supply shocks and demand shocks

This is for all the inflation mongers out there who think that today's oil price spikes are a prelude to hyperinflation.

Up until a few weeks ago, demand factors were driving oil and other commodity prices.  When oil prices went up, so did the stock market and interest rates.  Aggregate demand shocks, the earlier drivers, were mainly good news about growth, and this drove up interest rates and the stock market.  They also signaled higher prospective inflation, which I saw as good news.  We could use a bit more inflation, given we remain well below target and there is lots of labor to sop up before wage increases (the main part of inflation) kick in.

Over the last couple weeks, prices have spiked more sharply, but for very different reasons, mainly the quickly unfolding events in the Middle East.  Markets are speculating about a possible, if unlikely, major disruption in supply.  While oil prices have spiked, other commodity prices have generally softened, and the stock market and long-term interest rates have declined.  Anticipated downward shifts in supply are clearly bad news for the economy and growth.

And the decline in interest rates shows how little we should be concerned about hyperinflation.

Short-run inflation and perhaps even stagflation are real possibilities in this fragile economy.  This is just a textbook shift in aggregate supply.  Whether or not you're a Keynesian, the textbook says a supply shock to a fundamental resource will cause the price level to increase and output to decline.  But it's not a monetary phenomenon.  It's not the kind of thing that could kickstart a vicious inflationary spiral.  Not with unemployment at 9%.  This is bad news, not too much good news about an overheated economy.  Markets realize this and that's why stocks and interest rates are down.

Hyperinflation remains a truly remote prognostication. The only risk here may be if we shutdown the government indefinitely and default on our debt.  But that has nothing to do with oil prices either.

Update : I'm not the only one who thinks these kinds of changing covariances are interesting.