This seems to be true going back to at least the 1950s, as the graph above illustrates. Note: I put the ratio of the oil spot price in a ratio over NGDP in the graph above, to account for the mutual effects of oil rising oil prices and falling NGDP.
So, what gives? Well, there are two things going on. First, negative oil supply shocks lower real GDP, of course. Second, historically, the Fed has responded by tightening monetary policy as prices rise, particularly since inflation targeting began in the US in the 90s.
Ideally, the Fed would actually expand the money supply during negative supply shocks, which would increase inflation even more than would be the case on the sole basis of oil price increases. This would keep NGDP, which is also aggregate demand, growing on a steady path, minimizing the effects on real GDP and employment.
This may seem counter-intuitive, but consider what causes unemployment during recessions. It’s sticky wages and other prices. If not for sticky wages and other prices, most unemployment in most recessions would not occur. Real GDP contractions would be far less severe, on average, if the Fed weren’t compounding the problem with tight money. Notice how sticky wages are relative to nominal GDP in the graph below.
Inflation targeting is the problem. Particularly when the Fed is targeting inflation at 2%, as it has for a generation, any rise above 2% means that it should raise rates. But, since many prices are relatively sticky, real GDP falls more sharply than inflation, and hence output and employment end up much lower than need be. Notice in the graph below how much stickier core PCE inflation is than real GDP.
It makes sense that as GDP falls faster than wages, then real wages increase, leading employers to cut hours and jobs, which also cuts output, of course.
Why cut hours and jobs instead of just cutting pay-per-unit time across-the-board? No one really knows, but it could be because a firm that tried to do this would have its most productive employees poached by competitors who were willing to lay off their least productive employees. Also, pay cuts, even if only nominal, are believed to reduce morale in the workplace, due to money illusion.
So, can we expect oil price shocks, or other commodity price shocks to predict recessions in the future, and related stock market crashes and dollar index run-ups? Perhaps, but the situation is somewhat less clear now that the Fed announced it began targeting an average inflation rate of 2% last year. The idea is that if inflation runs below 2% for a time, the Fed will let it run higher than 2% for a time thereafter, until the average inflation rate over that period of time is 2%.
Unfortunately, the Fed hasn’t specified what that period of time will be, so it raises questions as to the Fed’s commitment regarding this form of price “level targeting”. This is seemingly a step in the right direction, but without a specific timeframe as to the average target, it may be little better than inflation targeting.
Hence, for now, I will be concerned about tight money when there are significant negative supply shocks, unless and until the Fed gives me solid reasons to think otherwise.