Who’s Afraid of a Big Bad Oil Shock?

Abstract

WHEN THE UNITED STATES INVADED Iraq in March 2003, many economists feared that the war would lead to a sharp decline in Iraqi oil production, a spike in oil prices, and a woeful U.S. economy that would follow the scripts of the oil shocks of 1973, 1978, and 1990. Real oil prices did increase, indeed more than tripled, from $20 in 2001:Q4 to $62 in 2006:Q3 (in 2007 dollars). But the ailments associated with earlier oil-price increases did not appear. Instead output grew rapidly, inflation was moderate, unemployment fell, and consumers remained reasonably happy. (1) Macroeconomists would be out of business if there were no surprises. The business of this paper is to inquire into the explanations for the surprising oil noncrisis of the early to mid-2000s. The robustness of the economy following the latest oil shock can perhaps be seen in the context of an important historical development in macroeconomics, namely, the Great Moderation. Over the past half-century, the economy has shown declining volatility of inflation, unemployment, and output growth. (2) Perhaps the moderated response of the economy to the latest oil shock should be understood as part of this overall decline in volatility. Although much has been written about the macroeconomic impacts of price shocks, little analysis is available on the impact of the most recent oil price shock. (3) Most research focuses on the role of monetary policy and inflation dynamics in the wake of a shock. There seems to be no consensus as to whether the macroeconomic fear of oil-price shocks remains warranted. Comparing Shocks Past and Present To begin, I define an oil-price shock as an inward shift in the supply curve for crude oil triggered by political events exogenous to the oil market and the macroeconomy. The four oil-price shocks considered here, summary statistics for which are shown in table 1, are associated with the 1973 Arab-Israeli war, the 1978 Iranian revolution, the 1990 Iraqi invasion of Kuwait, and the 2002 run-up to the American invasion of Iraq. Other periods might be considered, such as the price blip from 1998 to 2000 or the recovery from the OPEC price-cutting war of 1986, but these probably reflected cartel dynamics and were not clearly triggered by exogenous political events. The first date shown for each event in the table is that when the oil shock began, and the second is that when oil prices peaked. The first numerical column shows the corresponding percentage increase in the real price of oil from start to peak. (The oil price measure used here is the average refiner acquisition cost of crude oil, except during the period of price controls, 1973:Q3-1982:Q4, when it is the import cost. The oil price is deflated by the price index of personal consumption expenditures, or PCE.) The most recent event involved a real oil-price increase of 125 percent, close to the size of the 1978-81 increase but smaller than that in 1973-75. The next two columns show the macroeconomic response in terms of the unemployment rate and output (the latter expressed as the ratio of actual to potential GDP). The latest shock did not appear to have the recessionary impact on the economy seen in the three earlier shocks. The unemployment rate actually fell by slightly more than ! percentage point, and output grew slightly relative to potential. This contrasts sharply with the shocks of the 1970s, when the unemployment rate rose by between 1 1/2 and 3 1/2 percentage points and output declined sharply relative to potential. Although the latest oil shock did not lead to a recession, the impacts on inflation and productivity were in the same direction as those from the shocks of the 1970s. But the impacts were different in magnitude: inflation during the 2002-06 shock rose about 1 1/4 percentage points--much less than the sharp rise in the 1970s, and labor productivity growth declined by about 1/2 percentage point, less than the average in the 1970s as well. …