VoxEU Column Energy Monetary Policy

Oil shocks redux

Why didn’t the most recent run-up in oil prices have dramatic effects as in the 1970s? Here one of the world’s leading macroeconomists surveys a variety of explanations: i) developed countries are now less energy-intensive, ii) wages are more flexible, iii) the US auto industry is relatively smaller, iv) monetary policy now targets core inflation, and recent shocks were to industrial demand, not oil supply.

From the end of 2002 to the middle of 2008, the US economy was in the throes of a significant oil price shock. The dollar price of oil rose fivefold, with spot prices briefly hitting $145/barrel. Even adjusting for inflation, the rise in oil prices was stunning. At their peak, real oil prices stood about 50% above their previous record high – reached following the second OPEC oil shock of 1979-80. (After hitting its 2008 peak, the price of oil fell rapidly, tumbling over the past six months into the $30-$50/barrel range.)

Although the recent run-up in oil prices is comparable in magnitude to the first two OPEC shocks, its effects on the economy seem to have been very different. Textbook accounts of the 1970s and early 1980s blame “supply shocks” (which included sharp rises in the price of food as well as oil) for the prolonged periods of both high unemployment and high inflation, or “stagflation,” that followed. By contrast, the most recent increase in oil prices appeared to have very little effect on the expansion that followed the 2001 recession. (While the US economy did enter a recession at the end of 2007, this was widely attributed to the collapse in consumer and business confidence that attended the subprime crisis and subsequent financial panic.) Similarly, core consumer price inflation – inflation excluding food and energy prices – was relatively stable over this period, which again contrasts sharply with the earlier episodes.

One interpretation of the experience of the past several years is that it vindicates “revisionist” views of the role played by oil shocks (and other supply shocks) in precipitating the stagflation of the 1970s. According to this view – variants of which have been propounded by DeLong (1997), Barsky and Kilian (2002), and Cecchetti et al. (2007) – the root cause of the abysmal macroeconomic performance from 1973 to 1983 was poor monetary policy, not the oil shocks. For example, in DeLong’s account of the period, bad memories of the Great Depression (which left the Fed fearful of using tight money to fight inflation) and the Fed’s attempt to exploit what it viewed as a non-vertical long-run Phillips curve created a situation that made a burst of inflation inevitable. Likewise, Barsky and Kilian see a “stop and go” monetary policy as driving both the higher inflation and the higher unemployment of the 1970s and early 1980s. Indeed, Barsky and Kilian even go so far as to argue that expansionary monetary policies in the US and elsewhere led to both the rise in overall inflation and the rise in the price of oil and other commodities that were observed over this period – implying that the supply shocks themselves were merely symptoms of an underlying policy failure.

In recent work (Blinder and Rudd 2008), we revisit the textbook account of the stagflation of the 1970s, in which sharp increases in oil and food prices (and, in the early 1970s, the removal of price controls) are the principal causes of the spikes in inflation and unemployment that characterised this unhappy period. We argue that this conventional explanation holds up remarkably well to the accumulation of new data, new theories, and new econometric evidence since it was first advanced more than thirty years ago. And we provide a critical assessment of the revisionist interpretations that have been proposed more recently.

A comparatively painless oil shock?

But that still leaves us with a puzzle. If supply shocks were the key factor behind the poor macroeconomic outcomes of the 1970s and early 1980s, why didn’t the most recent run-up in oil prices have similarly dramatic effects? As has been documented by a number of authors – including Hooker (1996, 2002), Blanchard and Gali (2007), and Nordhaus (2007) – oil shocks have had smaller macroeconomic effects since the early 1980s. The basic stylised facts seem to be that the positive response of core inflation has diminished sharply over time and the negative responses of output and employment have nearly vanished.

Why might that be? One reason is obvious. Thanks largely to an array of market reactions to higher energy prices after OPEC I and II, the US and other industrialised countries are now far less energy-intensive than they were in 1973. In the case of the US, the energy content of GDP (measured as the number of BTUs consumed per dollar of real output) has fallen dramatically since 1973 and is now about half of what it was then. By itself, this halving of the US economy’s energy intensity would also halve the macroeconomic impacts of oil shocks, with the reductions roughly equal for prices and quantities.

However, Hooker (2002) finds that pass-through from oil prices to other prices has diminished to negligible proportions over time – which is about twice the change that can be explained by energy’s shrinking share. Similarly, from 2002 to 2007, we find that there appears to be essentially no positive relationship between the energy intensity of consumption goods (measured using input-output coefficients) and their price change. So there must be more to the story.

Recent work by William Nordhaus (2007) explores three possibilities. The first is that the more gradual nature of the most recent oil price increases muted their effects. While the cumulative price increase was huge, the rolling oil shock of 2002-2008 was far smaller than either OPEC I or OPEC II when viewed on an annualised basis – just 0.7% of GDP per annum (through the second quarter of 2006, when Nordhaus’ study ends) versus roughly 2% of GDP per annum for both OPEC I and II. It is easier to cope with more gradual oil price increases.

Perhaps more important, Nordhaus uses econometric Taylor rules to estimate that the Federal Reserve responded more to headline inflation until 1980 but more to core inflation afterward. If so – and if the Fed’s attempt to keep inflation in check following an oil price increase accounts for a large portion of the effect that higher oil prices have on output, as Bernanke et al. (1997) argue – then we might expect substantially smaller contractionary effects following the more recent oil shocks because of the limited effect that oil prices now appear to have on core inflation.

Finally, Nordhaus finds modest evidence that wages have absorbed more of the recent oil shocks than was true in the 1970s. Greater wage flexibility makes the responses to an oil shock more neoclassical (based on factor substitution in the context of full employment) and less Keynesian (based on demand-side effects like the “oil tax”). And since simple calculations suggest that neoclassical effects on output are far smaller than Keynesian effects, that change would make oil shocks less powerful.

Blanchard and Gali (2007) also adduce some modest evidence in favour of greater wage flexibility in recent years. But their more speculative hypothesis is that the anti-inflation credibility of monetary policy has increased since the 1970s – which would reduce both the inflationary impacts and the output losses from an oil shock, presumably by limiting the reaction of inflationary expectations. Blanchard and Gali (2007) do find smaller recent impacts on expected inflation; but they caution against putting too much weight on the conclusions from their model (which they themselves characterise as “primitive”).

Kilian (2007) adds two other empirically appealing ideas to the list, both connected with international trade. First, the changed structure of the US automobile industry since 1973 – arguably itself a reaction to the OPEC shocks – means that Americans no longer turn only to imports when they seek smaller, more fuel-efficient vehicles. As a result, domestic aggregate demand falls by less after an oil shock than it formerly did. (The SUV craze clearly represented some backsliding in this regard, and the US auto industry – and the rest of the economy – is now paying the price.) In addition, the domestic auto industry, which is of course especially vulnerable to higher gasoline prices, is a much smaller share of the economy now than in the 1970s.

Second, the rolling 2002-2008 oil shock seems to have been driven by strong global demand for industrial output, not by supply or demand shocks specific to the oil market. While rising oil prices still constitute an “oil shock” to importing nations like the US, the recent shock came accompanied by stronger export performance, which cushioned the blow to aggregate demand.

Conclusions

In sum, the search for an explanation of why oil shocks have smaller impacts now than they did in the 1970s has not come up empty. Rather, it has turned up a long list of factors, no one of which appears to be dominant, but each of which may play some role. If that is correct, the supply-shock explanation of stagflation remains qualitatively relevant today, but it is less important quantitatively than it used to be. Thus with luck and sensible policy, food and energy shocks need not have the devastating effects that the supply shocks of the 1970s and early 1980s did.

References

Barsky, Robert B., and Lutz Kilian. 2002. Do we really know that oil caused the Great Stagflation? A monetary alternative. In NBER Macroeconomics Annual 2001, eds. Ben S. Bernanke and Kenneth Rogoff, 137-183.
Bernanke, Ben S., Mark Gertler, and Mark Watson. 1997. Systematic monetary policy and the effects of oil price shocks. Brookings Papers on Economic Activity 1: 91-142.
Blanchard, Olivier J., and Jordi Gali. 2007. The macroeconomic effects of oil shocks: Why are the 2000s so different from the 1970s? NBER Working Paper no. 13368, September.
Blinder, Alan S., and Jeremy B. Rudd. 2008. The supply-shock explanation of the Great Stagflation revisited. NBER Working Paper no. 14563, December.
Cecchetti, Stephen G., Peter Hooper, Bruce C. Kasman, Kermit L. Schoenholtz, and Mark W. Watson. 2007. Understanding the evolving inflation process. US Monetary Policy Forum working paper, July.
DeLong, J. Bradford. 1997. America’s peacetime inflation: The 1970s. In Reducing inflation: Motivation and strategy, eds. Christina D. Romer and David H. Romer, 247-280. Chicago: University of Chicago Press.
Hooker, Mark A. 1996. What happened to the oil price-macroeconomy relationship? Journal of Monetary Economics 38 (October): 195-213
Hooker, Mark A. 2002. Are oil shocks inflationary? Asymmetric and nonlinear specifications versus changes in regime. Journal of Money, Credit, and Banking 34 (May): 540-561.
Kilian, Lutz. 2007. The economic effects of energy price shocks. University of Michigan, October. Mimeo.
Nordhaus, William D. 2007. Who’s afraid of a big bad oil shock? Brookings Papers on Economic Activity 2: 219-240.

3,465 Reads