Arnold Kling  

Why No Oil Shock Effect?

Rodrik's Question... Unemployment...

Olivier Blanchard and Jordi Gali write,

Since the 1970s, and at least until recently, macroeconomists have viewed changes in the price of oil as as an important source of economic fluctuations, as well as a paradigm of a global shock, likely to affect many economies simultaneously. Such a perception is largely due to the two episodes of low growth, high unemployment, and high inflation that characterized most industrialized economies in the mid and late 1970s.

...The events of the past decade, however, seem to call into question the relevance of oil price changes as a signficant source of economic fluctuations. The reason: Since the late 1990s, the global economy has experienced two oil shocks of sign and magnitude comparable to those of the 1970s but, in contrast with the latter episodes, GDP growth and inflation have remained relatively stable in much of the industrialized world.

In trying to explain the difference, what is the first hypothesis that comes to mind? For me, it is the fact that this time around we are not making things worse through the disorder of price controls. But the authors do not even consider that hypothesis. Instead, they consider better luck (meaning fewer other shocks), less wage rigidity, better monetary policy, and a less oil-sensitive economy.

All perfectly reasonable hypotheses, but not enough to justify overlooking the order vs. disorder hypothesis.

And the empirical methodology--vector autoregressions. For my money, you cannot do anything less convincing.

Blanchard (5 votes in the Rodrik poll) was, like Krugman, a couple years ahead of me at MIT. Krugman's success as a columnist reflects a lay readership with tastes than differ from mine. Blanchard's success reflects an economics profession with tastes that differ from mine. Most top economists value mathematical grinding more highly than I do.

COMMENTS (8 to date)
General Specific writes:

"not making things worse through the disorder of price controls."

I certainly wouldn't propose price controls, but in order to better substantiate your hypothesis you should at least qualify the implications better. Economists with their noses in the dirt of numbers may be missing the big picture, but a general hypothesis of order versus disorder is flying at about the 30,000 level.

What are the specific implications of the price controls that affected GDP back in the 1970s? And I think it should be more than "I remember standing in gas lines." I do as well, but the number of times I really had to wait in a long line was limited. I could probably count them on my two hands.

In summary, price controls aren't efficient, but without better qualifying--and yes quantifying--how inefficient they are, it's difficult to consider it.

What about monetary policy? What about the increased use of debt? Are we just looking at an economy that is richer--and borrows more--and a fed with at least a slightly better grip on monetary policy? And what about all the other issues in the 1970s? The US was in the process of losing its competitive or manufacturing edge after dominating the post WWII world, requiring the start of a restructuring process that has continued to date. Perhaps that had implications as well. We've not acclimated to continuous restructing.

Also, on a related but side note, I'm not a huge fan of the energy is a smaller part of the GDP and therefore we're less dependent on it because, in fact, increased efficiency could make short-falls worse. E.g. we have an inverted pyramid, with energy at the bottom of the economy fueling much else, in which we use energy more efficiently--which means pulling out lower runs of that energy pyramid could cause greater disruption, not less. There are less efficiencies to gain. Perhaps.

Finally, we have not had an energy shock. The 1970s did (I'm pretty sure). Production dropped significantly. Supply was removed from the market almost overnight. We've not seen that yet. And we didn't have the strategic reserve to calm markets when supply falls a bit short (e.g. Katrina).

In summary, it's not clear to me whether price controls explains the difference at all, and if so, it may be in the noise. Or the dirt, using the above metaphor. I think the critical issue is that effect of supply shocks.

James Hamilton would be a good voice on this issue.

spencer writes:

In the 1970s consumer spending on energy rose from some 6% of nominal consumer spending to over 9%. So far the cycle the increase has been from 4% to 6% --still under the 1970s bottom.

I think one of the big difference is productivity and the wage price spiral. In the 1970s unit labor cost were running at near 10% annual rates and firms had no choice but to pass higher energy prices through to sustain profits. Now unit labor cost increases are generally significantly less then general inflation so most firms can absorb the higher energy cost and still sustain profits.
Consequently, the secondary impact of higher energy prices on the prices of other products is much smaller this time -- the pass though is small.

But maybe all of this can be subsumed under the heading of lowered inflation expectations. In the 1970s inflation expectations were high and people saw higher oil prices as just another part of high generalized inflation. Now inflation expectations are low and the increased energy prices has not generated a rise in overall inflation expectations.

Another way to look at the differences is that the 1970s was a supply shock -- we did have physical shortages because Arab suppliers cut supply. US imports plunged but there were no price controls on imported oil. Price controls did impact distribution, but they had no significant impact on supplies. This time the price rise is a demand pull phenomena. Prices are being pulled higher by strong demand rather than being pushed higher by supply shocks. Yes, the strong demand is largely in other countries, but it is still a different phenomena.

I would also suggest a third factor and that is higher income inequality. Because of rising income inequality the more affluent account for a much larger share of total consumption then in the 1970s. The lower income quantiles are being squeezed but the affluent can better absorb the higher oil prices. But since the top income quantile now accounts for some 60% of retail sales --significantly higher than in the 1970s -- the secondary impact of higher oil prices on demand for other products is sharply curtailed.

General Specific writes:

I agree with Spencer's comments. Combining his and mine, I think we have to take into account the improved adaptibility of the US economy, which was forced upon us after we no longer had a free ride--no real competitors. And monetary theory and a greater reliance on the market--a good thing--help us adapt.

But let's be honest here. We've not had a supply shock. Yet.

TGGP writes:

Income & consumption inequality are different. This very blog had a post about a paper saying "we show that there has been a large increase in income inequality but no concurrent increase in consumption inequality in the 1990s. Conversely, income mobility and consumption mobility are similar during this time period.". More from Marginal Revolution here.

General Specific writes:

There is merit in the idea that, in terms of consumption, we have not seen near the gap as we have in terms of income. And the increased income of the wealthy obviously results in increased savings & capital as well as some job-producing consumption.

But, anecdotal evidence tells me that those at the lower economic end are now making decisions they didn't have to make in the past. Gas prices are hurting, and they are cutting back elsewhere. I think this is demonstrated by the recent health in the upper echelon consumer stores while those serving the lower economics classes have seen growth stagnate.

Anecdotal and off the top of my head, but I think we are seeing evidence of downward consumption pressure among the lower incomes, and I think the end of the housing ATM will accelerate this trend.

Rough seas ahead. Batten down your hatches.

Stan writes:

Watching traffic on such sites as it is clear that those on fixed incomes have been in troble for some time. This administration only cares about oil people

TGGP writes:

General Specific, I trust the data economists present over your anecdotal evidence.

General Specific writes:

TGGP: Asolutely. One should always look at the data. Then, because we are dealing with extremely complex systems, we have to make a gut decision.


E.g. Bloomberg "Aug. 14 (Bloomberg) -- Wal-Mart Stores Inc. and Home Depot Inc., the two largest U.S. retailers, said the housing slump, rising mortgage defaults and high energy prices will depress earnings for the year. " (link)

E.g. "Sept. 20 (Bloomberg) -- U.S. retailers may post the smallest holiday sales gain in five years as consumers cope with tighter credit and a housing slump, the National Retail Federation said." (link)

E.g. "Higher end retailers posted gains, as the affluent appear to be unconcerned with the current macroeconomic issues that continue to loom. Saks Inc. saw an 18% increase in August same-store sales, double analysts' expectations" (link)

I understand this only anecdotal. I try to read broadly and then, finally, make a decision. We're only at the beginning of the housing problems. ATM resets reach a peak in March '08. That's when the real pain will start. By that time, housing will likely be in a much greater decline.

My read: Much slower growth for the lower end retailers. Continued higher or higher growth for the upper end retailers.

Comments for this entry have been closed
Return to top