October 16, 2011
Could monetary policy mitigate the real effects of oil shocks?
Michael Levi (hat tip:Marginal Revolution) and Jeremy Kahn are among those who recently rediscovered some earlier research by Ben Bernanke and others that concluded that the economic downturns that followed historical oil price shocks could have been avoided if the Fed had followed a more expansionary monetary policy at the time. Here I call attention to some subsequent research that took another look at their evidence and reached a different conclusion.
Big oil price increases such as occurred in 1973-74, 1979, and 1990 were each followed by economic recessions in the United States.
In a study published in Brookings Papers on Economic Activity in 1997, Ben Bernanke and Mark Watson (professors at Princeton University at the time) and Mark Gertler of New York University concluded that these economic downturns could have been avoided if the Fed had not allowed the fed funds rate to rise following the oil shocks. The figure below reproduces one of the important findings from their paper. In these figures, the price of oil is assumed to rise 10% above its previous 12-month high at month k = 0, with the horizontal axis registering the number of months k after the oil price increase. The solid line on the vertical axis plots how that increase in oil prices at date 0 would cause you to revise your forecast of real GDP k months into the future. The calculation is based on a historical summary of the dynamic relations among a set of 7 different macroeconomic variables assuming that you could get a reasonable forecast of each of those variables by looking at their values over the previous 7 months. These changes in forecasts were calculated using methods developed by recent Nobel laureate Christopher Sims that I briefly described last week. The figure suggests that we might see real GDP 0.25% lower about two years after the shock. The dashed line represents a counterfactual simulation that Bernanke and co-authors conducted assuming that the Fed deviated from the policy that it followed over this period and instead kept the fed funds rate from rising following the shock. This counterfactual simulation suggested that the Fed may have been able to avoid any contractionary effect of the oil price increase.
Wayne State University Professor Ana Herrera and I took a closer look at those results in a paper published in the Journal of Money, Credit and Banking in 2004. One thing we noticed was that the counterfactual policy being proposed would have been a pretty radical departure from what actually happened. For example, the Bernanke simulation would call for the Fed to have kept the fed funds rate at 4% all through 1973 and 1974, whereas in the actual event it rose above 13%. Ana and I applied some ideas proposed by Chris Sims in 1982, and elaborated on in 2003 by two of his students, Eric Leeper and Tao Zha (each of whom were to go on to become quite famous in their own right). Their framework allowed us to construct confidence intervals to summarize how far out of the ordinary a proposed counterfactual policy is relative to the data. The solid line in the figure below indicates the amount by which the proposed counterfactual policy intervention was supposed to have been able to change U.S. real GDP according to the Bernanke, Gertler, and Watson dynamic relations, indicating that real GDP could have been 6% higher by 1977. This effect turns out to be well outside the Leeper-Zha 95% confidence intervals, indicated by dashed lines in the figure below,
A more important detail that Ana and I brought out was that the Bernanke, Gertler, and Watson simulations used a vector autoregression with 7 monthly lags, whereas most of the rest of the literature on oil shocks has recognized that some of the biggest effects take longer than this to be observed. It turns out that using the Bernanke, Gertler, and Watson data, one would reject the null hypothesis that 7 lags are enough, favoring an alternative of 12 lags with a p-value as small as one in a billion.
Ana and I then redid the Bernanke, Gertler, and Watson counterfactual simulations, doing everything exactly as in their original study, except this time using 12 lags instead of 7. The results are summarized in the figure below. The estimated effect of the oil shock itself (the solid line) is significantly bigger than one would conclude if one assumed that the effects can all be seen within 7 months. More importantly, even the very aggressively counterfactual monetary policy would appear to have made relatively little difference for the outcome.
Here's the inference that Ana and I drew from the exercise:
We conclude that the potential of monetary policy to avert the contractionary consequences of an oil price shock is not as great as suggested by the analysis of Bernanke, Gertler, and Watson. Oil shocks appear to have a greater effect on the economy than suggested by their VAR, and we are unpersuaded of the feasibility of implementing the monetary policy needed to offset even their small shocks.
Posted by James Hamilton at October 16, 2011 06:40 AMdigg this | reddit
It is possible to borrow money to borrow oil consumption from other countries, but at some point, you have to pay it back. That's not adjustment smoothing, it's intertemporal transfers.
The argument for lower interest rates could make sense if you look at the elasticity of GDP wrt oil consumption, specifically related to investment in fixed assets. If the economy takes time to adjust to an oil shock, then lower interest rates might help by making it easier to purchase more fuel-efficient equipment. But then you need a view of how that adjustment actually occurs--there needs to be some micro level narrative in which elasticities are central. Does anyone have that? It certainly has not popped up in the blogosphere, and I am under the impression that the media and much of the economics community is assiduously avoiding the issue.
In any event, Carol Dahl, from the Colorado School of Mines, is the expert on all things elasticity (which apparently everyone knew but me). I'll look her up in the next couple of weeks for a chat on the matter.
Posted by: Steven Kopits at October 16, 2011 08:12 AM
Let's not forget the "stagflation" of the 1970s.
Posted by: bmz at October 16, 2011 08:21 AM
JDH This is interesting. Unfortunately the papers are gated. You said that the Bernanke paper used a VAR approach similar to the one described in your post of Sims winning the Nobel, but in that post you referenced Sims 1980 paper rather than his 1986 paper. Since Bernanke did parallel work with Sims in 1986 on structural VARs I'm assuming that Bernanke's paper on oil shocks also used a structural VAR approach??? But then you said that you used 12 lags rather than 7. Was there any reason why the VARs had to be balanced?
Your impulse/response charts...are these cumulative or step-ahead shocks?
Finally, were there any interesting results from the variance decompositions? In other words, even if the impulse/response charts didn't support the Bernanke paper, was there anything in the way one variable affected the variance of another variable that might be of interest?
See what you get when you post something and all of the references are gated? You get a bunch of questions!
Posted by: 2slugbaits at October 16, 2011 08:23 AM
Posted by: Adam P at October 16, 2011 09:39 AM
Interpreting the answer to the forecasting question (if oil prices are higher than I expected, how does that cause me to revise my forecast of GDP) is often accepted to be a question directly of interest, for which the original Sims' methodology is adequate. To address the counterfactual (if Fed had behaved differently) requires some structural assumptions about the Fed behavior, though it is possible to answer the question of interest with only these and without spelling out a full-blown structural model as in Sims (1986) or Bernanke (1986)-- I believe Ben Bernanke himself may well have been the first to recognize this point. The second figure in my post above is actually making use of the Leeper-Zha idea that is not itself in any of Sims' papers.
Figures 1 and 3 above indicate effects on the growth rate, whereas Figure 2 accumulates to get an effect on the level of real GDP. I've amended the figure captions to make this more clear.
I don't recall if we did separate calculations of variance decompositions; in any case, these didn't get into our write-up of the results.
Posted by: JDH at October 16, 2011 10:53 AM
The global oil supply situation, post-2005, is materially different from prior oil shocks, although discrete regions have certainly shown similar patterns to what we are seeing globally.
Following is a first draft of one of the graphs that I will be showing at the ASPO-USA conference in Washington, D.C., in early November:
More info: www.aspousa.org
Three curves are shown for 2002 to 2010: Production by top 33 net oil exporters; Global Net Exports (GNE) and Available Net Exports (ANE). The shaded regions correspond to Top 33 consumption, to Chindia's combined net oil imports and to ANE.
GNE = Net Exports from top 33 net oil exporters in 2005, BP + Minor EIA data
ANE = GNE, less Chindia's combined net oil imports
The 2010 to 2020 data are extrapolated, based on 2005 to 2010 rates of change.
ANE fell from about 40 mbpd in 2005 to about 35 mbpd in 2010. If we extrapolate the 2005 to 2010 rates of change in top 33 production and consumption and in Chinda's net oil imports, ANE would be down to about 21 mbpd in 2020. In other words, for every two barrels of oil that non-Chindia importers net imported in 2005, they would have to make do with one barrel in 2020, while Top 33 total petroleum liquids production in 2020 would be basically the same as 2005, under this scenario.
If we assume a 1%/year Top 33 production decline rate from 2010 to 2020, and leave the other assumptions unchanged, ANE would be down to about 15 mbpd in 2020, versus 40 mbpd in 2005. I plan to do another chart showing this scenario.
Posted by: Jeffrey J. Brown at October 16, 2011 12:45 PM
Bernanke has the typical Keynesian problem of being to obsessed with interest rates. FFR don't mean that much. Oil "shocks" are a opinion. 1973 is much stronger, than 1990, when the shock occurred when the economy was already in recession and didn't last long at all. I don't see it as a oil shock. Of course Bernanke is also such influenced by traditional keynesianism, he rejects a strong central bank instead for fiscal policy, boy I understand why the mainstream republican party feels slighted, which I think is coming out in this campaign(and not for the reason they say). Via Hayek/Friedman(the mainstream Republican party economists) where is the NGDP targeting Ben lol? Even though they won't admit it. I think alot of Ben's remarks about "Milton" and such were kiss ups in name only.
Yes, the decline of the north sea and the Saudi's unwillingness to produce at a greater rate to take up China/India huge intake is the key behind the move in oil and really commodity prices over the last 10 years, actually hurting the economy and creating a "inflation" global condition. The only way to drop them farther is destroy enough money where demand drops back down to current production levels. The key and fight going on for the last 5 years is who, when and how. Chininda's think the US is a decadent gluttonous pig. And the US thinks Chininda is a undeserving, jealous losers.
Posted by: The Rage at October 16, 2011 01:17 PM
"Yes, the decline of the north sea and the Saudi's unwillingness to produce at a greater rate to take up China/India huge intake is the key behind the move in oil and really commodity prices over the last 10 years"
What if the reality is better expressed as follows:
"Yes, the decline of the north sea and of Saudi Arabia, versus China/India huge intake is the key behind the move in oil and really commodity prices over the last 10 years"
In any case, the key problem is not that Saudi Arabia has not increased production. Based on annual data, the problem is that Saudi net oil exports have dropped from 9.1 mpbd in 2005 to 7.2 mbpd in 2010 (BP). While there are reports of increases in recent monthly production, this may have come from inventory draw downs, and in any case it appears that little if any additional oil made it into the export market.
If we extrapolate the 2005 to 2010 rate of increase in the ratio of Saudi oil consumption to production, they would approach 100%, and thus zero net oil exports, in about 14 years.
Posted by: Jeffrey J. Brown at October 16, 2011 02:37 PM
To Bernanke, the answer to all of world's problems was and is always - lower interest rates. It is hard to find such bankruptcy of economically useful ideas among the central bankers of even today. This is precisely the reason why we are in such a mess today. And the economy is refusing to respond to his bogus theories and ministrations. He should resign immediately.
Posted by: ReturnFreeRisk at October 17, 2011 03:58 AM
External account CAB imbalances have to be rebalanced through the external trade and balance of payments accounts.
Lower interest rates give the wrong signal to net importing countries, which need to cut consumption and cut imports, even as they shift some of the cost of the adjustment to the net exporting countries.
What is needed is a big increase in spreads and intermediation margins, to protect the banks which intermediate between the net surplus and net deficit countries.
And small countries are simply steam-rolled.
How can trade midgets rebalance the entire intra-Eurozone trade imbalances?
Posted by: Mariana Abrantes at October 17, 2011 05:04 AM
Raher than denying the existence of a potential variable that is monetary policy,one may check from the existing set of exogeneous or endogeneous variables wether or not monetary policy (ies) may have been of influence in prices variances or prices volatility.
Oil Price Volatility and U.S. Macroeconomic Activity
Hui Guo and Kevin L. Kliesen
"A narrative approach that relates Wall Street Journal news accounts with the 10 largest daily price movements of the 12-month futures contracts over the period April 1983–December 2004.
Most of the events associated with the largest percentage changes are related to developments among the Organization of the Petroleum Exporting Countries (OPEC) or political instabilities in the Middle East. Interestingly, among the 10 largest price changes, half occurred during 1986,when crude oil prices plunged. We also found similar results using the next 40 largest price movements (which are available upon request).We confirmed Hamilton’s results using higherfrequency data."
"The second method relies on formal statistical tests. Table 2 measures whether standard macrovariables forecast one-quarter-ahead realized oil futures variance. The predictive variables include past realized oil variance, RV_O; the oil price change, RET_O; realized stock market variance,RV_S; stock market return, RET_S; the default premium, DEF; the term premium, TERM; andthe growth rate of real GDP, D_GDP. The default premium is the difference between the yield on Baa- and Aaa-rated corporate bonds, and the term The term premium is the difference between the yield on 10-year Treasury notes and 3-month Treasury bills.A sizable literature suggests that yield spreads like these contain valuable information about current and prospective business conditions "
Above batch of data are closer to monetary policy through term premium 10 year treasury notes 3 months treasury bills (even real monetary policy during this last decade)
One conclusion on both tests CQFD (Ce qu'il fallait démontrer)
"Therefore, our results are consistent with the evidence in Table 1 that oil price volatility originates mainly from exogenous shocks to the U.S. economy rather
than endogenous responses to these shocks."
Posted by: ppcm at October 17, 2011 05:33 AM
Sober conclusion in an era of peak cheap oil. The third world now wants to use more oil, but production cannot ramp up significantly. There is plenty of non conventional oil, but it takes too many resources and man hours to extract. Canadian oil shale is an example. It takes 25% of Canada's natural gas to extract a small, but steady, amount of oil from the shale. You just can't ramp this up enough to meet third world demand.
When the central bank printed the housing Minsky, this helped drive oil to $150. This slowed the world economy, and the subsequent credit crisis slowed it even more. Building Keynesian pyramids is counter productive in an era of peak cheap resources. That is, printing more credit to "stimulate" the economy is counter productive when resource extraction cannot ramp up significantly.
Posted by: Mike at October 17, 2011 10:25 AM
More of this will not simply mitigate "oil shocks," but help eliminate them:
"North Dakota will likely leapfrog California and may even overtake Alaska in the next year - far outpacing earlier industry predictions - to become one of the nation's three biggest oil-producing states, a government regulator said. Government and industry officials had predicted that North Dakota likely would hit the No. 2 spot within the decade but the explosion of drilling activity has accelerated the timeline.
Not considered a big oil state until recently, North Dakota went from the ninth-biggest producer in 2006 to fourth in 2009, where it currently stands. This boom is thanks to advances in drilling and hydraulic fracturing techniques and a rise in oil prices that made it more profitable for companies to tap into the vast reserves trapped in the Bakken and Three Forks shale formations."
But fear not... the EPA will find a way to reverse this.
Posted by: Bruce Hall at October 17, 2011 03:52 PM
Here's a topic for you, Jim: Should we now replace the oil drawn from the SPR earlier this year? Or should we wait?
Platt's take on it: http://www.platts.com/weblog/oilblog/2011/10/17/regulation_the_8.html
Posted by: Steven Kopits at October 17, 2011 03:53 PM
Bruce Hall: Here are some numbers from a paper I'm writing at the moment that might interest you. In 2010, the combined production from North Dakota and Montana was 1/2 that of the state of Oklahoma in 1927 (the year when Oklahoma production peaked), and 1/5 that of the state of Alaska in 1988 (the date of the Alaska peak).
Posted by: JDH at October 17, 2011 04:21 PM
Of course ND's absolute numbers are still small given the cost of start-up and the general antipathy of the Obama administration. Regardless, the progress being made there ... and in Canada ... is strong evidence that newer technology can deliver the goods [or the oil as the case may be] as long as the government gets out of the way a little rather than being an active impediment.
Additionally, natural gas from the same sources can and will provide a much greater degree of freedom from energy deprivation so much desired by the "alternative" crowd.
Now if we can only get moving on the 4th generation nuclear reactors to provide electricity so all of those Chevy Volts being sold will have enough juice to power them... ;-}
Posted by: Bruce Hall at October 17, 2011 06:22 PM
Do you agree with the market's pressure today on Halliburton? Sounds like drilling (fracturing) for gas would increase rather than decrease, although HAL's management expressed concern about slowing of gas drilling. If I understood correctly, this sentiment was repeated on the Nightly Business Report on PBS. as a reason for the stock decline of HAL today
Reuters reported on HAL's management opinion quoted below.
"Oct 17 (Reuters) - Halliburton Co , the world's second-largest oilfield services company, posted a higher-than-expected profit, but its stock reversed Friday's gain due to the less exuberant outlook of its management.
Chief Executive Dave Lesar talked of the risk of decreased U.S. gas-directed drilling, and expected some rigs to be redeployed to liquids-rich regions, though he also noted such shifts can weigh on efficiency and financial performance."
Posted by: AS at October 17, 2011 06:36 PM
Bruce Hall, States with declining oil production are easier to overtake. You aren't comparing Alaska in its oil hey day. You are comparing Alaska as its oil production continues its long descent.
Posted by: Randall Parker at October 17, 2011 08:36 PM
Two different issues: short-term market fluctuations in the price of stocks related to oil exploration and the long-term trend of oil exploration and technology improvements in extraction.
Unless the world drops into a major depression and demand tanks, the break-even point for oil extraction will continue to drop as methods improve.
Look at the cost/time for DNA analysis which now approaches 3 minutes with on-site analysis. It wasn't that many years ago when the cost was prohibitive. http://arstechnica.com/science/news/2011/10/-new-hardware-powers-through-dna-tests-in-under-3-minutes.ars
For some reason(s), the government has been a reluctant party in taking advantage of both resources and the technology to access them... and the positive economic impact that comes with it. http://money.cnn.com/2011/09/28/pf/north_dakota_jobs/index.htm
Posted by: Bruce Hall at October 18, 2011 03:35 AM
Re: Bruce Hall
It's true that there are some good stories about rising US Mid-continent oil production, but we should keep it in perspective, and the industry is facing very real infrastructure and personnel bottlenecks.
It's a safe assumption that at least 90% of all Bakken well that are currently producing will probably be down to less than 10 BOPD by the year 2020.
Average US crude oil production for 2010 was 5.5 mbpd (EIA). And the average rate through September for 2011 was 5.6 mbpd.
In effect, we are seeing a race between slowly rising US crude oil production and the declining global supply of Available Net Exports (ANE), which fell from 40 mbpd in 2005 to 35 mbpd in 2010. ANE = Global Net Exports of oil, less Chindia's combined net oil imports.
Posted by: Jeffrey J. Brown at October 18, 2011 05:07 AM
... and the point is that we should put up more roadblocks to resource development?
You seem to have forgotten the past 4 decades of government regulation and obstruction regarding oil exploration and development. In contrast, Canada's production has increased by about 28% [1998-2008] http://www.neb-one.gc.ca/clf-nsi/rnrgynfmtn/sttstc/crdlndptrlmprdct/stmtdprdctn-eng.html
It's a matter of will and resolve, neither of which has come out of Washington, D.C.
I suppose we could wait until the condition is Stage 4 and then declare there is no hope for the patient. Oh, wait. That's another government program.
Posted by: Bruce Hall at October 18, 2011 06:43 AM