January 15, 2011
Oil shocks and economic recessions
I've just completed a new research paper that surveys the history of the oil industry with a particular focus on the events associated with significant changes in the price of oil. Here I report the paper's summary of oil market disruptions and economic downturns since the Second World War. Every recession (with one exception) was preceded by an increase in oil prices, and every oil market disruption (with one exception) was followed by an economic recession.
|Business cycle |
|Nov 47-Dec 47||Nov 47-Jan 48|
|May 52||Jun 53|
|Nov 56-Dec 56|
|Jan 57-Feb 57|
|Suez Crisis||Aug 57|
|none||Feb 69 (7%)|
Nov 70 (8%)
|no||strike, strong demand,|
Dec 73-Mar 74
|Apr 73-Sep 73|
Nov 73-Feb 74
|May 79-Jul 79||May 79-Jan 80|
|yes||Iranian revolution||Jan 80|
|none||Nov 80-Feb 81|
|none||Aug 90-Oct 90|
|no||Gulf War I||Jul 90|
|none||Dec 99-Nov 00|
|no||strong demand||Mar 01|
|none||Nov 02-Mar 03|
Gulf War II
|none||Feb 07-Jun 08|
The table above itemizes the particular postwar events that are reviewed in detail in my paper. The paper also provides the following summary discussion:
The first column indicates months in which there were contemporary accounts of consumer rationing of gasoline. Ramey and Vine have emphasized that non-price rationing can significantly amplify the economic dislocations associated with oil shocks. There were at least some such accounts for 5 of the 7 episodes prior to 1980, but none since then.
The third column indicates whether price controls on crude oil or gasoline were in place at the time. This is relevant for a number of reasons. First, price controls are of course a major explanation for why non-price rationing such as reported in column 1 would be observed. And although there were no explicit price controls in effect in 1947, the threat that they might be imposed at any time was quite significant (Goodwin and Herren, 1975), and this is presumably one reason why reports of rationing are also associated with this episode. No price controls were in effect in the United States in 1956, but they do appear to have been in use in Europe, where the rationing at the time was reported.
Second, price controls were sometimes an important factor contributing to the episode itself. Controls can inhibit markets from responding efficiently to the challenges and can be one cause of inadequate or misallocated supply. In addition, the lifting of price controls was often the explanation for the discrete jump eventually observed in prices, as was the case for example in June 1953 and February 1981. The gradual lifting of price ceilings was likewise a reason that events such as the exile of the Shah of Iran in January of 1979 showed up in oil prices only gradually over time.
Price controls also complicate what one means by the magnitude of the observed price change associated with a given episode. Particularly during the 1970s, there was a very involved set of regulations with elaborate rules for different categories of crude oil. Commonly used measures of oil prices look quite different from each other over this period. Hamilton (2010) found that the producer price index for crude petroleum has a better correlation over this period with the prices consumers actually paid for gasoline than do other popular measures such as the price of West Texas Intermediate or the refiner acquisition cost. I have for this reason used the crude petroleum PPI over the period 1973-1981 as the basis for calculating the magnitude of the price change reported in the second column of Table 1. For all other dates the reported price change is based on the monthly WTI.
The fourth column of Table 1 summarizes key contributing factors in each episode. Many of these episodes were associated with dramatic geopolitical developments arising out of conflicts in the Middle East. Strong demand confronting a limited supply response also contributed to many of these episodes. The table collects the price increases of 1973-74 together, though in many respects the shortages in the spring of 1973 and the winter of 1973-74 were distinct events with distinct causes. The modest price spikes of 1969 and 1970 have likewise been grouped together for purposes of the summary....
These historical episodes were often followed by economic recessions in the United States. The last column of Table 1 reports the starting date of U.S. recessions as determined by the National Bureau of Economic Research. All but one of the 11 postwar recessions were associated with an increase in the price of oil, the single exception being the recession of 1960. Likewise, all but one of the 12 oil price episodes listed in Table 1 were accompanied by U.S. recessions, the single exception being the 2003 oil price increase associated with the Venezuelan unrest and second Persian Gulf War.
The correlation between oil shocks and economic recessions appears to be too strong to be just a coincidence (Hamilton, 1983a, 1985). And although demand pressure associated with the later stages of a business cycle expansion seems to have been a contributing factor in a number of these episodes, statistically one cannot predict the oil price changes prior to 1973 on the basis of prior developments in the U.S. economy (Hamilton, 1983a). Moreover, supply disruptions arising from dramatic geopolitical events are prominent causes of a number of the most important episodes. Insofar as events such as the Suez Crisis and first Persian Gulf War were not caused by U.S. business cycle dynamics, a correlation between these events and subsequent economic downturns should be viewed as causal. This is not to claim that the oil price increases themselves were the sole cause of most postwar recessions. Instead the indicated conclusion is that oil shocks were a contributing factor in at least some postwar recessions.
Posted by James Hamilton at January 15, 2011 08:18 AMdigg this | reddit
What do you make of the doubling in oil price since early 2009?
Given the growth rate in developing countries and the recovering of developed countries, oil is probably going to head higher.
Posted by: Joe at January 15, 2011 08:57 AM
Joe: According to the model developed in my 2003 Journal of Econometrics paper (working paper version here), oil prices wouldn't have the potential to induce another recession until after June of 2011. I do believe they can start to exert a drag on growth, however, as I described here.
Posted by: JDH at January 15, 2011 09:09 AM
How many times were there major price increases (however you define major) that were not followed by a recession?
Posted by: Art at January 15, 2011 09:59 AM
Loosely related, Freakonomics weblog has a post up on Peak Driving.
You already know my beliefs regarding this subject. The aging baby boom population and more people driving with children are probably contributing.
But I think there are oil/gasoline price responses that compound these problems and create a positive feedback loop. People driving more during more congested times for business to make up for shortfalls in income and driving less for recreation and leisure in off-peak times. People moving closer together overwhelming benefits of shorter distances...
Here is my comment I submitted:
I believe the reason is congestion. Miles per gallon started dropping in 2005 (data has since been revised and this very significant drop no longer shows, I suspect this is due to the influx of stimulus dollars introducing bias. The biennial report done in 2010 brought miles travelled up almost 3% from the TVT numbers published monthly. Normally the TVT numbers are within 1%.)
Three things I believe are the primary causes are:
Inadequate road building and maintenance.
Cell phone usage.
Rising gas prices.
Rising gasoline prices decreasing our fleet efficiency is counter-intuitive, but it is very logical when thought through thoroughly.
Prior to the 2008 price spike, I believe there was a sort of Giffen Behavior happening, where people were driving more during congested times for business to make up for their shortfalls in income and driving less during off-peak times for recreation and leisure.
Accelerating too slowly. The common advice to drive economically is to drive as if there is a glass of water on the dash board. This is good advice for an individual driver, but it is flawed for a couple reasons.
One, it is great advice because it minimizes braking, which is the primary energy wasting behavior in driving. However, this advice is overly conservative on the acceleration side. When accelerating, smooth and quick acceleration is most efficient. Getting RPM a little above 3000 (for typical gasoline vehicles) is most efficient. Accelerating smoothly and quickly will be more efficient, provided the driver does not accelerate too much and then need to brake.
Second, while this advice is good for an individual driver, it does not consider the driver’s effect on the system as a whole. Slow accerlation decrease through-put at bottlenecks. This is also why cellphone/smart-phone usage probably makes a large contribution to the increase in cogenstion we see, despite less driving overall. We end up with more stop-and-go. It's sort of like the tragedy of the Tragedy of the the Commons. Individuals trying to optimize their fuel economy (or time, in the case of phone users) degrades the economy of the system as a whole.
Posted by: aaron at January 15, 2011 11:32 AM
Art, he says in the post just one, the 2003 price rise. I think the reason is that before 2008 people still had hope their income situation would improve and borrowing was cheap.
Posted by: aaron at January 15, 2011 11:38 AM
Upward drive-time spikes would likely have an effect similar to energy price effects.
Posted by: aaron at January 15, 2011 11:43 AM
Good and timely job. Not al countries and not all economic sectors depend equally on oil prices. It should be nice to know a bit more specifically about the putative effects of oil hikes on households who in many cases are debt-strained. Another question is if, given that many manufactures are done in countries with low labor costs, increasing oil prices would potentially translate in relatively higher price increases in those manufactured goods (manufacturing energy costs may be relatively more important now than two decades ago).
Posted by: Ignacio at January 15, 2011 02:52 PM
Great work, thanks!
Posted by: Carl Lumma at January 15, 2011 03:08 PM
Dr. Hamilton, I recommend the Brookings Institute paper linked at the end of the Freakonomics post.
Posted by: aaron at January 15, 2011 03:49 PM
"Every recession (with one exception) was preceded by an increase in oil prices,"
There would seem to be some simultaneity at work in this result - namely, recessions are typically prededed by booms, which would tend to drive up the price of oil, since it has rather steep short run supply and demand curves.
Posted by: don at January 15, 2011 04:26 PM
The importance of oil to the modern economy can not be overstated ... Roughly 95% of transportation is driven with oil. There are NO scalable substitutes.
Both the supply and demand are very inelastic.
The supply takes years to develop and because of the huge investments in exploration, development and transportation there are created two phenomena ...
The picking of first, the low hanging fruit, the best and largest deposits closest to easy transportation and markets. The second is the rapid depletion of developed oil to increase the ROI on these enormous investments.
The demand is also very inelastic ... especially transportation of goods. Huge investment both personally in cars and commercially in trucks, rail and ships are long term sunk costs. In the very short term dissavings will bridge this gap. In the medium and long term demand destruction will occur.
What we are facing now is threefold. First the underinvestment and corner cutting in oil production infrastructure as witnessed by the Alaskan Pipeline leak, the fires at Canadian tar sands production and the BP blowout in the gulf.
Second the depletion of all of the world's "low hanging fruit" oil deposits. New oil fields are smaller, contain less desirable oil,in locations where transportation is difficult and \or politically unstable.
Third, we have now moved oil futures onto markets where it is no longer producers and refiners that play but huge financial firms that have no stake in the actual oil itself ...
According to the IEA world crude oil production peaked in 2006. The graphs that you see are "liquids" that can be anything from nat gas liquids, to ethanol, to tar sands etc. The point is that these substitutes contain far less energy and are not scalable to substitute for crude oil.
Going forward there will have to be enormous demand destruction in the economy ...
Posted by: mmckinl at January 15, 2011 04:53 PM
Posted by: Dave at January 15, 2011 05:26 PM
Yes speculation ...
"Third, we have now moved oil futures onto markets where it is no longer producers and refiners that play but huge financial firms that have no stake in the actual oil itself ..."
I was trying to be polite and let people draw their own conclusions ... But when money plays markets where they trade on bets mischief is a given ... I say if they buy it they should take delivery ... Wall Street would be awash in oil LOL!
Posted by: mmckinl at January 15, 2011 06:00 PM
Wow, that last line of your paper is a casual little bombshell, just another row to your table...
What's clear is that neither market forces nor central planning "price controls" and the like can prevent the boom and bust economics of oil availability. Geopolitical and geological events seem to carry the day regardless. Such suggests an inherently unstable dependence on oil as a means of human sustenance.
Thanks so much for your paper on the history of oil prices and shocks.
Posted by: Dave at January 15, 2011 06:06 PM
well, you should also note which ones were associated with monetary tightening (interest rates or money supply or both). its not the oil shock itself, its the ensuing inflation and subsequent policy response.
Posted by: dwb at January 15, 2011 06:30 PM
There are far bigger efficiency gains to be made on the road than in the power-train.
Posted by: aaron at January 15, 2011 07:10 PM
"There are far bigger efficiency gains to be made on the road than in the power-train." ~ aaron
Indeed there are, but sunk costs don't go away. Very troubling is the fact that the best selling models are once again trucks and SUVs.
When we look at electric vehicles they are on average twice as much as gas models ... they are now being subsidized.
Fact is there is not enough lithium or any other battery component that can replace oil powered equipment.
Posted by: mmckinl at January 15, 2011 08:38 PM
For Joe, who wrote "What do you make of the doubling in oil price since early 2009?"
My assessment (details below) is that the oil price runup (boom phase) can go on for quite a while before the recession hits. A major decline in the oil price would be a bearish sign. Of course, in 2008 the economy had been in recession for some time before the price peaked... but once it peaked, the recession took a deeper turn.
I'm also of the opinion that the "2001" recession went on longer than NBER would like to admit (or was in some ways a double-dipper), so the 2002-2003 oil price episode isn't really separate from that recession IMHO.
It would be nice if Professor Hamilton were to include in his paper a chart of the historical oil price, indicating both when recessions occurred and when price controls were either in effect or were threatened.
I've attempted to generate such a chart (using the info from the table plus various FRED Graph proxies for energy costs), and my conclusion is that a price runup doesn't tell you a recession is imminent, but a price drop after a runup is likely to indicate a recession is in progress.
Posted by: Wisdom Seeker at January 15, 2011 10:46 PM
I also wonder about the role of speculation. Speculators bought almost a trillion barrels of oil between 2003 and 2008, almost as much as the increase in Chinese oil consumption in the same time period.But speculation doesn't even figure into this?
Posted by: Michael Haley at January 15, 2011 10:46 PM
Also see the work of Andrew Oswald, who correctly predicted the 2000 recession using oil prices and interest rates.
One key consideration is the relationship between energy and labor. As Oswald said somewhere, everything is composed of energy and labor, and in order for general prices to remain stable, when energy goes up, labor must come down.
Much of the success of the economy during the Clinton years had to do with the unprecedented low oil prices while OPEC was in disarray.
It is interesting to speculate on why oil prices trigger recessions. It is easy to see that oil price increases act as a tax on consumers, reducing their spending on consumer discretionaries and other commodities. But what is also interesting is to see the low number of jobs per dollar in the resource extraction industries. Spending on oil does not support the jobs that spending on manufactures supports.
Good work. Thank you.
Posted by: demandside at January 15, 2011 11:29 PM
June July 2011 as a start of peak oil that will start the next recession in Q3 of Q4 of 2011 is a very accurate observation. Oil will reach about 130 USD in late June 2011, and never really look back from that number during the next 3 years. More so, it will continue to rise with 15% fluctuations, reaching 155 USD before the end of 2011 and 190 USD during spring of 2012, with a fall-back after that down to 135USD till Q3 2012. After that, increase again throughout 2012 and 2013. Before the USA takes military action (successful????) to protect the oil supply and the value of USD in 2014.
Posted by: Ivars at January 16, 2011 08:03 AM
"Oil will reach about 130 USD in late June 2011, and never really look back from that number during the next 3 years."
A lot of people thought that before 2008 and they were wrong. They saw the price drop precipitously and were surprised.
The price also depends on how the financial system responds. There are two systems that play against each other, the energy system and the financial system. If the financial system breaks then all bets are off.
If it doesn't break, businesses have lower sales and shed workers or go bankrupt because they can't meet their debt obligations. (If you think public debt is bad, private debt dwarfs it. See Steven Keen's excellent work on that.)
In other words, we are more likely to see oil prices oscillate from now on, no reason it can't go down to $40/barrel then back up to $180. Oscillation is the "natural" behavior of the price of a resource when it becomes scarce.
The oscillation may center around $130 but it's hard to know in advance.
Posted by: Andre Angelantoni at January 16, 2011 08:55 AM
A very important contribution to our understanding of the impact of oil shocks on the economy. And, of course, it sets up our understanding of future oil shocks.
Posted by: Steven Kopits at January 16, 2011 09:01 AM
But which came first, the chicken or the egg? Money flows expanded first, then oil prices responded, all of which culminated in collapsing production.
Posted by: flow5 at January 16, 2011 09:14 AM
actually there were a lot of price controls in 2008. all the arab world, venezuela, china, russia, etc. of course the prices varied significantly, but, at the margin, the straight economic effects of high prices occurred mostly in the "developed" world (lower usage, substitution, etc).
Posted by: clipb at January 16, 2011 10:10 AM
Mr Hamiltyon, very interesting. But what I see always in te US postwar recessions, is a some great contracion of money supplay (M2 for example). See
So, it is possible that teh contribution of oil prices rise to recessions is via an contrative monetary policy to reduce the excess of AD.
Posted by: Luis H Arroyo at January 16, 2011 10:40 AM
sorry, i´d mean
Posted by: Luis H Arroyo at January 16, 2011 11:25 AM
"The price also depends on how the financial system responds. There are two systems that play against each other, the energy system and the financial system. If the financial system breaks then all bets are off.
If it doesn't break, businesses have lower sales and shed workers or go bankrupt because they can't meet their debt obligations. (If you think public debt is bad, private debt dwarfs it. See Steven Keen's excellent work on that.)"
It seems more likely to me that the USD will break first. And more of them will seep into oil futures. It will happen relatively slowly after the lessons of 2008. So both up and downswings will be smaller in amplitude, say 15%.
Posted by: Ivars at January 16, 2011 03:13 PM
Doesn't it make sense that the demand for oil peaks at and then after the business cycle has itself peaked? Perhaps the rise in oil prices marks the inflection stage or point rather than being the necessary cause. In other words, the two should correlate. I'm sorry I don't have time to read the paper to know if you addressed this or if other commenters have brought it up.
Posted by: jonathan at January 16, 2011 07:10 PM
Do not think that supply/demand is directly related to the oil price, and therefore I do not agree with your key factors and your results.
Would rather say the oil price increases due to speculative forces .. Yes James, speculation does not care about market forces, indeed.
Posted by: Johannes at January 17, 2011 12:41 AM
Dave, mmckinl, and Johannes: In your opinion, is the price of oil such that the physical quantity supplied is greater than the physical quantity demanded? If so, what happens to the difference? If not, in what sense do you say that the price is inconsistent with fundamentals?
Posted by: JDH at January 17, 2011 07:29 AM
Although global annual crude oil production has so far not exceeded the 2005 annual rate, despite annual oil prices exceeding, for five years, the $57 level that we saw in 2005, with four of the five years showing year over year increases in oil prices, the real story is the global net export market.
Here is a slide showing key global net export data, along with "Chindia's" net imports:
Our work suggests that the US, and many other developed countries, are well on our way to "freedom" from our dependence on foreign sources of oil, as we are gradually outbid--by the developing countries--for declining global net oil exports.
Posted by: Jeffrey J Brown at January 17, 2011 09:16 AM
@ JDH at January 17, 2011 07:29 AM :
James, there are some financial vehicles that are, by definition, speculation. For instance, trading commodity futures contracts, such as for oil and gold, is, by definition, speculation.
Speculation can cause prices to deviate from their intrinsic value if speculators trade on misinformation, or if they are just plain wrong. Yes James, this can also happen ! Anyway, this creates a positive feedback loop in which prices rise dramatically above the underlying value or worth of the items. This is known as an economic bubble. Such a period of increasing speculative purchasing is typically followed by one of speculative selling in which the price falls significantly, in extreme cases this may lead to crashes. But supply/demand of physical quantities is not related to such circumstances, believe it or not, James. May someone call it psychological or whatsever ...Ah yes, James, you may ask why I know this futures contract stuff so well. Yes, I was a trader ..
Posted by: Johannes at January 17, 2011 09:48 AM
Johannes: You did not answer my question, and I therefore repeat it. Is the price of oil such that the physical quantity supplied is greater than the physical quantity demanded?
Posted by: JDH at January 17, 2011 09:50 AM
"Is the price of oil such that the physical quantity supplied is greater than the physical quantity demanded ?"
James, the price of oil does not really care about physical quantities, that's what I think.
Therefore if someone says "The price of oil is such that the physical quantity supplied is greater than the physical quantity demanded" or someone says "The price of oil is such that the physical quantity demanded is greater than the physical quantity supplied" is irrelevant, in my opinion. And why I think so, I told you already.
Posted by: Johannes at January 17, 2011 10:14 AM
JDH: To answer your question, yes. If the 'excess' quantity oil supplied can be stored in offshore tankers and sold at a later date, oil supplied can, at least temporarily, exceed oil consumed.
Posted by: westslope at January 17, 2011 10:22 AM
Thanks JDH once again for another insightful paper on oil.
I wasn't aware that the supply impacts of OAPEC's oil embargo in late 1973 lasted so long as the official embargo was over within a few months. Yet between the oil shocks of 1973 and 1979, oil prices continued to rise steadily and the US avoided recession although it experienced near zero per capita real GDP growth. Perhaps the '73-'79 period hints at how the US economy will perform going forward?
Is the global economy as vulnerable to oil shocks as the US economy appears to be? It strikes me that the recent US recession that began in December 2007 was insufficient to push the world economy into recession by itself. A global recession was achieved only once the US financial sector imploded.
I ask because the structure of US excise taxes leaves US consumers more vulernable to oil shocks than consumers in other rich OECD countries.
Posted by: westslope at January 17, 2011 10:51 AM
Johannes: You may consider my question irrelevant, but if you cannot answer it, I am unable to understand your claim.
I suspect that our lack of communication is a two-way street, and that you have perhaps still not grasped what it is I have asked. For what it's worth, here is my question again, rephrased: do you claim that the physical quantity produced equals the physical quantity consumed no matter what the price?
West Slope: Yes, but offshore storage did not become significant until the second half of 2008. I presume that the comments of others were in response to the summary in the above table, which referred to the period 2007:Q4-2008:Q2. At this time EIA inventory data were also significantly below normal and falling.
Subsequently, inventories and tanker storage moved above normal, consistent with the assertion that there may be a speculative contribution to the current price.
Posted by: JDH at January 17, 2011 11:03 AM
As I understand, buying oil futures does not require anything to be delivered. Where there oil futures markets during previous shocks?
March 30, 1983: Crude-oil futures begin trading in New York
Posted by: Ivars at January 17, 2011 11:05 AM
Heating oil futures first started trading in 1978 on the New York Mercantile Exchange (NYMEX). This was the first of the energy-related commodities or by-products to have futures trading. It was quickly followed by crude oil in 1983, then unleaded gasoline in 1984.
Posted by: Ivars at January 17, 2011 11:07 AM
Excellent comments, demandside and clipb.
As a BSE in Industrial and Operation Engineering, in Ergonomics we are taught to see things in terms of Work. Conversion of potential energy to a desired outcome.
Beyond energy, I would add just one more input to economic production: Information
I see information as the organization of work. Ergonomics studies the processes and energy that generate specific desired outcomes. IOEs focussing on ergonomics see efficiency in terms of converting PE to useful KE. For IOEs efficiency in information is generally the effective use of materials in addition to energy in producing those outcomes. IOEs also look at informational efficiency by optimizing financial outcomes.
Economists would benefit greatly from some cross-training in IOE (specifically an introduction to Ergonomics, Optimization Method, Engineering Economy and Risk Management, and Financial Engineering).
To be continued... (typing on kindle and need a break)
Posted by: aaron at January 17, 2011 11:25 AM
As I have understood from prof.Shiller's lectures on net, oil that is bought and sold in futures markets has not been extracted yet- its stored in the oil wells. Futures deal with this stored oil, not the one currently consumed, however, spot prices of really consumed and delivered oil are influenced by futures prices as oil is almost always in backwardation due to perceived supply shortages as causes for temporarily high oil prices, so high future prices mean even higher spot prices. If someone speculates on future prices, spot prices move up by the character of oil trade.
The tankers started to appear when it was clear oil has moved into contagon and it made sense to really store oil as future prices were higher than spot and would cover storage costs. But that was already closer to the end of the spike.
Posted by: Ivars at January 17, 2011 11:39 AM
JDH, thank you for the analysis and for summarizing it so succinctly. It is interesting that pre-Cartel price increases of roughly 9% (1 outlier) had as much impact as post-Cartel price increases that are roughly 5 to 10 times larger. I wonder if the post-Cartel conservation efforts (e.g. popular cars now get ~3 times the MPG as cars in the '50s & '60s, homes have more insulation now, etc.) changed the micro-economics of consumers costs in ways that change the 'breaking point' where petroleum cost makes a significant budget impact in their overall consumption pattern - and hence the 'trigger point' for recession . And if so, can this be empirically modeled? Maybe you've done this - I didn't read the full paper. Anyway, its an interesting question to me.
Posted by: Mike Laird at January 17, 2011 02:27 PM
In regards to information, IOEs also see it in terms of error rates, processing time, accessability, integrity and security (adequate backup, also sometimes confidentiality).
Economists see information efficiency similarly, but more generally or as more of an abstraction. Basically a macro vs micro perspective.
Anyway, I digress.
Dr. Hamilton, in regard to speculation, I think there is a combination of informational inefficiencies that have allowed it to drive up prices above where they should be. Many political factors have resulted in a very inefficient market in information on potential reserves and potential output capacity.
Also, the lagging effect of prices on consumer behavior probably lead oil buyers to accept higher prices than are ultimately unsustainable. That, and the high cost of shutdown and start-up.
There is visible in ground hoarding, particularly in the US. This sends a powerful signal to other producers. It is also a clear indication that alternatives will not be viable for at least several decades.
Considering clipb's comment and demandside's comment in this context, it looks like we are not attempting to expand production to meet current demand. It also looks that the current price in many places prevents supply from reaching demand (supported by Jeffrey Brown's link).
Conditions seem perfect for the dynamic Krugman describes here to set in.
Posted by: aaron at January 17, 2011 02:42 PM
I think some people miss the point about speculation: it responds to the business cycle too. Is there anyone who believes that speculation in oil drives up prices when demand is dropping versus when demand is peaking and many people can thus convince themselves that demand will continue to grow?
Posted by: jonathan at January 17, 2011 02:44 PM
Dr. Hamilton, in your earlier post, linked in one of the earlier comments, you mention that Steve Koptis found that economy stumbles at oil at 4% of GDP. Did he also look at oil as a percentage of GDP growth?
Posted by: aaron at January 17, 2011 03:01 PM
One means to identify oil shocks is sharp uptick in the oil-to-gas price ratio (because gas is domestically produced and traded only regionally--hence not entirely exposed to the same pressures as oil).
Using the ratio, from 1949, we can see shocks in '57 (Suez), '74 (Arab Embargo), '79 (Iran-Iraq War), '91 (Gulf I), and from '05 ('07-'08, Great Recession). All of these were supply shocks except '07--which was a demand shock, as no adverse event affected the oil supply.
Posted by: Steven Kopits at January 17, 2011 03:18 PM
Ivars, futures prices can only influence the spot prices. Buyers will not buy once cost reduces margins below required return on investment. Bubbles in consumable and degradable commodities like oil can only happen a few ways; artificial constraint on production, by producer nations buying their own product and selling it below cost to an isolated market, and buyers with ever increasing storage capacity hoarding.
If futures prices raise spot prices,without one of the three happening, it means suppliers were under charging or buyers are willing to take a loss.
Posted by: aaron at January 17, 2011 04:42 PM
Steven Kopits: I like the idea of using energy price ratios to indicate oil price shocks. Note that natural gas is not only regionally priced but infrastructure-restricted. It is the costly barriers to adding new customers that explain in part why natural gas prices remain so low.
Inuitively, one might expect a supply-driven oil price shock to have a greater macroeconomic impact than a demand-driven oil price shock. Most supply shocks are surprises, hard to forecast. Demand shocks on the other hand should lead to slower steadier increases in price. For those prone to using simple momentum or time series models, trending higher prices might suggest even higher prices. In the end, consumers and producers have a longer time frame within which to adjust and thus the coordination problems caused by unexpectedly higher energy prices are to some extent avoided.
If Americans fully anticipate US$4/gallon gasoline, then maybe it won't cause so much economic havoc when it finally arrives..
Posted by: westslope at January 18, 2011 01:10 AM
I think I need to watch Shiller lectures about futures again. He was explaining quite simply, starting with corn examples, but the subject seems to involve more variables simultaneously that my mind can process:). Scary.
Posted by: Ivars at January 18, 2011 05:41 AM
Through what channel do you suggest oil shocks induce recessions? Is it an inflation pass-through?
Posted by: Pepe at January 18, 2011 06:37 AM
Pepe asked: Through what channel do you suggest oil shocks induce recessions? Is it an inflation pass-through?"
No. Forget inflation which has a very specific meaning in economics. Think in terms of real costs facing consumers and producers. Positive oil price shocks render existing capital (including infrastructure) obsolete by making it cost-ineffective. Sunk costs in the choice of transportation, and home location make adapting to the unexpected energy price shocks difficult. Consumers trapped in the short-term in a high energy-use lifestyle may continue to consume the same amount energy at higher and higher prices which means consumption of other items declines.
As the US is net-importer of oil, the exporters benefit and non-energy activity in the USA slows down. Retooling for higher energy prices may eventually stimulate private investment and GDP growth but the immediate consequence is slower growing or negative GDP growth that coincides with the loss of real wealth.
Hope this helps.
Posted by: westslope at January 18, 2011 11:40 AM
1) Many economic studies of climate assume an X% GDP/capita growth rate through 2100, which at 3% (a common number) makes everyone about 14X richer. that sometimes leads to the conclusion that one should defer spending money on climate change mitigation, because 2100AD folks will be much richer and can afford it much better.
2) Robert Ayres and his frequent coauthor Benjamin Warr think that energy (or more precisely, work = energy*efficiency) is a major factor in GDP growth. See The Economic Growth Engine, or especially, the last page of Economic growth and cheap oil. If he is right, in the face of Peaks, GDP is not a nice 3% CAGR, but curves down.
(Now, I understand this is a minority position, but I've also talked with Ayres over dinner and he is certainly sharp and what he says makes intuitive sense.)
3) But, as in JDH's article, the economy does seem to ignore energy supplies, especially in the form of oil. So, can someone familiar with Ayres&Warr explain to me why they are wrong, so that one can ignore any impact from peak fossil fuels? I would sleep better. I've asked, but I have yet to get a clear response.
Posted by: John Mashey at January 19, 2011 12:49 AM
"But, as in JDH's article, the economy does seem to ignore energy supplies, especially in the form of oil. "
Did I misread JDH's article? I ask because I drew the opposite conclusion.
Posted by: westslope at January 19, 2011 07:27 AM
Pepe: I attribute the effects to changes in the composition of demand.
Mike Laird: My claim is that several of the earlier price spikes would have been bigger in the absence of price controls.
John Mashey: Can you please rephrase your question?
Posted by: JDH at January 19, 2011 07:43 AM
Sorry, I was off to bed and omitted the NOT:
"the economy does NOT seem to ignore energy supplies"
(That';s the whole point of Ayres&Warr, that energy matters.)
Posted by: John Mashey at January 19, 2011 09:23 AM
So, let me try again:
Given Peak Oil/gas this century, is it plausible that GDP (in US, world or per per capita) will average anything like 2-3% CAGR over the next 90 years, as many models seem to assume?
I.e., is GDP growth plausibly independent of the availability and price of these fuels?
Posted by: John Mashey at January 19, 2011 10:37 PM
John Mashey: I don't know what GDP growth is going to be over the next 9 months, let alone the next 90 years, let alone how that number changes based on different contingencies. I understand the need for purposes of advising policy to make estimates or assumptions about these numbers, but for my part I'm just too intimidated by the challenge to give it a try.
I will say this, however, in answer to your question. I am persuaded that the mechanism behind the historical correlation that I discuss here operates through the spending channel, and that economic recessions represent temporary episodes in which output falls below a longer-run potential as a result of insufficient demand in particular sectors. This mechanism is not relevant for a 90-year horizon.
However, a key reason that I think this must be the historical mechanism is that historically the dollar share of energy costs out of total expenditures has been fairly modest. The key issue for questions such as yours is what would happen to that share under forces that put significant upward pressure on the price, which comes down to the long-run price-elasticity of energy demand. This again is another critical parameter that I'm not persuaded I know. If the elasticity is low enough, it would be hard to maintain historical growth rates if energy supplies no longer increase.
Posted by: JDH at January 20, 2011 08:29 AM
Thanks: I am happy to see someone express clear uncertainty!
Many people seem to be making assumptions ion long-term GDP growth rates, do much analysis, and come out with various conclusions about the fraction of GDP lost to climate change mitigation or lack thereof. Fierce arguments ensue regarding discount rates ... but lost from view is fact that there were pretty firm GDP growth rates buried underneath.
I would be happier if more of the modelers were saying "We don't know, we'll show several points in the range, including possibility that peak oil causes serious impacts."
Anyway, thanks for the clear answer.
Posted by: John Mashey at January 20, 2011 07:56 PM
I would like your comments regarding the role of the financial (speculative) markets over the price. I would like to share with you this numbers. Total volume of oil transactions in financial markets. Open Interest (Numbers of contracts) 2010: 163.801.627, Long Positions: 22.504.444, Net Positions: 10.461.686, Value in dolars (Billions): 13.022/2000: Open interest: 33.571.214, LP: 4.956.984, NP: 1.400.841, V$: 1.019
It is not naive to think that this amount of money would not have any impact at all?. There are several anecdotical cases where prices were driven by anything but the fundamentals. For instance, in 2008, the trader who purchased the first oil contract over $ 100, admited that he bought for the record. After that he sold it and recorded losses on this transaction.
Posted by: Hermes Pérez at January 25, 2011 06:43 AM