That’s the topic of the latest Wall Street Journal Econoblog, in which I was pleased to participate along with Stephen Brown, who is Director of Energy Economics and Microeconomic Policy Analysis at the Federal Reserve Bank of Dallas.
Here were my opening thoughts:
The price of oil has more than doubled in the last three years, going from $30 a barrel to now above $75. On previous occasions when we saw oil price moves of this magnitude, such as 1974, 1979, and 1990, the world economy went into a recession. What’s different this time?
In my view, part of the answer has to do with the cause and timing of the oil shocks. In each of the previous episodes, oil prices made their move within the space of a few months and were caused by war-related cuts in oil production. By contrast, the current run-up has been a more gradual process extended over several years, caused by surging world demand running up against limits to what global producers can supply.
One option available to an individual consumer or producer facing higher oil prices is just to go about business as before, saving or spending a little less on other items in order to pay the energy bill. The size of what you lose economically if you do that should be limited by the size of the increased energy bill itself. While painful, this is actually a fairly small number relative to the size of the GDP decline associated with an economic recession. One of the questions in explaining the recessions associated with earlier oil shocks is what accounted for that magnification of economic losses.
The answer seems to be that, in those previous episodes, consumers and businesses did not go about things as before, but instead made some pretty abrupt changes in their spending patterns. These showed up, for example, in a sudden big drop in U.S. sales of the larger, less fuel-efficient vehicles. A rapid drop in consumer confidence also led to significant drops in spending for key sectors. Because workers and capital cannot immediately relocate, the result was unemployment and idle capacity which greatly magnified the economic losses associated with the energy bill itself.
For the most part, those sudden shifts in spending haven’t been observed this time around. U.S. gasoline use has in fact continued to increase during the last three years, with the major manifestation of the higher oil prices being a decrease in the U.S. personal saving rate and increase in the extent to which Americans borrow from foreigners.
The point at which we came closest to a replay of the historical pattern was September of last year. Hurricane Katrina had a fairly dramatic effect on oil and gasoline supplies and prices, causing SUV sales and consumer sentiment to plunge. If auto sales had held steady, we would have seen healthy 3.3% growth in 2005:Q4 instead of the anemic 1.8% actually observed.
While I believe that episode had the potential to turn into an economic recession, in the event the disturbance to oil supplies was resolved more quickly than in earlier episodes such as the 1974 OAPEC oil embargo, the aftermath of the 1978 Iranian revolution, or the first Persian Gulf War in 1990. In 2005:Q4, oil prices came down as quickly as they had gone up, and consumer confidence was restored. That in my mind is a key reason why we’re not talking (yet) about the recession of 2006.
You can read the whole discussion at the Wall Street Journal’s free link, and add your own thoughts on these issues either in the comments section here or at the Wall Street Journal’s discussion board.
Do we know to what extent consumers carry the higher energy costs as (new) debt?
That’s what pessimists tell me, and if they are right, I’d say we are building in some hurt to be felt later.
Over at TOD, Stuart Staniford has been graphing a general plateau in worldwide oil consumption (he calls it oil production, but you can just as easily see it as a plateau in consumption). See http://www.theoildrum.com/tag/plateau . It would be interesting to see if the same thing applies specifically to the U.S. If U.S. oil consumption is continuing to grow while the world overall is flat, then that would suggest that other parts of the world might be declining.
I note that in prior oil shocks there were substantial decreases in oil suppy, in the range of 5-10%. In the events of 2005-2006, there has been only a failure to increase, not an actual decrease (so far). Thus the actual physical limitations on what economic activity can be carried out are far less stringent than in prior shocks. Therefore, it’s not surprising that the economic impact has been smaller.
JDH: I am interested in your point in the WSJ article that personal (dis)savings are a function of higher oil prices. In turn, for how long can such dissavings (going on for 15 months straight now) be sustained? Is it truly a function of how far up interest rates go and how far down house prices go?
Just when all the economics blogs had written off the consumer, retail sales are up 1.4% in July (boosted especially by higher oil prices). To me it seems mad, but for how much longer can this assault on fiscal prudence continue at this rate?
Emmanuel, that’s the question of the hour, isn’t it? Wish I had a fully satisfying answer.
The saving rate referred to is calculated by subtracting consumption spending from personal disposable income as measured by the national income and product accounts. That definition of income does not take into account capital gains, such as consumers may have made on houses they own. Since any capital gains are of course part of what makes you solvent, it is conceptually possible that we could maintain the current negative saving rates.
There are a number of suggestions that capital gains on stocks and real estate may be hard to come by over the next year, in which case something would seem to have to give, although even then some might argue consumers can still keep going on the strength of cashing in previous capital gains. I am not certain what to expect, other than to highlight the potential for a significant correction.
Regarding Stuart’s comment above, it is an interesting possibility that consumption is affected not just by price, but by people’s knowledge that supply has been reduced. JDH suggested a similar point a few weeks ago, at:
https://econbrowser.com/archives/2006/07/can_the_economy.html
In that case we could say that an X% increase in prices has one effect if people know that it is due to supply interruption, but a different effect if it is, as now, due to a general supply squeeze and (perhaps) anticipated future shortages.
It still seems like faulty logic to me, requiring actions based on the good of the collective rather than the good of the individual. If people know that oil supply has fallen 10%, they know that society as a whole must reduce its consumption by 10%. So, what is their response? By collective logic, they would reduce their own consumption by roughly 10%. Prices would still go up but not by as much as would normally be necessary to get that degree of reduction. People’s decisions, in this model, are guided by their desire to act for the benefit of society rather than for their own benefit.
If people were only motivated by individual benefit, we would have a group-scale prisoner’s dilemma situation. Each individual would hope that everyone else would cut back 10%, so that price increases were marginal, then that individual would only cut back by the amount that would normally be triggered by such a price increase. Maybe that is only a 3% cut, for example. Now, he knows that if everyone tries to cut back by only 3% it won’t work, and in that case prices will spike to astronomical levels. But at the same time he knows that his own individual decision to cut back by less than 10% makes no effective difference to the total world supply of oil, hence there is no reason for him to take on the costs of an additional cutback, for no benefit.
That latter reasoning is what one would expect, I think, from traditional economic analysis. However, it’s possible that in the face of an oil crisis, people are more motivated than usual to act in a collective manner and to downplay their personal well-being in the face of social necessity. Certainly we have seen such behavior during wartime and other crises. In that case, the real difference between this oil crunch and those of the past is that this one has crept up on us, without a sense of crisis (except immediately post Katrina). Perhaps this perception is beginning to change, and we may see more of a collective rather than individual response to the new peak oil crisis. In that case we should see greater cutbacks than would be motivated by price alone, therefore price will not climb as high as some predict.
Cherchez le Petrole
RE: “…the current [OIL price] run-up has been a more gradual process extended over several years, caused by surging world [energy] demand running up against limits to what global [energy] producers can supply.”
This is the conventional wisdom. That PRICES have something to do with rational SUPPLY and DEMAND, as well as irrational MARKET FORCES, is certainly true. However, the presumption of the “conventional wisdom” cited here, and echo’d elsewhere ~ bears scrutiny.
[1] So let us take OIL prices, “over several years,” up to MAR-2003; adjust these data for “inflation” or, any other measure of “buying power of USD” in real terms, and strip them of their seasonality; then, throw these results, thus up on our screens: there is no “gradual process of OIL price run-up;” ie., there is no UpTrend apparent.
[2] However, there is a long-wave industrial cycle, apparent. Accounting for this business cycle, by further de-composition, or any other means ~ the residual, thus is interpreted by many financial economists as “near stationary,” and by a few, energy economists as de facto deflation in the real Cost of Energy.
[3] Now let us add-on, or analyze apart the known OIL price data from MAR-2003: this is, to be sure ~ an OIL price “shock” apparent. Why?
Not withstanding rational SUPPLY and DEMAND, as well as irrational MARKET FORCES ~ for this is not a reason, rather, it is a rhetorical tautology ~ the “conventional wisdom” presumed and echo’d almost everywhere goes something like this: “emerging markets, such as China and India, et al are sopping up all the OIL.”
While it is certainly true that China; eg., is growing like gangbusters (from a near-dormant Third World basis), and we can clearly see in their growing demand and limited production capacity a cause for their Cost of Energy to increase ~ it is a bit of a stretch for us to conclude that the sleeping giant suddenly awoke circa MAR-2003 with a mighty thirst that “shocked” global OIL prices.
Digging around in other dirt since MAR-2003, eventually we do find a “shocking” ~ I daresay hideous, and abrupt new demand for OIL=FUEL in the most un-likely of places: The Middle East! Why…?
…War.
The AngloAmerican invasions and occupations of Afghanistan and Iraq (and now, Lebanon) DEMAND the consumption of mind-boggling quantities of OIL to achieve ~ and the logistics = “add-on cost burdens” of this SUPPLY chain are equally mind-boggling, and bloody, to sustain.
We could describe further how; eg., an Abrams tank (~1 mpg) or Bradley armoured turtle (2~3mpg), or 3 Naval Fleets offshore, or all those Combat Air Patrol and heavy, supply aircraft overhead, not to mention the obscene dynamic of increasing vehicle & personal armour demands more and more energy to move about, or even stand at-ease at the many off-grid, GreenZone Forward Operating Bases ~ demands phenomenal quantities of OIL=ENERGY; it is perhaps THE great untold story of The War…
…but consider, too, the incredible journey, say, that a barrel of oil goes through ~ from Middle-East OilWell SOURCEs, pipe-lined or tanker-shipped to refineries in the region or all the way from Persian Gulf terminals to refineries in the Gulf of Mexico; and back again (!) An endless train, from a bird’s eye view, of tanker-truck vehicles from Kuwait on up to DESTINATIONs Baghdad, ground and air bases: The overwhelming cargo of these convoys is not ammo, food & water, but FUEL: almost ~50% of FUEL cargo is consumed by these FUEL convoys, themselves :-/
The COST OF ENERGY in dollars, (and Treasury Debt) of the AngloAmerican Wars is, by far ~ far greater than any other Treasury expenditure: “shocking”.
If the AngloAmerican populace knew, as a handul of MoD economists do, the OIL PRICE SHOCK of Borrow & Bleed, perhaps more would opt to Cut & Run; or, at the very least: tie those yellow “Support The Troops” ribbons around the hoses of their petrol station pumps.
I have an idea about the negative savings rates. Since 1996 when the Fed changed CPI formulas, SS pymts have been reduced by up to 25% taking into account REAL inflation & reduced payments all compounded over the last ten yrs. I suppose it may be higher, but we’ll stick with the 25%.
The retirees are growing every year and for those who have already been on SS, those checks haven’t grown as much as inflation. They are now emptying their bank accounts slowly. US BANK leveraged retiree savings to build their bank. They can depend on retirees, but those retiress are worrying. They don’t have enough money and as they grow we all know we don’t have enough new blood to keep SS going.
Now the younger generations including younger baby boomers and Gen X are further aggravating the trend due to excess consumption financed by debt. So we’ve got swaths of America who can’t save, don’t save, or who are spending what savings they have.
Oil shocks in the past were pure politics. The coming oil shocks are pure geological limits that can not be changed not matter how much we “play with the numbers.”
le cout / avantage de paix
If sudden WAR is at least a, one of many root causes of apparent OIL PRICE SHOCK, Then let us consider the COST:BENEFIT effects of sudden PEACE.
Peace, we shall say for sake of argument, is a decimation of AngloAmerican occupation forces; that is: a reduction to one-tenth of current personnel (labour) and (capital) equipment requirements; and, energy consumption, thus.
[1] OIL PRICE SHOCK : [A] The mind-boggling DEMAND for OIL currently directed to AngloAmerican occupation forces is converted to SUPPLY. [B] The equally mind-boggling logistics of moving and refining OIL to sustain AngloAmerican occupation forces, currently re-directed from “free market,” becomes SURPLUS TRANSPORT and REFINING CAPACITY. [C] The so-called “War Premium,” or speculative BID support for OIL PRICES, OIL & FUEL TRANSPORT & REFINING CAPACITIES, abates.
[2] DESTRUCTION becomes CONSTRUCTION : Whether the regimes-in-residence in the war theatres remain (historic) Dictatorial, become (hideous) Fundamentalist Theocracies, or do evolve into some semblance of Social Democracy ~ as a result of sudden peace ~ is economically moot: CAPITAL WILL FLOW INTO THE INVESTMENT DEFICIT REGION.
[3] Monetary Policy in the AngloAmerican states will, in my opinion remain re-actionary to perceived “inflation,” treasury Debt-Loading, and “economic well-being.”
=> The other shoe to drop, for Western economies in general, and North America in particular ~ is the “stealth” effect that years of Borrow & Bleed war-time FISCAL POLICY has hitherto been a benefit, now evaporates.
America… England… will not be “re-constructed” after the war: There will be no PEACE DIVIDEND, only WAR DEBT burden.
Though there will be some effort to replace lost war-making capacity, we predict re-armament scale will be politically limited, and entirely different in scope, than the traditional (ie., 20th century) war-making capital equipment. Rather, investment will grow “security services” and SOFTWARE infra-structures to a greater extent than military HARDWARE. The Military-Industrial Complex will become the Military-Information Network.
One of the most common mistakes of businesses, and their economists ~ is that they fail to plan for success.
Rather than PEAK OIL, I ask you, consider… PEACE : What do you think that means ?
The subject line referred to “oil price shocks” but my take is that you are really arguing that this time around there is a NON-shock. In the past oil prices surged quickly and unexpectedly due to a shock in the supply curve. This time there is no shock per se.
Your argument is that consumers respond to a temporary supply shock by making abrupt changes in spending patterns, but consumers respond to longer run shocks (or non-shocks) by adjusting consumption/savings decisions. In a way isn’t this contrary to what the Permanent Income Hypothesis (PIH) would suggest? The PIH says that people adjust their consumption/savings behavior to the long run, but do not respond to short run changes. Wouldn’t the PIH predict that a short run supply shock (e.g., due to war) should not result in changes to consumer spending patterns? But you are arguing that previous oil supply shocks did result in altered spending. And wouldn’t the PIH predict that a permanent increase in the price of oil should result in a decrease in consumer spending rather than dissaving?
Great points, Slugbaits. To a first approximation, I’d treat all oil price changes as permanent, regardless of the cause, so that if prices move a little this month you shouldn’t change much this month, whereas if prices change a lot this month, you should change a lot this month. And my view is that macroeconomic frictions make those big changes the thing to worry about.
This is also related to the issue that Stuart and Hal were discussing. A rational consumer is making forecasts of both the price of gasoline and the future level of income. If, as a result of those frictions, “shocks” depress the level of income and “non-shocks” don’t, it would be rational for consumers to have a bigger response when there is a sudden big disruption to supply.
Having said all that, I’m still a bit puzzled that the cumulative response of U.S. gasoline use to the creeping price increases has been as modest as it has, when then all of a sudden last fall there’s such a huge hit for the SUV sales. I’m not going to insist that swings in consumer sentiment can always be rationally explained. The correlation between sentiment and subsequent downturns could be because consumers rationally anticipated the downturn, or could be because an emotional response by consumers contributed causally to the downturn.
I’m surprised that in all the polling on gas prices, no one has ever polled consumers on next year’s gas prices, and the year after. At least, I’ve never seen such a thing.
Remember that quote from the SUV salesman (I think it was here) who said “my customers know that gas prices will come back down.” Yeah, and maybe Prius drivers have a different idea.
And with respect to the unknown(?) of how much is being put on the credit card … that might vary between those who view this as a short term blip and those who do not.
The IEA’s World Energy Oulook 2005 is still their most recent annual report, and I think a lot of news sources are still putting out a story based on:
“The average IEA crude oil import price is
now assumed to ease to around $35 per barrel in 2010 (in year-2004 dollars) as new crude oil production and refining capacity come on stream. It is then assumed to rise slowly to $37 in 2020 and $39 in 2030. In nominal terms, the price will reach $65 in 2030.”
Sweet, eh?
Odograph
I think the price elasticity of gasoline for the US driver is very low?
Why? Because after food and mortgage, gas is the bill that gets paid. Most americans who drive cars, commute to work. You can’t work if you can’t drive there.
In fact, you can trade down to cheaper food, before you trade down your gasoline bill.
And cheaper holidays involve, normally, more driving, not less? You don’t fly to England, you go up to the lake in Michigan, 350 miles away.
The cost of gasoline is still only c. 20% of the cost of ownership of a car. Depreciation and insurance are the big costs.
So if someone offers you a big discount on a new, big car (GM and Ford did), it’s not necessarily irrational to take it, *even if* you expect gas prices to stay where they are (ie up).
From all I read, a Prius without the ‘electric button’ feature (which European and Japanese Priuses have, but American ones lack for regulatory reasons) doesn’t deliver that superior a gas mileage in any case. And except for the VW, there are no diesel cars available in America (v. Europe where they are half of new vehicles).
You’ll know gas prices are hitting when people slow down on freeways.
I totally agree with all who observe that short-term price elasticity of gasoline for the US driver is very low.
I was just thinking that people’s responses, both in the way they respond to a price surge, and the way they make long term plans, must be shaped by some long term price expectation. Who do they believe: their buddy the peak oiler, or the government spokesman with IEA numbers?
Maybe the reason so many of us drift along with old consumption patterns is that we don’t know which way the trend will really go.
(For what it’s worth, I think the real-world MPG figures from GreenHybrid hold up pretty well. My Prius racks up 50 mpg a fair amount of the time. My worst mpg over a whole tank was probably 47 mpg and change, but a lot of it depends on the driving mix. I went into the nerdy details in this old blog enrgy.)
Is it a coincidence that both gold and oil have doubled in price in the past three years?
Odograph
My guess is that people make a forecast of the future, incorporating all past information but with a strong overweight to the recent past. I believe this is consistent with Nobel-prize winning psychologists Daniel Kahneman and Amos Tversky’s ‘prospect theory’.
(in economics this is called ‘adaptive forecasting’ I believe, as opposed to the pure rational expectations model, where economic agents show perfect foresight and rationally weight all past and current information)
So a ‘blip’ rise in gas prices doesn’t change behaviour much. Because the last 3 months data, say, doesn’t have a big impact on behaviour.
However when it has been a whole year, then there is likely to be more of a change– most of us can remember what we paid for something a year ago. At a guess, when gas has been $3/gal for a whole year, the individual forecast of gas prices moves a lot closer to $3/ gal.
I notice this in stock market boards. You can almost predict when the wave of discussants will move on to commodity prices, or emerging markets say: after the underlying has doubled or trebled, there is suddenly a great interest in commodity or emerging market funds.
In those cases, where asset prices tend to regress towards the mean, this creates an observable lag in behaviour: the small guy piles in last.
Whether gas prices are the same I don’t know– but I suspect so. I also suspect something similar in housing prices (the last buyers in the cycle are convinced that housing prices never go down, they can only go up from here, etc.).
Will
odo
ps will check fuel economy blog. Thank you!
Jim Hamilton argues that one reason the current run-up in oil prices has had a smaller effect on the real economy is that the current run-up has been a “gradual process extended over several years”.
There is both theoretical and empirical evidence to support that view. With regard to the empirical evidence, my paper Oil Price Uncertainty (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=908675)
finds that the standard deviation of oil prices has had a significant and negative effect on industrial production. At a monthly frequency, the standard deviation of oil prices has remained tame relative to, for example, the episode in 1990. (See figure 4 in this paper.)
Interestingly, there is also reason to believe that a rapid decrease in oil prices may not induce immediate economic expansion. A rapid decrease in oil prices will likely increase uncertainty about future prices, inducing firms to postpone irreversible investment decisions. This may explain, in part, why the rapid decrease in oil prices in early 1986 was not accompanied by an immediate increase in economic activity.
John Elder
Thank you for your explanation of the asymmetry of oil supply movements and their impact on the economy.
Up fast is bad, down fast (or slow) doesn’t necessarily have the same effect because of the expectation effects.
I think William Nordhaus had a paper on oil price and productivity. What I found striking was that the ‘productivity slowdown’ which bedeviled the US economy in the 70s and 80s fell most strongly on oil-consuming sectors of the economy.
This is intuitively appealing. It was not that the US economy somehow lost the magic of productivity growth (although there were trends to regulation in the late 60s and early 70s which I believe may have slowed it down) but rather that the efforts and investment of the US economy was focused on a massive shift in relative prices (oil v. everything else) which took a long time to overcome.
Which you would expect, because in capital intensive industries, machinery lasts 20,30 years so if you need to replace it, it can be a long term process. The industry I am most familiar with (electric power generation) had to spend billions switching from oil fired generation to coal and nuclear generation. Basically those oil fired plants were scrapped or replaced.
So much of the 1990s-00s productivity boom has been about more efficient distribution and spreading out (eg the ‘big box’ retailers like WalMart which have shown such phenomenal productivity gains) and the lengthened nature of supply chains (think Dell’s global supply chain) that I wonder whether a similar effect might again come into play?
ie the US has moved to an economy which is centred around distance, and fast road and air transport. If that suddenly becomes less economic, does it get harder to have the (labour) productivity miracles?
Other reasons:
1. The ’79 shock peaked at $100 (current) and stayed above $70 for 2-3 years. Heck, the ’74 run-up did not end before the ’79 shock took us to the next level. Getting hit by a baseball hurts more than getting hit by a nerf ball.
2. I have read that the economy overall is a good bit less oil-dependent than in the past. Less oil in the price of goods.
3. The fact that the price run-up was due to demand meant that we were in a economic growth phase. Where were we in those earlier price shocks in the economic cycle? Seems to me that the economy would tend to hold up better, and be easier to control by the Fed, if it was topping when the oil price run-up occurred. Is the recent run-up in oil prices just the typical shortage/inflation that one whould see at the top of an economic cycle that is perhaps a bit over-extended? The other price shocks may have hit us an economy that was more fragile.