August 17, 2005
Talk of recession
Nine out of the ten recessions in the United States since World War II were preceded by a spike in oil prices. Nevertheless, for the past year, I've been telling people that this time it's going to be different-- the economy could weather the rising price of oil without a downturn. Developments of the last couple of weeks make me a little more concerned.
Expenditures on energy are a sufficiently small share of GDP that if the only thing that changes is the price of oil, we really shouldn't expect to see that big a change in total output. The way that previous oil spikes seem to have contributed to broad economic downturns was by helping to precipitate sudden shifts in the pattern of spending by consumers and firms, which demand shifts led to underutilization of labor and capital by the suppliers in those secondary markets.
In my opinion, the reason that the oil price increases of the last two years have not caused a recession yet is that they have built up gradually, and resulted not from a drop in supply but instead from strong global demand. Faced with a gradual price increase and rising incomes, most people have been able to adapt to the higher prices and make adjustments in an orderly way that does not cause serious economic dislocations.
On the other hand, just within the last couple of weeks, I've been hearing a lot more expressions of anxiety and concern-- the sort of psychological factors that produce abrupt spending changes. That is to some degree a purely subjective judgment on my part. Quantitatively, I suppose I could point to the drop in the University of Michigan's index of consumer sentiment from 96.5 in July to 92.7 in August, or the spike in phone calls I've been getting this last week from worried reporters.
Is there a rational basis for new anxiety? The price of gasoline in much of the U.S. has gone up 30 cents a gallon over the last month. For somebody who plans on driving a 15 mpg gas-guzzler 12,000 miles over the next year, that's $240 they thought they had available to spend but now realize they don't. It's more than twice as bad if you want to make the comparison with a point earlier this year rather than just the last month, and of course would only get worse if gas prices keep climbing. That's more than enough to make somebody on a tight budget anxious, particularly if they've already stretched with an adjustable rate mortgage on a house and now see those monthly payments about to shoot up.
In most historical recessions, changes in auto sales have been a big factor in the downturn. That's been avoided this go-round in part because of the big dealer incentive programs. But GM's North American division posted a $1.2 billion loss in the second quarter, so something's got to give. This week Automotive News reports:
Ford wants to cut as much as 30 percent of its 35,000 salaried workers in the North American automotive business. Assembly plants also could be closed. Many of the salaried job cuts will unfold this year, though timing will vary by department.
Airlines are certainly also struggling with higher fuel costs. Delta Airlines remains worried that it will be forced into bankruptcy, a reasonable enough concern for anybody who's lost $10 billion in the last five years.
Third-quarter GDP figures will likely be strong, but concerns about housing and oil have led both the Big Picture and Calculated Risk (writing at Angry Bear) to predict that the U.S. economy could easily go into a recession in 2006.
I am not prepared to join them in making that prediction just yet. For one thing, September gasoline futures fell almost 10 cents a gallon today on NYMEX, though that may be too late to prevent job cuts at Delta and Ford. And furthermore, even though the market seems to expect the Fed to push the funds rate up to 4.25% by early next year, the folks in Washington don't really have to do that, do they? Unless they want to make sure we hit 10 for 11.
Posted by James Hamilton at August 17, 2005 07:15 PMdigg this | reddit
Listed below are links to weblogs that reference Talk of recession:
» Recession Predictor from The Big Picture
There's an interesting dynamic swirling around the impact of Oil and the likelihood (or not) of a possible recession. Some economists are saying that, well, its different this time, and Oil won't precipitate a recession, that Expenditures on energy are... [Read More]
Tracked on August 18, 2005 03:30 AM
» Psychology and Sentiment Starting to Turn South from The Stalwart
James Hamilton:In my opinion, the reason that the oil price increases of the last two years have not caused a recession yet is that they have built up gradually, and resulted not from a drop in supply but instead from [Read More]
Tracked on August 18, 2005 05:05 AM
» Will the Latest Oil Price Shock Lead to a US and Global Recession? from RGE Monitor
With oil prices recently rising above $65 dollars a barrel there is now a growing concern that the latest oil price spike may lead to a U.S. and global economic slowdown and perhaps even a recession. How likely is it that a recession will occur in 2006? [Read More]
Tracked on August 18, 2005 02:13 PM
Tracked on August 22, 2005 02:57 PM
» How badly is the Gulf oil infrastructure damaged? from Pacific Views
And how much of a hit will the US economy take as a result of that damage? Watching CNN last night and reading various press accounts of the hurricane aftermath, we've noticed that little of the coverage has dealt with... [Read More]
Tracked on September 1, 2005 10:43 AM
So some of the above is the kind of stuff that tends to make the jaws drop of those of us with a physical science background when we see economists write it. Let me state how it seems to me, and you economists can respond.
"The way that previous oil spikes seem to have contributed to broad economic downturns was by helping to precipitate sudden shifts in the pattern of spending by consumers and firms, which demand shifts led to underutilization of labor and capital by the suppliers in those secondary markets."
Ok - energy is what make everything in the economy (and the underlying natural ecology) go. Not a vehicle moves, not a machine hums, not a computer boots, not a plant grows, not a sugar molecule gets burned in a muscle cell, unless the system has a source of highly ordered energy which it can process into low ordered energy and sustain its own highly ordered state in the process. The economy *is* a system for processing energy. It takes in high-quality energy, that energy flows through every single part of the system being transformed from one form to another, powering absolutely every activity that occurs, and then it finally gets dumped out as low grade heat into the environment. (Energy is invariably conserved - what we normally refer to as "consuming energy" is really transforming it from a low entropy form to a higher entropy form).
In an oil shock, there's suddenly 3% or 5% or whatever it is less total flow of energy available in a suitable state as input to the system. So the economy is, almost overnight, of absolute physical necessity, 3% or 5% smaller than it was before the shock. Somehow or other, that much less processing of materials and movements of persons, and manufacture of things has to happen. Therefore, there is going to be unused capacity (labor and capital) of necessity - we suddenly don't have enough energy to run the economy quite at the rate we are used to.
The fact that raw energy is a small part of GDP is irrelevant to this - it's still a critical input that is required in every single economic transaction or the transaction cannot occur. And there's nothing, in the very short term, the economic system can do about this reduction - all it gets to do is decide how the reductions get allocated. What has mainly tended to happen in the past is that the money supply doesn't fall in line with the reduction in size of the underlying material economy, and so there's a lot of inflation.
In the longer term, all kinds of adaptations can happen to reduce the energy intensity of the economy so that humans can have more perceived value for the same amount of energy. But none of these can occur overnight, since they require changes to various kinds of durable goods, and so can only be changed on the margins as long-lived durable goods get replaced.
I'm not saying that consumer psychology, and bad policy responses, don't amplify and extend the problem. But surely the core difference between current events and past oil shocks has to be that there's been, as yet, no physical reduction in available energy, just not quite as much available to power growth as the economy would have liked to have used.
Posted by: Stuart Staniford at August 17, 2005 08:50 PM
If the oil price keeps rising, and the resulting OPEC revenues are recycled into Treasuries, can this expansion be lengthened via low rates?
Also, Calculated Risk said:
"Leamer said eight of the last 10 housing corrections since World War II have precipitated recessions, the two exceptions being in the early 1950s and early 1960s, when spending on wars in Korea and Vietnam sustained growth."
*Edward Leamer is director of UCLA's Anderson Forecast.
Martin Barnes of BCA Research said this in a Barron's interview (Aug 1, 2005, p.34-35):
"Down the road, there is a U.S. consumer recession lurking and a big housing downturn. It could be five years down the road. The world does not move to a precise rhythm, but it is interesting how well the four-to-five-year cycle has worked the last few decades. If we continue with that pattern, you would expect another recession in 2009-10."
Any thoughts on the fiscal component to this expansion? Like wartime spending?
Also, if Greenspan manages a "soft landing", can we avoid the reckoning that Barnes anticipates?
Posted by: SoCal Chris (formerly chris) at August 17, 2005 09:26 PM
Nice comments, guys. Gives me lots of ideas for things to write about in later posts. Can't really do much justice to your ideas in a brief paragraph here, except to maybe touch on the Barnes-Greenspan question. I think yes, there's some path for slowing real estate without precipitating a crisis, but it's a pretty tricky business to figure out exactly where that is. Also, I don't find the evidence for a regular cycle of 5-year duration (or any other) is too compelling.
Posted by: JDH at August 17, 2005 09:52 PM
Professor, thanks for the mention. I've been watching for any signs of slackening oil demand and I haven't found anything significant yet.
The threat of political turmoil in several oil producing states remains, but supplies appear solid, so the current $60 spot prices might not hold. For those interested, the DOE provides some great information weekly on price and supplies:
I haven't quite predicted a recession: "My view (not Angry Bear) is a combination of a housing slowdown and high energy prices will probably lead to an economic slowdown next year, and possibly a recession."
Just a slowdown prediction so far ... I'll save the recession prediction for when I have more data.
Posted by: CalculatedRisk at August 18, 2005 12:13 AM
To follow up on the Physics comments, if you believe in periodicy (or the cyclical nature of business ), than a recession in the 2006-07 time frame is hardly a stretch.
Consider the past few contractions, and then fill in the blank: 1990, 1994, 2000, ______.
Its even easier to presume a potnetial contraction when one looks at how stimulus driven this post-recession period has been, and the net results of what happens as that stimulus attenuates.
Posted by: Barry Ritholtz at August 18, 2005 03:39 AM
Don't forget the monetary side of things. M2 went through a major deceleration for most of 2H '04/1H 05 (actually contracted for a brief period), which has only just ended it seems. A monetarist would argue this slowdown bodes well for a recession no matter what happens to the price of oil.
Posted by: Jim Miller at August 18, 2005 04:39 AM
Right now the price of oil mirrors mostly growing demand, not tight supply. The 60+ $ dollar oil is mostly the result of recent economic recovery, so it is quite natural that we have not seen the signs of oil-related recession yet. The supply has been able to grow and the economy with it.
The real problem is in the growth pace of oil production. At 2 - 2.5% it is unsustainable. This means that we either will get a recession that cuts demand enough or the oil price will rise enough to do that and cause a recession.
Posted by: TI at August 18, 2005 05:42 AM
If you are looking for signs of weakness in the economy the oil pump is the wrong place to look. The most present danger of a cataclysmic financial event is in the lowest level of residential finance. For the past half dozen years we have seen a dramatic increase in the use of zero down financing and loans that do not require principal reduction. These vehicles obviously have a higher risk profile than past lending practices, but for the most part we haven't had any significant amount of default yet because the appreciation of real estate has allowed the most marginal borrowers to continue a pattern of refinancing to bail out their otherwise weak financial positions.
With rising interest rates, the refi as a solution for a bad loan is eliminated, with the ultimate result either being a larger number of repossesions or an increase supply of house for sale as these are the exit paths from this delima. With little equity, sales may not be possible with out lender participation, and as losses grow the financial markets (and the government) will demand an improvement in loan quality and financial institution capitalization.
Tightening credit for the most exposed borrowers will farther exacerbate the problem as the potential pool of buyers will shrink thus reducing demand, lowering property prices, and forcing more sellers to default, pushing farther tightening of credit, and lowering home values as lendors attempt to unload their REO.
Ultimately lendor losses will increase the cost of credit for other industries, and thus the availability of adequate funding to finance growth becomes the weak link in our current economy.
Posted by: Bill Ellis at August 18, 2005 06:44 AM
I think Bill makes an excellent point. In automotive retail the captive lenders are using the "fog a mirror standard" for approvals. Those that are declined frequently are capable of executing a puchase anyway through a home equity loan. Any reduction in home values may bring significant distress to the lending community.
Posted by: Rick at August 18, 2005 08:41 AM
"Ultimately lendor losses will increase the cost of credit for other industries, and thus the availability of adequate funding to finance growth becomes the weak link in our current economy."
But ultimately our funding comes from foreigners. How does that effect things?
Posted by: camille roy at August 18, 2005 09:02 AM
Bill Ellis: one caveat to your point is that any decrease in corporate bank lending, one of the major areas that could see terms tighten due to real estate lending losses, would have less of an impact than usual due to the unusually healthy overall state of corporate balance sheets. Those balance sheets could provide another cushion.
Posted by: jult52 at August 18, 2005 10:12 AM
"the reason that the oil price increases have not caused a recession yet... and resulted not from a drop in supply but instead from strong global demand"
Does it matter if the increase in demand is external to the U.S.? If demand in the U.S. stays the same, but a shift in global demand raises the price, how is the impact in the U.S. any different than that of a supply shock?
So why not a greater impact from oil? Oil accounts for a smaller portion of the U.S. economy than it did thirty years ago. Futures markets are well developed (cushioning the impact, spreading it out over time). There are fewer labor contracts tied to the CPI (a weaker wage-price inflation mechanism) -- so less of a need for the Fed to be agressive in raising rates (and eventually choking off growth). Consumers have easy access to credit (consumers begin to scale back only when the credit card bills start to pile up -- it's the debt service burden that matters). All of this suggests a lagged impact. But how much?
Posted by: Scott Brown at August 18, 2005 10:57 AM
Your comments remind me of the boiled frog syndrome: if you put a frog in boiling water, it will jump right out; if you put it in lukewarm water and slowly heat it up, you can boil the frog before it recognizes the danger. True, the increase in oil prices has been slow, giving the economy time to respond. However, the impact on inflation has been muted by the extensive use of hedonism and the practice of emphasizing the CPI excluding food and energy. Do you think this is another reason for economy's equanimity to high oil prices?
Posted by: RayJ at August 18, 2005 10:59 AM
NYTimes will come out with 10K word article on peak oil Sunday by ace investigative reporter Peter Maas who has done good work on the Mob and national security. Maybe drop in oil in the last two days anticipates this article??
Posted by: Jonathan at August 18, 2005 11:49 AM
Interesting theory re the NY Times article, Jonathan, but isn't the saying "buy on the rumor, sell on the news?" This story could be a major introduction to the Peak Oil concept for a big segment of the public. It's well known in the news business that once the Times covers a story that makes it OK for all the other news outlets to talk about it. If the markets anticipated this I'd expect them to be moving up this week as the pros anticipate selling to the flood of newbies who will be stampeding into the oil market after reading the article.
Unless of course Maas is going to pooh-pooh PO?
Posted by: Hal at August 18, 2005 05:39 PM
JDH: That kind of tea leaf reading macro stuff is not my cup of tea. Please talk about something like incentives for R and D, amount of innovation possible, how it is captured effectively (versus free riders), whether market is efficient in taking geology into account, etc.
EVery couple months someone predicts a recession. Sooner or later, they'll be right...
Posted by: TCO at August 18, 2005 07:06 PM
Our economy has grown much more than our energy consumption has.
Economists often say this is evidence that energy is less important to our economic growth.
However, couldn't it -also- mean that each unit of energy is responsible for a much bigger piece of economic pie, and that losing each unit energy now takes out a bigger chunk of the economy?
Posted by: Dan at August 18, 2005 07:55 PM
I'll try to explain why a 5% decrease in energy available won't result in a 5% decrease in the size of the economy. Everyone please feel free to jump down my throat if I bollocks this up. First I'll try to explain why if the price of oil goes up by 5%, I am not 5% poorer. This is because I don't spend 100% of my income on oil. On average in the United States people only spend about $4 on oil per person per day. This is for the total amount of oil used in the country, not just what is used for gasoline. That means that on average a family of four spends about $16 a day on oil, both for their cars and as part of the cost of things they buy. If this family makes a median income of about $43,000 then a 5% increase in the price of oil makes them .0068% poorer.
Now if the total amount of energy we had to use was reduced by five percent, things would be a little more difficult, but we wouldn't automatically have a five percent smaller economy because we don't value all uses of energy equally. An experienced neurosurgeon and a boy scout with a pen knife might use the same amount of energy when operating on your brain, but you probably wouldn't value their services equally. If we had to reduce energy use we would cut it in wasteful marginal areas first. For example, if I had to personally reduce my energy use by five percent I would just adjust my hot water thermostat and drive a little more carefully. I wouldn't consider this as bad as a five percent pay cut. Eventually I would replace my appliances and perhaps car with more efficent models and save the 5% that way.
Posted by: Ronald Brak at August 18, 2005 07:57 PM
I agree I'm oversimplifying, but I think only slightly and I don't think it much matters to my basic argument.
Obviously, since we are talking about oil, the major things that oil is used for are transport and industry/agriculture. See the graph here:
So in an oil shock there will be less transport, and less of whatever agricultural/industrial activity depended on oil. I suggest that it will not actually be wasteful uses that are eliminated, but rather uses by individuals or firms with the least marginal ability to pay more. As far as individuals, that's going to mean poor people. As far as firms, that's going to mean firms who have oil as a large part of their cost structure. Eg, if you do searches in news.google.com for "diesel prices" or "fertilizer prices" you'll see that the recent pain in the US has mainly been going to farmers, truckers, taxi drivers. Oh, and airlines. Regardless, it's all economic activity. If poorer people go to the store less because they can't manage the gas and the cost of the goods at the store, the economy is smaller by their lost purchases. If the trucking industry or the airline industry shrinks, there is less economic activity in very obvious ways (fewer goods are being delivered to customers, fewer sales people are calling on their accounts). If some more marginal farmers go out of business, there is less food being grown.
I think the pattern for firms is that the weakest go out of business, which reduces the capacity of the sector, that allows the rest to pass on their increased costs to consumers and become profitable again. Lowered economic activity and inflation at the same time is known as stagflation.
The neurosurgeon and the boy scout do *not* use the same amount of energy. The neurosurgeon spends his fat fee on a big house (built with materials mined out of the ground and transported long distances with fossile fuels), several large cars (imported from Germany and using lots of gas), etc, etc. The boy scout spends his pocket money on candy.
See "Embodied energy and economic valuation" COSTANZA, R
Science. Vol. 210, pp. 1219-1224. 12 Dec. 1980, for an argument that embodied energy and $$ value roughly track each other (which is what is required to justify my assertion that X% drop in available energy overnight requires X% less economic activity in a GDP/$$ sense).
I do agree with you that eventually the economy adapts by becoming more energy efficient (as very clear following the 1978/1980 shocks). But it can only do so at a certain rate, based on the turnover in capital/consumer durables. In the short term, there's no choice but to simply do less economic activity if there's less energy available.
An issue I don't fully understand is why the resulting inflation is typically much larger than the one time 3% or 5% (give-or-take) that the economy has to shrink (in historical oil shocks). My hypothesis is as follows, but please correct me: when people's lifestyle starts to contract, their first instinct is to maintain it by borrowing more. So the demand for credit increases. However, on an economy-wide basis this cannot succeed, since there's physically less stuff available to be shared around. However, the attempt to borrow more tends to cause the money supply to keep trying to grow even more than normal, when what it needs to do is shrink. Hence inflation goes up a lot for an extended period - people/firms are all trying to out-borrow each other to get at the remaining resources as prices continue to rise (look at airline balance sheets right now for an illustration of the instinct). Thus the money supply keeps bloating, which eventually has to be reined in by viciously high interest rates to get it back in line with the contraction in the material economy. Does that make sense?
Posted by: Stuart Staniford at August 18, 2005 11:30 PM
I agree with Stuart. The oil or gasoline are not expensive in the US even at 60+$ oil barrel. Measured by purchasing power they are cheaper than in the '50s. But it is different elsewhere. Americans can afford to buy oil at this or much higher price and increase their consumption - even some time after the supply starts shrinking.
It is quite possible that we will not see a real "oil crisis" - ie. supply crisis. Rather we would see a deep recession or depression that destroys demand so much that we might in fact get oil oversupply for some time. Lets's face it: how many of you know that the Argentine crisis was caused by the peaking of their oil production and the simultaneous fall of oil prices? Look at the oil production statistics and you will understand.
How many of you see that the Chinese coal is now the real engine of the world economy? The Chinese coal production has been growing at or over 10% a year for years now - it is now at the level of 2 billion tons a year. This is unprecendented. This and the 2+% growth of oil production has kept the world energy supply and economy growing.
It is easy to see that this growth has a real limit. Can the Chinese get to the 3 billion tons level? May be, but with 10% growth rate this will take only 5 years. Can the world oil production reach the 95 million bbld/d level in five years (if we have 2.5% growth, it should)? Possibly but not likely. In a sense the world has this energy euphoria like the Argentinians had before the collapse. Everything seems fine and we will never know what hit us.
Posted by: TI at August 19, 2005 03:18 AM
Boiled frogs cut both ways (to torture the metaphor savagely). The frog gets boiled, but a person who ascends a mountain gradually acclimatises to the thinner air - they don't suffocate slowly.
Posted by: Alex at August 19, 2005 03:26 AM
One question: how many of you think that the first oil crisis in the beginning of '70s was a genuine physical supply crisis?
Yes, in essence it was. Look at the hyperbolic growth of oil consumtion and extrapolate the accelerating growth of 1960 - 1970 to 1980. If you do that you will see that the growth rate was absolutely unsustainable. After the oil crisis the growth never returned to the same path. And no, the oil production did not hit the ceiling then. We already know something about how oil and economy interact.
Posted by: TI at August 19, 2005 03:52 AM
One thing I don't understand about the bullish outlook for the US consumer: if the savings rate is zero, debt burdens are peaking, home equity is cashed out, wages are not rising, interest rates are rising, where does the money come from to drive the bullish scenario for the US consumer (who obviously makes up 2/3 of the US economy) that we hear so much about in what bulls are calling this "Goldilocks" economy?
Am I missing something basic?
Thanks for your help.
Posted by: Roger at August 19, 2005 06:14 AM
"The frog gets boiled, but a person who ascends a mountain gradually acclimatises to the thinner air - they don't suffocate slowly."
Yes they do, if they climb high enough! And this is the point; without a self-limiting feedback mechanism to warn early enough, a system may head blithely for disaster without rrealizing it. In the long-long-run, this applies to population (overshoot) or the economy (exponential growth leading to exhaustion of resources).
Posted by: RayJ at August 19, 2005 09:09 AM
Okay, so 9 out of 10 recessions started with an oil spike. Interesting. I'm wondering if anybody knows how many oil spikes result in recessions? Thanks.
Posted by: chris at August 19, 2005 09:10 AM
Chris, it depends on how you want to define an "oil spike." You might be interested in my paper on this:
Posted by: JDH at August 19, 2005 09:14 AM
Stuart, I don't think I want to go with you to the conclusions about the effects of the oil prices on the money supply. The money supply is ultimately controlled by the Federal Reserve. It's really up to the Fed where that goes.
Posted by: JDH at August 19, 2005 09:18 AM
"One thing I don't understand about the bullish outlook for the US consumer ... where does the money come from to drive the bullish scenario for the US consumer (who obviously makes up 2/3 of the US economy) [if]....
"the savings rate is zero..."
Well, this excludes many forms of wealth gains to households, like gains in the value of retirement accounts, private businesses, investments, etc...
"debt burdens are peaking"
They're not, they are declining. Remember that debt "burden" must be relative to something, such as income or wealth.
Household financial obligations as a % of disposable income have been declining, not rising, for three years:
Note that if you take out a $20K home loan at 5% interest and use half of it to pay off $10k of credit card debt incurring 20% interest (while using the other half to buy toys) you reduce your debt payment burden as a % of your income by a good 50% (even as your total debt increases). The Fed says half of home refinancings have been used to pay down higher-cost consumer debt in this way.
Moreover, even after such borrowing, household net worth (net of all the new debt) has been increasing since 2002.
"home equity is cashed out,"
Hardly -- $10 trillion net positive not cashed out.
"wages are not rising",
Real employee total compensation has been for three years, since the recession ("wages" are only a subset of it).
Posted by: Jim Glass at August 19, 2005 10:08 AM
I'm generally inclined to agree with Stuart that the role of readily-available energy to fuel growth has not been given enough attention by economists in general. We can take a sample of countries, some rich and some poor, take the log of per-capita GDP and the log of per-capita energy use and fit a simple linear model:
It suggests that decreasing energy use by 1.3% (all other things remaining unchanged) should yield a 1% decline in GDP. All things don't have to remain the same, of course. Japan only uses about 60% as much energy per dollar of GDP as the US. They do have some geographical advantages, and given their lack of domestic energy resources, have been more focused on efficiency than we have.
The causal arrow isn't all one-way. Some of the energy that we consume is because we're rich enough to be able to make the choice. But it seems clear that to have a wealthy economy requires a lot of external energy. If your economy consumes energy at the level of a subsistence farmer, you're as rich as... a subsistence farmer. No amount of specialization or human capital is going to let you, say, build a skyscraper at that level of energy use.
Posted by: Michael Cain at August 19, 2005 10:35 AM
I understand the fed has the final say. But what it controls is the cost of borrowing, not the position of the demand curve for credit, which can move for other reasons. It seems to me there's something to be explained in why central banks have to push interest rates *so* high to control inflation after oil shocks. There's interest rate data at
and an inflation rate graph at
In 1980,1981 we had 20,21% interest rates to control 14%,15% inflation, even as the economy was in recession.
There's obviously something inherently inflationary about oil shocks, and a lot more than a one time 3% or 5% recalibration as the physical economy shrinks by that much. I was trying to think of a hypothesis that would explain that. I still like my idea (the race to outborrow one another to get an inadequately supplied quantity seems to describe well what airlines are now doing, and if the whole economy does the same thing, it's got be very inflationary).
Posted by: Stuart Staniford at August 19, 2005 10:46 AM
Stuart, I wanted to thank you for your comments here, especially the first one. It seems right on the (you will pardon the expression) money.
Posted by: Avo at August 19, 2005 11:23 AM
Re oil prices and consumption:
"Look at the hyperbolic growth of oil consumption and extrapolate the accelerating growth of 1960 - 1970..."
That growth rate being virtually exactly the same as the GDP growth rate. Both US oil consumption and real GDP grew 50% 1960 to 1970.
"... to 1980. If you do that you will see that the growth rate was absolutely unsustainable."
Well, if there'd been a competitive market among suppliers with the Saudis producing at their marginal cost (under $10 of today's dollars) as the low-cost supplier to take the market, being willing to put the higher-price suppliers out of business, that growth rate would have been sustainable to 1980. But cartel economics and politics had their say.
"So [today] why not a greater impact from oil? Oil accounts for a smaller portion of the U.S. economy than it did thirty years ago."
Correct -- big changes in oil use since the 1973 "crisis".
From 1972 through last year real GDP increased 162% while petroleum consumption increased only 25%. So GDP growth-to-oil consumption growth went from 1-to-1 to 6.5-to-1.
Or, to look at it a couple of other ways, oil consumption per real dollar of GDP has declined 52% ... oil consumption per person in the US has declined 11% while GDP per per person has increased 87%
With the declining role of oil in the economy it is only logical that oil price rises have less effect than comparable rises in the past. Also, of course, today's price is still about one-third below the historical high in real terms.
Posted by: Jim Glass at August 19, 2005 11:56 AM
The economy is like an upside-down pyramid in which the entire GDP ultimately depends on oil. Your argument of relative unimportance is false.
Posted by: Max at August 19, 2005 12:18 PM
Oil does not have a "declining role in the economy" . Hardly any plane, train, truck, or car moves without oil. No oil = no economy (to a close approximation for the US). What you mean is that the economy uses oil more efficiently (I agree). But a comparable oil shock percentage-wise would be just as bad today as it was in 1973 or 1980. Right now, we don't have an oil shock (supply has been increasing so far), though it is a period of exceptional vulnerability to one, and perhaps will turn into one in coming years.
Again, I reiterate since I sense that at least some people with economic training seem to misconceptualize the situation somewhat. Any movement of any person or goods, great or small, or any transformation of any material from one form to another involves doing "work" in a physics sense, which requires expending energy. Thus the economy is, in the most fundamental sense, a giant energy finding-and-using machine. Energy will *never* be anything less than absolutely critical to its operation. An economic system that cannot locate energy will be without motion (I mean this quite literally - no part of it will move or transform), which I hope we could agree equates to GDP=0.
 With one exception - oil used to be used for electricity generation somewhat and that is now negligible - the market realized it was better to do that with coal, gas, hydro, and nuclear and save oil for transportation.
 I use the common everyday terminology of "using up" energy, rather than the technically correct but ponderous description of transforming energy from more highly ordered (low entropy)forms to less highly ordered (higher entropy) forms.
Posted by: Stuart Staniford at August 19, 2005 12:44 PM
"the savings rate is zero..."
"Well, this excludes many forms of wealth gains to households, like gains in the value of retirement accounts, private businesses, investments, etc..."
- Maybe these sorts of wealth gains may be present for the wealthy, but for the average consumer, gains in these areas are negligible if not negative. Further, most retirement accounts are not spendable, few if any consumers own private businesses, and most have few investment gains these days. Beyond that, wealth gains have precious little to do with the savings rate.
"debt burdens are peaking"
"They're not, they are declining. Remember that debt "burden" must be relative to something, such as income or wealth."
- This ignores the fact that housing prices, by far the biggest source of "wealth" gain among the average American consumer who's seen big wealth gains, may not retrench as interest rates rise and as long term Price/Rent ratios equalize.
"Household financial obligations as a % of disposable income have been declining, not rising, for three years:
- Well, a brief look at the broad picture (http://research.stlouisfed.org/fred2/series/FODSP/97/Max) shows a much different story, doesn't it.
"Note that if you take out a $20K home loan at 5% interest and use half of it to pay off $10k of credit card debt incurring 20% interest (while using the other half to buy toys) you reduce your debt payment burden as a % of your income by a good 50% (even as your total debt increases). The Fed says half of home refinancings have been used to pay down higher-cost consumer debt in this way."
- This would be relevant if credit card debt levels were not also at historically high levels. However as you state it, it's a red herring.
Moreover, even after such borrowing, household net worth (net of all the new debt) has been increasing since 2002.
- Again, largely on the back of record-breaking and likely bubble-caused home prices caused by unsustainbly low interest rates.
"home equity is cashed out,"
Hardly -- $10 trillion net positive not cashed out.
- Another red herring: that doesn't help the millions of people who HAVE cashed out equity against ARMS or negative amortization instruments in a rising interest rate environment.
"wages are not rising",
Real employee total compensation has been for three years, since the recession ("wages" are only a subset of it).
- That may be true, but has again little relevance if the savings rate has hit zero.
In sum, are you simply denying that our savings rate, consumer debt levels and rising interest rate environments are an issue at all?
Rubin, Volcker, Buffett, Krugman, Peterson and any number of other serious economic minds beg to differ.
Posted by: Roger at August 19, 2005 12:52 PM
"An issue I don't fully understand is why the resulting inflation is typically much larger than the one time 3% or 5% (give-or-take) that the economy has to shrink (in historical oil shocks)... There's obviously something inherently inflationary about oil shocks..."
That's not obvious at all -- to the contrary, especially when one looks at the data for the major economies during the oil shocks of the 70s and early 80s.
The US incurred big inflation. Germany got no inflation. Japan before the oil shock had major inflation but eliminated it during the oil shock years.
Yet the US was the most cushioned from the shock of the three by far, since in 1973 it still obtained most if its oil from domestic production while the other two were plumb out of luck on that score. So how did only the US of the three have an inflation problem?
As our host noted, significant inflation is always the result of monetary policy. The US began a too loose "guns and butter" policy during the Vietnam years, the effects of which were masked by Nixon's price controls, and compounded it with a looser money policy that was mistakenly believed would offset the real effect of the oil shock.
Germany has famously stuck to "no inflation" monetary policy since its Weimar hyperinflation. Japan had adopted a monetary policy to reduce its high inflation before the oil shocks hit, and stuck with it.
There's no logic saying the rise in price of any commodity must result in general inflation -- if consumers are forced to pay more for A, then they have less to spend on B to Z, and the price of B to Z will thus fall correspondingly, and offsettingly.
In fact, it is rather obvious that to say something is both "recessionary" and "inflationary" is a contradiction in terms. Recessions are deflationary. If something -- say, the rising price of oil -- puts truckers and airlines and retailers out of business as shoppers stop shopping and employees become unemployed, then prices generally will be heading down instead of up.
Note that the US, for all its loose monetary policy before and during the oil shocks, did not do better in real terms than Germany and Japan with their no- and declining-inflation monetary policies. Rather, in the end it did worse, being all alone in also having to endure the "Volcker shock" of 20% interest rates and 10% inflation to get the double-digit inflation it alone experienced out of its system.
This episode is what finally put the old-style Keynesians pretty much out of business. They'd thought that lowering interest rates would goose the economy in real terms in a way to offset the cost of the oil shock.
M. Friedman to the contrary predicted (always difficult in the real world) this would result only in the economy still getting hit by the real cost of the oil price rise (effectively a tax hike to the extent oil prices were paid out of the country to OPEC) while compounding matters by creating accelerating general price increases, inflation, not seen elsewhere.
The ensuing "stagflation" in the US showed him to be right on the money. And the Keynesians moved on to become NeoKeynesians.
Posted by: Jim Glass at August 19, 2005 12:59 PM
Stuart, Max, and others-- This has produced some very interesting discussion and I hope to get another post up about some of these issues soon. But let me just raise one thought now. Might you not equally well argue that water is the basis for all life, and each of us would perish within days if we had none? As statements of biology, those certainly appear to be correct. Yet it does not follow from those truths that, if the people of the world were forced to get by with 10% less water, we would be decimated (in the literal historical meaning of that term).
As energy supplies shrink, I would expect the fraction of GDP spent on energy to go up, and the cost of going without another Btu would become greater and greater as we have less and less of it. But that is not the question at hand. The question here is, given the amount of energy we're using now, and its current market value, what are the likely short-run economic consequences of a price increase of the magnitude that we've recently observed?
Posted by: JDH at August 19, 2005 01:18 PM
Jim: Your ideology is clouding your judgement. Alternate view:
The Quantity Theory of Money neglects wage increases and tax increases which also affect inflation. Wages are the single most important cost in the economy. Nominal wages shot up in the 70s and contributed to inflation.
As for stagflation, economic theory broke down because the price mechanism cannot send the correct signals today about the availability of resources in the future. This is where the relationship between economic theory and the physical world breaks down. The wealth that is distributed in the markets must be produced in the real world where processes must obey the laws of physics. But production models of economics arenít based on the laws of physics. As Stuart goes to pains to point out, nothing happens in the real world without energy conversion and entropy (waste) production. But the basic neoclassical model is a perpetual motion machine with no inputs or limits. The only reason that decreasing concentrations and qualities of resources were not translated into higher prices for constant quality for so long was because of the decreasing price of energy and its increasing use in the exploitation of increasingly lower grade reserves. Remove the cheap energy and the price of all productive processes goes up. Thatís about to happen or may already be happening. Growth is beginning to cost more than it is worth at the margin. Classical theories do not work well in regions close to limits.
Posted by: Max at August 19, 2005 01:20 PM
"Jim: Your ideology is clouding your judgement..."
Many things cloud my judgment, but my ideology did not cause inflation to decline in Japan and remain stable in Germany while the oil crisis was supposed to be inherently driving it up.
"As for stagflation, economic theory broke down because..."
Whose theory? ;-)
"In sum, are you simply denying that our savings rate, consumer debt levels and rising interest rate environments are an issue at all?"
Not at all. Where would you get that?
I've posted in comments right here that the Fed raising rates is a concern IMHO, as our host says, and at many sites (including my own) that I'd much rather see a far higher savings rate, among many other good things.
OTOH, are you saying that the decline in household obligations as a % of income over the last three years, increasing real employee compensation, increasing net household wealth, the large items of value that are omitted from the savings rates, the fact that interest rates continue at a historic low level even if rising on the short end, are all to be totally disregarded?
Look, you asked what I took as an honest question and I tried to give an honest, empirical, useful answer. Perhaps I was mistaken and you were only being rhetorical.
But a classic error -- often subterfuge -- of rhetoric is the fallacy of the excluded middle. The fact that many things could be better does automatically not mean things on the whole are so bad. E.g, If you look at the data at the same length of time after the '91 recession as today is after the '01 recession, you'll find everything then was about the same or worse than today -- and that was leading right into the "miracle economy".
So don't over-invest in pessimism unless you really want to. The future is always uncertain. There's always a risk of good things happening.
Posted by: Jim Glass at August 19, 2005 02:51 PM
"Oil does not have a 'declining role in the economy'. Hardly any plane, train, truck, or car moves without oil. No oil = no economy"... a comparable oil shock percentage-wise would be just as bad today as it was in 1973 or 1980.
If you want to argue that with oil use per $ of GDP down 50%, and increase of oil use per unit of GDP growth down 85%, a price rise of $X or supply interruption of X barrels will have no smaller effect on GDP and the economy as a whole today than before, be my guest. Others can form their own opinions on that.
But the idea that no car, plane or truck would run if there is no oil is classic false dilemma.
There is no issue of being physically unable to obtain oil. Every single person and business in the US can always get more oil if willing to pay the price for it: $60, $80, $120, whatever. The issue is price allocation and how it increases efficiency and productivity (e.g. reducing oil use per $ of GDP by 50%).
And there's no reason why price allocation should cause any disruption to other parts of the economy as you've described in earlier comments, except in the very shortest run. If it's too costly for airlines to fly four flights with 60% loads they can fly three with 80%, just as many people fly. Or an inefficient legacy airline can go broke so Southwest ends up with its planes, and even more people fly. Why would people buy less at the store because the cost of driving there goes up? Why not just cut out the trips to buy just a few things, buy more on average, or car pool? Then buy a more fuel efficient car next time around?
Your example of the marginal farmer going broke with the farm sector producing less as a result is right on point. The marginal farmer has been going broke for 150 years as farmers dropped from >50% of the population to
Those marginal farmers weren't driven out of business by low crop prices, the high cost of fertilizer, or usurious interest rates -- though most of them certainly saw it that way. It was their relatively low productivity that culled them from the economy -- making farms hugely productive, food cheap and Americans fat as a result.
Increasing productivity is what has gotten us out of the muck in front of our caves and given us the Internet world -- enabling us the get more welfare from fewer inputs. The marginal producer goes broke, yes, everyone sees that -- then his assets are shifted to use by the more productive. Why do so few see this follow-up? When it's what's created our 75-year life expectancies.
I'm always surprised by how many "physical oil" types totally disregard productivity increases. After all, what's the oil industry's productivity increase from 1880s Pennsylvania to deep sea drilling? Almost as great as that of the aeronautic engineers from Kitty Hawk to the Moon in 60-odd years.
Why would productivity of both production and use of oil stop now? If they both progress at a mere 2% a year, and so does that of use of alternative energy sources, and you compound that, then like Sheik Yamani said, the oil age won't end for lack of oil any more than the stone age ended for lack of stone.
Posted by: Jim Glass at August 19, 2005 03:05 PM
Re the Times article on peak oil: Oil had been going up fairly steadily for some time, til midweek last week. SOMEBODY had it right and was making money on their long futures position up to 17 points, or $17,000 per contract if they caught the move precisely. If you are a short term trader - especially in a leveraged vehicle like a futures contract -- the longer a trend goes in your favor, the more pressure you feel psychologically to nail at least some profits down. That is why the Times article could be an excuse to sell on the news.(This is a matter of behavioral finance to be sure, and not what microtheory would suggest).
Furthermore it has been my experience (and James Cramer notes this is as well in his new book) that when a big market move gets reported on the front page or in the Sunday magazine where everyone sees it (rather than tucked away in the business section) that is a classic tell for the move to reverse or at least get dampened.
For Stu: you asked if the Russians would get the same amount of revenue if the they withheld 4.5 million barrels of output. Of course JDH is right that the answer depends on the elasticity of demand. Now, we can find out easily enough, what elasticity of demand would have to be in force for this to happen:
For the Russians' revenue to remain constant when the Russian supply were cut in half, the world price would have to double. So if we apply the formula for price elasticity of demand and use round numbers, the world price is $60, the world consumption is 80 mm/bd : then dP=60 (price must move an additional 60 to double), and dQ=4.5, as we cut supply in Stu's query by 4.5 mm/bd. so the elasticity required for this to happen (assuming no replacement supply) is:100*(dP/P)/(dQ/Q)=100*(60/60)/(4.5/80)=100*1/(4.5/80)=100*(80/4.5) = 100*17.8 = 17,800%. I make no guess as to how reasonable or not that might be.
Posted by: Jonathan at August 19, 2005 03:44 PM
misplaced the decimal, that should be 1,780%
Posted by: Jonathan at August 19, 2005 03:47 PM
Excellent question! I had to stop, make some tea, and think hard before responding. My first instinct was to throw a lot of thermodynamics around as a way of blowing smoke, but let me try to formulate things more intuitively to get at the distinction.
Let us first think about under what circumstances it *would* be true that a 10% reduction in available water flow would result in a 10% reduction in the economy. Let us imagine a primitive agricultural economy which has been operating for quite some time in a hot dry climate, and then experiences a sudden climatic fluctuation which causes 10% less rainfall in the catchment area available to that economy (let's suppose, implausibly but for the sake of the discussion, that the annual rainfall had hitherto been very steady and predictable).
In this economy, economic activity is limited by available water. There is land available which is uncultivated, because there is not enough water to irrigate the plants and give the people and draft animals required for cultivation enough to drink. In general, the total amount of biomass produced each year is going to be pretty much proportional to the water supply, so more food cannot be grown. More economic activity of other kinds cannot occur because working longer in the sun, whether for beast or man, would require more water. Thus the 10% constriction in water will result in 10% less economic activity (and after a while, assuming the population has reached its carrying capacity, in 10% less humanity, given the same social structure and available technology).
However, our economy is different. It is true that, artificial intelligence and robotics not having progressed as some had hoped, all economic activity is dependent on humans, and therefore zero water = zero economic activity. However, it is not the case that any given economic activity is proportional to the amount of water used in it. If we wish to truck around 10% more goods, the truckers, shaded in their air conditioned cabs, do not need 10% more water. Water is only peripherally required by the truck itself at all (some other coolant would work just fine and anyway we only very rarely have to put more water in the truck). Similarly, if we wish to put on an extra shift in the factory, the workers, checking on the CNC controlled machines, do not need more water to be in the factory than at home, and the machines use none. Ditto for the software company that developed the software that controls the machines.
In short, there is no proportionality between water use and economic activity for most economic activities in the Western world.
Nor are we, in most parts of the world, limited by the water supply even for those parts of the economy that do need water (say farming). If we want more, we suck it out of the aquifer or the river faster. 
Now, energy is different. This is true for very fundamental reasons of physics, but let us try to think about the everyday examples above. If we wish to truck 10% more goods, we are going to have to use 10% more diesel in the trucks. If the factory wants to put on 10% more hours, it's going to need to buy 10% more electricity. If we want 10% more steel, the miners are going to need to use 10% more dynamite (chemical energy) to blast open 10% more rock, scoop 10% more times with their mechanical shovels (diesel again), lift the extra rock to the surface with 10% more electricity in the mine elevators, use 10% more energy to smelt this extra steel, and then hand it over to the 10%-more-diesel-using truckers. The farmer producing more food by cultivating a larger area of soil is going to use extra electricity to pump the extra water, extra diesel for the extra tractor miles, extra natural gas or oil in the form of fertilizer, etc.
Even things like software programs or movies, if we want more of them, we will have to have more square feet of office with lighting, computers, and air conditioning consuming electricity, and highly paid technical professionals spending their resources on large energy consuming and embodying houses and cars.
Even for me to think harder to answer this question requires more glucose and oxygen to the brain - more energy use. Even purely mental activity requires energy, and the more of it, the more energy.
In short, there *is* a direct proportionality between the amount of any economic activity I can think of, and the amount of energy required in the doing . If you can think of an economic activity where the amount of it produced and the energy used in producing it are not proportional, I'd be interested.
Now, I am not saying that energy use (at least since the industrial revolution till now) has generally been energy supply constrained, in the way that the desert dwellers are water constrained. Our economy could have chosen to exploit more energy faster than it did. But with any given supply setup, there is somewhat limited instantaneous extra supply (extremely limited right now). Oil shocks exceed this buffer, and so society suddenly is energy constrained for the types of activity it is in the middle of doing, with the technology it has. So, somewhere inevitably, some factories must work fewer hours, or some machines must be shut off, or some mines must produce less, or some trucks must deliver less. Probably some of all of the above - the weakest economic actors will go to the wall first.
An additional difference between water and energy is as follows. A robot economy might well be entirely free of dependence on water - water is a requirement for biological creatures like humans, but perhaps the robots would require only metal, plastic, electricity etc. However, no conceivable economy could not have energy proportionality of this kind. No intelligent agent can cause any material or system to move in some direction or transform in some fashion that it wouldn't otherwise have done without the use of energy.
 There are localities where this is not true, and that may spread in the future, but it's broadly true in the Western world at present.
 Absent changes of technology, obviously, which can enter the picture at longer timescales, as I think we all agree.
 Caveat: we could get more complex here and distinguish the operating energy required for each marginal unit of production, and the amortized embodied energy in the capital goods required for production (the machine, trucks etc). The per-unit amount of the latter obviously varies depending on total production for the lifetime of the capital good in question. But this actually work against us in an oil shock.
Posted by: Stuart Staniford at August 19, 2005 04:25 PM
You write: "Why would productivity of both production and use of oil stop now? If they both progress at a mere 2% a year,..."
That is the $200/barrel question. "Physical oil types" posit that productivity of oil production will slow and indeed fall as we are forced to produce oil from difficult sources ( deepwater, oil sands, etc.) which require much higher investment and energy input.
So, if the real cost of producing oil increases at a mere 2% a year, what do you think will happen?
Posted by: RayJ at August 19, 2005 04:32 PM
Your points about Germany and Japan are interesting. I could respond better if I had good data for German and Japanese inflation, interest rates, etc in the 70s and 80s. I can't quickly find them via Google. Can you point to them?
In general, I'm not asserting monetary policy is irrelevant. I can well imagine that a prompt fierce interest rate hike in response to an oil shock will constrain inflation much quicker than not taking that policy step. My point is that it is going to take higher real interest rates to contain inflation right after an oil shock because the physical economy has just contracted noticeably of necessity.
Posted by: Stuart Staniford at August 19, 2005 04:35 PM
It's worth reading the work of Robert Ayres if you want to further examine the relationship between energy (or specifically 'exergy') and economic growth.
"Exergy, power and work in the US economy, 1900-1998"
"Accounting for growth: The role of physical work"
Posted by: FTX at August 19, 2005 05:00 PM
"Why would productivity of both production and use of oil stop now? If they both progress at a mere 2% a year, and so does that of use of alternative energy sources, and you compound that, then like Sheik Yamani said, the oil age won't end for lack of oil any more than the stone age ended for lack of stone."
It's a humorous proverb but of course ridiculous on its face. The literal reason we progressed beyond the stone age is that we found something better. None of the proposed energy alternatives to oil can rival oil, particularly for transportation which is where it is used most. Add to that the problem of replacing the "stuff" oil is made of such as petrochemicals, plastics and pesticides. The transition to the next energy source may well be a transition to a lower quality one with all the negative economic implications that implies.
Posted by: Max at August 19, 2005 05:06 PM
Jonathan, the question about elasticity and Russia actually came up under the China thread, but I'll go ahead and address it here where you bring it up again. I stand by my claim there that the answer to whether Russia's revenue goes up or down depends on the ratio of k (Russia's share of total world production) to e (the elasticity of global demand. The reason why your numbers seem to be giving a different answer is that you are talking about a very large change, for which the exact answer from calculus which I gave is not as good an approximation as for a small change.
To look at your example for a small change, suppose that Russia cuts production by 0.001 mbd. To earn the same revenue, the price would need to go up to 60.0066674 (60.0066674 x 8.999 = 60 x 9). Then (0.0066674/60)(80/0.001) = 8.8899. The elasticity is actually the reciprocal of this magnitude, or 0.11249. Russia's share of world output for your example is 9/80 = 0.1125.
Posted by: JDH at August 19, 2005 05:16 PM
A substantial portion of the energy we expend comes from solar radiation via photosynthesis and the food chain. It isn't all from the petrochemical route.
Maybe with another cup of tea and some further glycolysis you could work out the economic implications of all of the raw inputs to the economy. ;)
Posted by: STS at August 19, 2005 05:30 PM
Stuart, what did you think about the examples that Jim Glass gave, of flying fewer planes closer to capacity, car-pooling, and fewer trips to the store? Why can't I teleconference or phone instead of flying or driving at all?
Posted by: JDH at August 19, 2005 06:04 PM
Jim (Glass) - I'm really enjoying your comments. Your analyses seem very clear and well backed up by facts. They shed a lot of light on these long term trends. Keep 'em coming!
Posted by: Hal at August 19, 2005 06:28 PM
Thank you for your comments on elasticity, I thought they were great. RE teleconferencing vs. travel: I've been thinking that as real energy prices go up, there will be an incentive to replace transport with communication. But when we need groceries, for example (or any other item we shop for), we will need to go and get them or pay to have them shipped to us. Even in the realm of science fiction, if Scottie could beam groceries up to us, that would presumably entail energy expenditure too.
And though travel may become more of a luxury, face to face contact yields intangible extras in human communication and experience that electronics can't yet convey with perfect fidelity, whether the relationship is business or personal. The phone is still only the next best thing to being there.
Nor will television ever put you in the same experience as visiting the Taj Mahal or hiking up Mount Whitney. For such experiences, you really do have to be there.
Posted by: Jonathan at August 19, 2005 07:15 PM
So these are purely behavioral ways to do more with less energy. My gut reaction is, yes this will happen, but it's a second order effect. (I have no useful quantitative data, only gut reaction).
I think the first question to be asked is why weren't they already being taken? If we understand that, we're in a better position to assess whether they are likely to be taken now in response to this oil shock. (In a free market context, I'm assuming - obviously the issues with government policy mandates such as rationing would be quite different).
I think there's two general classes of situation.
One is, I wasn't already doing this economization in energy use, because this energy use (let's say gas to go to the store) was a very small deal to me. After there's less gas, and it doubles in price, I don't care, I'm still going to the store as often as I feel like because it's convenient and I can afford it. Gas is going to have to get a lot more expensive before I'm willing to conserve. Someone else is going to have to do the conserving, not me. Since a lot of the largest energy consumers are wealthy and respond inelastically to changes in price, the burden of conserving is going to fall on a class of users who don't have a choice. So that class of consumers are going to have to conserve more than 5%. For them, gas is a non-trivial expense, they already weren't wasting it any more than they could help, and now it's doubled in price, that is going to directly come off of how much they can spend at Walmart (ie, reduced economic activity).
The other case is exemplified by the airline fullness example. Fuel has been a pretty significant cost for airlines for a long time, as has the capital cost of owning airplanes. They had excellent reasons to fly their planes as full as possible all along. Presumably, to the degree they had spare capacity on flights, it was because they felt that a fewer-fuller-plane schedule would cause inconvenience to their customers, such that fewer customers might fly, and that this drop in revenue would be enough to offset the reduced expenses and cut their profit.
Let's assume they know their business, so now when they adopt the new schedule in response to the oil shock, indeed they have less customers. This is the new right equilibrium point for them, but there is less flying going on in consequence.
As to flying versus teleconference, we get into fuzzy factors, but nonetheless important ones. The accumulated experience of the worlds sales forces is that they are more effective if they work face to face than over the phone, and that's why they fly such a lot. Given they are on commission, they have excellent reasons to figure out the correct balance between flying and phoning. If they only phone, customers will be less persuaded and take longer to make up their minds. Again let's assume the sales people have this about right. Enter the oil shock, flying becomes a bigger hassle and more expensive, the travel department clamps down on it, the equilibrium shifts and the sales folks do less of it. Less flying on the part of sales people will, not all at once, result in fewer deals closed and less economic activity. The arguments for the productivity of economics professors are even more indirect, but presumably similar :-)
I think this is an economist type argument right? If that was really $100 there on the pavement, it would already be picked up. Those conservation measures must have costs that meant they weren't already being done. In most cases, pretty unquantifiable, but presumably on aggregate, the costs of doing the conservation measures must have been equivalent to the price of the gas that would have been saved (at the pre-shock prices). Now that we have done them, we are going to pay those costs, and all those effects go in the direction of reducing physical economic activity.
I sense I still need one or two more pieces to make this argument solid, but my brain needs some more efficient technology first. :-)
BTW, the Ayres' et al papers that FTX posted do seem extremely relevant to this discussion, though I don't have all the necessary economics prerequisites to understand them in full. At any rate they only make one mistake in their thermodynamics that I saw and it is not material to their argument. They argue that increasing energy use (strictly, delivered useful energy post conversion losses) is the major factor explaining US GDP growth - something that I think seems obvious to a physicist or an ecologist, but obviously not to many economists.
Posted by: Stuart Staniford at August 19, 2005 07:22 PM
"I reiterate since I sense that at least some people with economic training seem to misconceptualize the situation somewhat. Any movement of any person or goods, great or small, or any transformation of any material from one form to another involves doing 'work' in a physics sense, which requires expending energy.
"Thus the economy is, in the most fundamental sense, a giant energy finding-and-using machine. Energy will *never* be anything less than absolutely critical to its operation.
"An economic system that cannot locate energy will be without motion (I mean this quite literally - no part of it will move or transform), which I hope we could agree equates to GDP=0."
Well, yes, sure. Similarly, the NY Times science section once started a story on green economics (5/20/97):
"Virtually everyone agrees that without the natural world, the human economy, and indeed human life, could not exist."
Gotta love that "virtually" -- they always have to allow for the possibility of dissenting opinion. ;-)
But in fact the natural world exists, so we move on to more interesting questions.
By the same token, while economic types all know full well that when specifying available energy as = 0 then GDP =$0, and in fact human life = 0, they also realize that in reality available energy is and always will remain far, far > 0. They further realize that accessibility to energy is constantly increasing at diminishing cost over time (sunlight, firewood, coal, whale oil, petroleum, nuclear, photovoltaic, etc ... ??) and the use of energy also increases in efficiency so that less accomplishes more in response to cost.
Thus, the interesting questions involve the most efficient development of energy resources, increasing their productivity of use, and maximizing productivity of allocation.
"What if there was no energy?" is just a weak form of "what if there was no natural world?". Not very enlightening.
Now if someone wants to make a case that the amount of total energy available for use by humanity will decline -- move *towards* zero -- over the next 100 years, reversing its course for the first time in history, that would be interesting.
Posted by: Jim Glass at August 19, 2005 07:52 PM
"Your example of the marginal farmer going broke with the farm sector producing less as a result is right on point. The marginal farmer has been going broke for 150 years as farmers dropped from >50% of the population to
You are conflating long term effects (where I agree efficiency improvements are important to the discussion), and short term, where they are very much harder to do. The discussion at hand is of oil shocks, which by definition are very sudden onset reductions in supply.
In any case, farmers are *not* producing more for the same amount of energy, they are using vastly *more* energy, than farmers ever used to. The EROEI in farming used to be positive, now it is hugely negative. We can afford that because we have had such high EROEI fossil fuels, but that's the only reason.
I get the feeling you are not understanding my point at all. I don't know how to explain it in a clearer way.
Posted by: Stuart Staniford at August 19, 2005 07:53 PM
"Jim Glass, You write: 'Why would productivity of both production and use of oil stop now? If they both progress at a mere 2% a year,...'
"That is the $200/barrel question. 'Physical oil types' posit that productivity of oil production will slow and indeed fall as we are forced to produce oil from difficult sources ( deepwater, oil sands, etc.) which require much higher investment and energy input.
"So, if the real cost of producing oil increases at a mere 2% a year, what do you think will happen?"
First I'd ask the interesting question of why increasing producer productivity would fail to keep up with the increasingly difficult challenges of production only starting just now, today -- when the industry has been "forced to produce oil from difficult sources" from back when it very first began drilling wells for oil, considered an extraordinary undertaking at the time.
E.g. how costly is today's deep-ocean drilling in terms of the tech of, say, 1880? Well, they couldn't imagine it in 1880, so that's one very nice productivity gain.
It's easy to say, 'difficulty of production will go up so cost will go up'. But when one notes that production productivity gains have in fact at least offset those costs increases for well over 100 years, then one has to have a reason as to *why* this would change *now*, as opposed to in 1940 or 2060. Otherwise one is just repeating over and over a prediction that's always been wrong until the day it comes true, someday, maybe, or not. Which is not good foundation for policy decisions.
Second, I'd try to find some empirical baseline estimates for real cases. For example, take the Athabasca Oil Sands, 1.6 trillion barrels, several times what Saudi Arabia has. Of that, 177 billion is "proven reserves" at the $25 prices of a few years ago, second only to the Saudis. Will it become more difficult to produce beyond that? Sure. How much might reasoably be guessed to be recoverable?
Well, if producer productivity increases 2% annually (which is quite modest) in 30 years we have an 80% productivity gain (what would be economic to produce at $45 today will be economic at $25 then). If real prices increase on top of that, things compound. If real price is $60 in 2030 then it will be able to produce as much as at $108 today.
If Athabasca's proven reserves are 177 billion at $25, what would they be at $108? One heck of a lot more, I'd guess. (I note for the record that at this year's $50+ prices the industry says they are > 300 billion).
A heck of a lot *less* than 1.6 trillion, to be sure, so there is indeed a limit, but a real whole lot more than today. Hundreds of billions is a lot of oil. Saudi scale. And $60, 30 years from now, 2% production productivity gains, are all pretty modest numbers.
Posted by: Jim Glass at August 19, 2005 08:01 PM
"None of the proposed energy alternatives to oil can rival oil, particularly for transportation "
That's exactly what the whale oil people were saying back when it was foreseen as running out and petroleum was still the "farmer's curse", an unburnable pollutant that destroyed the value of farmland when it seeped to the surface of some unlucky Pennsylvanian's property. I could quote them. Then, oops, that Canadian figured out kerosene. It really seems rather conceited to think "human history ends now, we know it all".
"Oil products" have substitutes in all directions. First, they don't come out of the ground, they are manufactured in plants. The manufacturing feed stock can be light oil, heavy oil, gas, coal, tar sands, plants, whatever, even garbage. This stuff exists in vast amounts. Germany ran its WWII economy on oil from coal (using 1930s tech), and South Africa did too, and both did rather well, neither did anything like collapse from "no oil", so this is not theoretical, it is real. The preference for light oil is *relative price*. But the cost of all the others drops constantly in real, absolute terms. If Saudi light and Texas tea never existed we'd be manufacturing the same products from other feedstocks at declining cost all the time.
Second, oil itself can be replaced in many cases. Not a few people prefer electric and hydrogen vehicles right now. The necessity behind them on a large scale is nuclear power, that's all. China is going ahead with pebble bed reactors today.
In 100 years what will be the best proven energy options? Who can say? If you can, you you must think Jevons and all those back in later 19th Century Britain were pretty dim to have been as concerned about "Peak Coal" as they were. Why didn't they see internal combustion and nuclear? And all that coal still left in the ground today not worth digging up?
"The Coal Question: An Inquiry Concerning the Progress of the Nation, and the Probable Exhaustion of Our Coal-Mines"
"Day by day it becomes more evident that the Coal we happily possess in excellent quality and abundance is the mainspring of modern material civilization ... It is the material energy of the country ó the universal aid ó the factor in everything we do. With coal almost any feat is possible or easy; without it we are thrown back into the laborious poverty of early times.
"With such facts familiarly before us, it can be no matter of surprise that year by year we make larger draughts upon a material of such myriad qualities ó of such miraculous powers.
"But it is at the same time impossible that men of foresight should not turn to compare with some anxiety the masses yearly drawn with the quantities known or supposed to lie within these islands..."
Reads pretty familiar to me. But as I'm getting pretty far from the original topic of current recession risk, I'll bail out of the conversation now.
Posted by: Jim Glass at August 19, 2005 08:33 PM
Ah, with recession the topic I bail back in!
"Jim: You are conflating long term effects (where I agree efficiency improvements are important to the discussion), and short term, where they are very much harder to do. The discussion at hand is of oil shocks, which by definition are very sudden onset reductions in supply."
Well, I noted my impression that with oil use per $ of GDP down 50% since the first oil shocks, the effect of a shock on GDP would be proportionately less today. Feel free to differ.
And I don't see any risk of sudden actual reduction of supply -- outside of war or meteor hit or some other non-economic cause -- only of future price rises (and falls!) by the normal mechanism.
Also, long-run effects are much more important than short-term ones, in the long run. ;-)
"I get the feeling you are not understanding my point at all. I don't know how to explain it in a clearer way."
Well, maybe not, but blog comments are still a good cut above usenet -- here, anyway.
You'll have to explain to me someday why EROEI does not appear in prices, and how it can be expected to affect transactions and economic behavior as long as it doesn't. But another time, that's a long way from the thread topic too.
Posted by: Jim Glass at August 19, 2005 08:55 PM
I'd just like to suggest a mental exercise that some people might find helpful in getting their minds around the effects of oil shortages. What would your country do if all foreign sources of oil were cut off? I find that trying to imagine what will happen if the world starts to run out of oil is very difficult, but thinking about your own country is much easier. Do you think civilization would collapse around you without outside oil, or would you and your fellow citizens be able to adapt? I've considerd the problem with the two countries I'm most familiar with, Australia and Japan, and decided that people in both would survive okay after a difficult period of adjustment.
Australia wouldn't have a huge problem. They produce most of the oil they consume and have large supplies of gas, coal, oil sands, sunlight, etc.
Japan would be much worse off, but they would still be able to adapt. Japan does have a large strategic oil reserve and some very marginal oil fields that could be exploited to ease the pain as their economy adapts. Japan already produces most hybrid cars in the world and would find it easy to start producing all electric cars that could be powered by an expansion of their nuclear power industry (or other means). Japan also has a lot of scope to save energy through improved housing stock. Many houses in Japan feel as drafty as packing crates.
I haven't really thought about what would happen to the United States, but I guess its situation would be somewhere between Australia's and Japan's. Although the U.S. might not have the social capital that Japan can utilize when dealing with disasters, Americans are often considered to be very creative and inventive, which are traits that could be very useful in overcoming the challenges they would face.
Posted by: Ronald Brak at August 20, 2005 02:23 AM
"Well, I noted my impression that with oil use per $ of GDP down 50% since the first oil shocks, the effect of a shock on GDP would be proportionately less today. Feel free to differ."
It is the correlation, not the oil use per $ of GDP. Logically, the situation is worse now than before: a smaller reduction of the oil (ie. in barrells) will cause a larger change in the GDP!
The growth of oil consumption in 1950 - 1970: the physical supply constraint applies only in world scale. The world cannot import oil and you must look the global numbers.
And you gave yourself en example to prove you are wrong: "And there's no reason why price allocation should cause any disruption to other parts of the economy as you've described in earlier comments, except in the very shortest run. If it's too costly for airlines to fly four flights with 60% loads they can fly three with 80%, just as many people fly." Well, three planes for four - one crew less, fewer planes ordered... Yes, there you have the disruptions in other parts of the economy.
And the oil depletion: yes it really happens. The Peak Oil happened in the US in 1970. It is irreversible. No matter what you say about the increasing productivity of the oil production. Increased productivity can not bring more oil (barrels). It has happened in all Europe. Also the European coal has depleted. The depletion of fossile fuels is a fact. So we don't speak only about a temprorary oil shock but about a long term phenomenon.
But this is an economics blog. So try to remember there are also investments and industry. The oil and energy prices don't affect only the ordinary consumer at the pump. The price of all energy is rising: oil, natural gas, coal, uranium. All this affects the long term investment decisions. This is also one reason why it is so difficult to see the effects of energy availability and pricing directly.
And for all those alternatives. We must understand the numbers. Make some oil from grabage. Fine. Make 1 million barrels/day oil from garbage. How much garbage is needed? I don't know but 1 barrel of synthetic oil is 0.4 ton of coal. 1 million barrels is 400,000 tons. 1 million bbl/day means 146 million tons of coal in a year. This is 5% of the US oil consumption.
Posted by: TI at August 20, 2005 08:44 AM
So, TI, are you in the "civilization is over" school, then?
Posted by: Mandos at August 20, 2005 08:46 AM
Jim Glass writes: "E.g. how costly is today's deep-ocean drilling in terms of the tech of, say, 1880?"
The comparison of productivity should not be between current technology and technology at the start of the oil age. Instead one should compare productivity now with productivity at the time it was cheapest to find/develop/extract oil, i.e. the time, say in the 30's when most oil was produced from gushers. I have not been able to find the data on-line but I am sure if you graph productivity in the oil business you will find it to be in the shape of an inverted bathtub; rapid increases in productivity at the beginning, to more slowly rising plateau and then a slow falling off as more expensive technolgy has to be deployed for harder to find/develop/extract oil. This fall off in productivity is probably not a recent thing.
Note that when I talk about productivity, I am looking at the global productivity of oil production. Even though each technology class is improving, each step up in technology to handle difficult to get oil reduces overall productivity which is why higher oil prices are required to justify those projects. For example, deepsea technology has been improving since the North Sea, but it is still much more expansive that onshore technology. Similarly, extracting oil from tar sands is very expensive compare to conventional onshore and even deepsea technology.
You mention the Athabasca project. Well, Shell just announced a huge increase in the project development budget for an additional 300,000 bopd. The original project cost $4.7b for 155,000 bopd compared to BP/Exxon's Thunder Horse deepwater platform which cost $5b for 250,000 bopd + 5.7 mcm/d of gas.
With tar sands and many other unconventional sources, the isssue is not the total recoverable reserves but how fast it can be extracted. Athabasca reserves may be similar to the size of Saudi Arabia's, but can Athabasca produce 10 million bpd? Also, not only are the upfront developments costs huge, but the ongoing cost of extraction will be high because of the high energy input required (estimates for the EROEI of tar sands range from 3-6 compared to about 30 for Middle Oil).
Finally, a case in point of resource depletion is North American Natural gas. When prices spiked in 1999, a drilling boom ensued and the number of wells completed doubled from 10k to 20k. The resulting amount of gas found was only enough to keep NA gas production flat by offsetting depletion. In 2001 CERA reported that there were enough gas reserves for the forseeable future. A month ago, CEO Raymond of Exxon announced that NA gas production has peaked. Only LNG imports will allow us to meet current gas demand at reasonable prices.
Hopefully the above answers your question about why "peak oilers" are concerned. I'll rephrase my original question; what would happen if the productivity of oil production decreases by 2% a year?
Posted by: RayJ at August 20, 2005 12:33 PM
"You'll have to explain to me someday why EROEI does not appear in prices, and how it can be expected to affect transactions and economic behavior as long as it doesn't."
Well, it will increasingly in due course, and has already.
Back in the Spindletop days, it's been estimated that oil EROEI was around 50. What this means is that only 2% of the energy you find needs to be reinvested to keep the supply flowing at a steady rate in the future. The other 98% can be used for fun stuff like making big cars and houses, and to subsidize energy intensive farming practices. This also means that it's completely trivial to go out and get more oil any time you need it. Also, oil was extremely cheap back then, because it flowed out of the ground in ample quantities with little effort, and it became an increasingly tiny proportion of GDP precisely because so little GDP needed to used for going and getting more. This is not because oil is unimportant to the economy, it's just so profitable to get it. With tons of super-cheap energy to throw around, labor productivity goes through the roof because we can power machines to do things instead of just using people.
Nowaday, EROEI in US oil is a little better than 10. Meaning a bit less than 10% of the energy found has to be reinvested to maintain the supply. Oil is significantly more expensive than it used to be (even before the recent price spike).
When (or if) we're all on oil sands, let's say with EROEI of 3, the oil industry is going to consume about 30% of it's own output to maintain supply. The industry is going to be a massively greater proportion of GDP, and prices are going to be much higher. The available energy for making big houses and big cars is going to be harder to come by.
In short: the reason that oil is now a small proportion of GDP is not because energy is not that important to the economy, it's because it's been incredibly energy profitable to extract oil, so not that much GDP needs to be reinvested to maintain supply That has been changing, and will change for the much worse in the future. The key question is whether we can improve our efficiency in using energy at a faster or slower rate than the EROEI is going to decline as conventional oil depletes. This is a very hard question to answer since the available relevant information mostly sucks.
One of the reasons I find it easier to think about the middle distant future in EROEI terms rather than cash terms is because present cash prices assume that investments (eg of steel and concrete to build plants) will be subsidized by still fairly cheap (high EROEI) conventional oil. Those prices might go up a lot if energy gets a lot more expensive in the future, causing the cash economics of projects to change dramatically. However, EROEI will probably change relatively little (somewhat with improving technology in the project itself). If something is highly positive in EROEI, we can be confident it will always be cash profitable. If something has lousy energetics (can you say "solar panels"), we can be confident it will never get cash profitable absent a technological breakthrough.
 In an industry wide sense - not to say individual entrepreneurs and companies didn't face big risks of drilling dry holes and going bust.
Posted by: Stuart Staniford at August 20, 2005 08:00 PM
And here's the NYTimes article. I haven't read it past the first page yet; I stopped when I read "there may not be enough oil to go around" and did a double-take about how that doesn't make sense in an economics sense.
Posted by: Dan at August 20, 2005 08:01 PM
Mandos: "So, TI, are you in the "civilization is over" school, then?". No, the civilization is not over yet - but we have a real problem here.
On EROEI: it is not directly connected to profitability. Profitability is connected to price and costs. And costs are not the same as EROEI. I see that we are all the time mixing energy and money, geology and economy, barrels and dollars.
The EROEI of producing oil is decreasing and that is connected to depletion. But I bet that in ten years time the "peak oilers" and "economists" will still be arguing. The physical oil people will say: we were right, the oil supply is down. The economists will say, no we were right: the supply is down because the demand is down and the demand is down because we have recession caused by Fed policy / debt / trade imbalance / whatever.
Can we ever know who has right? Yes, we can. The US government and oil companies have tried to stop the depletion of the US oil for over three decades now. For no avail. New oil has been found but it has not been able to stop the decrease of oil production. And nothing will be able to stop the world depletion.
But the alternatives? Just do the numbers. Both for the zero total energy consumption growth and 2% - 3% growth. Natural gas is depleting. Coal? How much more can be produced (it is the top production, not how long we will have coal left). Nuclear energy? Check the world uranium production. Biomass? How many acres is needed (remember the EROEI here)? Wind, solar? How many megawatts? It is not only the technology, it is the volume, too.
Increasing energy efficiency? Yes, but there is a limit and the rising costs (read: more energy input). Hopeless? No. Just start finding out how they survive in Europe with much lower energy consumption per capita than in the US. Just try it.
Posted by: TI at August 21, 2005 02:43 AM
Why should we be like Europe? I think America is a much better place in every regard (except coffee) than Europe. Freer, braver, better music.
WRT coal, there is no reason to think that if "production rate" is a bottleneck, more mines can't be constructed/expanded. Why didn't you think of/mention that. BTW, there is a story in the WSJ about how coal mines are hiring like crazy. Guess why? To raise production. duh!
Posted by: TCO at August 21, 2005 06:34 AM
I know exactly what Stuart means when he says he gets the sense Jim isn't getting his arguments. And I think I have an inkling as to the reason.
The situation boils down to a fairly simple picture: can man outhink the inevitable decline in natural resources? It seems no one argues there's no decline in resources, and no one argues mankind isn't getting smarter and more productive. Jim's "everything will adjust" picture is accurate from a purely theoretical sense, and Stuart is of course right too.
But Stuart is much more right from a policy standpoint, because Jim's inherent argument - productivity keeps increasing, ingenuity applied to energy substitution won't stop - is merely restating a natural law: evolution. We evolve to meet the challenges at hand.
But the heart of the matter, I believe, is who is "we". Evolution is a painful process that involves progress for some but extinction and death for others. Policy is about avoiding pain before it comes, managing the details, and most importantly, (and I suspect when you get down to it this is really where Jim and James most differ from Stuart) doing it in a way that is the most equitable for the most people.
In Jim's world, paying $6.00 for oil sand extracted fuel is just the way productive humanity deals with natural challenges. In Stuart's world, this scenario means painful adjustments for real people. I suspect Jim personally largely won't be affected by $6.00 gas prices. The dangerous thing from a policy standpoint is, he sees no relevance to the fact that many other less fortunate people WILL be deeply negatively impacted. But truly, at the end of the day, that's all that -is- relevant: quality of life and who gets to have it.
Morality of course isn't important to theory, but it is eminently important to the state of humankind. And isn't that what we're most interested in? Isn't improving such the prime directive of economic thought? For Jim to end his arguments at pricing adjustment and human productivity gains is to hide in an ivory tower without any care for how things shuffle themselves out on the ground. Good theory, terrible policy. He cites Germany in the 30s as a case of ingenuity in alternate energy use and but fails to mention that WWII was largely a war built of inequitable economic conditions at the same time. His theory holds, but he conveniently fails to mention the consequences.
So we would do well to be careful how much actual policy we base on that sort of thinking. Who wins and who loses may not matter in economic theory land, but it sure as heck matters in the real world. And isn't that ultimately where we all live?
Posted by: Rosenblatt at August 21, 2005 08:38 AM
For those wondering about converting coal to oil, the two largest coal mines in the US produce about 140 million tons per year. If the numbers provided by TI are correct, those two mines produce enough coal, which when converted to oil, would supply 5% of US needs.
Posted by: Scott at August 21, 2005 09:15 AM
scott, (1) what percent of total production are the "two top mines"? (2) what is the coal reserve situation (how much available for more mining)?
The market is obviously torn between the increased productivity possibility winning versus the running out possibility winning. Why do you think that a government reaction would have any more correctness than the market opinion. Why do you think we need a government action (that things will be so awful if we don't have one or that a government reaction will be able to fix a problem that might be unfixable?)
Posted by: TCO at August 21, 2005 10:22 AM
Don't get me wrong, I certainly DON'T think government will get this one right, epecially the current government which seems to create a fiasco out of everything it touches, outside of fattening the superwealthy.
The problem is that contrary to libertarian belief, the market is not always "right"; the market is simply the positions or bets of all the actors involved. Further, when applied to things like energy pricing, a great many of the market participants have little short term ability to "trade" their position. If I bought a $35,000 SUV and get hit by rising gas prices, I can't as simply dispose of my "position" as I could if I had shorted Chevron stock. Some markets can be terribly inefficient when it comes to moving quickly enough to new information. And that has consequences.
The market represents the best guesses of its actors, but it can also be terribly misinformed on occasion. The internet "boom" is a perfect example. All sorts of new economic theory came along to justify valuations, much of it "productivity gain"-based, like Jim Glass' arguments. Now certainly some of that theory was defensible at some level, but the truth is millions took huge losses, and many lives were massively affected. At some level that's the price of capitalism, yes. But also at some level there comes a point when you start to talk about things like crime, wars, even survival of the species. Will the markets get global warming right?  Will this energy issue get an Iranian nuclear weapon delivered by Zarqawi to Boston someday? It is largely believed that Sept. 11 was a result of our presence in the Middle East which aims to keep oil flowing to us. How is something like 9/11 factored into the "market" price of oil? Talk about externalities. How do "markets" deal with these very real effects? And does it make sense at some point to try to intervene through policy at some point and not just trust in the "markets"? By the time all the information is there for markets to adjust, it may be too late.
I don't know what the answer to the oil situation is, and frankly I doubt the government has a prayer of really getting it right. But I do know the real solution is much, much more complicated than Jim and James' arguments try to have us believe.
 As Stuart is no doubt aware, global warming can lead to what's called the "runaway" greenhouse effect where an atmosphere, densely filled with CO2 and water vapor from more quickly evaporating oceans absorbs infrared light, which causes temperatures to rise and hence more evaporation, and so on in a kind of feedback mechanism, out of our, and yes our "markets' ", control.
Posted by: Rosenblatt at August 21, 2005 12:19 PM
Rosenblatt: The German synthetic oil production in '30s was unsustainable (remember, they lost). It was clear from the outset that it was not possible to increase the German coal production enough to meet the demand of a growing, motorizing economy for oil. You see, all the coal produced was already used for other purposes. It was not the idea, either. The idea was to produce just enough synthetic oil for the war that was to get the control of oil in Romania, Baku and Persian Gulf. Just look at the coal production numbers at that time.
German Anton Zischka wrote a book in the end of '30s named "The Fight for Oil". He wrote that synthetic oil "will guarantee peace" and gave a lot of details and numbers about the synthetic oil production and the importance of oil for the economy. The point here is that the reader will soon see that, on the contrary, all these facts show that the war will be inevitable. The synthetic oil is really a sign of desperation. The book was clearly intended to tell informed readers why Germany had to go to war. But of course the informed reader could also see why Germany would probably lose - there would not be enough oil to win...
But one problem in the discussion on oil and recession is that peaking oil is a global phenomenon. The world oil consumption will diminish after that but Americans could import more for some time. The effects will vary greatly. We have seen what happened in Soviet Union: their oil production peaked, economy (and the political system) collapsed, demand was crushed - and they could export more oil even after the peak! Collapsing Russian economy saved the world economy. Something like this could happen again.
Posted by: TI at August 21, 2005 12:33 PM
Thanks for that elaboration on the German synthetic oil production scenario. Very illuminating and relevant to this issue, it seems.
Posted by: Rosenblatt at August 21, 2005 12:45 PM
I would like to throw in one more fact. The mankind has never really replaced any older energy source. It has only added new ones. The world uses now more energy biomass, hydropower, wind, solar, coal, natural gas, nuclear energy and oil than never before. (OK, whale oil depleted and was really replaced.) The world never "moved on" to better fuels. It uses more of everything. After nuclear energy (beginning of '50s) there have not been any genuinely new energy sources.
This is important to note. Individual countries have met the depletion of a domestic energy source (ie. oil) mostly by importing it (or sometimes another form of energy ie. uranium) or products made with it. The world cannot do that. Many countries have adapted to changing energy structure but very few to diminishing total energy usage. These are North Korea, Cuba, Russia and some other former Soviet states. Ask them what it is like.
Posted by: TI at August 21, 2005 02:01 PM
I don't know that you or James are so far apart...or that you're so far from me. (I guess James should speak for himself.) While, there are people running around saying the market can fix anything and invent anything when needed (sidenote: then where the heck is my table-top fusion and my room temperature superconductor), that is not the position of anyone who is sophisticated. Instead, we (I) say that we don't know! But that the market guess is better than a government one. And that market response (even if we're "screwed", possibly even inevetiably) is a better means of coping with it than a government one.
This is at least in contravention to the people saying "look I don't believe in econ (as if believing in classical microecomics equates to being an evil Republican) and the market is stupid (and don't bother me with arbitrage arguments...I don't follow them) and I know better than it that there is an iceberg ahead".
Posted by: TCO at August 22, 2005 08:33 AM
M. Friedman to the contrary predicted (always difficult in the real world) this would result only in the economy still getting hit by the real cost of the oil price rise (effectively a tax hike to the extent oil prices were paid out of the country to OPEC) while compounding matters by creating accelerating general price increases, inflation, not seen elsewhere.
I don't understand this comparison. If we raise taxes on oil by $20 billion, someone else in America's economy gets to keep that money. Whereas a price hike is more like buying your oil and then dumping the extra $20 billion into the ocean.
Posted by: Noumenon at August 22, 2005 01:58 PM
The comments don't recognize HTML tags like , I guess. The first paragraph of the above was supposed to appear like a quote.
Posted by: Noumenon at August 22, 2005 02:01 PM
I'm spending too much time here. I've decided that I will only read and respond once a week (on the weekend). c y'all.
Posted by: TCO at August 22, 2005 05:14 PM
I note that Steven "Freakonomics" Levitt gives this blog's discussion of peak oil a nice plug, while being rather tough on the Peter Maass article on it in the Times.
Maass certainly impressed me with such insights as...
"Few people imagined a time when supply would dry up because of demand alone."
... but *now* we know supply will dry up because of demand alone! ;-)
Posted by: Jim Glass at August 22, 2005 08:10 PM
Or, as MaxSpeak puts the same argument:
"We then get a textbook rehash of the principle that when scarcity grows, prices go up, people buy less of the scarce good, and substitutes come online. Investment flows to alternatives, which get less expensive in consequence.
So don't worry about the world's dependence on finite fossil fuel resources. Who needs expertise in resource economics? Markets solve all problems. The universe is a convex set whose boundary is smooth and everywhere differentiable, may God strike me dead.
The principles make perfect sense, but they are completely beside the point. They typify the incompleteness and intellectual arrogance that pollutes academic economic discourse.
Do markets really solve all problems? Most problems? Nothing ever goes wrong? I guess it depends on how you define "wrong." For instance, Amartya Sen wrote a book about poverty and famines which describes how markets solved the problems of people having no money to buy food: they die of starvation, the ultimate steady state.
If oil runs out, sure there will be substitutes. How fast will these come online, if they do? How much will they cost? What will be the costs of adjustment? Will that be fun? Who knows? Markets solve problems. Solutions do not exclude freezing in the dark, a new kind of equilbrium.
Nobody should be let loose in the wild with a Ph.D. in economics unless they've been required to take three or four courses in history, preferably taught by non-economists."
Posted by: Anonymous at August 23, 2005 06:12 AM
"Nobody should be let loose in the wild with a Ph.D. in economics unless they've been required to take three or four courses in history, preferably taught by non-economists."
Couldn't agree more with my friend Max about this. The dropping of history from economics courses of study is a real bad thing, IMHO.
E.g., on this particular point we could look at the historical record of the long list of all the other depletable mineral resources that have had their prices relentlessly rise through time as humans inexorably depleted them down to the nothingness that we so suffer from due to their lack today.
I myself linked to Jevons's paper on the inevitable exhaustion of coal and how it would bring the industrial revolution to a grinding halt, upward in these very comments. That's history to learn from, surely.
And John Tierney today places great faith in the lessons of history in making a cash money "Simon bet" against peak oil...
Too much faith in history, perhaps?
Posted by: Jim Glass at August 23, 2005 09:02 AM
One of the things to remember about PO is that the economy has to become more efficient in energy use more or less according to the depletion rate.
Most discussions in this thread talk about the impact of high oil price (the economist point of view) or the system view (physics).
Point I am missing and which I think is the key to the issue is: What happens to the economy when oil supply is reduced 3-5% every year?
So not a one-time adjustment, but every year new adjustments.
Posted by: richard at August 23, 2005 11:15 PM