July 29, 2008
Oil prices and demand
More evidence of significant changes in the behavior of American consumers.
U.S. motorists drove less for a seventh consecutive month in May, as vehicle-miles traveled on all U.S. roads fell 3.7 percent during the month from a year earlier, the Federal Highway Administration said in a report yesterday. The seven-month slide is the longest downward streak since 1979.
Demand for oil and petroleum products dropped 4.3 percent in May from a year earlier to 19.7 million barrels a day, according to Energy Department data released yesterday. That's 889,000 barrels a day less on average for the first five months of the year, compared with the same period a year before.
U.S. gasoline demand appeared to be quite unresponsive to price during 2001-2006. My interpretation is that the primary reason we ignored $3.00 gasoline is because we could afford to. But as gas prices rose and the number of gallons purchased held steady, the budget share for energy has gone up pretty dramatically over the last year. Fewer people seem to be ignoring $4.00.
Posted by James Hamilton at July 29, 2008 02:37 PMdigg this | reddit
I think velocity has something to do with it. Increases the sticker shock.
Posted by: Dave Schuler at July 29, 2008 03:03 PM
Traffic deaths and injuries have also plummeted in America, as have pollution, traffic jams, and competition for parking spaces.
Expensive oil is a net positive for society.
Posted by: GK at July 29, 2008 03:07 PM
I'd love to see regular public opinion surveys: "How much do you think a gallon of gas will cost two years from now?" My hunch -- and since I haven't seen any such data, it's only a hunch -- is that last year people were still expecting high prices to be temporary, but now they're expecting them to continue or even rise more. I think that the expected future price is at least as important as the current price in changing behavior.
Posted by: lilnev at July 29, 2008 03:31 PM
I assume that less people getting out in boats means fewer fish taken.
Posted by: odograph at July 29, 2008 03:55 PM
GK, you're crazy.
Fuel Efficiency is higher now than back in 1998. Efficiency Declined from 1998 unitl the end of 1999 and then improved quickly. It remained relatively flat until 2004 and then began improving until the middle of 2006. Since 2006 fuel efficiency has been in steep decline.
(Data is from EIA and DOT. The graph is of % change from one year prior in vehicle miles driven minus gasoline consumption % change summed for each month since Jan 1998)
I've noted some possibe reasons why as well.
Posted by: aaron at July 29, 2008 03:56 PM
lilnev, I was pleased that my prediction for the high in my area(*), for summer '07, was spot on (I guessed $3.50 and got $3.48). I only guessed $3.75 for summer of '08, and we've seen $4.59 ... off by an amazing 79 cents. So much for my skills.
I'm sure not ready to guess two years out.
* - http://www.orangecountygasprices.com
Posted by: odograph at July 29, 2008 04:01 PM
JDH, it seems like you have enough info to estimate the price elasticity of demand for American consumers. Care to have a go at it?
Posted by: W T F at July 29, 2008 04:02 PM
Logic is not your strong suit. Your links consist of incoherent rambling.
Demand contraction of oil has very clear benefits (such as a drop in traffic fatalities). Only a fool would dismiss this.
Posted by: GK at July 29, 2008 04:43 PM
This new paper shows that gasoline price elasticities are probably higher than previously thought:
Disclosure: this paper is another reason why I am still short the price of crude.
Posted by: Charlie Stromeyer Jr at July 29, 2008 04:45 PM
Last night I did my own VMT chart and added data from the 70's plus also the percent change. Seeing the period to period change figures is interesting.
I suppose I should overlay periods of recession with this.
Posted by: Mike at July 29, 2008 05:13 PM
As an astute poster wrote upon hearing US had destroyed about 900bk/d over the last year..."The first million is the easiest. It gets harder from here."
BP said today in their report also, that OECD demand for petrol had fallen about 4.00%. And yet, they said, all that demand is being taken up by non OECD demand. In the US and Germany, two of the world's biggest fuel markets, gasoline
sales are down by around 4-5%, said BP's downstream head Iain Conn, while
consumers and industry in China and India are buying about 10% more fuel.
And also.... Speaking to analysts during a quarterly results presentation, Hayward
said, however, that fuel demand from the fast-growing economies of China and
India is more than offsetting OECD demand.
Welcome to peak.
Posted by: Gregor Macdonald at July 29, 2008 06:21 PM
lilnev, Are you a student of Robert Barro?
To my lay friends and acquaintances, I find myself arguing one-time relative price changes while most of them argue "inflation".
That could be generously interpreted as a temporary price hike. For the relative importance of petroleum fuel price increases since 1999, inflation is still tame by 1970s and 1980s standards.
It would be an interesting empirical exercise to compute how much of productivity growth in the US economy during the 2001-2006 period was effectively neutralized by oil price increases.
Incidentally, this oil price shock reaffirms the value of price theory in specific and freemarket economics in general. Bullish for economists. Bearish for neomarxists.
Posted by: GNP at July 29, 2008 06:35 PM
The principal is that if you push the price up to high not even the most loyal consumer will be able to buy it. It is a simple real demand curve, and perhaps we've hit that price..
Posted by: ideas to make money at July 29, 2008 07:13 PM
fuel demand from the fast-growing economies of China and India is more than offsetting OECD demand.
China subsidizes oil consumption by $40 billion a year says the the NY Times. It's still a poor country per capita. How much can its govt spend on this (and on subsidizing the exchange rate, and investing ever more in dollar-denominated assets as the dollar loses value...)?
The same story covers how other Asian nations do the same. They aren't helping get the world oil price down that way.
With that attitude, they aren't going to be helping on greenhouse gases either, though I guess that's another topic.
OTOH, for all the talk about China's oil consumption, California consumes more gasoline that China does, at least according to the California gov't.
Certainly the US and Europe consume a whole lot more oil than China does. If US consumption now really is heading down as not seen since 1979, and Europe is heading is into a recession as some say, while producers are ramping up due to $100+ prices, that could take care of the price of oil for a few years...
Posted by: Jim Glass at July 29, 2008 07:49 PM
Who reading this blog is *really* cutting down their driving? We are, a little, but we always do in the nice weather. Bikes become our favored mode of transport.
But when the nasty weather returns to Vancouver, and early darkness, there is no way my wife is heading to work across the bridge (covered with debris) in the dark and rain. Our gasoline usage will return to normal by about October. This summer we perhaps drove even more, to date, thanks to a 2700km road trip.
Man, the roads seemed quiet.
My point is its easy to rid the marginal use of petroleum. The rest is oh so much harder.
As others have pointed out, whatever slack we enlightened bikers and new found corner store Twinkie-buyin' walkers might have put back into the system, others are sucking it right up.
Domestic demand in most exporting countries (those of any real significance) is also up.
Whatever slack is in the system is literally mere months away from being sucked up by demand growth elsewhere.
Posted by: Mike at July 29, 2008 08:00 PM
I think the most useful way to extrapolate gas consumption, and demand destruction, is by looking not at prices per se but rather what percentage of household income is is going to fuels.
Back in the early 80's when there was a significant decline in consumption (about 10-12% I believe), the tipping point seemed to be when over 6% of median household income was going to pay for gas. When gas went over $3.7 or so, that percentage occurred again. An interesting recapitulation.
Also, that decline in consumption I strongly suspect slopes down as incomes go up - it would be interesting, if the data was available, to chart changes in consumption rates by income quintiles.
Posted by: SecondLook at July 29, 2008 10:38 PM
Mike, you make a good point about automobile utilization rates which are accounted for in the paper I refer to above.
Note that since the start of JDH's first chart above that the U.S. vehicle fleet switched in relative terms to a more truck based vehicle fleet due to the advent of SUVs. The paper I refer to above should be able to account for a change in the fuel efficiency of the U.S. vehicle fleet.
Posted by: Charlie Stromeyer Jr at July 30, 2008 04:03 AM
Higher gasoline prices have reduced road deaths by about 5,000 a year (8%) - teenagers don't have as much money, so the most risky drivers are driving less; people are switching out of the bigger, older more dangerous vehicles, and people is driving slower to save gasoline. A bonanza.
Posted by: j at July 30, 2008 06:30 AM
SecondLook, you may be correct. There is also an obvious demographic distinction: retirees should have more elastic demand than workers who have to commute regularly, but I don't know what percentage of drivers are retirees.
Posted by: Charlie Stromeyer Jr at July 30, 2008 06:31 AM
Actually, I now see from this recent U.S. DOT form that over 14% of U.S. drivers are age 65 plus:
Posted by: Charlie Stromeyer Jr at July 30, 2008 06:59 AM
Here's an interesting article about fossil fuels and possible alternatives to oil that can decrease our costs.
Posted by: Nathan at July 30, 2008 08:09 AM
GK, is this clearer?
Posted by: aaron at July 30, 2008 08:17 AM
I doubt traffic fatalities are down more than miles traveled. Congestion seems to have gotten worse here in Michigan. And pollution is probably up realive to the amount of driving we are doing.
Posted by: aaron at July 30, 2008 08:31 AM
fuel demand from the fast-growing economies of China and India is more than offsetting OECD demand.
It would be interesting to compare the fraction of fuel used for "recreational" purposes in the two groups. As an example, I suspect very little is used in China and India, relatively speaking, to take the family camping in the mountains. OTOH, China and India still generate significant amounts of their electrical power from petroleum -- a practice the OECD countries gave up in the '70s. If the statistics were available, I think they would show a significant amount of OECD fuel consumption on activities for which there are easy substitutes that require less fuel. Replacing an oil-fired generating plant takes much longer, and requires either heavy investments in efficiency or the availability of an alternate fuel.
Posted by: Michael Cain at July 30, 2008 08:50 AM
Nathan, please also take a look at Amyris Biotechnologies because they have developed ways to ferment sugar into industrial chemicals, plastics, diesel, jet fuel and a pure hydrocarbon fuel which is as energy dense as gasoline and can be shipped through existing pipelines and pumped into almost any car:
Disclosure: I covered my short position in crude oil this morning at 121.4 (which was not the exact low) because I suspect there will soon be more strikes on pipelines by Nigerian militants, and because the Q2 U.S. GDP report might be higher than consensus due primarily to the effects of the stimulus checks.
Also, I am a professional energy expert who is also an adviser to this renewable energy investment nonprofit, and we will be investing in biomass solutions in addition to other renewables:
Please note that this website is still quite preliminary.
Posted by: Charlie Stromeyer Jr at July 30, 2008 08:56 AM
a 20% increase in Chinese gasoline usage is still less than a 3% decrease in US usage.
Petroleum in China is used for electricity generation, and the primary use of kerosene in India is home cooking oil. Neither have anything to do with cars. It is easy to drive down car usage (don't drive!) hard to stop generating electricity or stop cooking.
Posted by: charlie at July 30, 2008 09:34 AM
Just for kicks, I plotted MPG vs Regular Gasoline Prices, 1976 to present. I don't know how to get dates into there. Anyone know a way to assign colors to data points based on date?
Posted by: aaron at July 30, 2008 09:44 AM
aaron, do you agree with this proposal from the Union of Concerned Scientists (UCS) about the future of the relationship between MPG and regular gasoline prices:
Disclosure: One member of the Board of Advisors of the nonprofit I am advising works for UCS.
Posted by: Charlie Stromeyer Jr at July 30, 2008 10:37 AM
I find Aaron's data here quite compelling. The paradox is that as vehicle miles travelled have dropped 4% in the past year, domestic gasoline consumption is only down 1%. Aaron discusses various reasons but none of them seem compelling to me.
My guess is that drivers of more efficient vehicles are cutting back more than guzzler and truck drivers. Maybe the reason they drive higher MPG vehicles is because they are more price sensitive. Some may be environmentally conscious and be trying to cut back in order to help with the carbon crisis. Commercial drivers can't afford to cut back. Guzzler drivers tend to be rich and are less affected by high prices. Of course these are all speculations; had the data shown the opposite effect, I would have had no trouble coming up with equally plausible justifications in the opposite direction.
Posted by: Hal at July 30, 2008 11:29 AM
the good news is... that last week Americans drove more and gasoline demand increased as seasonal patterns predicted,
but the bad news is... that today OIL goes up and rebate checks ended !!!
It might be to early to eat confetti...???
As long as energy prices keep going up the US economy will never be out of the woods.
Posted by: the procastinator at July 30, 2008 12:43 PM
I thought of that too, but the difference is too great. There would need to be an increase in inefficient driving as well as a decrease in efficient driving.
Add to that population growth (increased congestion) and communities neglecting traffic managment such as light timing and adding and improving roads.
Posted by: aaron at July 30, 2008 12:55 PM
Charlie, I'm all for higher standards, though I'm weary of regulation. As for the prices, I have no idea. I think gas and oil are overpriced now, but I don't see anything wrong with using higher prices for our future planning. Current prices are our best estimate of future prices.
I'm very weary of anything that says "Global Warming" on it. And $41 a ton is absurdly high carbon cost. I don't think there should be any (I think people are confusing risk with uncertainty and mistreating uncertainty), but serious people who believe that there are carbon costs estimate them between $2 and $14 a ton.
I for efficiency for it's own sake.
Posted by: aaron at July 30, 2008 01:19 PM
aaron, I am inclined to agree with your first paragraph and would rather see higher standards achieved via intelligent competition more than via new regulations.
As for your second paragraph, I would rather see markets set a price for a metric ton of CO2 than see politicians set a price. The largest market now for trading carbon is the European Climate Exchange which currently shows a December 2012 futures price of $55.4 U.S. dollars per ton of CO2:
However, you should know that I am biased in favor of a markets based approach because I am long both the global price of carbon and shares of Climate Exchange Plc.
Posted by: Charlie Stromeyer Jr at July 30, 2008 01:56 PM
The 12 mo moving avg number has essentially been in a flat range since December 2004. So US drivers were modifying their driving well before gas prices got to $4.00 per gallon. The detail data from the FHA shows that between Jan 1, 2005 and May 31, 2008 the change in this number has been essentially zero.
With respect to China, the percentage increase is off of a much smaller base volume of usage. So the actual increase of gallons of gas consumed in China doesn't necessarily offset the decrease in Western gas consumption.
Posted by: Ravenor at July 30, 2008 03:35 PM
a 20% increase in Chinese gasoline usage is still less than a 3% decrease in US usage.
One is still on an upward trajectory while the other has flat-lined and is declining. Its unlikely that comps next year are going to show as dramatic a decrease. This is easy to believe if one believes that people have already made all the easy cuts to their petroleum usage.
Posted by: Mike at July 30, 2008 04:28 PM
"fuel demand from the fast-growing economies of China and India is more than offsetting OECD demand."
Not so at all. See http://peakoildebunked.blogspot.com/
"a year-on-year drop in U.S demand of 1,103,000 barrels/day. Huge. Roughly equal to the production of a super-giant oil field, or a small producing country like Qatar, Indonesia or Azerbaijan....A 1.1 million barrel/day drop in US demand is gigantic -- literally enough to wipe out all growth in oil consumption for the entire world, which at the moment is running at about 0.9mbd. (The current IEA forecast for 2008 demand growth is 0.89mbd. Source: July 10 OMR).....There is simply no way for China and India (who combined have average annual growth of around 0.5mbd) to overcome a 1.1mbd drop in US consumption. They simply can't grow that fast."
Peak oil theory is b.s. To attempt to blame the developing world for rising oil prices is not in anyway based in reality. Speculation and a falling dollar, not neo-malthusianism explains most of the oil surge.
Posted by: brad at August 1, 2008 06:26 AM
My pet peeve is the term "demand destruction". We don't talk about "demand construction" do we? Do we say that demand destruction in the U.S. is balanced against demand construction in China? It sounds ridiculous.
Demand reduction, or decreasing demand, does the job just fine, I would say.
Posted by: Hal at August 1, 2008 02:22 PM
help me with something. I have always assumed when we talked about BP, or I saw BP ads(the ones with all the windmills, we were talking about British Petroleum. Now I wonder if it is Boone Pickens. Which is right?
Posted by: randymiller at August 2, 2008 09:43 AM
One of the age old discussions in economics is what you do about external costs, particularly waste. Growing up in the midwest in the fifties, no hog or cattle feeder worried about his manure running off into a creek, he dumped his trash in a ditch. If there was a cost to that waste stream, the guy downstream who was damaged had to pay for it.
How do we calculate the external cost of a ton of CO2 going into the atmosphere from a power plant?
Posted by: randymiller at August 2, 2008 09:55 AM
randymiller, what you ask is a complicated question but you can see some attempts at estimating the external costs about halfway down on this webpage:
Posted by: Charlie Stromeyer Jr at August 2, 2008 11:47 AM
CS, I read the webpage, and if I read it correctly, the cost of generation of electricity in Europe from coal is 4 cents per kwh, and the external cost created by the coal fired generators is 4 cents per KWH.
But that external cost calculation did not even bring in the effect on global warming.
I know how all of us free market types hate regulation, but if external costs are borne by the taxpayers, is it not the responsibility of the government to protect the taxpayer and recover those externalized costs?
Posted by: randymiller at August 2, 2008 02:35 PM
I think oil will eventually be worthless as demand will continue to drop. Exxon will be another Enron for thier GREED.
Posted by: gil at August 13, 2008 10:23 AM