a linear extrapolation of oil consumption by demand growth, production growth, and demand elasticity
Posted by Prof. Goose on January 22, 2006 - 1:19am
Topic: Supply/Production
Tags: energy, oil, oil prices, peak oil, prices [list all tags]
Demand has increased since 1983 in a very linear fashion, with a growth of 1.08MBD(Million Barrels per Day) per Year. This can be represented in the supply-demand graphs below by a steady shift in the demand curve from left to right as we move forward in time. Also, at any given time the demand for oil is very inflexible; price has little effect on consumption within a certain range around what, at the time, is considered a “normal” price. In my own experience, I can cut out the impulsive run to the hardware store on the weekend, but I still have to get to work every day. This makes the demand “curves” quite steep for any point in time. To try to get a ballpark WAG on what the slope might be, I looked at the data around Katrina, when we had a price spike of close to $10 and a consumption reduction of about 0.87 MBD (Million Barrels per Day), for a slope of about $-11.50/MBD. In this simple, linear model this slope, shown below, is one of three very important parameters in trying to predict future prices, and as I mentioned, $-11.5/MBD is a WAG.
The second important parameter is the rate of demand growth. As we saw in the first graph, demand has grown at 1.08MBD/Year since 1983 (least squares fit). But if we fit the data from 1993 forward,
we get growth of 1.32MBD/Year. And looking at even more recently 2002-2005, we get growth of 1.95MBD/Year. Interesting that growth seems to be accelerating even in the face of the recent steep price increases (think India, China, and everyone else who wants what we have). The third parameter is either the value for peak production, or the rate of increase of production per year if peak is not imminent. Let’s pick my WAG slope of $–11.50/MBD for demand, a middle-of-the-road demand growth of 1.5MBD/Year, assume production can grow at the same rate as 2004-2005, and try to plot the future.
The supply curve (labeled Historical Data), until the late 1990’s, has been pretty flat – producers have been willing to just pump more oil with little or no increase in price, to appease the shift in the demand curve. But if Peak Oil exists, and we are close to or at that point, then supply has an upper limit, so the supply curve starts to go vertical. In the middle-of-the-road example above, the average price of oil is predicted to increase about $12 per barrel per year.
Now, how about playing with the parameters. Some have argued that demand cannot be as inflexible as computed above. Economists put short-term elasticity around 0.2. By my amateur calculations, for $56 per barrel and 84MBD, that equates to a slope of $-3.36/MBD.
Regardless of the more elastic short-term demand, we are looking at $100 per barrel sometime in 2008, if production increases at the current rate. What is not shown or predicted, is how high the price must go before demand GROWTH begins to falter.
Now let’s see what happens if 85MBD really is peak production. We’ll keep the same demand slope, and assume that demand growth will be impacted and reduced to 1MBD (it has to if 85MBD is peak):
If production peaks at 85MBD, we’re looking at $100 oil this year. Even if demand grows by only 0.5MBD per year, we’re looking at $78 this year and $100 next year, again assuming a steady demand growth.
Finally, a look at the Cornucopian view, where the price of oil actually drops to $35 per barrel next year, in the face of 2MBD demand growth per year (Steve Forbes, I believe it was, predicted this late last year):
The world would have to increase production this year to about 92 MBD this year, or an increase of 8MBD. I don’t think anyone anywhere is suggesting that sort of increase.
So, pick your favorite scenario. I think this next year will really start to reveal the true status of Peak Oil.
Great post, but is there anyway to
make the graphs, so users can click
them and they enlarge?
I can only see the left 3/4 using
Explorer and junk monitor.
I can't imagine how high oil would be right now if we had a cold winter.
If I was your average Joe, that knew nothing about the coming oil problem and saw these graphs I would be horrified. Even someone saying we will have 70 dollar oil for the next 10 years and peak won't hit till 2015, would be considered a Cornucopian in my mind.
Recommended for anyone who can't see all the graphs. (maybe I'm the only one)
I also like the LinkChecker and CustomizeGoolgle extensions.
Get the Adblock extension. The Internet without ads is a beautiful thing.
And get Launchy, for the small number of sites/pages that don't display properly in Firefox.
BBCode is very handy, too. It allows you to insert links, images, formatting, etc., into HTML and PHP boards very easily.
This isn't a hard constraint. There are lots of easy ways to commute with much less gasoline: ride a scooter, van pool, take the bus, bicycle, telecommute etc. Look at NY during the transit strike. They forced every car coming into Manhattan to carry at least 3 people, and people arranged ways of dealing with it in a day or two. If a similar system were announced nationwide in the U.S., the oil speculators would be trampling each other, running for the exits. There are lot of non-price ways of adjusting elasticity. See Saving Oil in a Hurry from the IEA. So I don't think we can really make the assumption that demand will always be inflexible. How flexible will it be if the government makes carpooling mandatory?
I would add to your comments that any talk of the price elasticity of oil demand has to be extremely specific in terms of three things, two of which are very hard, if not impossible, to model:
The easy one is the size of the price increase. The demand response will be non-linear with respect to the price change (hence the fact that they're demand curves and not straight lines). Over a pretty broad range, the larger the increase the ever larger the non-linear demand response.
The timing of the price increase and the demand response. If price creeps up over several months, people will adjust much more slowly and less in absolute terms than if the same percentage price increase happens in a week.
Market psychology. After the hurricanes we had a lot of very spooked drivers in the US. People were scared that they wouldn't have gasoline to get to work or the food market. Just as people started to get serious about minimizing their fuel use, prices came back down.
I'd also toss in a reminder that we simply don't know how people will react to truly high prices in the US. The price level in 1981 and 2005 were, adjusted for inflation, almost identical. But we still haven't seen what the US consumers will do when faced with months of gasoline prices above $5/gallon. (That price, in 2006 dollars, is my personal, seat-of-the-pants level for the lower bound of genuinely high gasoline prices.) IMO, at that price level we're deep into unchartered waters, and we could very well see a level of commitment to conservation rivaling or even surpassing anything the US did in WW II.
IMO demand reduction in US will not come through high prices but through shortages. All things being equal people will continue driving even with 10$ gallon because there is simply no real alternative. Yes they will drive less but not that less than you think now.
The SUVs must go, and before too long, one way or another. After all, they (and their associated mentality) are the real reason the 'american way of life' is in peril.
If you have an SUV sell it now, before it becomes worthless.
Business as usual will soon not be an option, the planet has said so and is not going to negotiate.
Eco-terrorists will not be needed. The people who are stuck owning the things will find them to be such albatrosses that they will torch them themselves and try to collect the insurance. In the end eco-terrorists would get the blame, I suppose.
Sleep on friend's couch to be within walking or biking distance - OK for three day strike, gets old and intolerable really fast.
Bus - slower than molasses even when it deigns to show up. In most places, a ten minute drive becomes an endless ninety minute ordeal. One's time must be utterly worthless (but why, then would one be commuting?), or else gas must exceed maybe $15/gallon.
Scooter - generally incredibly dangerous. Useless in winter.
Bicycle - generally too dangerous. Often useless because banned from U.S. bridges and tunnels. Useless in winter snow and ice. Socially useless in summer, few places have any showers, much less enough showers.
Van pool - very practical in densely populated Manhattan. Elsewhere, endless ordeal like bus.
Telecommute - can be practical, but it's not our decision, it's our boss's decision.
Car pool - OK short term, but bosses will not tolerate the scheduling constraints for long. Only the tiny unionized sector still has any rigid 7:30 to 4:30 jobs.
These alternatives are generally impractical without forceful legislation. That would many businesses, taxpayers, and transit providers very, very unhappy. Employers told to install showers, to allow telecommuting, or to allow European-style uniform working hours (I once witnessed Belgian labor police swarming a small business at 6PM to make sure all the employees were out of the building), will yowl to the heavens. So will shiftless bus agencies asked to become actually useful by doing better than "just show up when you happen to feel like it".
Without making the alternatives practical, just stranding everyone who doesn't have a child or two to plop into the car to meet a police-enforced minimum headcount is not the most workable or equitable long-term plan.
They can also move back to the city, where there are lots of eager workers, all in walking distance.
In most areas, employment is more centrally located than housing. There is usually no single location that can be described as "near the workers."
Fly In the Oil
I'm just an engineer too, so maybe I don't get it. You international economists may be able to straighten it out.
As long as the US doesn't represent 100% of world consumption, I can't help thinking that the devaluation of the dollar (both historic and projected) must somehow be considered in any scenario involving worldwide price and demand calculations for oil or anything else, but I don't usually see any consideration of that at all. For example, a high oil price in USD should have no immediate effect at all on China's, Japan's or Europe's consumption of oil, if the values of their curriencies increase accordingly, until (if, unless) their dollar sales of export of goods fall enough to cause readjustment of their oil consumption levels. It would be interesting to see the analysis based on (assumed) constant (1970?) dollars.
Runaway Train
Oil price adjustment through the international economic path might not be seen for 1 to 2 years, which might result in a very high overshoot in the USD price/BBL, making recovery from all after-effects all that much more difficult. In fact, would it not be entirely inconceiveable that, with an increasing Yuan <Rupee, Euro, gold, whatever>, China <whomever> may be able to supplant a significant amount of the present US demand for Japanese exports to such an extent that Japan's loss of a sizeable proportion of the US market might not present too much of a difficulty for them to hurdle ...over time.
With consistant US double deficits with (apparently) no end in sight, mustn't the $/BBL price convention be fading before our very own eyes? Guess we'll have to wait for 20 March 2006 to see.
As you have stated, the real question is what price is required to cause people to actually use less taking into consideration the additional demand of China and India. My guess is that it is a lot higher than anyone now contemplates and well north of $200 by 2010.
On the bright side, $400 oil would certainly encourage conservation.
If you look at demand, for gasoline, or oil, it is still growing as "normal" - even with the steady long-term increases of the last 3.5 years. Sales at GM and Ford are down, but Americans are still buying SUV's. When gasoline spiked to $3.25 this fall Americans reacted, but we cannot tell how much they would have gotten used to the price and reverted to old behavior - because the price went back down.
Until gasoline sustains $3.50 (putting oil at roughly $100-$120) we won't be able to see much. Comparisons to earlier points in history are largely irrelevent, since it is a different world in so many ways.
I encourage those of you who are doing these excellent analysis/graphs of price/supply/demand to stop looking at the daily price of oil, since it contains so much temporary "noise." Try using a 26-week moving average of the Nymex-Futures/WTI Spot price. It provides a nice, smooth line and I believe has a very healthy built-in measurement of the long-term, economy-effecting component of price.
Over the last 3.5 years this price has basically never gone down, except for a couple of two- or three-week periods. It also shows that the rate-of-increase (acceleration) has gone through several phases, and is currently not at a high.
I'd be interested to know how the dollar being held up artificially by China, Japan and others plays into these price scenarios.
Thanks for any info you have!
http://www.gold-eagle.com/editorials_04/norcini082704.html
Jim
Euro history in pre 1998 years exists as an average of several comprising basket currencies, so you're not out of luck there either.
I've plotted oil price against the Euro and gold since 1998 and it maintains a pretty good, although a little rough, linear relationship, up until about 4 months ago, when gold appears to break away not only from the Euro, but from oil too.
This much is clear. When I take dollars from my paycheck to buy gas for my car, I don't need to worry about "foreign exchange risk" because the imported oil is purchased with dollars. The price I pay for gasoline may fluctuate because of various supply and demand factors (seasonal, hurricanes, Nigeria) but "foreign exchange risk" is not a factor.
Imagine a world where oil is traded for a basket of currencies reflecting the countries oil exporters buy their manufactured goods from (more Europe and Japan than USA). When I use my dollar paycheck to buy gas, I now must worry about how much yen and euro my dollars will buy because those curriencies will be needed to import oil. Suddenly, the dollar-euro and dollar-yen ratios affect how much it costs Joe Sixpack to get to work. Such a shift would have a dramatic impact on politics and culture as well as economy and finance.
Even as oil prices rise, the fact that oil continues to be traded in dollars is a major source of stability and benefit to the U.S. economy.
But does the present system encourage maintaining the soundness of the dollar or does it encourage devaluation? As my seeing eye tells me, hyper-$-intrest is coming unless,
1.) a vast quantity of dark matter Does exist
2.) a vast quantity of dark matter will be added to what exists now.
If oil is priced against a basket of currencies, it may cause the USD to stabilize as it becomes apparent that ever-increasing quantities can not be procured at oil's increasing (real worldwide) value, rather than just keep on heating up the Ben Franklin PhotoShop to pay the higher prices in inflated $. The US just gets to the energy savings scenarios quicker. <I mean like hey, its never going to happen???>
The perspective from this side is that actually we don't care how much the oil price is in USD, and technically as long as long as its the same (+/-) in Euros, its no problem until the US stops buying our exports. If we can replace your market with China-India, bad news for you. If not, I think oil will go higher in Euros too, but it'll probably be in a lot more gradual over time within a series of shoot ups and slides back down as the dollar inflates each time.
As for price converted to Euros, we're probably close to $6.00/gal consequently, if we're going more than 100 miles, we simply take the train or fly. If > 3.X adults are traveling, driving can be cheaper /P, but I'd still rather pay the difference and travel train. Club car beats driving any day. I've seen about 2 HMVs (both with those nasty British plates) in the last 6 months, so I call it... encouraging.