July 09, 2007
IEA: Oil Supply Crunch After 2010

Peak Oil anyone? The latest Medium-Term Oil Market Report from the International Energy Agency (IEA) paints a moderately bleak picture on oil availability in the next 5 years.

World oil demand will rise faster than expected to 2012 while production lags, leading to a supply crunch, the International Energy Agency said on Monday.


"Despite four years of high oil prices, this report sees increasing market tightness beyond 2010," the IEA said.

"It is possible that the supply crunch could be deferred -- but not by much."

Why is this picture only moderately bleak? They aren't yet predicting a decline in production. But the growing ranks of those affluent enough to buy gasoline and other petroleum products are creating market conditions similar to those we'd expect to see once oil production peaks. The price keeps going up.

Prices will keep rising until demand stops growing. How high will prices have go to go stop oil demand growth? $100 per barrel? $120 per barrel? We are going to find out.

The IEA expects assorted messed up places to stay messed up.

The report assumes no net expansion of capacity from Iran, Iraq and Venezuela and that 500,000 barrels a day of Nigerian oil - shut for a year - will not reopen in the next five.

Iran's own internal demand growth is going to cause Iranian oil exports to decline. So even if Nigeria's government puts down insurgencies an increase there will probably get cancelled out by a decline in Iranian exports. The civil war in Iraq still has some legs too. Though a US pull-out might accelerate the civil war and bring stability sooner.

The IEA projects declining oil extraction from OPEC countries by 2009.

But with forecasts of world economic growth of 4.5pc a year, the report argued that oil demand was likely to soar to 95.8m barrels a day in 2012 from 81.6m barrels this year.

At the same time it predicted production from the international oil cartel Opec would fall, slipping by 2m barrels a day in 2009, and it also cut supply forecasts for non-Opec countries by 800,000 barrels.

I am skeptical of claims that OPEC will substantially increase production. Probably most OPEC members prefer higher prices to higher production. Plus, the ranks of post-peak nations keeps growing. Pemex in Mexico can no longer keep up Mexican oil production.

Petroleos Mexicanos, the state-owned oil monopoly, said crude production fell 6.6 percent in May from a year earlier and dropped to its lowest this year as the company struggles with declining output from its Cantarell field.

We are not lacking in energy so much as we are lacking in energy storage for the types of energy which we can afford to create more of. We can build more nuclear power plants or wind towers to get lots of affordable electricity. But we do not have good enough means to store that electricity for use in transportation.

We need pluggable hybrid electric vehicles (PHEVs) and pure electric cars too. The energy cost per mile of electrically powered travel is much lower than the cost of gasoline to travel the same distance. But for electric vehicles to work out we need much better batteries. Some battery makers (most notably A123Systems) claim they have figured out how to make long lasting and affordable lithium batteries for cars. We are going to find out in 2 or 3 years whether this is the case. Those batteries are going to come not a moment too soon.

Short of PHEVs we have plenty of other ways to adjust: Get cars with smaller engines, diesel engines, or conventional hybrids. Also, live closer to work and take fewer car trips. Also, if you use oil or natural gas for heating then make your house much more energy efficient.

Share |      Randall Parker, 2007 July 09 10:55 PM  Energy Fossil Fuels

Wolf-Dog said at July 10, 2007 6:06 AM:

It turns out that there is enough uranium in the world for a thousand years even if the entire world uses nuclear power for everything. The nuclear waste can also be used as fuel for more advanced reactors, meaning that the remaining final waste would have a half-life of less than 300 years, which means that nuclear waste would not accumulate too much in the long run.

But how many nuclear reactors does the U.S. need in order to charge 250 million pure-electric cars? Assuming that the average person drives less than 75 miles per day, probably 200 reactors of capacity 1,000 megawatts each, would be sufficient for this task.

But can the U.S. build 200 new reactors within 10 years? Since it would cost only between $1 to $2 billion,
it follows that building 20 reactors per year, would cost "only" approximately $30billion per year due to the
large number. Given that the annual Government Deficit Spending is much more than $250 billion
in average, this would be a very small, since it would increase the annual deficit only by 10 % approximately.
And yes, this time the task should be done by the government, since energy is national security, and in
most countries, utilities are owned by the government.(The conservative Randall Parker might object
to Government getting involved in this business, but if the Manhattan Project was a government project,
energy will be also strategically important.)



David A. Young said at July 10, 2007 9:44 AM:

Prize contests have worked well in some areas in promoting innovation (i.e., the X-Prize). Do you think a "Battery Prize" would be useful? The government would stipulate the conditions (operating parameters, per-unit manufacturing costs, recyclability, etc.) then pay a billion or two to the first company/person to produce a manufacturing-ready product. I've always thought the beauty of this approach was that it doesn't cost you anything until you have product in-hand; or conversely, if you don't get the product in a reasonable time, it's probably because it can't be done efficiently with available technology -- which is also useful information, which you've been given for free.

TTT said at July 10, 2007 11:20 AM:

"Though a US pull-out might accelerate the civil war and bring stability sooner."

So genocide worse than the Cambodian killing fields would increase 'stability' and reduce oil prices?

Nick said at July 10, 2007 12:13 PM:

Actually, Wolf-dog, the average person in the US only drives 23 miles per day, so you only need about 72GW of average production, or about 80 reactors at 1GW and 90% capacity factor.

Or, 80,000 wind turbines at 3GW and 30% capacity factor, which would probably be more doable. A higher capital cost, but lower operating cost, and much faster installation. Plus, no need for pesky uranium imports, prices for which are skyrocketing, due to lagging capital expenditures.

Undoubtedly we'll have a mix of the two, plus very fast growing solar from a relatively small base (maybe .2GW per year in 2006, growing to 15GW in 10 years).

rsilvetz said at July 10, 2007 12:48 PM:

We could build 500 reactors in 4 years if the government did the following:

1) Ban environmental challenges to nuclear reactors

2) Move to a "Must Grant" licensing system with a maximum of 6 months wait time with one permit.

3) Guarantee full cost recovery at whatever the market can bear prices -- e.g. an actual free market in energy and its distribution without the incessant political meddling in energy prices. (Are you listening California...?)

4) Ban "Not in My Back Yard" legislation. This absurd concept that Tom, Dick and Harry can tell Joe, who owns the land and capital for the nuclear plant, what he can do with his property has to go the way of the dodo.

Until this happens -- you won't see a nuclear industry worth mentioning in the US of A.

Libertarian Joe said at July 10, 2007 1:21 PM:

Price increases don't "stop the increase in demand" for oil or any other product. Demand increases increase the price. Price is output, not input. Prices will keep going up until that point where prices justify the production of sufficient supply to meet the quantity demanded given the price. The price of oil will ultimately be based on where the demand curve actually is: In other words, the price the marginal global consumer is willing to pay for gas is the maximum amount the price can be. How much is that? Now that, we are going to find out.

TTT said at July 10, 2007 1:22 PM:

One thing I never understood was why there was opposition to nuclear plants if they are made very far from any inhabited area.

All of Alaska should be powered by nuclear and wind. There is enough open land that no fossil fuels for electricity should be used at all.

Why can't all of Las vegas be powered by nuclear, if the plant is 50 miles from the city?

Reality Czech said at July 10, 2007 3:56 PM:

It would be impossible to build 125 1 GW reactors per year in the USA.  It would be impossible to build 12 1 GW reactors per year in the USA!  We do not have enough workers with the required expertise to build them that fast.

Randall Parker said at July 10, 2007 6:29 PM:

David A. Young,

5 or 10 years ago I would have said to offer the prize for batteries. But now the demand for batteries is high enough that I do not think prizes are necessary. Venture capitalists are chasing much bigger prizes.

We are at a stage now where if a company can build batteries needed to make a 200 mile range electric car feasible then millions of such cars would get sold per year. The continuing rise in oil prices will make the demand for PHEVs and pure electric cars even higher next year and each year afterward.


How many watt-hours per mile are you assuming? Based on what size car?

I'm expecting once electric cars become feasible that people won't restrict their demand to Prius-sized cars. I'm expecting to see lots of PHEV Ford Expeditions and PHEV Caddie Escalades. I'd like to know what are the watt-hours of energy use for various size existing vehicles. Anyone have a good source for this?

Randall Parker said at July 10, 2007 6:53 PM:

Reality Czech,

I had an old high school history teacher who said he sat on an island in the Pacific in 1944 and watched the US fleet steam by with ships as far as the eye could see and it took them 3 days to all pass by. He said that in 1940 all that steel was iron ore in the ground in Minnesota. Somehow in 4 years massive quantities of material was extracted from the ground, processed, and built into all manner of devices on an enormous scale.

Productivity and capital available has skyrocketed since then. Given sufficient will I fail to see why we couldn't build dozens of nuclear reactors per year.

What I'd like to know:

1) How much steel and concrete goes into a single nuclear reactor?

2) How much steel and concrete do US producers currently produce?

Randall Parker said at July 10, 2007 7:07 PM:


The threat of genocide is greatly exaggerated. Genocide is an attempt to kill everyone. The killing of civilians by the religious factions in Iraq is mostly used to terrorize so that opposing groups will flee.

If we stay they'll keep killing since we hold off a resolution of the conflict. The Sunnis either must submit to Shia rule or secede. We are preventing both outcomes. So the Sunnis keep opposing Shia rule.

Wolf-Dog said at July 10, 2007 9:35 PM:

The amount of steel and concrete that goes into a reactor is a very small fraction of what is produced in the US, and don't forget the newer designs such as the Pebble Bed Reactor, which has very little heavy infrastructure.

It is true that we do not yet have a large enough work force to build many reactors per year, but but once the training is done in a few years, then we can certainly build 20 reactors per year for 10 years.

Although the US is currently importing uranium, this is because the American mines were abandoned due to the lower prices during the previous decades. Otherwise there is plenty of uranium in the US that can be extracted below the current high prices. Also, once the price of Uranium goes over $200 per pound, uranium can be extracted from sea water, and this kind of uranium would last for thousands of years even if the whole world used nuclear power for everything. Note that even $400 per pound uranium would only be a small fraction of the electricity generated by a nuclear plant, since most of the cost of nuclear power is due to operating expenses.

Kralizec said at July 10, 2007 9:42 PM:

It seems a protracted time of resource scarcity would be a fine occasion for natural selection to continue its work.

Nick said at July 11, 2007 9:54 AM:

Randall, the Volt is spec'd at .25 KWH per mile, and the Tesla at .215. Larger vehicles will use more, but not as much more as you might think, because electrical motors actually gain efficiency with size & power, where the reverse is true for ICE's, and regenerative braking greatly reduces the efficiency penalty of increased weight. Also, PHEV's will gain efficiency with time.

So, I used .25, which I think is good for all passenger cars. I'd estimate a max of .5 for the heavy SUV's. If the SUV CAFE loophole is actually elimininated, and fuel prices stay high, we can expect the mix of vehicles to gravitate towards perhaps 2:1 in favor of lighter vehicles, so the average would be about .333. That would mean an average of 96GW, about 105 nukes, and 105,000 turbines.

The steel, and tech, used in WWII was off the shelf. Nukes require much higher grade, specialized steel, concrete, etc. Not that it can't be done, but doing it very quickly would be quite expensive, much more so than wind, which uses mostly off the shelf materials. Some wind components are specialized (i.e., blades, generator), but they don't have the safety concerns which slow down nuclear, just your normal commercial reliability & cost concerns, and they have much faster development & implementation cycles than nuclear, so the experience/cost curve can ride down much faster.

I'd estimate that 3.5GW of wind will be installed this year. That's a 40% increase over 2006, and that growth rate is likely to continue, now that wind's future is becoming much clearer. At that rate, wind can handle new generation needs of about 7MW (average) in about 6 years, well before any new nuclear plants can start coming online. I would note that none of the new nuclear applications yet have a firm commitment from their utilities - they're all hedging their bets.

Wolf-dog, do you have a link, or source for 1) the US uranium resource estimate, broken down by type of ore, and 2) cost of seawater extraction? I would note that $400 uranium raises the cost of fuel to about 1.75 cents per KWH, which is pretty significant on top of the other operating costs of about 1.5 cents per KWH, in addition to capital costs.

Reality Czech said at July 11, 2007 10:54 AM:

Nukes may not use much steel, but they are competing with everything and everyone else.  The Oil Drum has carried notices about skyrocketing prices of even coal-fired powerplants, because the materials costs are going through the roof.  Just about everything is going to China.

It makes one wonder what would happen if the US demanded right of first refusal on all scrap shipments out, and purchased everything for a Strategic Metals Reserve.

Wolf-Dog said at July 11, 2007 1:30 PM:


The price estimates of Uranium extracted from sea water vary, but it seems that the Japanese scientists have conducted some detailed studies, as mentioned in this web site:
If I understood the comments at the bottom of the latter web page, the high estimate in this paper, is 88,000 yens per kilogram of uranium, and the lowest estimate is 25,000 yens per kilogram of uranium. Taking into account the current exchange rate for 1 U.S. Dollar = 118 Japanese Yen, and the fact that 1 kg is approximately 2.2 pounds, it follows that the high estimate (in this Japanese research paper) for 1 pound of uranium from sea water is 88,000X(1/118)X(1/2.2) = $338 per pound of uranium, and the low estimate is 25,000X(1/118)X(1/2.2) = $96 per per pound of uranium. The current price for one pound of uranium (in the form of U3O8 oxide) is $133.

I would say that in the distant future, the regular mined uranium is becoming more expensive in 50-100 years, then the sea water uranium can almost certainly be extracted at less than $200 per pound, which is reasonably economical. Meanwhile, the operating cost of reactors will also decline in 25 years, since science will be much more advanced. Note carefully that the Pebble Bed reactor is a very simple design that does not require the heavy construction. There will be liquid radioactive salt reactors also.

In any case, nuclear power will become competitive with coal. The following web site says that the amount of uranium contained in the coal that is burned in the US every year, is enough to power the current 100 reactors that we have:)


Will Blevins said at July 11, 2007 2:35 PM:

Reality Czech is right that there aren't enough trained people to build 12 1 GW reactors per year in the US. But Randall is correct that the people can be trained quite quickly, if it is results we are looking for--not political correctness and other extraneous elements of conventional education.

The military knows how to focus on essential information in its training programs. A crash program in the skills necessary for nuclear plant construction could also learn that focus. It's amazing how fast you can do something when the powers that be get behind it.

Randall Parker said at July 11, 2007 9:35 PM:

Reality Czech,

I agree with Will Blevins that we could train construction workers for nuclear plants very quickly.

As for the price of steel and concrete: But as long as the marginal costs of new steel and concrete plants are not high and as long as the marginal cost of more iron ore is not high then I do not see the problem.

The question I have: How fast will steel and concrete production expand to satisfy growing Chinese demand? Will prices fall in a few years time?

Randall Parker said at July 11, 2007 9:40 PM:


Would a Ford Expedition or Cadillac Escalade really use only 500 watt-hours per mile?

I'd like to find a few kinds of information from reliable sources:

1) watt-hours per mile for different sized vehicles.

2) Percentages of driving trips of various lengths.

3) Purposes of miles driven split up into a few categories (e.g. commuting to/from work, vacation trips, trips to stores).

Nick said at July 12, 2007 9:00 AM:

"Would a Ford Expedition or Cadillac Escalade really use only 500 watt-hours per mile?"

Yes, I think they could for the reasons I gave, that electrical motors actually gain efficiency with size & power, where the reverse is true for ICE's, and regenerative braking greatly reduces the efficiency penalty of increased weight. The largest SUV highway MPG might be worse, because of high wind resistance.

Ultimately, these things depend on design: much of the Prius efficiency comes from careful attention to reduced wind resistance, improved tire rolling resistance, electric controls which are much more efficient than mechanical controls, reduced coolant heat loss, etc. So...you could make consumption higher if you didn't make it a priority. OTOH, there are a lot of opportunities for efficiency that pop up when you go electric, so decent efficiency is a reasonable expectation.

There is a fair amount of old information out there about EV power consumption, and it appears to be largely outdated. Li-ion batteries have a much higher round-trip efficiency than NIMH and lead-acid (90-95% vs 70% or worse), and electric motors and power electronics are improving.

Randall Parker said at July 13, 2007 10:24 PM:


I need something more than intuition on watt-hours per mile with different sized cars. If you come across some references on the web please let me know. I want to do a post on gasoline versus electricity for transportation and how the shift to electric power will affect car design.

Nick said at July 15, 2007 8:12 PM:

Well, a few thoughts. The Ford Expedition is 300HP, and 5,600 lbs. The Tesla is 250HP, and 2,700 lbs (not much more than it's sister car the Elise, due to weight savings due to the simplicity of an EV). So, based on the weight ratio, .5 kwh/mile is about right.

The shift to electric power doesn't force a change in external car design, but it does give many opportunities. You eliminate a lot of moving parts and a lot of space consumption. GM refers to their basic design of an EV as a skateboard, because it takes up so little space. You can build whatever you want on top of it. Finally, an EV is fundamentally more powerful and gives better performance than an ICE, so there are no compromises there.

The shift to greater efficiency, on the other hand, does push you in certain ways: lower wind resistance in particular, and lower size to some extent, though PHEV/EV's have a lower penalty for weight (regenerative breaking captures the energy of acceleration, regardless of mass), so you actually have a lower incentive to downsize with a PHEV.

Nick said at July 16, 2007 9:14 AM:

Randall, here's DOE/EPRI data from page 9 of http://www.pnl.gov/energy/eed/etd/pdfs/phev_feasibility_analysis_combined.pdf
This appears to include all efficiency losses. Efficiencies have increased since then, so these would appear to be conservatively high:

Vehicle Class kWh/mile
Compact sedan 0.26
Mid-size sedan 0.30
Mid-size SUV 0.38
Full-size SUV 0.46

Randall Parker said at July 16, 2007 8:32 PM:


Thank you much for finding that.

Well, at .5 kwh/mile that's only 5 cents a mile to drive a big SUV around. You could drive across the United States (about 3000 miles) for $150 worth of electricity. Am I missing something? Will high energy density batteries enable travel that is really that cheap?

Nick said at July 16, 2007 9:36 PM:

"Am I missing something? "

No, it's really that good. In fact, it's even better, as night time electricity is cheaper. Check out www.thewattspot.com and, at the right, https://il.thewattspot.com/login.do?method=showChart .

1-3 cents per night-time KWH!! $30 cross country!

Eventually, I lookforward to really cheap & high efficiency PV on cars. The average car probably has 10 sq meters of cross-section, more on an angle: that's 6 KW peak power (at 60%), and 24 MPH fully on solar power. That's a long way in the future, though. But cheap PHEV transportation? That looks very likely to be as close as 2-3 years away.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright