Today Econbrowser is pleased to host this guest contribution from Steven Kopits, who heads the New York office of Douglas-Westwood, energy business consultants.
EIA: Hard Core Peak Oil Forecast
by Steven R. Kopits
The EIA has gone hard core.
The EIA, the statistics arm of the US Department of Energy, recently released its International Energy Outlook (IEO) for 2010. This is an important document for forecasters, as it represents the EIA’s integrated view of the global energy markets in the years to come and contains a long term forecast on the range of energy sources and CO2. Like it or hate it, the IEO is a touchstone for the energy industry and is treated as the authoritative government forecast in the press and in capital raising documents like prospectuses. It influences policy-makers, the media, public opinion and investors. What it says matters.
And what does it say?
That peak oil is all but on us. And that’s new.
As recently as 2007, the EIA saw a rosy future of oil supplies increasing with demand. It predicted oil consumption would rise by 15 mbpd to 2020, an ample amount to cover most eventualities. By 2030, the oil supply would reach nearly 118 mbpd, or 23 mbpd more than in 2006. But over time, this optimism has faded, with each succeeding year forecast lower than the year before. For 2030, the oil supply forecast has declined by 14 mbpd in only the last three years. This drop is as much as the combined output of Saudi Arabia and China.
 |
But that’s not the interesting part. The more salient development is the reduction in the forecast to 2020. Forecasts beyond ten years are highly uncertain and more subject to massaging and interpretation. Shorter term forecasts are more definite, and the forecasters are more accountable. As a consequence, the outlook for the short to medium term warrants greater attention. In the case of the EIA, the forecast changes most dramatically here.
For the remainder of the decade, even though China would be expected to hit its stride for increased oil demand, the EIA sees no year in which liquids production will increase by even 1%. Petroleum liquids supply increases by an average of 0.6% per year from 2011 to 2020. In other words, the EIA is expecting the oil supply to be essentially flat for the rest of the decade. The supply will creep up from 86 mbpd today to approximately 92 mbpd to 2020, but that is not much growth, and indeed, is about the same as current global liquids production capacity. Moreover, it represents a reduction of nearly 4 mbpd from last year’s forecast for 2020. On paper, the output of China has disappeared over the course of the last year.
In its forecast, the EIA, normally the cheerleader for production growth, has become amongst the most pessimistic forecasters around. For example, its forecasts to 2020 are 2-3 mbpd lower than that of traditionally dour Total, the French oil major. And they are below our own forecasts at Douglas-Westwood through 2020. As we are normally considered to be in the peak oil camp, the EIA’s forecast is nothing short of remarkable, and grim.
Is it right? In the last decade or so, the EIA’s forecast has inevitably proved too rosy by a margin. While SEC-approved prospectuses still routinely cite the EIA, those who deal with oil forecasts on a daily basis have come to discount the EIA as simply unreliable and inappropriate as a basis for investments or decision-making. But the EIA appears to have drawn a line in the sand with its new IEO and placed its fortunes firmly with the peak oil crowd. At least to 2020.
We’ll see how it plays out. But for now, the EIA appears to be making a statement. Perhaps we should sit up and pay attention.
Has any long term forecast for ANYTHING ever been correct, other than by chance? Extrapolation of trends never seems any good beyond the very near term. Creating long term forecasts, thinking about long term forecasts, acting on long term forecasts is all a big waste of time.
2020 ? That’s EIA nonsense.
Unfortunately, we all let the corporations (foreign and domestic) spend as much money as they want to influence the elections by allowing limitless spending on propaganda. Compounding the problem is the fact that the American people have been dumbed down enough to allow 30 second commercials to drastically influence their perception of the truth.
I’m not sure you can say it shows ‘peak oil is upon us’ when by 2020s it has oil production rising as fast as ever. The shortfall in the 2010s is new, until now the forecasts look to be the same just pushed out a few years after a bad 2000-2009 period. But surely towards the end of this period they have it growing stronger (in % terms almost certainly, but looking at the curve even perhaps in absolute terms) than ever in the period shown?
It’s also worth note that most of the liquids production increase from now to 2020 is projected to be unconventional in the IEO. Most of this is biofuels and oil sands. They REALLY ARE projecting flat oil production.
This IEO forecast actually shows U.S. per capita energy consumption declining faster on a percentage basis (-0.5%/p.a.) than any other part of the OECD! Tee-hee.
It also appears transportation is projected to continue to be virtually totally fueled by liquids, with per capita energy use in transportation declining at the same 0.5%/p.a. of overall consumption.
“As recently as 2007, the EIA saw a rosy future of oil supplies increasing with demand.”
The EIA still does (supply will always increase with demand). It’s sort of stupid to suggest otherwise.
What about predictions for Peak Demand?
Not to be too much of a stickler, but couldn’t their forecast changes (in the short term) over the last four years simply be due to *demand* slack caused by the recession? The 2009 “actual” level of production still shows a spike immediately afterward. It seems to me they have to factor demand changes and economic recovery in their forecasts. Although the ability of demand to drive supply changes is constrained it isn’t negligible IMO.
Talk about having it both ways – note the comments about the EIA forecast in the first and penultimate paragraphs:
“…[the EIA] influences policy-makers, the media, public opinion and investors. What it says matters.” AND “…those who deal with oil forecasts on a daily basis have come to discount the EIA as simply unreliable and inappropriate as a basis for investments or decision-making.”
Long time peak oil follower. Does anyone think the timing of this new report “confirming” peak oil a little strange and convenient considering that BP is under fire for the spill? This report would seem to give producers a “Now we really need all the oil producers we can get” argument plus boost stock of any producers. Too cynical?
I am quite surprised to see this analysis by Kopits. It seems tossed off, and without the depth he is capable of. The EIA has once again conflated liquids–so the best analysis is one that addresses that obfuscation. Without that analysis first, the chart displayed here is without much value.
The inability to increase production since 2005 with oil prices averaging much higher than pre 2005 levels is consistent with these EIA estimates.
Steve, how much of the change in the forecast is due to a change in the model rather than a change in the inputs? In other words, if we used the same inputs as last year, would we get the same forecast as last year?
EIA changes the nature and structure of its forecasts all the time. If you look at a report by EIA’s own Renee Miller, who was the only one to look at EIA forecast errors, you will see this in detail. (Yes, of course, EIA calls their forecasts “projections”, but this is just another name for conditional forecasts.) So this isn’t that big a deal.
I haven’t checked if the situation has changed since 2009, but it used to be that EIA (as opposed to IEA recently) does only demand based forecasts.
Which means that they very little of actual upstream constraints go into modeling their future production forecasts.
That is a sad state of affairs.
To believe, that if there’s a demand, price will meet to produce enough quantity by the upstream industry and the downstreamers will somehow get it delivered, regardless of where it’s in the world, in what form, where it needs to be distilled into end products and where the end users are.
Further, they base most of their models on R/P ratios still, which is a disgrace to anybody who understands the difference of geological constraints for flow vs stock models.
But then again, I don’t get paid to be wrong, like the EIA people do.
They have been consistently *more* wrong for the past decade or so.
Why anybody listens to what they have to say, is beyond me.
Rubbish.
I don’t know what this guy is capable of, but in the words of the famous scientist, this analysis “isn’t even wrong” … at least based on the data shown.
What I see as an outsider is that each year the EIA has consistently predicted reasonable more-or-less linear growth for the next 2 decades. I see them adjusting the “start point” for the predicted trend downward each year to match the prior year’s reality. I don’t see any “peak” in the predictions posted.
More interestingly, for each of the last few years the EIA’s forecast has, within a year, been proven substantially wrong, not just in magnitude of change but even in the direction of change.
Accountability?? The method being used clearly needs improvement.
Given that the EIA cannot predict next year’s oil production changes, and that Kopits sees “peaks” on charts which show no peaks… why should one believe any of this? From either side? Much easier to predict that “next year will look a lot like last year”.
First, let me thank Jim and Menzie for the opportunity to post. To my mind, Econbrowser is the premier economics blog, so I consider it an honor to post here, whatever the shortcomings of my analysis.
To answer of couple of comments:
John Smith:
You’re correct. Long term forecasts are usually junk (although, I have to say, Douglas-Westwood forecasts in volume terms have tended to hold up). The issue, however, is mindset. Do we have an oil supply issue or not? If I’m a generalist policy maker, say, a Congressman, then I’ll check with the forecaster of record to see. That’s the EIA. Right now, the EIA appears to be saying we have a problem. If we have one, then maybe looking at potential solutions is a good idea. Maybe it should be on the political and legislative agenda. So, to my way of thinking, that’s the importance of the EIA forecast. It’s suggesting we have a problem.
Matthew and Bryce:
I see the IEO as a negotiated document that looks to straddle the gap between the technical analysis and political pressures. In recent years, the EIA has tried to avoid making waves; I believe that’s why its forecasts have been so weak. So I read the IEO as a compromise document: harder on the short end; while retaining softness on the long end. But it’s pretty hard on the short end. It’s actually taking a position. It appears that EIA Administrator Richard Newell deserves some credit for this. It feels like he’s giving the organization some teeth. But, of course, it makes no sense to forecast the oil supply increasing by 0.6% annually from 2011 to 2020, and 1.3% thereafter. The long stuff is still fluff; but the next ten year forecast is truly harsh.
I think the difference at the EIA is a change in management, not model. The motivation for the change can be two-fold: i) the analysts are gaining the upper hand, or ii) the administration is setting the stage for a post-oil energy strategy. I would prefer it is the former.
Vak001:
Jim Hamilton and I have both written about demand trends. China looks like a monster lurking on the horizon, so I don’t see any realistic peak demand on a global scale. I do, however, see the re-allocation of oil consumption from the OECD to the non-OECD countries. That started around 2006, and it’s continuing. The April Short Term Energy Outlook numbers (Table 3a) are almost a textbook example of this. A version of my article, Debunking Peak Demand, can be found here: http://www.aspousa.org/index.php/2010/04/debunking-peak-demand/
“As recently as 2007, the EIA saw a rosy future of oil supplies increasing with demand.”
To me, this was not rosy. Just about as bad as the recent come-back in demand for SUV’s by U.S. consumers, as gas prices have fallen.
Good observations, Steven. The EIA does seem to be following in the IEA’s footsteps as it slowly backs into the ugly truth about future oil supply. Like you, I have observed and chronicled their falling expectations for years. Even so, I believe their current outlook is still too rosy. I will be surprised if global “all liquids” supply ever breaks 90 mbpd.
I similarly critiqued the early release of the EIA’s 2010 Annual Energy Outlook back in December, and concluded that the EIA should be statutorily barred from making forecasts.
Should I be concerned that the change in the EIA forecasts from ‘rosy’ to ‘grim’, are consistent with the administration’s views on global climate control and the views of the newly appointed (AUG 2009) EIA administrator?
Are EIA forecasts supposed to be objective? Or is there one of those disclaimers “may reflect the views of …”
I just googled Ricard Newell to see if his research had a slant to it. I am sure you can go to the source, but here is a synopsis,
http://www.whorunsgov.com/Profiles/Richard_G._Newell
Don’t get me wrong. I agree that peak oil and climate change are important topics. And, I don’t want to make a blanket statement that all government statistics are biased. However, in this particular case, we know congress is going to point to the new EIA report as the climate bill is debated over the summer. We know the view the administration is promoting. We know the views of the new EIA administrator, Richard Newell. We see the change in perspective from ‘rosy’ to ‘grim’ in the first annual report under Newell.
You’re right, I’m sure it’s all coincidence.
I don’t see any point in trying to project quantity without also projecting price. And the EIA does project price: reference case, rising above $100/barrel soon, then more slowly towards $135 by 2030. And as their quantity forecasts have been revised downwards in recent years, their price forecasts have increased — as recently as 2007, they were forecasting $59/barrel in 2030.
The way to think about peak oil is the movement of the supply curve over time. Historically, the curve shifted to the right — more people willing to supply oil at any given price — as technology improved and the geographic reach of the oil business expanded. Naturally the demand curve also shifted to the right with growing world population and wealth. Thus the point of intersection shifted to higher volumes, more-or-less constant prices, over the last 100 years or so.
But now the easy oil has almost all been found and tapped, and the supply curve is stalling, soon to shift back left — fewer people willing to sell oil at any given price. With the demand curve still shifting right, we can expect rising prices and more-or-less constant volumes.
My own guess is that a price above about $150 will be sufficient to keep supply from falling in the medium term (old fields will be worked over, smaller or lower-quality fields will be worth developing, unconventionals will be profitable) and to keep demand from rising (conservation and switching to either electricity or CNG in the developed world offsetting growth in the developing world).
I assume after 2020 we will be manufacturing it as opposed to discovering it. Liquids are highly convenient storage even if not natural.
I thought “Limits to Growth” more than 40 years ago showed the limitations of straight linear projections. Supply and demand will be very interactive thus modifying peak oils general impact on demand. That said petroleum demand is relatively inelastic precisely because it is such a high quality chemical – despite being frequently wasted on gas guzzling vehicles and similar wasteful uses. Ultimately, energy efficiency and conservation is the only real future but nobody ever wants to talk about that.
a silver lining in this cloud. If we really are at the oil peak, then we can stop worrying about AGW.
“CalTech professor Dave Rutledge … Cal Tech Chair for the Division of Engineering and Applied Science, says the UN Intergovernmental Panel on Climate Change … models for future climate change assume fossil fuels supplies available to raise atmospheric CO2 which overstate future hydrocarbon burning by a factor of 3 or 4 or more.”
http://www.futurepundit.com/archives/004350.html
Let me comment on the EIA. In a number of reponses here, the EIA has been much derided, and rightly so, based on its track record in the last five years. But to accept that it should be prohibited from forecasting or otherwise ignored is to accept a vision of ourselves and our country as limited, flawed and incapable of improvement. I reject that view.
Instead, this should be the mission statement of the EIA: “To be the premier and unrivaled data gatherer, analyst and strategic forecaster of global energy markets”
“Premier”: the best, meaning the most comprehensive and usually the most accurate
“Unrivaled”: acknowledged as the best and willing to demonstrate and defend its competence (for example, on blogs like this)
“Gatherer”: charged with obtaining all necessary information from around the globe
“Analyst”: converting data into information
“Strategic forecaster”: long-term. First and foremost, the EIA should drive policy-making, and that will tend to be strategic. I could have included “tactical” here, but there are a number of very, very good traders in this business. If you can beat Andy Hall over at Phibro in predicting short term oil market movements, then you should work for him, not the EIA,and increase your salary by a factor of 20.
“Global”: global, not just US.
The EIA already performs a number of these functions, and pretty well. For a bureaucracy, remarkably well. What it lacks is independence and aggression. The IEO is important to me, because for the first time in years, the EIA appears to be showing some aggression. It is now firmly placed at the low end of the oil supply forecasts to 2020. This gives the organization an opening to be more direct in its communications and start providing some real input into policy-making, rather than just glossing over challenges in the oil supply.
Consider: In the recovery from the last recession, oil demand increased by 7 mbpd in the following three years. Today, after a deeper drop in oil demand and with China far readier to take oil in quantity, global spare production capacity is 5 mbpd. So, taking the simple scenario that this recovery should be like the past, well blow through all of global spare production capacity by 2012, and probably not later than 2014, with an oil shock, recession, or war all possible outcomes within that time frame.
Where is the EIA in all this? It should be out front talking about the simple math, even if that is disturbing. This means being analyst and data-led. If the Obama administration wants it to be fun to work for government again, this is the way forward. Let the technocrats do their job and the chips fall where they will.
H/T to Chris Nelder. His work is well worth reading, for those interested in oil topics.
Amazing how people think oil will magically appear if the price is far enough. Even this EIA forecast is among the more “rosy” analysis available at this time. Are you folks aware that the US military is predicting a global shortage of potentially 10mbpd by 2016?
http://www.jfcom.mil/newslink/storyarchive/2010/JOE_2010_o.pdf
Yes there were many underlying factors to the economic crash of 2008 but oil price is what blew down the house of cards. This was the first in what will probably be a series of spike/crash cycles down the backside of Hubbert’s curve.
From what I remember, Hubbert predicted the U.S. peak oil scenario correctly (around 1971) as well as global peak oil by 2005:
http://www.youtube.com/watch?v=ImV1voi41YY
i.e., 1995 + 10-year shift due to an OPEC increase in production.
So far, global production has been relatively flat since 2005.
With regards to demand, factor in a 3-pct drop for OECD countries given economic crisis and 3 to 7 pct monthly increases in car and appliance sales in China, India, and other countries.
Finally, I think the U.S. military reports shortage between 2012 and 2015 or ’16.
Professor Hamilton,
If your theory of oil prices driving contractions is accurate shouldn’t lower oil prices be driving economic growth?
And if peak oil is a problem shouldn’t we be seeing the price of oil increasing faster than inflation?
My prediction:
The number of economists working on “Moore’s Law for Petroleum Engineers” will double every 1.5 years.
@Anonymous (at June 3, 2010 06:22 AM),
Bingo
Steven Kopits wrote:
.
“Petroleum liquids supply…will creep up from 86 mbpd today to approximately 92 mbpd to 2020, but that is not much growth, and indeed, is about the same as current global liquids production capacity.”
.
This is incorrect. These figures apply to USE, not supply:
.
“World use of liquids and other petroleum grows from 86.1 million barrels per day in 2007 to 92.1 million barrels per day in 2020, 103.9 million barrels per day in 2030, and 110.6 million barrels per day in 2035.”
.
http://www.eia.doe.gov/oiaf/ieo/highlights.html#2
.
Note also that “liquids and other petroleum” includes liquified natural gas (LNG) — see footnote #2 in the report — and that the report expects LNG production capacity to increase 240 percent by 2035 with significant gains in the next decade:
.
“World natural gas trade, both by pipeline and by shipment in the form of liquefied natural gas (LNG), is poised to increase in the future. Most of the projected increase in LNG supply comes from the Middle East and Australia, where a number of new liquefaction projects are expected to become operational within the next decade. In the IEO2010 Reference case, world liquefaction capacity increases 2.4-fold, from about 8 trillion cubic feet in 2007 to 19 trillion cubic feet in 2035. In addition, new pipelines currently under construction or planned will increase natural gas exports from Africa to European markets and from Eurasia to China.”
.
Emil: “Natural gas liquids” are included in total liquids. That’s the propane, butane, etc, that also comes out of a gas well and is separated from the methane during processing. “Liquified natural gas” is different, gas (methane) that has been condensed under high pressure to be transported by tanker.
“Are you folks aware that the US military is predicting a global shortage of potentially 10mbpd by 2016?”
At what price? We had oil ‘shortages’ in the ’70s, too, but they went away when the price controls did.
Don,
Maybe the US military won’t be able to afford it?
BTW, The Pentagon was a proponent of CTL (coal to liquids) as a way to guarantee National Security.
Which reminds me, w/CO2 capture, CTL in the US was estimated to be competitive with $100 oil. And you can make mostly clean diesel and/or jet fuel, tho you do get other saleable by-products.
Also, for about 10 years now, Norway has been sequestering CO2 in depleted oil wells in the North Sea. The low temps and high pressure down there compress and liquefy the gas and it stays put very reliably.
But we don’t do that here. Better to fret about unemployment, I guess.
@Lord, Re:
“I thought ‘Limits to Growth’ more than 40 years ago showed the limitations of straight linear projections.”
This may be the mythology, but this early-1970s study was based on nonlinear system dynamics (lots of stocks, flows, feedback loops…), and tested a range of parameters to look at general patterns of system behavior. Conclusion: the natural modes look like overshoot and collapse.
They had a standard best-guess reference case, and it looks a lot like what has actually happened to date.
”
highly uncertain and more subject to massaging and interpretation. Shorter term forecasts are
”
We are now left with few choices. Return to primitive life of hunt, scavenge, and die from disease and murder? Government led death squads? Sterilize with vasectomy every newborn male from every country? But not in that order. Sooner we start — easier transition will be.
lilnev, it also includes “gas to liquids” (e.g., liquid natural gas). Here’s the full footnote from the EIA report”
.
“Liquid fuels and other petroleum include petroleum-derived fuels and non-petroleum-derived liquid fuels, such as ethanol and biodiesel, coal-to-liquids, and gas-to-liquids. Petroleum coke, which is a solid, is included. Also included are natural gas liquids, crude oil consumed as a fuel, and liquid hydrogen.”
.
World oil production will never exceed the 88 M Barrels / d achieved in 2008. World oil production will have significantly declined by 2020. As somebody else mentioned here, M. King Hubbert predicted the US oil peak accurately within a few years and the JOE 2010 (Joint Operating Environment report)mentions a potential gap between demand and supply by 2015. I think the gap is already starting now.
The U.S has held its production almost flat the last few years by aggressively pursuing deep sea oil drilling as production from the mainland has decreased. About 35% of U.S oil production is now from deep sea drilling (source: EIA). Deep sea drilling is such more energy and resource expensive than drilling on land, but that’s where they went to hold output relatively constant. Deep sea reserves usually deplete fairly quickly.
Oil discoveries worldwide (by reserve estimates) peaked in 1964. Since then the size of discoveries has decreased by an order of magnitude. Oil is now on its way out.