Guest post: A 10-year oil supply retrospective shows unwarranted optimism

Our guest post today comes from Steve Andrews, who is  a retired energy consultant and contributor to the Peak Oil Review, reachable at We reached out to CERA to determine its interest in providing a response, but did not hear back.

“False optimism leads to very poor investment decisions.”
–Jeremy Grantham, co-founder and Chief Investment Strategist, GMO

Ten years ago this month the Oil & Gas Journal published a story from CERAWeek — an annual elite conference for the oil industry put on by Cambridge Energy Research Associates — that bears revisiting.

Why go back? Three reasons. First, CERA arguably has maintained the highest profile of any oil industry analytical shop since at least the turn of the century, thanks in large part to founder Daniel Yergin’s reputation. Every time there is a surprise in world oil supply, he’s the media’s go-to guru. When the National Petroleum Council convenes a world oil study, you can bet the ranch that CERA will play a lead role. When the US Senate or House convenes a committee hearing on oil, CERA often sits on the panel; they also deliver some of their key research papers free of charge to all US lawmakers. Their policy-oriented footprint is large and their strategic media outreach effective.

Second, at the time CERA’s 2004 forecast of seven years of history-breaking sustained growth in world oil production capacity struck many players as being an unreasonably if not outrageously optimistic headline. How does it look 10 years later? Way off base.

Third, if CERA’s oil forecast was that off base a decade ago, should we believe the current abundant-oil story line that CERA jump started in the fall of 2011 and that has been embraced by the press and policy makers alike? So let’s look back.

On CERA’s lead panel in February 2004, Robert W. Esser, senior consultant and director on global oil resources, predicted global oil production capacity would expand by 20 million barrels/day from 2004 through 2010. (CERA doesn’t forecast production. It forecasts production capacity, which is essentially unverifiable.) That’s nearly 3 million b/d of capacity growth every year for seven straight years from 2004 onward. It didn’t happen.

Per data from BP’s Statistical Review of World Energy, actual production of global petroleum liquids grew by 5.7 million b/d during that period. Then consider the 4 million b/d of spare OPEC capacity that the US EIA shows for 2010. But there were also at least 2 million b/d of spare OPEC capacity in January 2004, at the start of the forecast period. So net, CERA missed their forecast by well over two thirds.

Note that by CERA’s definition production capacity “…eliminates economic or political factors and temporary interruptions such as weather or labor strikes.” Note too that unused productive capacity is never intentionally present among non-OPEC nations, and unused and undamaged production capacity among OPEC nations was primarily limited to Saudi Arabia, Kuwait and United Arab Emirates during 2010.

Why did CERA stumble so badly?

First and foremost, CERA underestimated decline rates from existing oil fields. About the time of its 2004 conference, an oil industry analyst who knew Daniel Yergin asked him, during an elevator discussion, what decline rate for producing fields CERA used when calculating growth in world oil supply in their major studies. Mr. Yergin replied that it would be in the 1% to 2% range.

Chalk that up as a fatal flaw. Over seven years, a decline rate of 1.5% would mean having to replace only 8+ mbd of production capacity. It’s ironic that by late 2007, in what CERA called a groundbreaking study, they calculated the actual decline rate from 811 of the world’s major oil fields at 4.5% per year. Over those same 7 years, using their 4.5% decline rate would require 23 million b/d of capacity just to keep production flat. IEA estimates for decline rates rank even higher than CERA’s.

On the production side, CERA spoke optimistically about projected gains from the Gulf of Mexico, West Africa, Brazil, the Caspian area, Canada, Venezuela, Iraq, Nigeria, Algeria, Ecuador, Sudan and Russia. Indeed, production increased in 11 of those 13 nations for a net gain of 6.7 million b/d. But on the bad news front, the rest of the world lost 1 million b/d of oil production during CERA’s forecast period, hence the 5.7 million b/d net gain. Declines badly undercut forecast gains.

On the demand side, CERA actually worried that “should this spurt in output exceed projections of a very large increase in world oil demand this decade, then persistent downward pressure on oil prices might result.” For the record, when CERA made that comment, oil prices were upward of $30. But while nearly everyone was wrong about oil prices a decade ago, CERA was also wrong about the key demand driver: China. CERA forecast that China’s demand growth for oil would slow to 5% in 2004, compared to what eventually occurred: record-breaking growth of nearly 17%. And while CERA talked about volatility in Chinese demand going forward, China’s record-setting growth rate for oil demand continued throughout CERA’s 2004-10 forecast period.

If this was a personal forecast which I had blown this badly (and I’ve blown a couple), no one would notice. But this enormously flawed vision was widely circulated. CERA gets press coverage, but the press isn’t checking CERA’s track record, and this 2004 prediction is just the tip of the iceberg.

In 2005, CERA dialed back their optimism only slightly. They projected world oil supply capacity still growing 16.4 mbd from 2004 through 2010 — a reduction of 3.6 mbd from their original forecast. They projected world demand in 2010 at 94 million b/d, leaving 7.5 million b/d of idle capacity. Note this comment about related impacts on prices: “We generally expect supply to further outpace demand growth in the next few years, which could result in oil price weakness around 2007-08 or thereafter.”

In 2006, CERA projected potential world oil capacity growth of 21.3 million b/d — from 88.7 million b/d in 2006 to 110 million b/d in 2015. We’re less than two years away from year-end 2015, yet total petroleum liquids production will likely run in the 90 million b/d range. Clearly, CERA’s was an exceedingly Pollyanna-ish view, rather like a best case in a perfect world.

Now square CERA’s long-standing optimism bias on future world oil supplies with the recent spate of sobering news from Wall Street on the financial travails of the large investor-owned super-majors. Mark Lewis has done a nice job highlighting the near tripling of oil and gas capacity expansion costs since 2000 — from $250 vs. $700 billion; three quarters of that was spent on oil, yet oil supply rose only a modest 15% (BP data). Increasingly blunt reports from analyst shops like Sanford Bernstein add to the growing contrarian chatter. Solid coverage in the UK of Richard Miller’s recent paper “The Future of World Oil Production” opened a few eyes about limits. But those voices still have a steep hill to climb.

That’s in part because, starting in September 2011, CERA went on the offensive as chief cheerleader of an overly optimistic, US-led oil abundance storyline. It features the US’s record-breaking shale oil bonanza — an amazingly successful yet term-limited reality. Viewed at the global level, for the last few years the brouhaha about our shale oil bonanza has been the tail wagging the world oil supply dialogue.

CERA’s oil supply predictions should have earned deep skepticism from the press and policy makers. That hasn’t happened yet. It’s overdue.

But please keep the larger backdrop in mind: Without a serious revisiting of the questionable optimism that dominates any dialogue related to longer-term world oil supplies, without a harshly realistic scrub of the facts, we face unnecessarily large energy policy risks.

Share this:
Facebook Twitter Email

All blog comments are moderated before being published.


  1. Johne168 at August 3, 2014 10:31 am

    I really like your writing style, good information, thankyou for posting D. kagbededcddd

  2. Nate Hagens at February 23, 2014 9:50 am

    Humans are part of nature. In nature organisms and ecosystems self-organize so as to better access available energy gradients. Since around the dawn of agriculture, individual human wants and needs have been suppressed for growing surplus for the ‘hive’ that is greater society, and now civilization. The cost of high quality energy is our true cost of capital. As aggregate EROI declines, ‘the hive’ can still maintain/slightly increase gross energy production by adding more credit to the system, which has a catabolic effect on existing infrastructure – in effect we are living on marginal energy return as opposed to fixed/total. Money doesn’t create energy, but it does allow us access to more of the remaining cost-tiers, faster/sooner, and in the process disguised the limits by hiding the effects.

    From this (anthropological) perspective, those in this thread and in the broader Peak Oil community are operating on a flawed assumption. The system isn’t broken. Its working perfectly. Who is this crazy fringe group that is trying to move the hive away from rich feeding grounds of un-oxidized fossil carbon. That an organization like CERA would emerge and remain as a spokesperson for rationalizing continued access to our (declining) energy gradient is exactly what one would expect under the Maximum Power Principle. Because to listen to us (in aggregate) it means we make decisions to move our policies and infrastructure away from fossil magic before the magic goes away on its own accord. This is possible but extremely unlikely given our genetic leashes and ultrasocial cultural makeup. CERAs pronouncements rationalize what the majority of people with status and accomplishments want to hear. Cognitive dissonance too, has been adaptive

    • Vanjesh at April 19, 2014 8:21 am

      This is a valid concern. Natural gas and peeroltum have had boom-and-bust cycles since they became important fuels around 170 years ago.It would be a typical progressive outcome if the green pressure on coal led to a shift to natural gas, and then the progressives made fracking almost impossible and helped raise the price of natural gas. Or is that suspicious of me?

  3. John Hallock at February 20, 2014 8:19 pm

    Again Steve Andrews – nice job. As David Hughes mentioned here, Adam Brandt mentioned in his excellent review of oil forecasting models a few years ago, and Charles Hall has said for years in his classes and publications, model validation and testing are essential to improving model performance and to garnering credibility. Validation indicates how seriously one should take the results of a model, and if it needs to be modified. I had the honor of publishing on this very topic recently. We revisited a set of conventional oil model scenarios originally published in 2004 vs. 11 years of history with the express purpose of seeing which were still realistic. I recommend giving it a look since it is on topic.
    Hallock J, Wu W, Hall C, and M Jefferson. 2014.
    Forecasting the limits to the availability and diversity of global conventional oil supply: Validation. Energy (64): 130-153.

    12th paper down. It is open access, so no need for a journal subscription. The results to this point vindicate Campbell’s and Laherrere’s arguments made years ago. Many of you are not surprised, but this shows it systematically and with hard data.

    While I have great respect for your work on oil price carrying capacity Steve (Kopits), I differ a little with you on the merit of simply moving on and forgetting what certain forecasters said so vociferously years ago. It is not a matter of being sour grapes or mean spirited about things. Not at all. It is about trying to make sure that those making decisions listen to the right people. All else being equal, I would choose to listen to those with a good track record. If CERA were a Wall Street stock analyst, would you continue to seek out their advice????

  4. John Theobald at February 20, 2014 1:59 pm

    It’s absolutely true that the unconventional is going mainstream . . . which given our impressive new technology is exactly what one would expect when so much conventional production poops out every year and needs to be replaced by something. Of course, that something increasingly is a product that is more difficult and expensive to produce than the $20 oil we were burning 15 years ago. The race is on, and nobody knows how or when it will end. It would seem wise to second the Andrews call for more transparency, facts, and analysis.

  5. Steven Kopits at February 20, 2014 1:29 pm

    Well, what a panoply of commenters, not to mention the illustrious Mr. Andrews. A who’s who of peak oil.

    Although CERA was clearly wrong in its forecast, I think events have moved on, and critical issues are emerging elsewhere.

    It is certainly true that shale oil and oil sands production is up since 2005. However, the conventional system as it existed in 2005 also peaked in 2005, that is, everything excluding US shale oil and Canadian oil sands is producing less today than it was then. It can be debated how long shale can continue to deliver 1.2 mbpd / year growth, but probably not more than another two years or so, if the business is to be limited to the US.

    Meanwhile, the business model of the IOCs is imploding, with capex cuts and divestments throughout the industry. It is this which is the news, and where our focus should be now. Layoffs have begun, and we should expect a 20% decline in IOC capex, -$80 bn, and 200,000 jobs losses in the oil field services sector in the next 2-3 years, with the impact concentrated on Houston, Aberdeen and Stavanger (Norway). This is Peak Oil Phase III, where consumers are no longer willing to underwrite higher oil prices even as E&P costs continue their remorseless rise. The IOCs are being forced to abandon certain high cost projects and focus on near term cash generation.

    I think it is this disturbing trend which is bringing the peak oil analysts out of the woodwork. I would add that data sources like the EIA are not so sanguine. Take out shale oil and oil sands growth from the EIA’s Short term Energy Outlook, and you’ll find a pretty sober assessment of future prospects. Indeed, the outlooks of, say Exxon or BP, are not particularly rosy either, in terms of supply.

    I do not view the matter as primarily one of us-versus-them, but rather as an evolving narrative.

    For a more detailed review of this topic, please see the link below for my talk on the matter at Columbia University last week. Min 11-17 reflect on the current state of supply, and min 40-50 reflects on the situation of the oil majors. I personally think some of the best points are after min 58, during the Q&A.

    • Ussy at April 18, 2014 3:33 pm

      Repeating the same gibberish in conmmet thread after conmmet thread does not make it true. Their are NO costs associated with the production of shale gas – NONE! It’s currently a byproduct associated with the production of crude oil, ethanes and other liquids which are extremely profitable:…Try looking at the 10-K forms filed with the SEC and tell me that there are no costs. In the summer of 2010 the CEO of Devon was telling investors that the break-even price was between $6 and $7. Aubrey McClendon was telling investors that if he had his way he would only be running a drill or two because he could not make money at the low prices that his company was getting. (They are lower this year.) I just did a Google search and found that is making the same points. If you go to the Oil Drum and a few other places you will find exactly the same thing. The conference calls are certainly not very cheery. The shale companies keep talking about asset sales and funding gaps even as they gloss over results that show substantial borrowings and negative cash flows. All you have to do to prove me wrong is to cite the operational results from shale oil and gas by the producers. No hype please. Look at the 10-Ks for confirmation, not empty words by promoters. continued…

  6. Nick Grealy at February 20, 2014 6:47 am

    All this proves is that some peak oil people, conventional by definition, just simply don’t get what was once called unconventional and is now going mainstream.

    Nothing can convince those who simply don’t want to be convinced, including it seems, actual production numbers.

    Whoever commented that only good news sells, has obviously never read a UK newspaper 😉 which explains why the tin foil hat peak oil brigade is still taken seriously in the UK. To quote Grantham is ridiculous: he’s only talking his green investment book.

    Beat up CERA and you beat up the EIA as well:

  7. Charles Hall at February 20, 2014 12:22 am

    Right on Steve Andrews, although you are only doing what the business press should have been doing routinely. What happened to fact checking? Another issue, ultimately of perhaps even greater concern than the failure of a tripling in investments in the past decade to move the global production needle much, if any, is the decline in EROI (Energy Return on Investment). As any oil person knows it is getting more and more expensive to get the next replacement barrel of oil. What is less commonly known is that it is more and more expensive in terms of energy too, so that EROI has declined from 30:1 or more in the past to 5-18:1 today – depending where you look – as depletion trumps technology. A linear extrapolation suggests that in much of the world it might take a barrel of oil’s worth of energy to get the next barrel of oil within a decade or two. At that point the oil age, responsible for our enormous present wealth, will end.

  8. Chris Kuykendall at February 19, 2014 3:59 pm

    One could say that CERA has been getting a free ride for years in its analyses, except that they get paid a lot for what they do, so it’s not really free. The too-obedient population of mouth repeaters who echo CERA conclusions need to pause occasionally to explore what post-audits of CERA forecasts would reveal. CERA’s non-responsiveness to an invitation to comment on Andrews’ piece is telling. It’s indicative of non-engagement in the nitty-gritties of oil supply debate raised by CERA critics. CERA resembles a top-ranked basketball team who expects and/or is given a series of byes through successive playoff rounds without having to demonstrate whether they’re really any good or their reputation deserved. The most that’s positively verifiable about the CERA team is that they have great cheerleaders. Policy wonks should rely on scoreboard results, not the cheerleading.

  9. Vince Matthews at February 19, 2014 3:13 pm

    Energy policy? What energy policy?

    Thanks to Platts and Andrews for having the fortitude to step up against the CERA machine. If shale resources are so good, one might ask why Shell decided to exit the shale plays.

    And, it isn’t just production to be looked at in evaluating the rosy scenarios, even if CERA’s forecast had validity different than their track record.

    On the global scene, the pie of oil supply is increasingly affected by countries taking bigger slices for a variety of reasons, e.g. China because of GDP growth; Indonesia, Egypt, and UK because of less exports and increasing percentages of domestic consumption.

  10. Richard Miller at February 19, 2014 2:27 pm

    CERA doesn’t specialise in talking truth to power – that’s not its role, only its disguise. Its role is to make money for its shareholders. And they’re not alone in horrible past forecasting; the IEA for example has no better a track record in estimating future prices. Their current forecasts await the test of time, of course.

  11. David Hughes at February 19, 2014 12:36 pm

    Good news sells. Few people bother to do a retrospective like this excellent piece. Attention spans in the information age last a few hours or a few days, not 10 years. The risks of a business model biased on the optimistic side making people (clients) feel good are few, judging by CERA’s success. However the risks of relying on wildly optimistic projections of a crucial energy commodity like oil without a harshly realistic scrub of the facts, as Andrew’s points out, are many.

  12. Mason Inman at February 19, 2014 12:12 pm

    Kudos to Platts for running this piece, in the interest of generating a healthy dialogue.

    I hope Platts will also do more in the future to look back over old forecasts. Journalists who cover energy rarely write articles that look back over past forecasts to see who was right and who was wrong—and why.

  13. John L Langhus at February 19, 2014 12:00 pm

    Mr. Andrews’ is a welcome sober-minded voice amid the uninformed chatter of those proclaiming the latest game-changer in energy. The real game-changer is the stubborn resilience of the stall in oil production increases in the face of unrelenting declines in major fields, despite monumental financial and technological investments in generating new production. The “shale gale” has been revealed to be profitable only as an OPM play (i.e., Other People’s Money). While its downward effect on consumer prices has been the prime contributor to the US’ comparatively better recovery than Europe’s, the cost has been the decimation of dozens of small and mid-sized producers whose production from shale operations looks increasingly unlikely to ever generate a real return on capital. The time to reexamine our energy situation is well-past.


Your Comment