Every year, baseball starts up in the Spring and there are rosy forecasts for supply growth in the non-OPEC oil supply. 2007 is no exception. Wood Mackenzie’s Non-OPEC Increases to Continue in 2007 announces the good news.

The strong upward momentum for non-OPEC supply, seen in the fourth quarter of 2006 [when Dalia in offshore Angola came onstream] is expected to be maintained in 2007. The rate of increase is likely to accelerate in the fourth quarter of the year, when production is expected to be 1.6 million b/d higher than in the fourth quarter of 2006.

Total non-OPEC oil/NGL production, including Angola, is forecast to average 50.2 million b/d in 2007 according to Wood Mackenzie, up 1.5 million b/d from 2006. Patrick Gibson, Principal Oil Supply Analyst for Wood Mackenzie said “Our analysis shows that there will be significant increases in the FSU states, North America and Africa.  The main areas that will experience decline are the North Sea and the Asia Pacific region.“

Last week’s column discussed the power of declines in existing oil production using the Oil & Gas Journal (OGJ) world oil supply data along with Chris Skrebowski’s new projects database, as layed out in New capacity fails to boost 2006 production — delays or depletion? from Petroleum Review (henceforth, PR database). In accord with Wood Mackenzie’s forecast for 2007, Skrebowsky’s schedule, shown in the figure taken from PR, indicates that new non-OPEC production will increase in 2007, and each year until 2009. Thereafter, new oil production will decline steadily out to 2014.

Exxonmobil_non_opec_2The putative surge in new non-OPEC production in the 2007 — 2009 period is crucial to the world oil supply. ExxonMobil’s 2005 Edition of  The Outlook for Energy — A View to 2030 forecast that the non-OPEC share of world supply will plateau after 2010. (snapshot shown left). The IEA’s World Energy Outlook 2005 made a similar prediction. Any increases made between now and 2010 will most likely determine the high-water mark of non-OPEC oil production forever!

The apex of non-OPEC production may not be the global peak yet, but the importance of this event can not be overstated. From that point forward, the “Call on OPEC” must increase if world oil production is to show any net gains. Consequently, the security of the world’s oil supply is in greater jeopardy should OPEC production falter for any reason. Given ongoing events in Nigeria, Iraq and Iran, along with declines in Venezuela and Indonesia, there are plenty of reasons to worry. As usual, Saudi Arabia is the wildcard. A future column will address the OPEC issues.

From the perspective of eternity, delays in new non-OPEC production may not be a bad thing, in so far as postponed production will be available in the plateau or mild decline period after 2010. However, a perception problem arises if Wood Mackenzie makes overly optimistic estimates of the near-term supply additions from the non-OPEC producers. Policy makers and the public are misled into thinking that there are no serious problems with the oil supply, an atititude which leads, in turn, to complacency in the face of a unique crisis in the history of industrial civilization which is only 3 years away. Even worse, declines in existing non-OPEC oil production imply that new supply delays move the peak forward in time and lower the peak rate of production, although it may not matter at this point — it is late in the game.

The decline rate in existing production underlies the forthcoming peak (or plateau) of non-OPEC oil production. Despite Wood Mackenzie’s optimistic forecast of an average addition of 1.5 million b/d, with a surge in the 4th quarter, the Centre for Global Energy Studies (CGES)  disagrees. The executive summary of their report Non-OPEC Production: Rising Slow Output Growth — the title is a bit misleading — is worth quoting in its entirety, both for accuracy and clarity in identifying the issues.

“The oil industry is finding it harder and harder to expand upstream capacity. Development costs are up sharply, essential equipment and skilled labour are in short supply and host governments want a bigger share of the proceeds. As a result, projects take longer to complete and output is growing more slowly than predicted. In 2006, non- OPEC oil production rose by about 450,000 bpd and although this was better than 2005 — when a myriad of unanticipated problems kept output virtually flat — it was still less than expected by most industry forecasters. Once again, project delays, adverse weather, equipment failure and oil field problems combined to trim nearly 1 mbpd off non-OPEC supply growth projections for 2006 — leaving the market much tighter than expected.

While there is no doubt that the industry is working flat out to expand its production capacity in response to high oil prices, the problem is that new projects must first compensate for output declines at existing fields before they can add to overall capacity. With underlying oil well productivity declining at an average rate of 5% worldwide, the industry needs to drill enough wells to replace more than 2 mbpd of ageing non-OPEC capacity every year just for production to stand still. Any delay in starting up a new project therefore widens the gap that needs to be filled before overall non-OPEC production can begin to rise.

In recent years, a ‘rational’ approach to forecasting based on planned capacity additions and observed output decline rates for mature producing areas over-estimated the actual growth in non-OPEC supplies by an average of more than 400,000 bpd. Using the same publicly available information about new projects and oil field performance, both the CGES and the IEA made errors of the same magnitude, suggesting that the real problem is not accuracy but uncertainty.

For 2007, a ‘rational’ forecast indicates that non-OPEC oil supplies (including Angola) are expected to rise by [1.07 million b/d]. Like last year, the biggest growth is expected in the FSU, closely followed by Africa. Significant increases are also expected in North America and Latin America, and modest gains in the Asia Pacific region. Production is expected to decline in Europe and the non-OPEC Middle East countries. The CGES believes, however, that this rational forecast should be cut by 400,000 bpd to reflect the impact of uncertainty on output growth, implying a likely gain of only 700,000 bpd in non- OPEC supplies this year.

The summary highlights worrisome uncertainty about declines (and delays), casting doubt on using a “rational” approach to estimating the 2007 non-OPEC supply addition. A concrete example of such over-estimation comes from the EIA, whose analyst Michael Cohen predicted that non-OPEC supply growth would be 0.900 million b/d in 2006. The table shows some early release numbers for non-OPEC production increases for 2006 (in million b/d) made in 2007.

Centre for Global Energy Studies 0.456
Energy Information Administration 0.345
Oil & Gas Journal 0.622
Average 0.474

The IEA has also acknowledged past over-estimates in a special section of the current Oil Market Report (March, 2007) called “Clarifying the Call on OPEC”.  At long last, the IEA, “adjusting for a known trend in supply and the miscellaneous-to-balance” category, has explicity noted that non-OPEC supply additions have been later revised downward by an average of 0.350 million b/d each year. Current predictions are flagged with a warning to this effect.

CGES states that a rational forecasting approach is based on “planned capacity additions and observed output declines”. Using the EIA production data and the PR database additions, let’s do some number crunching to see what the rational approach predicts for new non-OPEC supply in 2007. Last week, a simple calculation based on the OGJ data revealed a “visible” global decline rate of 3.4%, with an overall additon of ≅ 0.495 million b/d (for crude + condensates + natural gas liquids).  Using 2005 data as the baseline and Skrebowski’s gross addition of 3.219 million b/d in 2006, the EIA data, which currently shows that global production was down 0.053 million b/d in 2006, yields a decline rate of 4%. This result is 0.6 greater than the OGJ percentage. Clearly, the fairly large discrepancies in the data sources can not be accounted for. But, let’s move on.

The PR database new capacity addition is 2.844 million b/d in 2007 from the non-OPEC side of ledger (including Angola, now an OPEC member). Again using the EIA data, it is possible to calculate the “visible” non-OPEC (as opposed to global) decline rate. Skrebowski indicates that 1.559 million b/d of new capacity was added to the non-OPEC supply in 2006. This works out to a rate of 2.56%. Applying this rate to the 2006 non-OPEC EIA production data, the rational approach yields a net gain of 1.662 million b/d in 2007 — this number is only about 10% higher than the Wood Mackenzie forecast of an additional 1.5 million b/d on average until the 4th quarter, when new supply climbs to 1.6 million b/d.

A reasonable working hypothesis postulates that what distinguishes the CGES forecast from Wood Mackenzie’s more optimistic prediction is the choice of a decline rate, although there may also be differences in the planned new capacity numbers. The underlying problem is that the decline rate in existing oil production can not be measured directly. CGES makes the reasonable inference that past over-estimates using a rational approach require an increase in the assumed decline rate, which they set at 5%. Remarkably, if a 5% decline rate in 2007 is applied to PR database capacity addition of 2.844 million b/d, the rise in supply is predicted to be 0.694 million b/d. The tally is essentially the same as the CGES estimate of 0.700 million b/d.

This number crunching exercise serves to demonstrate how analysts like Wood Mackenzie make their projections. It is not possible to determine precisely how Wood Mackenzie worked out the numbers internally, although they have left us some clues.

We have identified seven projects, led by BP’s Azeri-Chirag-Gunashli development in Azerbaijan, which will add an average of over 100,000 b/d each; all but one of which are already onstream. With the top 25 projects adding an aggregate 2.1 million b/d of capacity there is a broad base to the production growth expected.

Wood Mackenzie does take declines into account. Their press release indicates that they expect major declines in the North Sea and the Asia Pacific region. Using a decline rate just a little greater than the 2.56% number used here, it is easy to obtain their stated estimate of 1.5 million b/d if the planned capacity addition is close to the estimate given in Skrebowski’s PR database. This latter number represents an addition of about 0.744 million b/d above the figure given in the Wood Mackenzie press release. However, if Wood Mackenzie is using something closer to the lower planned capacity addition number quoted above, then they are also using a non-OPEC decline rate much lower than that which historical data would tell us is realistic.

So, we can conclude that Wood Mackenzie’s forecast is likely wrong on the high side. On the other hand, 2007 may be a stellar year for new non-OPEC production capacity additions, defying the historical trend.

The world is fast approaching a monumental historical shift. The impending peak of the non-OPEC oil supply circa 2010 illustrates the Red Queen Problem as posed in Lewis Carroll’s Through the Looking Glass and illustrated by last week’s Gator graph — one must run faster and faster merely to stay in place, let alone increase production. As CGES points out, and as we have shown here, there is no longer any historical basis for using a “rational” approach to estimate new capacity additions. Wood Mackenzie’s rosy forecast does not take the full power of declines in existing production into account. In so doing, they have sent an unrealistically reassuring message to the policy-makers and the public that peak oil is not a clear and present danger to the health of the world’s economies.