Peak Oil News: BP Amoco's magic curve

Tuesday, May 17, 2005

BP Amoco's magic curve

Andrew McKillop

Click read original and view figures and tables oil industry commentarist Andrew McKillop writes: In its ceaseless quest to present world supply-demand trends and factors as favouring Cheap Oil, BP Amoco and other consumer nation ‘players’ in the world oil market, formerly nicknamed the ‘Seven Sisters’ but now shrunk to 5 Anxious Dwarfs, necessarily has to provide apparent logic for this quest.

On the supply side, oil consumer nation agencies such as the OECD’s IEA and the US EIA, latterly joined by OPEC and Russian oil sector forecasting institutions, present ‘potential major increase’ in world oil supply as being feasible and likely, at least through 2005-06.

This has enabled some analysts to claim there can be a return to ‘structural oversupply’ as in the 1986-99 ‘cheap oil interval’, with recurring, if weakly credible supporting claims of ‘OPEC overproduction’.

* The underlying rationale is that world oil supply will always tend to increase faster than world oil demand, and that the OPEC group will always have unused production capacity.

On the demand side, to lend credibility to this fundamentally impossible situation (see below), BP Amoco has created what can be called its ‘Magic Curve’. This purports to show that world oil demand growth is always trending downwards. Over the 10 years 1994-2004, for example, the curve produced by researchers inside BP Amoco (see below) shows that world oil demand growth would shrink by about 65%. Complete zero growth of world oil demand would be attained in about the year 2007.

This can be called an absurd and preposterous fantasy.

It could also be called a belief in magic -- or a way to admit the certainty of Peak Oil without saying it out loud !

In the real world outside the realms of magic, both on the supply side and on the demand side, it is clear that any temporary ‘oversupply’ of the market will be precisely that – temporary. This is for a large number of reasons. The real trend for oil discoveries, replacement to lost capacity from both economic and geological depletion, and the net energy available from the overall effort to produce and deliver oil (and gas) will and can only fall. World oil demand, in the absence of intense economic recession or long-term energy transition, can only rise. Underlying the BP Amoco Magic Curve, which can be called petroleum’s answer to the Laffer Curve, is the preconception of intense and world wide ‘price elastic’ responses to what, in real terms, have been in fact limited and relatively slow increases in oil prices since the most recent low -- about US$10 billion in dollars of 2003, briefly attained in 1998-99.

Since then, as we know, Cheap Oil has become a fond but increasingly distant memory, while world oil demand grows at its fastest rate for 25 years. Through the period 1999-early 2005, there has been no trace of the ‘price elastic’ fall in oil demand that “extreme oil prices” are supposed, by BP Amoco’s well paid ‘experts’, and by an areopage of other ‘experts’, to automatically trigger as surely as day follows night.

Price elasticity or reverse elasticity?

Further contradicting and controverting ‘price elasticity’, and reducing its demand forecasting value and capability to nothing, world oil demand growth in a context of inevitably rising prices behaves in fact and in reality as a reverse elastic function. That is, in very simple terms, oil demand growth increases as oil prices increase. This has levered up, and will continue increasing world oil demand, exercising ‘high gain positive feedback’ on the oil price, until very high price levels are attained. An approximate ‘trend reversal price’ of perhaps US$90-per-barrel could be suggested. This would be effective through fiscal intervention or a round of ever larger interest rate hikes, rather than the economic impact of higher oil and gas prices.

The ‘reverse elastic function’ has been especially strong since late 2003. Through the period September 2003-September 2004, for example, oil prices increased more than 35% (from below US$35/barrel to about $50/bbl), while the rate of world oil demand growth also increased by about one-third (from about 2.25% annual, to more than 3% annual).

This ‘reverse elasticity’ occurs for several reasons, notably that much oil demand is difficult or impossible to substitute in the short-term (that is inelastic demand), but also because of global macroeconomic changes induced or triggered whenever oil prices break out of the artificially low levels they were kept at during the ‘cheap oil interval’ of 1986-99. In brief, the revenue effect outweighs the price effect.

Higher oil and energy prices positively impact world economic trends (prices and notably terms of trade adjusted prices) for “real resource exporters”, that is oil, gas, other minerals, metals and agrocommodity exporter countries.

The net effect of this, at the international economy level, is vastly more important than any purchasing power depressing effect, that is reduced spending power for non-oil/non-energy goods due to what A. Greenspan of the US Federal Reserve calls “an oil tax on consumers”.

This “tax” essentially or only concerns the richer, very oil intensive OECD ‘postindustrial’ nations, typically consuming 12-25 barrels of oil and petroleum products per head of population, per year, to purchase and operate their plethora of industrial consumer goods.

Higher revenues for the world’s many real resource exporters, due to higher oil and energy prices, completely swamps the price-elastic response by final consumers in some national economies (essentially the high-energy, but de-industrialised OECD service economies), and by some ‘primary’ users of fossil energy (e.g. certain producers of fertilizers, aluminum, etc). Fast growth of ‘real resource’ prices has for example completely transformed economic growth prospects and trends of Brazil (an oil importer but major exporter of minerals and agro-commodities), resulting in Brazil’s economy growing by over 4%/year in 2004, compared with contraction (-0.5%) in 2003

It is likely, or at least possible that BP Amoco’s Magic Curve integrates or takes account of possible but decreasingly likely cuts in oil and gas burn by Kyoto Treaty compliant countries (EU countries, Japan, Canada and others). In theory and in certain cases, Kyoto compliance would imply heroic 25-35% reductions in oil and gas burn through 2008-12, and coal burning where it is significant. Measures towards this end have started (February 2005), but the many loopholes and potentials for transfer of oil and energy demand from Kyoto compliant to non-Treaty countries (Clean Development Mechanism) will likely result in net increases in world energy demand.

Energy conservation and efficiency raising, however, can be completely discarded as any kind of supporting rationale in favor of BP Amoco’s claimed « trend rate » of falling annual oil demand growth at the world level (that is about 65% fall in oil demand growth, in 9 years).

Efficiency raising and conservation had little or no effect while oil remained a throwaway Sunset Commodity, and even with higher prices, consumption trends remain very strong in the most oil intensive societies.

All OECD countries, notably, have rising average car engine sizes (especially of 4WD vehicles), numbers of leisure motor vehicles, and annual kilometers run per vehicle.

Plastics and pharmaceuticals consumption per capita is increasing faster than national economic growth rates. Habitat construction and operation is, on balance, yet more oil intensive in most OECD nations than in the early 1980s, sole 25 years ago. Car production in most OECD countries is growing at a very fast rate -- the five-biggest EU car producers, for example, are expected to produce and sell 15 Million cars in 2005, up from about 13 million units in 2004.

Real world trends

The BP Amoco Magic Curve (see Figure, below) has a no doubt agreeable scientific look to it, through showing a sinusoidal down-trending pattern of annual increases of world oil demand. What is more important is that reality shows little or no compliance with the required ‘long term trend’, shown by the straight line. The ‘surprising’ growth of world oil – and energy -- demand through the period since 1999 to now has even been referred to even by BP Amoco, in the ‘Introduction’ to its 2003 edition of the Statistical Review.

Figure 1 BP Amoco Magic Curve
Red and blue lines/ BP Amoco claimed ‘trend rates for world oil demand growth’
Black line/ Logarithmic trend added by this author.

One way to show the extremely unsure predictive value of the so-called ‘trend rate’ is to utilize data from various editions of the BP Amoco Statistical Review in the 1994-2002 period, completed with data for 2002-2004, and compare the two. For the ‘trend’ data above, the July or mid-year figure is taken, while real growth figures in the Table, below, are for year average. Probably the most significant as well as largest variations between the projected or forecast ‘trend rate’, and reality, are found for the years 1999-2000 and 2002-2004, both periods being ones of which BP Amoco no doubt considers as ‘high oil price’ periods.

The year 2001 (September 11 attacks and the precipitous fall in world airline activity) can be taken as ‘special’, and not significant, while being the only recent year for which BP’s Magic Curve has any semblance of being related to, or in contact with reality!

Table 1 Projected ‘trend rate’ of oil demand growth as implied by BP Amoco’s ‘Magic Curve’ (July-July basis), Million barrels/day (Mbd) compared with real world outturns

Sources/ above Figure 1, and BP Statistical Review World Energy (to 2003)
Data for 2002, 03 and 04 from OECD IEA and US EIA. 2003-04 growth forecast by McKillop using bases of 2002 average demand 76.1 Mbd and 2003 average demand 77.9 Mbd
IEA data released February 2005 shows world oil demand increased 2.68 Mbd in 2004

The first conclusion is that whatever method is hidden below the BP Amoco forecasts (without doubt ‘price elastic’ based notions) the results are lamentably inaccurate and imprecise. Actual and real world outturns for oil demand growth diverge very strongly from the BP Amoco forecasts by as early as 1998. The year 2001 showed world oil demand growth of almost nil (0.08 Mbd) for the simple reason of one of the longest and deepest-ever contractions of equity numbers taking place on the world’s major bourses, and the ‘special events’ of 11 September 2001 terror attacks in the USA. The BP Magic Curve forecast for world oil demand growth in 2001 was about 0.7 Mbd.

The BP Amoco ‘Magic Curve’ is particularly useless for forecasting beyond about 1996-1998 because, very simply, the “trend rate of demand growth” is presented by BP Amoco as continually falling. However and in fact the trend rate of world oil demand growth begins to strongly recover from about 1996-1998. This new trend is especially clear if we eliminate 2001 data, and include 2004 forecasts (about 2.65 Mbd increase in world oil demand, real outturn 2.68 Mbd).

Why upward potential in demand growth is ‘unlimited’

Implicit in the BP Amoco forecasts is the illusory fantasy that, somehow, world oil demand growth will reduce to zero, by unspecified ways and means, necessarily including stronger, and faster-acting trends or measures than simple ‘price elastic’ reductions in oil demand due to ‘high prices’. As already noted, the ‘supporting rationale’ of price elastic falls in demand is itself totally controverted by actual world trends.

Whatever the economic, fiscal, technological, policy or legislative (e.g. Kyoto compliance) rationales and desires underpinning the ‘Magic Curve’, real world demographic and energy economic facts and trends make for almost unlimited upside potential. World oil demand, at the current US per capita rate of about 25.6 barrels/person/year (bpy), would for example run at about 445 Mbd if by some somber miracle the world’s population consumed oil at US rates. As shown in the Table, below, the simple fact of continuing world population growth likely ‘generates’ or leads to about 1.1 Mbd/year as a near incompressible minimum in the absence of severe, worldwide economic recession or major technology changes (that is unchanged energy economic infrastructures, zero economic growth worldwide).

Table 2 Demographic rate of oil demand, 2004
(average per capita oil demand, barrels/capita/year, for real world
and at rates based on national oil consumption rates)

Sources/ Population data from UN Population Information Network, Oil demand BP Amoco Statistical Review of World Energy, 2003, 2004

This ‘vital minimum demand growth’ (about 1.1 Mbd per year) is likely a floor level for any near-term reductions in world oil demand growth through as yet inexistant international cooperation and action for adjustment to Peak Oil and for a sustainable future, assuming there is no very intense international economic recession, caused by interest rate hikes. Achievement of this ‘vital minimum’, of about 1.1 Mbd increase-per-year, would require complete de-linking of oil from economic growth, and would obligatorily include action to produce strong, and continuing reduction in oil demand per capita in the OECD countries, due to incompressible growth of demand in New Industrial Countries.

Without intense economic recession, or slower-acting energy economic change, world oil demand growth potential is almost unlimited. We have only to consider the very fast growth of oil demand recorded by the Asian Tiger economies during their period of fastest industrial growth (about 1975-90), during which nominal oil prices increased 405%, and then apply those rates of growth, and cumulative per capita consumption levels attained, to eg. China, India, Brazil, Pakistan and Turkey. Such growth rates of national oil demand, running at an average of well above 6%/year and almost ‘impervious’ to price rises, were often exceeded by natural gas demand growth (typically running at 9%/year or more).

This growth due to industrialization, urbanization, the advent of the ‘consumer car-owning society’ and mass tourism will easily compensate stagnating or low-growth oil demand patterns and trends in the aging, services oriented, Kyoto compliant economies and societies of the OECD group (essentially EU-25, Japan, Canada).

Overall, very large growth of world oil demand is more likely than smaller. Any projection of low – or even zero - growth of world oil demand, as BP Amoco claims with its ‘Magic Curve’ must assume very significant and committed energy transition away from fossil fuels in the OECD countries and/or slowed economic growth, or a completely hypothetical ‘alternate model’ for economic growth in the large population fast industrializing countries. At the current time there is little or no sign that either of these ‘adventurous’ hypotheses is coming about in the real world. Annual growth of oil demand by the large population industrializing countries, by 2010, may itself attain a combined 1.75-2 Mbd, that is more than the annual average world annual oil demand growth of the late 1990s.

Probable impacts of coming oil price rises

The ‘only alternative’, on current trends, to oil price rising beyond US$60/barrel is that of severe fiscal reaction: in other words, use of the interest rate weapon to suppress and destroy economic growth through bankrupting businesses and destroying employment, resulting in a fall of oil demand.

This ‘strategy’ was applied in 1980-82, by politicians and central bankers of the OECD countries, reacting to oil prices (in 2003 dollars) that briefly attained about US$100/barrel, on the mostly erroneous basis that ‘high oil prices cause inflation’.

The underlying assumption or hope is that economic rout will depress oil demand enough for oil prices to fall. There is a subsidiary rationale or expectation -- that oil exporters will produce and supply their diminishing, non renewable oil resources at any price.

This may no longer be sure or certain, especially for oil producers exposed to rapid depletion of cheaply produced ‘conventional’ oil resources, such as Indonesia, Venezuela, Iran, Canada, Mexico.

Under almost any scenario oil prices will soon rise to, or beyond USD 60/bbl, and at this price level the net macroeconomic effect at the world economy level will at worst be ‘neutral’, and in fact will likely remain ‘pro-growth’.

* This in turn will further reduce the ‘window of opportunity’ or time available before Peak Oil results in annual falls of output.

In the 1975-78 period, following the 295% nominal price rise for oil at the time of the ‘first Oil Shock’, with oil prices in 2003 dollars at around US$38-55/barrel, the major OECD economies averaged about 3.5% annual growth of their economies on a real GDP base, and their annual oil demand increased at about 3-4%. The major reaction, at the time and in fact, was psychological. Sharp and rapid, large oil price rises (that is ‘oil shock’) always impact financial and corporate sentiment making for at least several months of so-called ‘crisis’, and inevitable sharp compression of stock market index numbers.

Belief in the popular myth that ‘high oil prices hurt growth’ is very much more important – for a certain period -- than the reality of global macroeconomic adjustment to higher oil and energy prices.

Depending, as said, on the fiscal environment, oil price rises in the 2006-2006 period could arise from further degradation of the Middle East situation, including a breakdown of the fragile Palestine -- Israel peace context, outright civil war (Sunnite-Chi’ite and Kurdish conflict) in occupied Iraq, or Israeli-US bombing of Iranian nuclear facilities, or regional destabilization of the Central Asian Republics.

Prices could easily attain US$100/barrel.

This price level may well be beyond the domestic economy adjustment capabilities of some very high-energy, oil-intensive economies (especially USA), but at the world economy level it is likely that US$100/barrel could be absorbed and adjusted to within at most 6 months, and even result in faster economic growth, rather than the opposite. The inflationary consequences would however be very large.


There is little need to enumerate the multiple reasons why the world’s increasingly fragile and slow growing world oil supply structure and system is faced by an emerging world supply-demand context that is radically different from the wishful thinking underlying the BP Amoco Magic Curve.

We can note the 2003 and 2004 statements by Lee Raymond and Jon Thompson of ExxonMobil, indicating that in their opinion some 4 Mbd of replacement oil production will be needed, each year on average, to compensate economic and geological depletion losses in the next 15-25 years, and cover rising demand. Their forecast is based on a ‘decay rate’ or depletion loss rate of world production capacities rising to about 2.5 Mbd-per-year, with demand growth at about 1.5 Mbd-per-year.

Current and recent exploration, drilling, proving and development effort is very far behind that rate -- suggesting that structural undersupply rather than the opposite is not a ‘long term menace’ but imminent.

Almost at any time, shortfalls can occur through conflict in the Middle East, or through simultaneous loss of Venezuelan and Nigerian supply, from an Iranian embargo resulting from US or Israeli bombing of its nuclear facilities or for any other force majeure. Any cumulative and total loss of about 3-4 Mbd supply would result in prices rising far above USD 100-per-barrel. This will be more than sufficient to result in oil supply and pricing being removed from current ‘market based’ procedures. Because of ‘intractable crisis’, therefore, oil will be treated ... as it should ... outside the market pricing system. This subject will very certainly come to the fore, within a rather short period of time.

Higher and much less volatile oil and energy prices underlying serious and committed energy conservation, transition to renewable energy and restructuring for a low energy economy, habitat and society are the real and only long-term solutions to emerging supply difficulties which will surely raise prices, but energy transition is discarded or rejected as utopian and unworkable by current political leaderships.

While claims are made that today’s economy is ‘less oil dependent than in the 1970s’ world oil consumption has risen by over 50% or 22 Mbd since 1983, and by about 22% since 1990.

Oil import dependence as a percentage of total oil consumption either remains very high, or continues to increase in a very large number of OECD economies. In the EU countries, in particular (unless demand is rapidly substituted) oil imports will soon show very fast growth.

* Unfortunately, the subject of oil prices is given benign neglect when they fall, and energetic propaganda treatment when they rise.

Most economic policy makers believe in a simple slogan: the lowest price is always the best.

This real world context ... paralyzing response and reducing options to zero ... is without doubt the real crisis we face.

Andrew McKillop is an energy economist and consultant who recently edited a book for Pluto Books, ISBN 0745320929, title 'The Final Energy Crisis' including articles by Colin Campbell and Edward R D Goldsmith. He has held posts in national, international and supranational (Euro Commission) energy, and energy policy divisions and agencies. These missions have for example included role of Energy policy coordinator, Dept Minerals & Energy, Govt of Papua NG, advisory and management at the AREC technology transfer subsidiary of OAPEC, Kuwait, study missions at the ILO and UNDP, in-house consulting to the Hydro & Power Authority of British Columbia, Canada, seminar presentations at the Administrative Staff College of India, Hyderabad, study and technology review at the Canada Science Council, and elsewhere. Andrew is a regular contributor to; he was first energy editor of the journal 'The Ecologist' and has co-authored published works with other analysts, e.g. 'Oil Crisis and Economic Adjustment.' Pinter Publishing, with Dr Salah al-Shaikhly, currently the Interim Iraqi government's Ambassador to London. He is actively seeking research, consulting or writing missions at this time. You may contact Mr. McKillop by email at -- telephone London UK +44/ (207) 288 0475


Post a Comment

<< Home