This paper was first presented by MBendi in October 2000, just
before the end of a cycle of rising prices and dwindling capacity. It
recognised the resource constraints and presented a scenario of the natural
consequences, foreseeing soaring prices, imminent shortage and recession. While
this general picture remains valid in resource terms, events did not unfold as
expected but took a different turn. The limits to production capacity were
indeed reached at the end of 2000 and prices did soar, but the economy reacted
more quickly than expected by plunging the world into recession, which cut oil
demand and reduced pressure on prices, which have remained weak.
Most of the points made in the paper remain valid, but an update
of the study to incorporate the latest data and new understandings does call
for some revision. A new scenario is proposed.
The paper has been revised to better reflect current
This paper is about Peak Oil. It truly is a turning point for
Mankind, which will affect everyone, although some more than others. Those
countries, which plan and prepare, will survive better than those that do not.
It is a large and difficult subject, but the essentials are clear.
In summary, these are the main points that have to be
- Conventional oil - and that will be defined - provides most
of the oil produced today, and is responsible for about 95% all oil that has
been produced so far.
- It will continue to dominate supply for a long time to come.
It is what matters most.
- Its discovery peaked in the 1960s. We now find one barrel for
every four we consume.
- Middle East share of production is set to rise. The rest of
the world peaked in 1997, and is therefore in terminal decline.
- Non-conventional oil delays peak only a few years, but will
ameliorate the subsequent decline.
- Gas, which is less depleted than oil, will likely peak around
- Capacity limits were breached late in 2000, causing prices to
soar leading to world recession.
- The recession may be permanent because any recovery would
lead to new oil demand until the limits were again breached which would lead to
new price shocks re-imposing recession in a vicious circle.
- World peak may prove to have been passed in 2000, if demand
is curtailed by recession.
- Prices may remain weak in such circumstances but since demand
is not infinitely elastic they must again rise from supply constraints when
essential needs are affected
Peak oil is a turning point for Mankind. The economic prosperity
of the 20th Century was driven by cheap, oil-based energy. Everyone had the
equivalent of several unpaid and unfed slaves to do his work for him, but now
these slaves are getting old and won't work much longer. We have an urgent
need to find how to live without them.
It is stressed that we are not facing a re-run of the Oil Shocks
of the 1970s. They were like the tremors before an earthquake, although serious
enough, tipping the World into recession. Now, we face the earthquake itself.
This shock is very different. It is driven by resource constraints, not
politics - although of course politics do enter into it. It is not a temporary
interruption but the onset of a permanent new condition. The warning signals
have been flying for a long time. They have been plain to see, but the world
turned a blind eye, and failed to read the message.
Our lack of preparedness is itself amazing, given the importance
of oil to our lives. The warnings were rejected and discredited as if they were
words of soothsayers and prophets. But the warning was not prophecy - it simply
recognised two undeniable facts:
- You have to find oil before you can produce it
- Production has to mirror discovery after a time lag
Discovery reached a peak in the 1960s - despite all the
technology we hear so much about, and a worldwide search for the best
prospects. It should surprise no one that the corresponding peak of production
is now upon us. This simple reasoning has, however, been rejected by flat-earth
economists and others with a blind faith in technology and markets forces.
Worse still, governments have listened to bad advice. There are many vested
interests bent on confusion and denial.
It is worth briefly recalling what occurred in Europe in late
2000, as a foretaste of what happens when oil supply becomes short and
expensive. The French fishermen blockaded the Channel Ports because their fuel
costs had doubled, even though their fuel was already tax-free. The dispute
spread rapidly to England and other countries. Schools were closed. Hospitals
had red alerts because staff and patients could not reach them. Supermarkets
started rationing bread. Trade and industry was seriously interrupted: the cost
was huge. People lost confidence in their governments, whose popular support
fell sharply. If an interruption in supply lasting only a few days could cause
such havoc, it surely demonstrates how utterly dependent on oil we have
Depletion is an easy concept to grasp. Think of an Irish pub
full of happy people. Think of their pleasure at the first sip from a full
glass. Think of the frowns that begin to cross their faces when their glasses
are half-empty. They know they have drunk more than is left. It is the turning
point. Watch them savour the last drops. While they can order another round of
drinks, they know in the back of their minds that eventually closing time will
come when there are no more to be had. That is the meaning of depletion. We
need to know how big each glass - or oilfield - is, and we need to think of
closing time, and judge how many oilfields are left to find.
We are not about to run out of oil, but production is about to
reach a peak, if it has not done so already. When peak comes depends on the
issue of Rates:
- Discovery Rate - we now find one barrel of conventional oil
for every four we consume
- Extraction Rate is controlled by the physics of the
Demand is driven by economic growth and price. Remember that
price is not the same as cost. The cost of producing oil remains low, but its
price has to reflect tax, scarcity and control of the main sources of
Before measuring something, the first step is to decide what
exactly to measure. It is a question every butcher asks. Does he weigh the meat
or the bones as well? There are many different kinds of oil. Each has its own
endowment in Nature, characteristics, costs, and rate of extraction that
follows a general and inevitable zero-peak-zero profile. Each type contributes
differently to peak. Some types rise to peak quickly, others slowly. We need to
identify and measure each type carefully.
It is convenient to identify so-called Conventional Oil. It is
the meat not the bones. It has contributed most oil to-date and will dominate
all supply long into the future. We may concentrate on it, as it controls the
date of peak. But there is no universal agreement on how to define it. Here, it
is defined to exclude:
- Oil from coal and “shale”
- Bitumen and Extra-Heavy Oil
- Heavy Oil
- Deepwater Oil
- Polar Oil
Natural Gas liquids from gasfields are also excluded because
they belong to the gas domain.
We should at least define what we try to measure, even if the
database is not up to doing it so accurately in all cases. We may start by
asking two simple questions:
- How much oil has been found? and
- When was it found?
They sound simple, but they are difficult to answer because the
data are weak. There is no consistency in what is reported. There is a large
range even for production, which is simply reading the meter. Reserve estimates
are still less reliable. The treatment of gas liquids ranges widely.
There are two main sources of public data: the Oil & Gas
Journal and World Oil, which are trade journals that compile information given
to them by governments and others. They are not in a position to assess the
validity of the information supplied to them. Another widely used source is the
BP Statistical Review of World Energy. BP is in a position to evaluate the
data, but prefers to reproduce the Oil and Gas Journal numbers, understandably
not wanting to involve itself with sensitive issues that might affect its
relationship with the host governments of the countries where it works. Lastly
is the industry database, which is relatively reliable, but too expensive for
most analysts to access. All these sources provide different numbers.
The industry is required to furnish estimates of so-called
Proved Reserves in its financial reports to governments and the
stock-exchanges. These estimates relate to what the wells in the current stage
of development are expected to produce, but say little about what the field as
a whole may eventually deliver. The industry has accordingly systematically
under-reported the size of discovery. It has good commercial reasons for doing
so rather than booking all their reserves up front because it smoothes their
assets, presenting a better image. It is not its job to forecast the future.
For most purposes, it does not matter, but we need to know the real record of
the past if we are to use the trend to forecast the future. Governments
variously under-report or over-report, or simply fail to update their
estimates. As many as 64 countries reported unchanged numbers in 2001, which is
We need the "best estimate" of the size of the field, namely its
Proved & Probable reserves, such that any revisions are statistically
An oilfield contains what it contains because it was filled in
the geological past, but knowledge of how much it contains evolves over time.
If we want a genuine discovery trend, we need to backdate revisions to the
discovery of the field. Failure to backdate gives the illusion that more is
being found than is the case. It is a cause of great misunderstanding
This demonstrates how BP reports reserves, failing to backdate
the revisions. It has misled many analysts. The large increases in the late
1980s were simply due to the OPEC quota wars. Nothing particular was actually
added. Kuwait added 50% in 1985 to increase its OPEC quota, which was based
partly on reserves. No corresponding new discoveries had been made. Nothing
particular changed in the reservoir. Venezuela doubled its reserves in 1987 by
the inclusion of large deposits of heavy oil that had been known for years,
forcing the other OPEC countries to retaliate with huge increases. Note too how
the numbers have changed little since despite production.
But it is not quite as simple as that, because the early numbers
were too low, having been inherited from the companies before they were
expropriated. Some of the increase was justified but it has to be backdated to
the discovery of the fields concerned that had been found up to 50 years
The failure to backdate gives this misleading popular image of
growing reserves. It is widely used by flat-earth economists in support of
classical economic theories of supply and demand
By no means all economists believe in a flat-earth. There are
enlightened economists who now relate economics with resources, and they are
coming to the fore. Financial institutions too are beginning to understand the
inevitable reality of the depletion of oil.
This shows the effect of proper backdating. The discovery trend
shown in yellow is falling not rising.
You will hear many claims for technology. No one disputes the
huge technological advances of the industry. But, what has been the impact? In
exploration, it shows better both where oil is and where it is NOT - thus
allowing better estimates of the potential to be made. In production, it keeps
production rate higher for longer, but has little impact on the reserves
themselves. Note that much of the oil in a reservoir cannot be extracted
because it is held there by capillary forces and natural constrictions. The
percentage recovered can be improved in some cases by injecting steam and other
well-tried methods, but by no means all fields are susceptible to treatment.
Most modern fields are produced to maximum efficiency from the outset.
This is well illustrated by the Prudhoe Bay field. It is the
largest field in North America. The Operator internally estimated its reserves
at 12.5 Gb in 1977, but reported 9 Gb to comply with stock-exchange rules.
Various techniques, such as gas injection followed by horizontal drilling, were
started in 1982, but decline commenced in 1988. Gas injection did arrest
decline for one year, but then the decline became steeper. It is now evident
that the field will barely make the original estimate of 1977. Nothing was
added by technology. This is a quite typical example, with many large fields
showing the same pattern. Such plots are incidentally a good way to estimate
Now let us turn to how much is yet-to-find. A geochemical
breakthrough in the 1980s made it possible to relate the oil in a well with the
rock from which it came. It became possible to identify and map the generating
belts. They are few and far between because prolific oil was formed only under
very rare geological circumstances. In fact, most of it comes from no more that
three or four epochs of intense global warming. We now know where most of the
generating areas are.
Great advances in seismic technology make it possible to see the
smallest and most subtle trap. In general, this better knowledge has reduced
the perceived potential, because it shows a dearth of large prospects. In other
words, we can find a needle in a haystack, but it is still a needle. We did not
need the resolution to find the giant fields of the past holding most of the
world's oil. It means we have a much better knowledge of the endowment in
Nature than we used to have.
Once we have secured valid data on the amounts and dates of
discovery, we can use it to project future discovery
This is the so-called creaming curve. It plots discovery against
exploration wildcats. They are the wells that either do - or do not - find a
new field. The larger fields are usually found first for obvious reasons, being
too large to miss. The curve flattens until new discoveries are too small to be
viable. It gives a good idea of how much is left to find. There are other
statistical techniques that evaluate the size distribution of fields and
correlate production with the corresponding earlier discovery trends.
The same applies to an individual oil company. Shell has found
about 60 Gb with almost 4000 exploration wells, drilled over its entire history
since 1895. If it drilled as many again, it could expect to find only 16 Gb.
Other companies have not had such a successful record.
|Current discovery rate
|Current depletion rate (ann.
prod. as % of Yet-to-Produce)
To sum up, these are the main parameters for Conventional oil.
The numbers are shown as computed but should be generously rounded. We have
produced almost half what is there, and we have found about 90%. We consume 22
Gb a year but find only 6 Gb. That is to say, we find one barrel for every four
we consume from our inheritance of past discovery. The current depletion rate
is about 2 % a year. These estimates are broadly consistent with those
presented in 2000, but we now take a different view of the treatment of
Condensate, including that from the gas-caps of oilfields with oil. It largely
explains the increase in the estimated ultimate recovery.
This shows the growing gap between discovery and consumption as
we move from surplus to deficit. The yellow curve shows exploration drilling.
Note that the level of activity barely affects the discovery trend. It destroys
the flat earth economists' claim that discovery is driven by market forces.
But in year 2000, we did have an exceptional discovery spike
from two large finds in the Caspian and Iran, which had hitherto been areas
closed to the industry. These exceptions apart, underlying general trend is
down to about 6 Gb, with perhaps as much again coming from new deepwater
discovery, here treated as Non-conventional. The new deepwater areas are
yielding an early crop of giant fields, as is to be expected, but discovery
there is set to decline too.
A few examples illustrate the nature of depletion. Remember that
the peak of discovery has to be followed by the peak of production, which
generally comes close to the midpoint of depletion when half the total has been
Let us start with the US-48, the most mature oil country of all.
It had plenty of money, every incentive with the oil rights in private hands
and soaring imports, and it had a large prospective territory. We can be sure
that if more could have been found, it would have been found. So what did
Nature deliver? Discovery, shown in green, peaked in 1930 at the edge of the
chart. Production peaked 40 years later. It is the same pattern in the North
Sea (UK, Norway and Denmark), which peaked in 2001. Advances in technology have
reduced the time lag from peak discovery in 1974 to 27 years. We are getting
better at depleting our resources.
Figure 7 shows the World as a whole. The oil shocks of the 1970s
cut demand so that the actual peak came later and lower than would otherwise
have been the case. It means that the decline is less steep than it would
otherwise have been. It reminds us that if we produce less today, there is more
left for tomorrow. It is a lesson we need to relearn as a matter of
This shows the distribution of oil. Note how North America has
consumed most of its oil, and how the Middle East has most of what is left. It
introduces the notion of Swing Share. The five Middle East major producers
countries have been forced into a certain swing role around peak, whereby for a
certain limited period, they can - at least in resource terms - make up the
difference between world demand and what the rest of the world can produce.
Swing Share was 38% in 1973 at the time of the first oil shock. It had fallen
to 18% by 1985 because new provinces in the North Sea, Alaska and elsewhere
started to deliver flush production from their giant fields, which are usually
found first. It is stressed that these new provinces had been found before the
shock and were not a consequence of it, as is so often claimed by flat-earth
economists. Swing share reached 29% in 2000, before falling to 25% in 2001 in
response to falling demand. Under the new scenario, discussed below, it is set
to reach 40% by 2010, which will likely represent the limit of capacity. Unlike
in the 1970s, this time there are no new major provinces waiting to deliver, or
even in sight.
This shows the depletion of the Middle East swing countries.
Production matched the theoretical unconstrained model well until the shocks of
the 1970s when it was artificially restricted by OPEC quota. Actual production
has been far below what was possible. Even though World demand is assumed to be
flat under the 2001 Scenario, discussed below, the Middle East swing countries
will be under pressure to increase their production rapidly to offset the
decline elsewhere. The scenario assumes that they can reach 24 Mb/d by 2010 but
it may prove to be beyond their ability, given that weak prices for the next
few years curb incentive.
It is worth digressing briefly to explain the impact of
expropriation. It started with BP in Iran in 1951 but had spread to the other
main producers by the 1970s. The major companies lost their main sources of
supply. Had they remained in control, they would have produced the cheap and
easy oil before turning to the expensive and difficult, offshore and in remote
areas. It would have given a gradual transition as depletion began to grip. But
when they lost their main supplies, they moved to the expensive and difficult
areas and they worked flat out. The main OPEC governments were left with the
cheap and easy stuff. It was contrary to normal economic practice and one of
the causes of the impending crisis
This is a very compelling graph. The red line is discovery smoothed
with a ten-year moving average. It shows a clear downward trend, easy
to extrapolate, as shown in orange. The green line is production,
extrapolated at a 2% growth in demand to match the past trend. The
inheritance from past discovery is the area between the red and green
lines. The inheritance is being increasingly consumed because future
discovery is insufficient, but like all inheritances, it does not
last for ever. There just is not enough to sustain growth, or even
hold current production for long. The blue line shows the inevitable
Previous studies evaluated alternative scenarios of supply and
demand, based on various assumptions of demand and oil price. Generally, they
depicted a plateau of production, starting when Swing Share reached a critical
threshold triggering an oil price shock, and ending when the Swing share
reached 50% of world demand, which was held to be maximum Swing capacity. But
as we approach closer to these critical times, we can see the unfolding picture
with greater clarity.
It now appears that the world capacity limits were about
breached at the end of 2000, and oil prices began to soar when it became clear
that the historic trend of growth at about 2% could not be maintained. As in
all previous cases, the high prices triggered economic recession, although
there may have been other contributory factors. A highly inflated stockmarket,
built on the cheap energy supply of the past and illusions of perpetual growth
was evidently due for a radical readjustment. The demand for oil plummeted,
falling 5% between 2000 and 2001 according to the Oil & Gas Journal, and
prices accordingly crumbled.
OPEC found itself unable to react lest any action were perceived
to be hostile to the United States in its conflict with Afghanistan, while the
US grand fleet was anchored off the coast of Middle East. Iran and Iraq have
been declared enemies by the US President, prompting fears that a new American
invasion in the region may be contemplated. It would be a brave man to forecast
the future in such circumstances, but the underlying resource constraints do
give a basis for a new scenario. It assumes that if the world economy were to
try to recover, the demand for oil would rise in parallel until it again hit
the ceiling of falling capacity. High prices in a volatile market would follow,
re-imposing recession. In these circumstances, it is reasonable to contemplate
flat average demand and production until the Swing countries reach their
assessed capacity limit of 24 Mb/d by 2010. Soaring prices and growing shortage
will then initiate the long-term decline of oil production at the then
depletion rate of about 2% a year. Although the model pictures a plateau of
production to 2010, it is unlikely to be a very flat one, as great fluctuations
in a highly volatile market may be anticipated. It seems therefore that we may
look back and find that world peak was passed in 2000, six years before the
midpoint of depletion.
The above scenarios refer to Conventional oil only, as herein
defined. Deepwater oil with an estimated endowment of about 60 Gb is expected
to peak at about 8 Mb/d around 2010, although weak prices in the meantime may
curtail investments somewhat. The production of heavy oils and tarsands of
Venezuela and Canada are expected to increase slowly with the added incentive
of strategic supplies in the United States. Gas is expected to peak around
These resource constraints are all very obvious from even a
cursory examination of the resource base and its depletion, which poses the
question of why this important subject is not better understood.
People once believed the earth was flat. Scientific observations
to the contrary were treated as blasphemy. The same pressures manifest
themselves to-day in a different guise. We might almost call some of them
conspiracies of denial and obfuscation. The United States seeks to exaggerate
the world's oil endowment to reduce OPEC's confidence. It pretends that it does
not depend on Middle East oil. It puts out very flawed studies by the US
Geological Survey and the Department of Energy. OPEC, for its part, exaggerates
its resource base to inhibit non-OPEC investments and moves to energy savings
or renewables. It fears losing its oil market on which it utterly depends, with
its rapidly rising population. Companies conceal depletion because it sits
badly on the investment community
The US Geological Survey has failed to live up to its scientific
reputation. It has assessed the Undiscovered Potential of each basin with a
range of subjective probabilities. It has a Low Case for the most sure and a
High Case for the least sure. The High Case itself has little meaning, being
little more than a wild guess. The Low Case is consistent with the discovery
trend, but The Mean value, which is the one publicised is meaningless because
it is influenced by the High Case. This has been confirmed by experience in the
real world because the Mean estimate is already 100 Gb short, five years into
the study period. Its notion of "reserve growth" is also flawed. It is depicted
as a technological dynamic when it is simply an artefact of reporting practice,
not to be extrapolated into the future.
Middle East (now 18 M/d)
Non-Middle East (now 45 M/d)
The International Energy Agency was established by the OECD
countries in the aftermath of the oil shocks of the 1970s. In 1998, it
succeeded in delivering a coded message. It showed how a "business as usual
scenario" could not be fulfilled without inventing a so-called balancing item
of Unidentified Unconventional, which miraculously rises from zero in 2010 to
19 Mb/d in 2020, when the identified makes a ceiling of only 2.4 by 2010. Since
the identified deposits are huge but constrained by low extraction rates, no
one needs to find more. The so-called Unidentified Unconventional is
accordingly a euphemism for rank shortage. It is also not realistic to imagine
that oil price will still be $25/b when the Middle East supplies 62% of the
world's needs. Now, two years later a new study appears in which the mythical
balancing item has disappeared, and non-Middle East production by 2020 is shown
to almost double the previous estimate. The IEA evidently was influenced by the
flawed study by the USGS. No credence can be given to such fluid
pronouncements, yet most governments build their energy policies on them.
Most companies have to sing to the stockmarket, but the Italian
national company is less concerned by stockmarket imagery. Its Chairman was
able to tell the truth when he reported "New reserves are failing to keep up
with growing output…… My forecast is that between 2000 and 2005 the
world will be reaching peak". The French company, Total-Fina-Elf, has also
published its view of a peak around 2010.
British Petroleum certainly wins the prize for the most oblique
reference to depletion when it changes its logo to a sunflower and says that BP
stands for Beyond Petroleum. But its executives sit on the board of Goldman
Sachs, the bankers. They should accordingly know what BP actually thinks behind
the lace curtains of corporate make-believe. What do the bankers say?
"The rig count over the last 12 years has reached bottom.
This is not because of low oil price. The oil companies are not going to keep
rigs employed to drill dry holes. They know it but are unable and willing to
admit it. The great merger mania is nothing more than a scaling down of a dying
industry in recognition that 90% of global conventional oil has already been
found" (Goldman Sachs - August 1999)
Actions speak louder than words. The major companies and many
others in the industry are merging and shedding staff. They are also buying
their own stock. They conspicuously fail to invest in new refining capacity,
which would surely be needed if production were set to rise as depicted. These
are moves to downsize because there are no major investment opportunities left.
Their past is worth more than their future - and they know it.
Some general comments may be offered in conclusion, starting
with a oil price
Oil outside the Middle East peaked in 1997, as was easily
foreseen. It should have heralded a gradual rise in price from growing Middle
East control. But instead there was an anomalous fall. Price collapsed in 1998
because of the interaction of warm weather, an Asian recession, the devaluation
of the rouble, events in Iraq, false supply estimates by the IEA that prompted
higher OPEC production and perhaps some manipulation by insiders. Then, prices
surged through 1999 in a staggering 300% increase, as the underlying capacity
limits were breached, triggering recession. Demand fell and prices slumped.
Spare capacity can mean many things. A closed flowing well is
the only form of spare capacity that can be restored at will. All the other
elements take investment, work and, above all, time to deliver. OPEC had very
little operational spare capacity, having to run ever faster to stand still, as
it desperately tried to offset the natural decline of its ageing fields. It
will be hard pressed to meet the demands made upon it even to maintain current
world production, never mind growth.
We may look back and find that the year 2000 was the peak: a
turning point when the prosperity of the past, driven by an abundant supply of
cheap oil-based energy, gave way to decline in the future. A discontinuity of
this magnitude is hard to grasp. The poor countries of the world will bear most
of the burden. But the United States will be in serious difficulties. There is
a danger of some ill-considered military intervention to try to secure oil, of
which the Afghan War may have been a foretaste. That affair may be seen to have
been more of an act of defiance to impose global economic hegemony by military
means than a calculated action to reduce the level of so-called terrorism. The
growing population pressures from declining wealth are manifested in new
migration trends as are already being felt in Europe and the United States with
human smuggling becoming a gruesome addition to the global market. As global
order disintegrates, self-sufficiency at the local level may become a priority
An oil crisis is bad for politicians. Blaming OPEC or the oil
companies will not wash much longer. It would be better to make a proper
analysis of the true position and inform the people at large. No one blames the
government for an earthquake. So they wouldn't blame it for an oil crisis
either, if they realised it was a natural phenomenon.
"If you don't deal with reality, reality will deal with
But let us not be too alarmist. The roof does not fall in at
peak. What changes are people's perceptions, as they come to realise that the
growth of the past is set to become the decline of the future. It may herald
the end of the US economic and cultural hegemony - which some people might
think was no bad thing. Climate concerns may recede as the emissions, held
responsible for change, dwindle. In the face of these pressures, we should use
our current high oil supply intelligently while it lasts to ease the
transition. For example, much more efficient vehicles have already been
designed, awaiting only a mass market to be introduced. More could be done to
penalise the wasteful use of energy.
Peak oil is a turning point for Mankind, when a hundred years of
easy growth ends. The population may be about to peak too for not unrelated
reasons. The transition to decline is a period of great tension when priorities
shift to self-sufficiency and sustainability. It may end up a better world,
freed from the widespread gross excesses of to-day.
The CWC Group is hosting an event concerning the rehabilitation and reconstructio of Iraq's petroleum sector.