Geopolitical Peak Oil Feedback Loops Revisited

The world has changed a great deal since this post was written in October 2007. I think we would all say it is a lot less stable. It is worthwhile to stop and think how the things Jeff points out still apply, and what has changed.- Gail

It is quite common to hear experts explain that the current tight oil markets are due to “above-ground factors,” and not a result of a global peaking in oil production. In reality, geological peaking is driving the geopolitical events that constitute the most significant “above-ground factors” such as the chaos in Iraq and Nigeria, the nationalization in Venezuela and Bolivia, etc. Geological peaking spawns positive feedback loops within the geopolitical system. Critically, these loops are not separable from the geological events—they are part of the broader “system” of Peak Oil.

Existing peaking models are based on the logistic curves demonstrated by past peaking in individual fields or oil producing regions. Global peaking is an entirely different phenomenon—the geology behind the logistic curves is the same, but global peaking will create far greater geopolitical side-effects, even in regions with stable or rising oil production. As a result, these geopolitical side-effects of peaking global production will accelerate the rate of production decline, as well as increase the impact of that production decline by simultaneously increasing marginal demand pressures. The result: the right side of the global oil production curve will not look like the left…whatever logistic curve is fit to the left side of the curve (where historical production increased), actual declines in the future will be sharper than that curve would predict.

Here are five geopolitical processes, each a positive-feedback loop, and each an accelerant of declining oil production:

1. Return on Investment: Increased scarcity of energy, as well as increased prices, increase the return on investment for attacks that target energy infrastructure. Whether the actor is an ideologically driven group (al-Qa’ida), or a privateer (youth gangs in the Niger Delta), the geologically-driven declines increase the ROI for attacks on energy, which will drive both decisions to act, as well as targeting decisions for that action. This is a positive feedback-loop because attacks on energy infrastructure and supply drive up the price, which further increases the ROI for such attacks. John Robb has calculated the Return on Investment for the most recent bombings of oil and natural gas pipelines in Mexico this September at as high as 1.4 million percent.

2. Mercantilism: To avoid the dawning “bidding cycles” between crude oil price increases and demand destruction, Nation-States are increasingly returning to a mercantilist paradigm on energy. This is the attitude of “there isn’t enough of it to go around, and we can’t afford to pay the market price, so we need to lock up our own supply.” Whether it’s the direction of a pipeline flow out of Central Asia, defending only specified sea lanes, or influencing an occupied nation’s laws on Production Sharing Agreements, there are signs of a new energy mercantilism all around us. This is a positive feedback-loop because, like an iterated “prisoner’s dilemma” game, once one power adopts or intensifies a mercantilist attitude all others must follow suit or lose energy share. It will act to accelerate oil production declines because mercantilism prevents the most economically efficient production of a resource, accelerating the underlying problem of diminishing marginal returns. This issue of energy mercantilism has recently hit the headlines again with the intensification of the race by several nations to to lay claim to the Arctic with its uncertain but possibly vast oil and gas potential.

3. “Export-Land” Model: Jeffrey Brown (westexas on The Oil Drum), has proposed a geopolitical feedback loop that he calls the export-land model (most recently discussed in his Iron Triangle post). In a regime of high or rising prices, a state’s existing oil exports brings in great revenues, which trickles into the state’s economy, and leads to increasing domestic oil consumption. This is exactly what is happening in most oil exporting states. The result, however, is that growth in domestic consumption reduces oil available for export. In states, such as Mexico, where oil production is also in decline, the “export-land” model predicts that oil exports will decline much faster than oil production—and this is exactly what is happening, with the latest PEMEX report showing 5% production decline year-on-year, but 11% export decline.

4. Nationalism: Because our Westphalian system is fundamentally broken, the territories of nations and states are rarely contiguous. As a result, it is often the case that a nation is cut out of the benefits from its host state’s oil exports. This will be especially apparent when the “export-land” effect reduces the total size of the pie to be divided. As a result, nations or sectarian groups within states will increasingly agitate for a larger share of the pie. We see this already within Iraq, Iran (Khuzestan), Nigeria (Delta State), Bolivia (indigenous groups), even places not normally associated with oil production such as Nagaland in India. This process will continue the spread and advancement of the tactics of infrastructure disruption, as well as desensitize energy firms to ever greater rents for the security of their facilities and personnel--both of which will drive the next loop…

5. Privateering: Nationalist insurgencies and economies ruined by the downslide of the “export-land” effect will leave huge populations with no conventional economic prospects. High oil prices, and the willingness to make high protection payments, will drive those people to become energy privateers. We are seeing exactly this effect in Nigeria, where a substantial portion of the infrastructure disruption is no longer carried out by politically-motivated insurgents, but by profit-motivated gangs. This is the ultimate positive feedback-loop: infrastructure disruption further degrades any remnants of a legitimate economy, increasing the incentive to engage in energy Privateering, and compensating for any diminishing marginal returns in Privateering caused by enhanced security or competition from other privateers.

We may see some or all of these effects in any given area, and are already seeing this in some trouble spots. Some states, like Iraq, have been thrown into full-fledged “Nationalism” and “Privateering”-driven geopolitical disruption by the actions of an outside power—in this case, the US invasion was itself largely the byproduct of a shift towards energy mercantilism. This is just one illustration of the synergistic interrelationship of these processes.

The big-picture effect of these geopolitical feedback-loops is this:

Peak Oil theory takes the logistic curve decline of oil from individual fields and producing regions and extrapolates those effects to the world. The result of that extrapolation is that world oil production will follow a geologically-driven logistic curve, and that it will peak and decline in a manner similar to individual fields or producing regions. The decline of a logistic curve gradually tails of in a "long tail" of oil production. The result is a phrase that has become virtual dogma: "Peak Oil is not the end of oil production, but rather the beginning of an inexorable decline in production."

Geopolitical positive feedback-loops, however, do not act like logistic curves. They are positive feedback loops that are both self-intensifying and intensified by geologically-driven declines in production. While the geologically-dictated baseline in oil production decline may exhibit a long tail of ongoing production, geopolitical forces may abruptly chop off that tail. Commercial oil production requires some threshold level of security, rule of law, etc. to operate at all. Below that threshold, oil production does not gradually decline, but rather stops completely. Will geopolitical forces, combined with geologically-driven decline, be sufficient to bring oil production to a total halt in the near-term, at least regionally?

if you put this in a graph then we would have a gradual downward slope with a sudden cliff at some point as I noticed in a news story about Alaskan production going to zero at 50% of current rate due to pipe pressure being too small. We could sort of generalize this as a metaphor that the spot markets,e tc. and finacial systme need a minimum pressure to work for the Exxons and NOC s to get the whole tanker and refineries and car manufacturers etc. profitabl so at a certain point a petroleum based economy loses economy of scale and collapses perhaps at 30% of current production so it just shuts down so we lose the last 30% of prodcution suddenly for all intents and purposes. This would mean effectively 700 billion barrels left with a normal looking decline curve suddenly going to zero in ten to 15 years out except for very local uses in primitive and old equipment.

Interesting concept - Economic MOL

Economic MOL seems like it makes sense. Part of the problem is systemic risk. At some point, one starts losing the electric supply, perhaps because of indirect impacts (financial (?)). Once the electric system start to go, everything else is much more difficult. Of course there is the pipeline issue as well, and the amount of overhead of the super majors.

I think the MOL is more for corporate business/globalisation. Eg at some point cellphones will [thankfully] become uneconomic. So far I have managed to avoid them and I'm still here.

This guy's apprentice will still make you a Samuria sword out of a few bags of iron sand a long time after that:

And, guess what, he's Chinese...

The sudden dropoff to 0 makes a lot sense on our big systems. That may give a better outcome than some other scenarios.

I wondered where that magic number was on AK's pipe, about 350,000 BpD seems right, that is about 1/6 of what full flow was. I have a feeling you will be seeing other fields filling that pipe a little, before the final steady decline. But none of the suspected good reserves are yet proven so that remains to be seen. Oil has been flowing in that thing 30 years now I doubt very much if enough oil is around up there to double that number up.

Re: Mexico & Net Oil Exports

The 2008 Pemex data show a -9.4%/year decline in production from 2007, and I estimate that net oil exports fell at -34%/year (from 1.4 mbpd to about 1.0 mbpd).

If Mexico wanted to maintain constant net oil exports at 1.0 mbpd for the next five years, assuming a -10%/year production decline rate, I estimate that they would have to cut consumption at about -18%/year (cutting consumption by half about every four years).

This is a link to a recent story about Saudi Arabia increasing their diesel imports:

I was struck by this excerpt:

Saudi demand for fuel for power generation and transport has risen rapidly over the past six years as record crude exports sparked an economic boom. But a plunge in global oil prices since last July, caused partly by the global economic slump, has seen the kingdom cut production to its lowest level in more than six years as Opec races to match supply with falling demand.

That said, the Saudi government, like many in the Gulf, hopes to sustain its growth momentum through increased public spending and incentives to weather the financial storm. For instance, the state-owned Public Investment Fund will extend project loan durations to 20 years from 15 years, and raise its cap on project lending to 40 percent of the value of the project instead of 30 per cent.

Consumption charts for Mexico & Saudi Arabia (EIA):


Saudi Arabia:

Take a look at net oil imports into the u.s. by eia.

we are importing about 2 million barrels per day less than the peak in 2005, and the trend started before the recession began.

Probably the most meaningful metric would be annual net oil imports, but if we look at the four week running average, through the end of January for the past few years, back to 2005, in 2009 (at 11.7 mbpd) we are about 800,000 bpd below the recent (2006) January peak and only slightly below the January, 2005 rate:

Also, I suspect that the January, 2006 number was affected by lingering production & refinery curtailments due to hurricane damage.

Here are the annual net import data for the US, through 2007 (EIA):

We have been between 12.0 mbpd and 12.5 mbpd, annual average, from 2004 to 2007. BTW, US crude oil production fell at -2.4%/year from 1997 to 2007. If crude production falls at the same rate from 2007 to 2017, the first 1.1 mbpd reduction in consumption over this time period would just offset the decline in domestic production.

If you "connect the dots" net imports should have been -14,000 in 2008. This is quite a relative drop. With oil prices at $145 one cannot argue this was due to drop in demand.

If we look at the US crude production, total liquids consumption and total liquids net imports, the 1997 to 2004 rate of changes were:

Crude Production: -2.5%/year
Liquids Consumption: +1.0%/year
Net Imports: +3.8%/year

(I think that US crude production is more meaningful than total liquids, since US total liquids is so affected by refinery gains and other non-crude liquids).

You are correct that if our net imports had kept increasing at +3.8%/year, we would have been importing 14 mbpd in 2009 (and 33 mbpd in 2031), but the conventional wisdom expectation of an infinite rate of increase in net oil imports collided with the reality of peak exports, and what we expect will be a long term accelerating decline in net oil exports.

If as I expect, we see a long term accelerating rate of decline in net oil exports, we will basically see importers bidding against each other for declining net oil exports, with a continuing pattern of a smaller number of consumers paying a higher unit price for a declining volume of net oil exports.

Recently, the decline in demand outpaced the long term export decline, but I suspect that this pattern will be reversed later this year--because of a combination voluntary + involuntary net export reductions, with subsequent net export reductions being mostly involuntary.

These are pretty chilling statistics especially in view of the fact that the trend seems to be continuing and even accelerating. We should probably be talking to our congressmen about "peak exports" instead of "peak oil"

Welcome to my world of Net Oil Exports Obsession (a still rare, but slowly growing clinical disorder which leads one to talk endlessly about obscure mathematical models at parties). Unfortunately, most politicians and most media types (with a handful of honorable exceptions) don't like to talk about finite energy supplies, and they especially don't want to talk about accelerating decline rates.

You do have to appreciate the irony of a group of politicians desperately trying--in effect--to increase our demand for exported petroleum and to bail out the auto/housing/finance industry, when I think that we are in the early stages of a long term accelerating decline in net oil exports.

If my candidate for office, Alan Drake, were running (on a platform based electrified rail) , his campaign slogan would be the following: "Vote for me and things will probably not be as bad as they would otherwise have been."

This is a little off the subject, but I've always had an interest in Great Depression and it's world wide repercussions. Lately I have been interested in coal production trends in the 1920s, specifically the decline of Anthracite coal production, throughout the prior decade even in the face of gradually increasing prices. This was THE major source of power and motive force during the decade. I know that the U.S. was a major exporter of coal at this time. Are you aware of any statistics regarding coal import/exports during and after this time period? If coal exports were dropping it could help explain why it's effects spread world wide.

see page 2 for a graph of production in USA from 1890-2005. Production drops from 600 million short tonnes down to 400 mt in 1930 and goes back up in 1940s. I can't find anything about global export stats over the whole time peiod.

Prodctivity per worker climbs by 4 times fom 70s through 90s due to open pit mining-shift to the west - and mechanization of underground pits- so a lack of cheap energy-diesel-post PO- will effect coal and therefore electricity production. If coal production falls back to 1960s levels due to PO then there will be a big problem for industrial economies. If millions of men have to work the pits as in previous generations then this will be expensive and inefficient and dangerous and will change the power balance towards labor.

a 1 page summary of global coal use and trade in last several decades.

It is hard to compare coal mining with today's statisics because the development of oil/petroleum powered mining equipment in the 1940s dramatically increased productivity and made a lot of coal which was not economical to produce in the 1920s, fairly profitable after it's introduction. The economic system also converted from using anthracite coal, which compared to bituminous coal, has a higher energy content and is much easier to work with in. If you want to read about it a good book is: The Face of Decline: The Pennsylvania Anthracite Region in the Twentieth Century

so maybe mechanization will reverse itself as oil gets expensive making coal production globally cut in half or less. This could stop the accleration of GW. If as the orginal post implies some break points will not allow a long tail for oil then it will certainly be the same for coal. Most of the carbon based fuels will never be used in the linear concept proposed by GW scientists so that the GW catastrophe might be slowed down due to above ground or other complications cutting off the tail in PO Peak NG, etc.

So maybe only 50% of coal and 70% of oil and NG ever get pumped out. We could adjust our climate models considerably and adjust our expectaions of collapse timming. We could also consider that strong union movements would return controlling mining, agriculture due to high levels of manpower involved. Most collapse scenarios do not consider labor movements but just say everyone will do theri own thing in some sort of chaos. At any rate lots of new and unpredictable elements could come into play. Since labor movemtn has become weakened in USA this could tip poitics into a real radical democratic direction. Obama and the dems are clearly just another neoliberal party slightly right of center. When people have to scrap and dig themselves and mechanical construction, etc. equipment and cars are gone then worker's rights and democrats will really mean class rights. The lazy rich will not sweat. Nowadays most people sit in offices or trucks or do service jobs which are not very hard physically. This will change. Life will go from all being rich upper class(in comparison to grandpa on the farm or in the mine ) to one of hard work and real class distinction based on sweat and dirt,etc. on the body. Our mechanical slaves will be gone(also in the household).

In Pennsylvania, as the mines became less and less profitable, they were simply shut down and abandoned. The price of coal continued to increase throughout the twenties but collapsed in the thirties. Basically this left a lot of people out of work and a whole region economically devestated. In Europe, unprofitable mines were nationalized and continued to produce, however, I'm afraid that no amount of manpower can replace mechanization, because all the cheap easy coal is now out of the ground and burned.

Regarding current coal production & consumption, the US is getting close to net importer status. US net coal exports (EIA):

This data is from 2006, so we might already be net importers. So much for the possibility that coal gassification could help us.

You can see the current data at

What that shows is that US net coal exports actually increased in 2007 and increased further in 2008. Looks like roughly extending the graph posted by Westexas would put 2007 at about 500 trillion BTU and 2008 at about 1000 trillion BTU

Here's my hand-drawn extension off the graph which I think is a rough estimate based on monthly data through November 2008. Someone can perform the precise conversions to BTU's from production/consumption figures if they really want.

And here's the latest EIA monthly graph in million short tons - although it is difficult to read the export graph line due to scale. Exact figures are at the website.

However interestingly these figures from the EIA don't seem to match those in the BP Energy review as shown by According to that net exports decreased from 2006 to 2007 and also shows a short period of net imports around 2003 not reflected at all in the EIA figures.

Any coal industry experts about?

I noticed something on the EIA country chart. Production & consumption are in terms of tons, but net exports are in terms of BTU's. Net Exported BTU's have fallen much more sharply than tons. My guess is that we are producing and exporting lower BTU per ton coal.


Net Exports (Million Short Tons): 60
Net Exports (Trillion BTU's): 1959


Net Exports (Million Short Tons): 47 (Down 22%)
Net Exports (Trillion BTU's): 296 (Down 85%)

If I've got my numbers right, a 1,000 Trillion BTU's is equivalent to 1,000 BCF of gas (about 18 days of current US dry gas production).

At first sight that would appear to suggest that a ton of coal in 2006 has only 1/5th the energy content as a ton in 1997. Now I know it's falling with time but there must be something else going on here. Perhaps there is some net energy being subtracted for coal production transportation or something which can make a big difference when the import/export volume is so small compared to total production as it has been for the last few years.

The different types of coal do have different amounts of heat potential.
Anthracite has 22 to 28 million Btu/ton (26 to 33 MJ/kg)
Bituminous coal ranges from 21 to 30 million Btu/ton (24 to 35 MJ/kg) on a moist, mineral-matter-free basis
Lignite coal contains 20 to 30 percent inherent moisture by weight and the heat content of lignite ranges from 9 to 17 million Btu/ton (10 to 20 MJ/kg).
The heat content of subbituminous coal ranges from 17 to 24 million Btu per ton on a moist, mineral-matter-free basis
The heat content of subbituminous coal consumed in the United States averages 17 to 18 million Btu/ton (20 to 21 MJ/kg), on the as-received basis.
So it is posssible that we might be exporting lower quality coal. It looks like total Btus might tell you more about production than total weight.

What a wonderful summation of fundamental forces at work in world oil markets.

Looking forward, consider a scenario of five to ten years of stagnant growth. The factors described by Vail will continue to operate during that period, albeit more slowly, further constricting the volume of petroleum available to world markets.

Any return to faster growth, then, will not be possible unless some way is found to avoid the petroleum price constraint we experienced last year (magnified by financial speculation). The choice we face is (1) find large-scale efficacious alternatives to traditional petroleum (not there yet) or (2) accept economic stagnation.

By the way, I'm not advocating any path here, just trying to get a clear statement of choices we face.

It is interesting that imports from Mexico to the u.s. have fallen off a cliff.

Here is a link to an essay I wrote on the subject in June, 2008. IMO, what prevented a serious Gulf Coast crude supply problem this summer (with demand met by supplies from the SPR) was the decline in demand (which is basically what Datamunger predicted). However, because of hurricane shutdowns, there were SPR releases toward the end of the summer.

There is a link to our (Khebab/Brown) paper on the top five net oil exporters in this article.

In any case, assuming some reduction in demand in Mexico, my "optimistic" scenario for Mexico is that they have already shipped about 80% of their post-2004 cumulative net oil exports and that they will approach zero net oil exports within four years. I estimate that three of our top four crude oil sources of imported oil--Canada; Mexico & Venezuela--went from combined net exports of 5.0 mbpd in 2004 to 3.9 mbpd in 2008. Their estimated four year combined decline slightly exceeds total current Canadian net oil exports.

Our middle case is that 10 years hence, at the end of 2018, the top five net oil exporters--Saudi Arabia; Russia; Norway; Iran and the UAE--will have shipped about 80% of their post-2005 cumulative net oil exports.

I heard rumors that Mexico was buying crude on the spot market last summer to fulfill their contracts in the U.S. Is there any truth to it?

I hadn't heard that, but it would not be surprising. Mexico is a prime example of "Net Export Math."

BTW, the largest producing fields in 2005 were North Ghawar (if we separate the north end, where the best production is, from the remainder of the structural trend) and Cantarell. According to Peter Wells, who gave an "optimistic" talk on world production as ASPO-USA, North Ghawar will be effectively watered out by the end of 2010, which will probably also be the case for Cantarell. So, how have net exports from these two countries, which had the two largest producing fields in the world in 2005, changed since 2005?

I estimate that their combined net oil exports fell from 10.8 mbpd in 2005 to about 9.4 mbpd in 2008 (despite a year over year increase in net exports from Saudi Arabia, to a level several hundred thousand barrels per day below their 2005 rate). This would be a net export decline rate of about -5%/year, but note that net export decline rates tend to accelerate with time (despite occasional year over year increases, e.g. Indonesia).

I think you might be interested in this article

It sure suggests we might see an accelerating decline rate in the future at least out of Russia

We need to get Jeff to update this post, to reflect the current situation. It seems like the financial situation of several of the oil exporters (Venezuela, Mexico, Russia) has been rapidly declining. So far, nothing has "snapped", but one cannot expect our good luck to continue indefinitely.

I find the mercantilism argument unpersuasive. If a major effect/motivation is to insulate the importer from world prices, it is clearly against the interests of the exporter. So some sort of pressure, -or sweateners must be part of the bargain. As the difference between mercantilist prices and world prices increases so does the pressure of the exporters to break out of the regime.

I think the argument for increased mercantilism is persuasive and it's based on the status quo. What else is China and the Sudan, or US and Iraq, or Saudi Arabia for that matter? In these situations the powerful countries have secured access to a dependable and regular oil supply which constitutes a significant portion of the powerful state's energy security portfolio. It may be that the going price for oil is set in international markets (as some would say), but regular supply contracts must be seen as strategic assets nonetheless. Given the historic trend toward conflict and imperialism in oil producing regions, and given the examples listed above, mercantile economics and militarism are deeply intertwined with the international energy system. I would say that this could be one of the most serious and destabilizing feedback loops that Jeff describes.

"So some sort of pressure, -or sweeteners must be part of the bargain."

The sweeteners to the bargain can be mutually reinforcing if vital goods are exchanged. Many [All?] MidEast FF-exporters are reliant upon food imports, as not much is grown in endless miles of sand dunes and desert. So a mercantilist contract where energy is traded for food can be beneficial to both parties.

As our Westexas has said many times before: the postPeak era is not a good time to be an net importer of both energy + food.

Alright, we're back on 'Peak Oil'!

The best methodological tool to consider feedbacks is system dynamics. So, I invite you to make models with this tool.
In ASPO VII meeting in Barcelona we have presented one such model. Our results are than ASPO and others oil peaks based solely in geological aspects are optimistic. J. Laherrere gives thanks to me because I am the first to tell him "you are optimistic".