Peak oil was thirty years ago

Taking a break from my war with Murdochracy, my most recent column in the Australian Financial Review (over the fold) was about Peak Oil. Partly for tactical reasons, but mainly because I believe it’s basically correct in this case, I’m wearing my hardest neoclassical hat.

One of the more intriguing sidelights to debates over climate change and energy policy is the idea of Peak Oil. On the face of it, the Peak Oil hypothesis is a straightforward claim. The amount of oil generated by any given field follows a bell-shaped curve, first rising as the field is developed and then declining as the oil becomes harder and harder to pump.

The curve is referred to as the Hubbert curve, after US geologist M King Hubbert[1] who used it to predict the peak of US oil output around 1970. Applying Hubbert’s analysis to the world as a whole yielded the prediction that the global peak in oil production should be happening around now.

On the evidence available, the predictions of the Peak Oil hypothesis don’t look too bad. Despite near-record prices for oil, the output of crude oil has remained broadly constant for the last seven years. Such an apparent plateau is exactly what the Hubbert curve would predict, bearing in mind that commercial production began 150 years ago.

The economic effects of the depletion of oil resources will be mixed. Clearly, since underlying demand is rising with population and income growth, the price of oil must rise to clear the market. That’s good for suppliers of oil, as well as competing energy sources, and bad for consumers. Overall because of the unpriced negative effects of burning oil, the most important of which is the release of carbon dioxide, a reduction in oil output is beneficial for the planet as a whole.

This is all straightforward: economists have been analysing markets for exhaustible resources ever since the pioneering work of Harold Hotelling in the 1930s. The observed outcomes fit Hotelling’s model pretty well – rising real prices are needed to sustain an optimal extraction path.

But discussions around Peak Oil are dominated, not by economic analysis, but by a range of more or less apocalyptic scenarios. In these scenarios, an end to ever-growing output of oil means an end to industrial civilisation as we know it.

There are a number of misunderstandings here. A lot of discussion seems to assume that Peak Oil means an immediate end to oil production, when the Hubbert curve implies a gradual decline over 100 years or more.

More importantly, though, the Peak Oil story is about production. But, if oil is essential to modern civilisation, what matters is not production but consumption.

The Oil Peak that actually mattered was the peak in consumption per person, which took place back in 1980 at 5.3 barrels per person per year. Since then, consumption per person has dropped to 4.4 barrels per person per year. Given the growth of demand in Asia, consumption per person in the countries that were already rich in 1980 has fallen much faster. Meanwhile living standards have risen substantially[2], unconstrained by declining consumption per person of oil, and of energy more generally.

Oddly enough, most people who worry about Peak Oil are also environmentalists concerned about climate change. From this viewpoint, which I share, Peak Oil looks like good news rather than bad. But the optimistic interpretation is trumped by the spurious idea that there is a 1-1 relationship between oil (or energy) and economic activity. This fallacious idea is held both by Peak Oil fans and by the rightwing doomsayers who suggest that reducing emissions of CO2 will destroy the economy.

A particularly interesting subgroup of Peak Oil fans are those who see nuclear energy as the only possible solution, a view that was mooted by Hubbert himself. This part of the discussion is dominated by a belief in something called ‘baseload power demand’ which must be met at all times if disaster is to be avoided. The idea that demand responds to prices and market structures seems entirely foreign to this discussion.

One of the few upsides of the disastrous Fukushima meltdown is that it has allowed a perfect test of this theory. Following the meltdown, Japan has taken 38 of its 54 reactors offline. It’s now midsummer there, and the blackouts predicted by the scaremongers have not occurred. Instead, the reduction in supply has been handled by (mostly voluntary) efficiency measures.

Energy is important, but it is no more ‘essential’ or ‘special’ than many other goods and services in a modern economy. If the supply is reduced, the market will respond to bring demand into line, especially if this response is facilitated by sensible government policy. No single source or technology, such as oil, nuclear or solar is essential, although none should be dismissed out of hand.

fn1. A fascinating guy, by the way. He was associated with the Technocracy movement, which briefly in the 1930s looked like a serious contender as an alternative form of government for the US. Wikipedia has lots on this.

fn2. In the rich world as a whole, and in most of Asia. Those in the bottom half of the US income distribution and in some very poor countries haven’t done so well, but that has nothing to do with oil.

7 Comments on "Peak oil was thirty years ago"

DragonSpawn on Sat, 6th Aug 2011 1:33 pm 



“Energy is important, but it is no more ‘essential’ or ‘special’ than many other goods and services in a modern economy”

Energy is the foundation of our modern society, period.

Another article based on ignorance and wishful thinking.

Miles on Sat, 6th Aug 2011 1:34 pm 



“No single source or technology, such as oil, nuclear or solar is essential…” We’ll see.

Ian Cooper on Sat, 6th Aug 2011 2:44 pm 



“Energy is important, but it is no more ‘essential’ or ‘special’ than many other goods and services in a modern economy. If the supply is reduced, the market will respond to bring demand into line”

This is, of course, true. But let’s not forget that in a situation in which oil became scarce, part of the market’s ‘correction’ would involve killing millions of people – people who will starve if they can no longer get the foods that abundant oil makes readily available.

The market is a great balancing force, but let’s never forget that it can use people’s lives to achieve that balance. The market is not what we should be striving to protect. The thing that must be safeguarded is life.

BS on Sat, 6th Aug 2011 5:01 pm 



A confusing article. Sounds like peak oil is accepted but economics will soften the blow.

Hubbert linearization on total world production says oil will be rare by mid century, not 100 years.

Perhaps population increase accounts for a decrease in per capita consumption from 5.3 to 4.4 bbls.

It is true that Japans consumption has decreased dramatically after Fukushima, but most of the worlds population are not Japanese.

A hard rains gonna fall.

BillT on Sun, 7th Aug 2011 1:52 am 



More vaporware from the “I don’t want change” group. It is simple. Less oil = less food, less people, less of everything. Technology is not going to find a way to do without what has become the engine of our modern world. Like taking the seeds from a farmer and telling him to grow a crop.

PM on Sun, 7th Aug 2011 2:56 am 



The author is correct in saying that usage is more important than availability. The problem however is that the example he provided is an example of increasing efficiencies or move to other energy sources and peak oil production may not coincide with such. In that case the lessening in usage wil be driven mainly by lack of supply (and higher costs as result) and in that case the same effect (lessening demand) may be rather catastrophic rather than benign. We can of course always hope for some new technology breakthroughs.

Mike on Sun, 7th Aug 2011 5:20 am 



“the Hubbert curve implies a gradual decline over 100 years or more.”

This is not correct. The Hubbert Curve is a model, not an inflexible timeline. While a close fit might be seen in the exhaustion of a field in the 1960’s, the same will not be seen on the downside of world production.

1) demand continues to rise, not fall in aggregate

2) technical methods of extracting oil faster exist today, conservation has its limits, substitutions are already maxing out

3) market mitigation of demand hurts economic activity no matter which way you choose to tally it up

4) long and ultimately severe market contractions will eventually render the new ‘hard to get’ fields unprofitable (steepening the decline)

Demand, price and extraction cost will play chicken with a succession of precipices. Each one will see someone priced out of oil. The poor in the West will forgo, the rising middle class in China will rise a little less. This is how “the market” will handle it – is handling it. And the affect on the world economy is amply evident through the proxy of currencies.

The day to watch for is that day when an oil company chooses to walk away from one of those new and difficult fields. Ghawar will produce for decades, but that day will signal a big precipice, a steepening decline and it will presage collapsing human systems to include the market mechanisms supposed to have the magic power of preventing such collapses in the first place.