Pannell Discussions

No. 172, 4 October 2010

Peak oil

This week's Pannell Discussion is a review of a book called 'Oil panic and the global crisis: predictions and myths' by Steven M. Gorelick. It's a fascinating book that reaches conclusions that might be surprising if you aren't familiar with details of the oil industry (as I wasn't). 

Before reading this book, my knowledge of the issues around peak oil was not great. I knew of the concept, of course, and understood the predicted consequences, but I had not read any of the arguments about the validity and usefulness of the idea. Steve Gorelick, a Professor in the Department of Environmental Earth System Science at Stanford University, has fixed that. This highly readable and interesting book presents both sides of a complex and important debate.

The central concept at issue is the idea, originally proposed by M. King Hubbert, that oil production over time is destined to follow a particular pattern. He predicted that it would take the shape of a classic bell-shaped curve (a normal distribution or logistic function), rising to a peak, turning over and then falling away to nothing. He predicted that the bell would be symmetrical, falling in the same pattern as it had previously risen.

After setting the scene in Chapter 1 by describing Hubbert’s idea, its appeal and its initial predictive success, Gorelick puts it aside in Chapter 2 while he provides us with detailed background on the global oil landscape. This was very valuable for a non-expert like me, covering issues such as the energy density of different fuels, the oil business, OPEC, assessments of how much oil there is, where oil is produced and consumed, oil quality, oil pricing, the price elasticity of petrol (gasoline) and petrol price variability. I was struck by the range of retail petrol prices in different countries (depending mainly on taxes and subsidies), ranging in March 2008 from US$2.30 per litre in Norway down to US$0.03 per litre in Venezuela!

Chapter 3 covers the historical resource depletion debate, starting with Malthus, via the Club of Rome and the 70s oil crises, and then returning to the ideas and predictions of Hubbert. A variety of evidence is presented supporting the argument that global oil is depleting. Indeed, the presentation of quantitative evidence is a very strong feature of the book. There are numerous graphs and tables of empirical data, almost one per page, looking at the issues from every possible angle. They show us, for example, that oil production has exceeded oil discovery since about 1980, that global oil discovery peaked in the early 1960s, that discovery of giant oil fields has particularly declined, and that projected growth in oil consumption by developing countries suggests great strain on future supplies. Using similar approaches to Hubbert’s, bodies such as the US Department of Energy (DOE) have predicted that the date of global peak oil production is getting close. Specifically, under its high-growth scenario, the DOE predicted in 2004 that the peak would occur around 2030, and would be followed by rapid declines in production.

By this stage of the book, the reader feels rather pessimistic – the evidence for imminent oil depletion looks so convincing – but then in Chapter 4 Gorelick presents the counter arguments. We learn that predictions of production rates based on Hubbert’s approach have consistently been too low, sometimes spectacularly so. “Analysts using Hubbert’s approach have the habit of ignoring their earlier predictions of the time of peak oil every so often and providing later and later dates based on larger and larger values of the global oil endowment” (p.98). Even in the US, which has done the most thorough job of finding and extracting its oil, official estimates of oil endowment have grown consistently over time, such that at any point in time there appears to be enough oil to last the next 35 years. The same applies at the global scale, except that the number is 45 years. Global oil reserves have consistently grown (e.g. they have doubled since 1980), mainly due to reassessment of how much oil there is in existing oil fields. There may be a lot more oil still to find; “The Middle East, Eastern Europe and Africa contain three-quarters of world oil reserves and yet account for only one-seventh of exploratory drilling” (p. 182). Lower recent rates of oil discovery are at least partly due to low levels of exploration during the 1980s and 1990s, due to low prices. Gorelick points out that resource economists have consistently been more optimistic about future resource scarcity than have neo-Malthusians, due to their accounting for factors such as price signals and technological change. And so on. The counter arguments to oil pessimism are many and varied.

Finally, in Chapter 5, titled “Beyond Panic”, Gorelick attempts to pull it all together and draw conclusions. He emphasises that the debate is not just about optimism and pessimism. “Informed positions transcend attitudes” (p. 221). The positions taken by different protagonists can be explained by the key assumptions they have made. Hubbert’s mass balance approach is based on the assumptions that the global oil endowment can be estimated accurately (which is palpably wrong), and that oil use is limited by production (ignoring the demand side of the market). On the other side of the debate, analysts have allowed for oil resources from unconventional sources, have considered the dynamics of oil demand, and have anticipated the role of technological change. Ultimately, Gorelick sides with the latter group. “If there is a peak and decline in global oil production during the next two decades, it is more likely that it will reflect a decrease in global oil demand, rather than production choked by critically low global availability” (p. 224). He concludes that we will probably move away from reliance on conventional oil long before the oil runs out.

A particularly nice feature of the book is the inclusion of key summary points at the end of each chapter. These sum up the essence of the material of the chapter in a set of well-selected dot points. A time-pressed or casually interested person could learn a lot just by reading these sections.

But that would be a shame. The book deserves to be read in full. For me, at least, it was a page turner. There is plenty of detailed information to allow readers to form their own judgements, but it is presented in a way that does not overwhelm the story. The lessons of the book are highly relevant to resource economists, and to anybody interested in energy.

David Pannell, The University of Western Australia,

Further reading

Gorelick, S.M. (2009). Oil Panic and the Global Crisis: Predictions and Myths, Wiley-Blackwell Publishers, Oxford, UK.


While I generally agree with your review, a casual reader would not realise that that US oil production did in fact peak, more or less as predicted by Hubbert (though he may well have picked the best of several predictions ex post). My take, for what it is worth is that conventional oil output probably has peaked, or will do so in the near future. The supply of non-conventional oil is more or less unbounded, except by price. John Quiggin.

Thanks John. Yes, US production did peak, in the 1970s. And yes, his early predictions (from 1956) were based on too conservative assumptions about oil endowment and their prediction of the peak was too early (Figure 1.6 in the book). Dave Pannell.

Pannell Discussions are brief pieces on issues and ideas in economics, science, the environment, natural resource management, politics, and agriculture.

171. Lifestyle landholders  27 Sep 2010

173. Indirect influences of policy  11 Oct 2010

URL for this page:

Home page:
Copyright © 2010 David J. Pannell
Last revised: June 01, 2011.