Twin Peaks

My topic today is outside the daily scope of economic news. While many overlooked details are too small to be seen, this is an example of the opposite. We don't normally perceive this trend because many of us have lived our entire lives inside it.

My story starts in the mid 1950s with a geophysicist named M. King Hubbert. (He didn't like to be called Marion.) He published a surprising prediction concerning not simply existing proven reserves, but even those to be discovered in the future. He used a mathematical model some people ridiculed as hopelessly inadequate for the task.

The problem was that his prediction of an all-time peak in U.S. domestic production in the late 1960s came true on schedule. Soon after 1970, U.S. imports of oil started a familiar climb. Imports of oil were paid for with a troubling shift in the balance of trade which remains with us.

The subtle point about Hubbert's model is that many details do not matter. The model is robust with respect to many alternatives. The central idea is that people motivated by profit will seek out the deposits easiest to exploit early on. As costs climb it will become worthwhile to exploit more difficult deposits. In addition to costing more, these will be harder to bring into production, taking more time. The end result is that proven reserves follow a pattern of growth and decline which allows an estimate of the ultimate amount of oil to be recovered even before you reach the peak.

Part of the reason people have trouble with this is that they don't normally think in terms of rates. The peak is not the point where oil runs out, it is the point where efforts to add to proven reserves fall behind the rate of consumption. You still have reserves, but you can only increase production rates for a limited time without being brought back down to the rate at which new reserves are discovered.

Yes, the model is oversimplified. You can get around it if you are allowed to ignore economics. You can also defeat it by behaving more intelligently. No, we have not made the changes necessary to invalidate the predictions. On the whole, people have behaved as if they were too dumb to beat Hubbert.

According to Hubbert's original model we are now close to the global Hubbert peak.

When questioned in the 1970s, Hubbert thought that OPEC might slow the arrival of that peak by raising prices and forcing conservation. He had no idea of the dramatic changes in demand about to appear in China, India or Russia. Prices have gone up, but demand has stayed high.

You may have heard about dramatic discoveries which will rewrite projections for future oil production. It helps to note that Hubbert did not know about the North Slope in Alaska, or the Overthrust Belt in the "lower 48". Even so, his predicted peak for U.S. oil production was never exceeded.

On a global scale there is no obvious potential for dramatic major discoveries that do not involve radically different technologies. Developing and deploying such technologies takes time. Don't expect a smooth transition between the global Hubbert peak and some endless future supply of energy.

Sometime in this decade global production will reach a rate it cannot sustain, let alone exceed. Demand will not suddenly slacken. The predictable result will be price instability leading to spikes. These will collapse when those with proven reserves temporarily increase production to grab immediate profits.

This will have major economic impacts both short-term, from the cost of such spikes, and long-term, from increased uncertainty. No existing economic control mechanisms are adequate to cope with the dramatic shift to a situation where supply becomes inelastic in response to constantly increasing demand.

This is the larger context in which the economic crisis of confidence starting in 2008 is now playing out.

A personal note: My introduction to Hubbert peaks came via Dr. Owen M. Phillips and his "The Last Chance Energy Book" published in 1979. I knew Dr. Phillips, though I could never approach his level of brilliance. He promoted a dangerous tendency in me to break down barriers between academic disciplines, and taught me to look at the big picture before I got lost in details. His use of dimensional analysis to find simple relations hidden in impossibly complicated physical problems became a mainstay of my intellectual toolbox.

One measure of the man is the vision he possessed, and the humility with which he conveyed it. When he was arguably the world expert on ocean waves he confessed that he could not sit watching the ocean for more than a few minutes without seeing something he could not explain. I have never since looked at water waves the way I did before.

At the time he published this book the U.S. energy crisis, and its economic consequences, were still fairly new. The title was then considered whimsical, at least by Physics Today. Today, sitting near the global Hubbert peak, I don't find it whimsical at all.


Hi anciendaze, I have been thinking a lot about limits to growth recently. Supply is finite, but growth leads to exponential demand. Exponential growth can suddenly and catastrophically lead to finding of a limit. It is not a question of if it will happen, but when it will happen.

I find your argument (and therefore Hubbert's) to be reasonable. I have long suspected that what we would see would be a slowly growing disparity between demand and supply - if for no other reason it can take five or more years to build a new mine, refinery etc, whereas exponential growth does not wait. It is nice to see this articulated better than I would have done.

We have centuries of oil left, but it will probably take centuries to build all the mines/wells - unless of course we can get an exponential increase in new mines/wells and refineries. I use the generic term mine because at a certain value of carbon fuels even low grade coal, gas or oil will become commercial. Even then there will be a finite limit, and we will hit it fast - only how fast is the point in debate.

So, to get to my point: resources are finite, even includng space exploration will only allow for a small increase in time before exponential growth consumes the increased resources (I am not just talking about fossil fuels here). There is however one factor which is an unknown. A factor which has an undetermined potential to change the equation - not refute it, but delay the time when the crunch happens, which gives us more time to develop and implement alternative growth models and alternative fuels. That is knowledge. Not mining harder, mining smarter, and using the fuels more intelligently. That however requires a much larger premium on science and technology than we currently see. I cannot see this coming from the west - our research priority is becoming pathetic as we have shifted to a policy of short term outcomes - science is funded much more often when it can be reasonably concluded it will produce short term benefit, and not otherwise.

So the rise of intellectual property based on research will be seen in China, India, and perhaps Brazil. The west will lose its intellectual supremacy, and the balance of power at all levels will move to the East. By all levels I mean financial, military, scientific and political.

I cannot see an alternative to this. We are slashing research budgets, tertiary education opportunities etc. all around the western world. There may be countries that defy the trend, I have not looked, but I doubt it will be enough. While Chinese citizens illegally use western research outside of patent to make products, in time it will be the west that illegally uses Chinese research to make products. Irony.

On another front I can see the rise in importance of carbon. I think it is almost criminal to burn carbon as fuel, it is far more valuable as industrial material than as fuel, or will be in the long run. So the mining sector will never be entirely left out in the cold, although there may be a sharp decline for a time if we transition to a non-carbon energy economy. Why is this the case? Plastics, and their potential to lead to cheap local industrial manufacture of almost everything.

You may not have read the book "Limits to Growth" published by the Club of Rome. Its arguments were partly based on the World 3 model. In some ways this was way too crude for detailed predictions. For example, it had a single variable called "pollution". I didn't see any option in it for a pollutant becoming a feedstock for new industries, as happened with coal tar, founding one of the first major chemical industries. There were also some rather heavy political messages based on the assumption that "socialist" countries had a better environmental record than "capitalist" countries. At the time, eastern Europe was considered a good example. Horrendous environmental problems there came to light later.

One problem in all such futurism is separating problem descriptions from proposed solutions. Everyone seems to have a particular axe to grind. A second problem goes back to that underlying model. It was based primarily on linear differential equations. This is adequate for short term projections in many cases, but fails badly when pushed beyond a given range. The model exhibited instability. If you took data from a point in the past, and did some small perturbation on the data, it would diverge from known data about changes past that point. This could put a predicted catastrophe in the present. The model's predictive power was low. The authors did not distinguish robust models, like Hubbert's, from "brittle" ones.

There are very robust arguments available for some aspects, though these do not dictate particular solutions. Here's a thought experiment.

Pretend, purely for the sake of argument, that the planet Venus has been transformed into the tropical paradise some once imagined it to be. Also pretend we have a huge space program capable of moving as many people from Earth to Venus as required. Assume these are hardworking people who will build the industrial infrastructure they need as rapidly as possible. How long would it take to fill that planet to capacity?

At present rates of population growth it would take about 25 years.

Population of developed nations tends to go through a demographic transition when health care improves and there is enough wealth to spend more on a few children with a good chance of survival than to place a large number of bets on many children with poor chances. This does not happen until people grow up with the benefits of improved health and relative prosperity all around them. For about 1/3 of the human population of the Earth, dire poverty and high infant mortality rates are present facts of life. Even if the rest of the human race settled into stable populations, this fraction would not immediately do so. In a single generation they would dominate population growth for the entire planet. If we continue to operate on the basis of everyone for themselves, this will happen. Issues of population, wealth/poverty and health are closely linked. If we do not address them together the resulting "natural" solution will involve billions of deaths. This is human suffering and death on a scale even this tragic planet has never seen.
I tend to agree with you anciendaze. Exponential growth will reach natural limits (that is limits confined by the technological and infrastructure and environmental boundaries) very fast. As we approach, death rates will rise. International stressors will rise. This may lead to wars, the traditional way to curb excess population - but at the current growth rates no war in history would be more than a blip on the curve. I have long been concerned that your prediction about billions of deaths looks like a reality. I began describing this in about 1989 - but nobody wanted to hear. "The world will sort it out", "all parties will come to the table and fix things", "we just need to go green" - I have heard many excuses over the years but nobody seems willing to realize that there is no collective will to sort this out, and without such an effort it cannot happen - we are headed to go beyond the limit of sustainable capacity on this planet, and we may already be there. If so, expect global death rates to soar. It is a sad observation that people tend to not want to fix things until after one or more disasters - and even then vested interests will deny that any action needs to be taken. We have to wait for a disaster, then another, then another, before we even start to get serious about issues.

I am not necessarily talking about climate change, or environmental polution either, although these can be factors. There are population, political, social, religious and financial aspects, as well as entrenched interests who stand to lose if we try to change things.

I am aware of the Club of Rome report, although I have never directly read it. The core principle is a mathematical certainty of limits, we just have poor methods to measure and predict those limits. However, no matter how much leeway you give to those limits, exponential growth will rapidly reach them, it just requires a few more years.

One of the points I wanted to make though is that physical limits are one thing, information limits are another. There are physical limits to increases in information/understanding too, but information limits are not as stark and immediate as resource/polution/population limits.
If you are talking about sustainability only on the scale of a century or so, I think the limit will be somewhere around 14 billion. The idea that we can slow both population growth and resource consumption starting immediately, leads to an optimistic projection of a stable population by 2050. The numbers I'm getting put the crisis about 15 years sooner. In this I am not making projections based on climate change.

My insight into the fundamental nature of the problem came in 1964, when I was in high school, and felt like a punch in the solar plexus. I've been looking for a way out ever since.

Blog entry information

Last update

More entries in User Blogs

More entries from anciendaze