We now stand at a crossroads in human history. One path leads to a new Golden Age wherein nine billion people might enjoy a high standard of living. The other leads back to the Stone Age.
All of the official models of future energy production — developed by economists within government agencies, NGOs, consultancies and the fossil fuel industry — are based on demand. They simply take current energy supply and apply a growth factor to it, informed by population growth trends and economic forecasts. Because they are essentially economic models, they are blind to real-world supply constraints (which was the essence of my critique two weeks ago). They assume that adequate energy resources will always be available at an affordable price, as the miracle of the market gradually leads the way to greater efficiency and fuels substitution, and project energy supply rising along with population, which is assumed to level off around nine billion people by 2050 or so.
Alternative models, which you will never see in the press, are based on supply. They apply mathematical formulas to the long record of fossil fuel production data and extrapolate into the future, taking into account some informed geological estimates of future production. Accordingly, these models do not anticipate new fuel breakthroughs or substitutions very well.
Over the last 150 years of industrialization, the demand- and supply-based models worked fairly well together, as the supply of available fuel gradually increased to meet demand. But now their forecasts have diverged dramatically.
Consider the following model of supply for all fossil fuels developed by some editors of The Oil Drum, in a study titled Olduvai Revisited 2008. It shows the aggregate peak of all fossil fuels (87 percent of the world’s primary energy supply) occurring around 2018, then proceeding into a long decline starting around 2025.
After a decade of intensive study on resource models, including the specific models of oil, gas and coal on which that chart is based, I believe this is a reasonably accurate forecast for fossil fuels. The actual production might be slightly higher or lower (say, plus or minus five percent) and the aggregate peak may occur slightly earlier or later, but I think it will be fairly accurate. (Those who are interested in the data are strongly encouraged to explore the cited sources.)
The enormous implication of this model is that by the end of this century, nearly all recoverable fossil fuels will be gone, and humanity will be on its way back to the Stone Age -- hence the title of that post. (The Olduvai Gorge in Tanzania is a major paleontological site, often referred to as the “Cradle of Mankind.”) The authors proceed to model how efficiency improvements and renewables can mitigate the loss of fossil fuels.
In order to do that, however, the world would have to begin an all-out effort immediately. Declining availability and increasing competition for the remaining fossil fuels will make it progressively more difficult to manufacture, transport, and install renewables and efficiency improvements. Within 25 years, the world could lose 25 percent or more of its oil supply, and nearly all of its available net exports. Any interruptions in oil supply will have immediate and far-reaching effects on our globalized world of resource production and manufacturing, and cause systemic dependencies to break down. Consider that a single large modern wind turbine has over 8,000 parts, with a supply chain that spans the globe. A few spot fuel shortages could easily create long delays in that chain, and sharply slow down the rate at which new capacity can be deployed.
For these and other reasons too complex to explore here, the vast majority of the mitigation efforts must be completed within the next 20 years. For a rough measure of how intense the effort must be, consider this: Starting at its current 1.3 percent fraction of primary energy supply, renewables would have to have an inconceivable compound annual growth rate of 25 percent to displace all fossil fuels within 20 years.
What we can be sure of is that the economic models are incorrect. Supply cannot continue rising for more than another 15 years or so.
Conventional oil supply has been flat around 74 million barrels a day since 2004. The decades-long history of enhanced recovery technology shows that it can thicken and extend the tail of production, but not move the peak. Marginal resources like tar sands and biofuels are not scalable much beyond their present levels. Shale gas (produced by “fracking”) has given the U.S. a boost in natural gas supply for the moment, but on a global scale the increase is minor. The history of shale gas production is extremely short, the data is poor, and most of the forecasts based on it are utter garbage. At this point I don’t believe its production curve can be shifted or raised significantly (and in any case, the gas curve shown in the chart above already takes unconventional gas from shale into account). Coal production could be slightly higher and the decline curve slightly more gradual than shown in the chart, but a careful survey of global coal reserves reveals a host of environmental, technical, and financial challenges that will preclude a radical departure from this model.
But this model only shows the gross magnitude of the problem. The really devilish issues are in the details.
One is the declining energy content of our fuels. The energy content of U.S. coal has been falling steadily for the last 47 years, and is now 20 percent lower than it was in 1949. We mined most of the best coal (anthracite, with 30 Mj/kg energy content) first, and we are now working our way into sub-bituminous coals and lignite with as little as 5 Mj/kg of energy. The volume of U.S. coal production may continue to creep up for another decade or more, but in terms of energy, it peaked over a decade ago in 1998. The gradual substitution of natural gas liquids and biofuels for crude in our liquid fuel mix has likewise resulted in a 6.2 percent loss in energy content per barrel over the same period.
Another is the falling net energy of our fuel supply. In the early 1930s, we got 100 barrels of energy out for every barrel’s worth of energy we invested in an oil well. Today, we get about 11. The net energy of most tar sands production is around 5. The net energy of natural gas is in decline. The net energy of all renewable sources is low: wind, 18; solar, 6.8; nuclear, 5 to 15; all biofuels, less than 2.
Then there is the cost factor. The growth of debt and leverage in the financial system has forced the prices of all commodities upward, which in turn has dampened demand. Many peakists (including me) used to think that oil would reach $200, $300, even $500 per barrel as it became increasingly scarce, but we saw significant demand destruction in the U.S. when oil crossed $120 a barrel in 2008, and we’ve seen it again this year when oil approached only $90. We now know that there is a descending limit to what people can afford.
Declining net energy has also exacerbated rising production costs, and paradoxically begun to exert a negative effect on supply. In contravention of economic theory, high prices have not made our vast endowment of untapped marginal resources like oil shale economically viable. Indeed, oil prices now appear to be trapped on a narrow ledge: Economic growth requires more oil, which requires high oil prices, which in turn undermine economic growth. It now costs $80 to $90 to bring a new barrel of supply online from marginal resources such as deepwater, tar sands, and the Arctic; almost exactly the price at which demand destruction occurs, leaving little room for profit. A boom in oil shale production has been anticipated since the 1940s, but its profitability remains a mirage, continually fading into a receding horizon. In 2005, I saw the CEO of Shell’s unconventional resources unit predict that its oil shale project in Colorado would be economically realistic with oil at $20 to $30 a barrel at an oil conference. Yet with an average oil price of $95 in 2011, there is still no commercial scale production from our trillions of barrels of oil shale resources, and the price at which they become profitable is now thought to be as much as $95.
The remaining alternatives have their own intractable issues. Hydroelectric production has been plagued by drought in recent years, and the trend in the U.S. is toward dam removal, not new dam construction. Nuclear power is beset with so many challenges — enormous up-front capital and loan guarantee requirements, unacceptable liability limits, extremely long lead times, the impending retirement of much of the existing U.S. fleet, and the externalized costs of waste handling and cleanup, to name just a few — that a nuclear renaissance seems extremely unlikely here. It is possible that the world might yet reach for a new generation of nuclear reactor technology, but there is no evidence that such a resurgence is under way in the U.S., certainly not within the next two decades. And hot fusion remains a hazy decade or more in the future, as it has always been.
The path forward
With this knowledge, there is no intellectually honest way to believe that the world can continue its near-total reliance on fossil fuels for much more than another decade — a paltry window of opportunity. We also know that we cannot wait until they go into decline before reaching for renewables and efficiency, simply because the scale of the challenge is so vast, and the alternatives are starting from such a low level that they will need decades of investment before they are ready to assume the load. The data is clear, and the mathematics are really quite straightforward.
The hard truth is that there are no good fuel substitutes anymore. Throughout human history, we have always been able to find not just a substitute fuel, but a better one: a cheaper, denser, more abundant one. That is simply no longer the case. One may hope for some miraculous technological breakthrough, and one may simply have faith that the invisible hand will solve our problems, but such thin threads are hardly a reasonable basis for policymaking and forecasting.
Between the supply-based and demand-based models lies a chasm of incomprehension. There are no good models that properly account for the feedback loops from GDP to oil demand, oil demand to price, price to supply, and supply to GDP. In the absence of any such models, we would do well to put aside the faith-based demand models and heed what the supply-based models are telling us: that we can't simply wait for the market to choose the right path. Time is not on our side. In the absence of any game-changing alternatives, the precautionary principle should be our guide. If we would see a new Golden Age for humanity, then we must choose the path toward renewables, and we must choose it now. Because the other road leads back to Olduvai.
Photo: Olduvai Gorge (jelite/Flickr)