A recent article by two Google engineers, Ross Koningstein and David Fork, in IEEE Spectrum has raised quite a discussion. The article entitled “What It Would Really Take to Reverse Climate Change” discusses Google’s investment in the “RE<C” project that sought to “…develop renewable energy sources that would generate electricity more cheaply than coal-fired power plants do”. The goal was to produce a gigawatt of power (presumably installed capacity). Google abandoned the project in 2011, according to the article because they believed it would not meet their cost goal and would also not avert significant impacts from climate change (they state the need to keep atmospheric concentrations of CO2 below 350 ppm as suggested by James Hansen).
I commend the two engineers for writing this article discussing their efforts and thoughts. However, I see this foray into energy as typical of the Silicon Valley mentality that is used to “solving” some technological problem quickly, selling the company or idea to a larger company, and then moving on to the next great app. Whether it is RE<C or making advanced biofuels from algae or cellulosic feedstocks, the Silicon Valley stereotype thinks the “energy problem” will be solvable just like cellular phones and that their “energy days” will be another line on their CV. Unfortunately, the realities of the energy production business are more difficult to change than realities on the energy consumption side of the business. Most innovative companies of the last several years are emerging to use information to consume energy more smartly because we no longer have the money and demographics to increase energy consumption. This is part of the new reality.
The Google engineers don’t mention the solution that will come about but needs no technology: consuming less energy. This will be the only solution that actually reduces CO2 emissions, but it will instead coincide with higher energy prices and costs, not “cheap zero-carbon energy” as is stated as a goal. The reason is because of the rebound effect, or Jevons Paradox (named after the British economist William Stanley Jevons). The cheaper energy becomes, the more the world consumes in the aggregate of all people consuming energy and not just a single device (refrigerator, car etc.) becoming more energy efficient.
Even divergent opinions on the limits of the planet and human endeavours agree that the effect of cheap energy is to increase total consumption compared to if energy were more expensive. I explain this concept via two books I use for my energy class at The University of Texas: The Bottomless Well (TBW) and Limits to Growth: The 30-year Update (LTG). I specifically use these two books (there are other possibilities) to force students to understand widely divergent opinions on how people interpret the past use of energy for guiding (or not) future energy policy and use of natural resources. TBW is optimistic on human ingenuity, the discovery of new technologies, and increased efficiency to provide the services we crave. LTG accepts that humans are clever animals, but also understands the physical constraints of a finite planet will eventually even trump gains in efficiency (so that production and consumption do not increase infinitely), forcing the reduction of consumption and physical stocks that we can maintain (largely people and industrial capital). TBW says it is best for the government to get out of the way of industry in improving technologies. LTG says that forward-looking policies are (really “would have needed to have been already”) necessary to minimize environmental damage and promote the necessary equity that will be needed after the world peaks in annual throughput (e.g. ~ GDP, but not exactly).
From the IEEE Spectrum article, I view Google as starting in the TBW camp, but never quite reaching the conclusion of the LTG authors. That is to say they no longer believe technology can solve the problem (they stopped their project), but they believe the solution is some new technology that we have yet to create. The Google authors state in their IEEE Spectrum article: “Our reckoning showed that reversing the trend [of increasing atmospheric CO2 concentration] would require both radical technological advances in cheap zero-carbon energy, as well as a method of extracting CO2 from the atmosphere and sequestering the carbon.” They further state: “Not only had RE<C failed to reach its goal of creating energy cheaper than coal, but that goal had not been ambitious enough to reverse climate change. That realization prompted us to reconsider the economics of energy. What’s needed, we concluded, are reliable zero-carbon energy sources so cheap that the operators of power plants and industrial facilities alike have an economic rationale for switching over soon—say, within the next 40 years.” Businesses choose the most economic solutions because those are the ones that give them the greatest chance of growing, not shrinking. If all businesses are growing, and storing, streaming, and beaming more and more information in the cloud servers that Google has provided us with, then this requires more resources, not less … more emissions, not less. Cheap low-carbon energy might coincide with cheap high-carbon energy too, because if it is really cheap enough, we might be growing enough to continue to afford fossil energy. Personally, I doubt this outcome because the large growth days are over. But how do we really assess how “cheap” energy really is? Let’s look to a time series from the UK.
Figure 1 shows a calculation from Roger Fouquet on the cost of energy in England and the United Kingdom. What a nice piece of work! (Note: The UK is perhaps the best example of understanding long-term energy costs and the transition to fossil fuel usage starting in earnest in the late 1700s.) If we use England and the UK as a proxy for the modern world, Fouquet’s calculations indicate that the last decade (2000-2010) was effectively the time of cheapest energy in the history of mankind (see Figure 1). It was cheap energy that enabled the human population to reach 7 billion. In other words, cheap energy enabled us to farm land more intensively with less human effort to produce more food such that it was possible to increase the population. Without modern farming (fertilizer inputs based on creating ammonia from the hydrogen in natural gas, liquid-fueled combustion engines in tractors, fossil fueled transport and storage of food) we would not have 7+ billion people on the planet. It is simply too expensive and physically impossible to feed 7 billion people via subsistence farming. More expensive food and energy (really, food is energy) puts downward pressure on population, and that in turn puts downward pressure on the environment.
Figure 1. The cost of energy for energy services as a percentage of England and United Kingdom gross domestic product [Data courtesy of Roger Fouquet].
At the end of the article, the Google team states: “We’re not trying to predict the winning technology here, but its cost needs to be vastly lower than that of fossil energy systems.” There are two mathematical ways for competing technologies to become vastly lower cost than fossil energy systems. Either the new technologies become cheaper while fossil energy stays roughly constant (or becomes cheaper more slowly), or fossil energy becomes more expensive while the competing technologies get cheaper, stay the same cost, or increase more slowly. The real curb on resource consumption and CO2 emissions will be indicated by aggregate energy costs per Figure 1. If energy spending as a fraction of GDP increases, it indicates we are reaching diminishing returns to consumption and our responses (e.g. research, new energy resource extraction) are inadequate to continue increasing consumption. This would be the interpretation if we are trying to make energy cheaper and cheaper (most people and governments want this). However, it is theoretically possible to purposely choose (e.g. by policy) to increase energy spending as a fraction of GDP. Putting a price or tax on CO2 emissions is an example policy (Note: Internalizing the cost of CO2 emissions makes fossil fuel consumption more expensive but does not make renewable energy cheaper.).
Energy has practically never been cheaper than during the time Google has existed as a company. If energy is already this cheap, how can we say it is not cheap enough to invest in technologies to mitigate fossil fuel impacts (carbon capture from coal-fired power plants and even capture of CO2 from the air)? The common statement is that we need (low-carbon) energy to be cheaper to mitigate climate change. This is tantamount to us “Waiting for Godot” to arrive. It’s as if we’re saying: “we’re so smart, but if only we were a little smarter, we’d have the cheap unobtainium we’ve been hoping for so that we can do as many things as we want with no environmental impact.” Unfortunately, all elements in the periodic table have mass and obey the laws of physics, not our social laws of economics. There are fundamental energetic (low energy return on energy invested) reasons why we have yet to be able to “policy induce” cellulosic liquid biofuels into existence.
The climate solution that Google could not find is not made of some more of some new kind of widget; it is made of less of all past and future widgets. We got into the climate predicament by millions of incremental advancements, and perhaps we’ll only reduce emissions rates in the same way. In terms of practically playing in the energy space, as a hybrid solution to “solving” the energy problem, Google ventured into energy management by buying Nest (thermostats that learn your habits and program your home climate control) earlier this year. This should pay dividends for all, with the tradeoff going further into an Orwellian future of increased mass information on citizen activities. It is unclear if these types of technologies will help Google (and the rest of us) decrease environmental impacts, increase use of low-carbon energy, or decrease greenhouse gas emissions rates. But we can be sure that Google owning NEST certainly follows their existing business model of gathering more information to continue selling targeted ads based on your habits.