The energy transition isn’t just sounding the death knell for fossil fuels. According to some experts, it has also revealed flaws in an idea that has bugged some academics for decades: As we move to less energy-dense fuels, could we end up without enough surplus for society?
At the heart of this debate is one of the most important physical metrics you’ve never heard of: energy return on investment, or EROI.
Devised in the 1980s by systems ecologist Charles A.S. Hall and others, the basic principle behind EROI, also called energy returned on energy invested, is simple: A source of energy is only useful if you can get more energy out of it than what you put in.
This obvious point has sustained debate over perpetual motion machines since the Middle Ages and continues to be a key issue for the commercial viability of technologies such as nuclear fusion. For Hall, an important application of the concept was in oil extraction.
Oil is an energy-dense fuel. If it can be obtained with relatively little energy expenditure, it can yield a big return in EROI terms. But as it gets harder to extract, and more energy is required for the process, the return on that energy investment diminishes — and at some point, this process may yield oil that’s not worth the energy put into it.
In his 2017 book on EROI, subtitled “A Unifying Principle for Biology, Economics and Sustainability,” Hall details how advanced societies require a surplus of energy to sustain the multitude of non-energy-producing activities that take place in modern life.
Society’s hierarchy of energetic needs can be represented graphically as a pyramid, with extracting and refining energy at the bottom and delivering education, health care, arts and so on at the top.
A key concept for the survival of society?
Hall and scholars such as Jessica Lambert of Next Generation Energy Initiative, a nongovernmental organization, calculated that the minimum EROI required for crude oil extraction would be 1.1:1. In other words, you would need to get 1.1 joules back from every joule invested to make it worthwhile.
But extraction is only the first step. To refine the oil as well, you would need an EROI of 1.2:1. To transport it, you would need to get 3 joules back from every joule invested, and so on.
Using this analysis, the researchers calculated that sustaining arts and similarly advanced societal activities would require an EROI of 14:1. In other words, only fuels that could deliver at least 14 times the amount of energy they consumed would be able to keep modern society afloat.
If that sounds alarming, there’s worse to come. Hall and others applied EROI calculations to a range of energy sources. The results for renewables did not look promising.
In a 2014 review, Hall, Lambert and Stephen Balogh of State University of New York found that PV had a mean EROI of around 10:1. Onshore wind fared better, at 18:1, but the authors noted that “the value in practice may be less due to the need for backup facilities.”
Of all the carbon-free generation options studied, only hydropower emerged as a promising energy source, with an EROI of 84:1. Even nuclear power appeared to offer marginal potential to support society, with an EROI of 14:1.
The authors admitted that challenges in collecting basic data could lead to inaccuracies in the EROI estimates. But the apparently low values for wind and particularly solar were worrying because they did not include the extra energy required for backup power or storage, which would further reduce EROI.
Growing doubts over the usefulness of EROI
Could this mean that policies aimed at getting most of society’s power from wind and solar would condemn humanity to an energy death spiral in which there would not be enough surplus to support modern civilization? If so, EROI should be on every energy policymaker’s agenda.
Yet EROI discussions have tended to remain academic in nature. That may be because investors are more concerned with financial measures such as levelized cost of energy. But it could also be due to the fact that the seemingly simple equation cannot be practically transferred to real-world applications, particularly in relation to renewables.
That’s the view of a number of distinguished energy thinkers, such as Schalk Cloete, of Norway’s Stiftelsen for Industriell og Teknisk Forskning research organization. “The true EROI of a technology is just too complex to quantify accurately, given the challenges with assessing full lifecycle energy inputs, converting between different energy sources, and accounting for effects on the rest of the energy system,” he said in an email.
Others see more fundamental issues with the concept of tying an energy resource’s value to an energy input when that input is theoretically almost limitless. It could be argued, for example, that a low EROI for solar is irrelevant because the resource is so abundant that even a small return could power modern society.
“The focus of [energy returned on energy invested] research is often with respect to whether the [energy returned on energy invested] of a particular type of energy production is ‘high enough’ relative to some goal, such as 3:1 or 10:1,” wrote energy researcher Gail Tverberg in a recent blog post. “I believe that there needs to be more focus on the total quantity of net energy produced.”
Adding a new dimension to the debate
Michael Liebreich, CEO of Liebreich Associates and the founder of BloombergNEF, highlights this flaw in EROI logic by citing the example of two solar companies producing identical PV panels. The first uses twice as much energy as the second, making its EROI much worse.
But that comparison fails to capture any information about the energy inputs in question. That makes it a “dimensionless” metric, and because of that, it is “absolutely useless,” Liebreich said.
“Maybe that company has a bad EROI because it has access to very cheap energy, so the other one is more efficient,” he told GTM in an interview. “All these [EROI] charts are founded on an assumption that energy is limited.”
To illustrate another apparent shortcoming of EROI, imagine two hypothetical energy sources with different EROIs, say, 10:1 and 20:1. Based on the EROI values alone, the second energy source seems twice as useful. You might say 1 megawatt-hour of energy invested in the second source gets twice as many megawatt-hours back.
But now assume the second source takes twice as long as the first to deliver the return on the energy invested. If the first source delivers 10 GWh over 10 years and the second delivers 20 GWh over 20 years then both produce 1 GWh per year, so they are of equal value on an annual basis.
Time is important in energy systems, said Liebreich, because even a modest EROI could be useful if the return on investment can be recouped in a brief period — days or weeks instead of years. The EROI concept, he said, “should never have got through peer review.”
Euan Mearns, a senior researcher at the Swiss Federal Institute of Technology in Zurich who formerly used EROI as the basis for a concept called the energy cliff, agrees with Liebreich’s analysis.
“I think it best to move the discussion to net energy since it is straightforward to talk about the rate of net energy delivery,” he said in an email. “That is another way of saying ‘energy payback time.’”
A largely academic question in today’s energy markets
Hall, who retired from a full-time professorship in 2012, said most attacks on the EROI concept “are specious and due to incomplete or deliberately distorted knowledge,” and have been rebutted in past research.
“Higher EROI allows most people to undertake other economic activities,” he said in an email. “A very low EROI means most people have to work in the energy-supply industries. One could reduce our EROI to 2.5:1 and survive, but there would be…[fewer] possibilities for other activities.”
At the same time, he conceded that “EROIs make a little more sense for a finite fossil fuel resource, like an oil field. We normally do sort of a rolling average, or more precisely the gain for one year over the cost for that year. We tend to see these values declining over time.”
Calculating the EROI of renewable energy technologies such as wind power was “somewhat more difficult to do” as “it depends on the time the facility would be running,” he said.
Hall also hinted at other limitations of EROI in real-life situations. “Fracked oil wells have had a decent EROI — 12:1 or more — when analyzed, but basically were a financial failure,” he said. “I am not sure why.”
Such misgivings may comfort those who see EROI as a fundamental value that could limit humanity’s ability to progress beyond fossil fuels. For the most part, though, debate over the usefulness of the concept will likely remain within academic circles for the foreseeable future.
“EROI’s main strength is that it considers the whole life costs of an asset,” said Tom Palmer, managing consultant at Cornwall Insight, in an email. “However, it is rather complex and not really relevant to the pricing of output. It is used very rarely [and] has limited value to decision-makers.”