Energy Return on Investment: The Ratio Behind the Argument
A source that returns less energy than it consumes is not an energy source at all, whatever else it may be. That makes EROI one of the few genuinely fundamental screens in energy analysis. It is also one of the most inconsistently applied, which is why the same technology appears in the literature with values differing by an order of magnitude.
What the Ratio Measures
The calculation is simple to state. Add up all energy delivered by an installation over its life, divide by all energy consumed to build, fuel, operate and decommission it. A ratio of 20 means twenty units out for one in.
The concept was developed in ecology before it entered energy analysis, where Charles Hall and others applied it to the question of whether an energy system supports the society that runs it. The underlying insight is that only the surplus - the energy left after energy production has taken its share - is available for everything else a society does.
That framing gives the ratio its significance. At an EROI of 20, five percent of gross energy goes back into producing energy. At 5, twenty percent does. At 2, half. The relationship is non-linear, which is the analytically important part: falling from 40 to 20 barely changes the surplus, while falling from 5 to 2 changes it enormously. Arguments that treat every decline in EROI as equally serious miss this.
Conventional oil illustrates the trajectory. Early twentieth-century fields that flowed under their own pressure returned well over 100 units per unit invested. Current global averages are much lower, and unconventional sources such as oil sands and tight oil are lower still, because best deposits are always exploited first and what remains requires more energy to extract.
Why the Numbers Disagree
The boundary problem dominates everything else. Does the energy invested include only the fuel used at the site? The energy embodied in the steel and concrete? The energy used to build the factory that made the steel? The energy consumed by the workers commuting to that factory?
Each of these is defensible and each produces a different answer. Narrow boundaries yield high values; extended boundaries yield low ones. Published EROI figures for solar have ranged from around 30 to below 5 in respectable literature, and most of that spread is boundary choice rather than genuine disagreement about the technology. Comparing figures from different studies is therefore usually meaningless, and comparing figures within one consistently applied study is usually informative.
The second problem is the quality of the energy being counted. A kilowatt-hour of electricity and a kilowatt-hour of coal are not equivalent: converting coal to electricity in a thermal plant discards roughly 60 percent as heat. Comparing an electricity-producing technology against a fuel-producing one without correcting for this systematically favours the fuel, and a great many published comparisons do exactly that.
The third is vintage. EROI is computed from data about how a technology was manufactured, and solar module manufacturing energy has fallen substantially over two decades while efficiency has risen. Any solar EROI figure older than a few years understates the current value, sometimes by a lot, which affects a surprising number of citations still in circulation.
The Minimum Society Needs
A well-known claim in this literature is that an industrial society requires a minimum EROI - values between 3 and 14 have been proposed - below which it cannot maintain the infrastructure, education and healthcare that distinguish it from a subsistence economy.
The reasoning is that everything beyond bare energy production must come from the surplus, and as the ratio falls, the surplus shrinks faster than the ratio does. There is real content here, and it is the strongest argument the framework offers.
The criticism is equally real. Proposed thresholds are derived rather than measured, they vary by a factor of four between authors, and they are computed for whole economies from a concept defined for individual installations. Aggregating installation-level ratios into a societal one requires assumptions that are doing much of the work.
The measured record also complicates the story. Global average EROI has fallen for decades while economic output has risen, mainly because efficiency in converting energy to useful work improved faster than the extraction ratio deteriorated. That does not refute the threshold argument, since a threshold could still lie ahead, but it does show that the ratio alone does not determine outcomes.
What It Settles and What It Does Not
EROI settles one question definitively. A process that returns less energy than it consumes is not an energy source, whatever its other merits. This screen has been genuinely useful, and it is why some proposed fuels are correctly described as energy carriers rather than sources - hydrogen most prominently, since it must be manufactured using more energy than it later delivers.
It does not settle whether a source is worth deploying. That depends on cost, on whether output arrives when needed, on siting relative to demand, on material requirements and on what the alternatives are. A technology with an excellent ratio that produces at the wrong time in the wrong place may be worth less than one with a mediocre ratio that produces reliably nearby.
It is also distinct from the financial calculation. Levelised cost and EROI are correlated, because energy inputs cost money, but they diverge whenever the price of energy differs from its quantity - which is most of the time, and especially where labour, capital cost or land dominate the economics.
The disciplined way to use the concept is as a screen and a sanity check rather than a ranking. Any claim that a source produces energy should be checkable against an honest accounting of what it consumes, with the boundary stated. For a genuinely new conversion method, that accounting is the first thing an informed reader will want and the first thing worth publishing, because it is the one test that a source either passes or fails outright.
Frequently asked questions
What is EROI?
Energy return on investment: the total energy an installation delivers over its life divided by the energy consumed to build, fuel, operate and decommission it. A ratio below 1 means the process consumes more energy than it produces, which disqualifies it as an energy source regardless of its other properties.
Why do published EROI values differ so much?
Mainly because of where the system boundary is drawn. Counting only site fuel gives high values; including the energy embodied in materials, in the factories that made them, and further back gives much lower ones. Published solar figures have ranged from around 30 to below 5, and most of that spread is boundary choice.
Has oil's EROI fallen?
Substantially. Early twentieth-century fields that flowed under their own pressure returned well over 100 units per unit invested. Current global averages are far lower and unconventional sources lower still, because the best deposits are always exploited first and what remains takes more energy to extract.
Is there a minimum EROI a society needs?
It has been proposed, with values between 3 and 14 depending on the author. The reasoning is that everything beyond energy production comes from the surplus, which shrinks faster than the ratio does. The thresholds are derived rather than measured and vary by a factor of four, so the argument is suggestive rather than settled.
Does a high EROI mean a source is worth building?
No. EROI is a necessary condition, not a sufficient one. Whether a source is worth deploying also depends on cost, on whether output arrives when it is needed, on location relative to demand, and on material requirements. A good ratio at the wrong time in the wrong place can be worth less than a mediocre one nearby.
