Energy return on investment

• 1. Introduction

  ›

  Energy return on investment - EROI - is a ratio, and for much of human history, that ratio was enormous. When oil was first discovered in the United States, it took roughly one barrel of oil to find, extract, and process about one hundred barrels in return. That is a 100:1 ratio, a windfall so vast it fueled an entire industrial civilization. By the 2010s, that same ratio for fossil fuel discovery in the United States had fallen to roughly 5:1. What happened in between, and what it means for everything from the price of gasoline to the fate of complex societies, is what this documentary explores. At its core, EROI asks a deceptively simple question: how much usable energy do you get back for every unit of energy you spend to obtain it? When the answer drops to 1:1, the energy source stops being a source at all. It becomes, in the language of energy economics, a net energy sink.

• 2. Origins of a Concept

  ›

  Charles A. S. Hall, a systems ecology and biophysical economics professor at the State University of New York, is credited with popularizing the field of energy analysis. Hall began by applying a biological methodology developed at an Ecosystems Marine Biological Laboratory, then adapted that method to study human industrial civilization. The concept reached its widest audience in 1984, when a paper by Hall appeared on the cover of the journal Science. That moment marked the point at which an idea rooted in ecology crossed over into the study of how societies are powered. A related measure, energy stored on energy invested - or ESOEI - was later developed specifically to analyze storage systems, where the question is not how much energy you extract but how efficiently you can bank it for later use.

• 3. The Threshold That Matters

  ›

  To be considered viable as a prominent fuel or energy source, any resource must clear an EROI ratio of at least 3:1. Below that threshold, too much energy is consumed in the process of obtaining energy to sustain the broader systems that depend on it. Murphy and Hall, writing in 2010, argued in their extended methodology that an EROI of 5 represents the minimum for sustainability. By Hall's own methodology, a value of 12 to 13 is considered the floor below which technological progress and a society capable of supporting high art cannot be maintained. These numbers carry real stakes. When oil sands in 1970 returned a net energy ratio of about 1.0, the resource was barely breaking even. By 2010, one study found that figure had climbed to about 5.23, crossing the sustainability line but still far below conventional oil.

• 4. Renewables Measured

  ›

  Photovoltaic solar is one of the harder energy sources to evaluate using EROI, because the energy invested in production varies so widely depending on technology, methodology, and where the system boundary is drawn. A 2013 meta-study found that figure ranged from a minimum of 300 kilowatt-hours per square meter of module area to a maximum of 2,000, with a median of 585. A 2015 review in Renewable and Sustainable Energy Reviews, using an insolation of 1,700 kilowatt-hours per square meter per year and a system lifetime of 30 years, found mean harmonized EROIs ranging from 8.7 to 34.2 across different photovoltaic module technologies. Energy payback time ranged from 1.0 to 4.1 years. By 2021, the Fraunhofer Institute for Solar Energy Systems calculated that European photovoltaic installations using wafer-based silicon PERC cells could pay back their energy investment in about one year - 0.9 years for Catania in southern Italy, 1.1 years for Brussels. Wind performs differently. Scientific literature places wind turbine EROI at around 16 unbuffered and 4 when storage buffering is included. Data collected in 2018 found that operational wind turbines averaged an EROI of 19.8, with significant variability tied to wind conditions and turbine size. Vestas reports an EROI of 31 for its V150 model. Hydropower, when run for about 100 years, averages an EROI of roughly 110, a figure that reflects both the longevity of the infrastructure and the relatively low ongoing energy cost of running water through a turbine.

• 5. Fossil Fuels in Decline

  ›

  Conventional oil from geological sources returns between 18 and 43 units of energy for each unit invested, a wide range that reflects differences in reservoir quality and depth. Oil shale sits at the opposite end of the spectrum. Extracting oil from kerogen in shale requires substantial process heat, typically supplied by natural gas burned directly or used to generate electricity for underground heating elements. The resulting EROI lands around 1.4 to 1.5, barely above the break-even point. Critics have noted that the natural gas consumed in shale heating could instead serve directly as transportation fuel, delivering a higher EROI and lower carbon emissions than the oil extracted from the shale it was used to heat. Looking at the broader picture, the weighted average EROI of all oil liquids - including coal-to-liquids, gas-to-liquids, and biofuels - is projected to fall from 44.4 in 1950 to 6.7 in 2050. Natural gas tells a parallel story: its standard EROI is estimated to drop from 141.5 in 1950 to a plateau near 16.8 in 2050. Nuclear power occupies a wide middle ground, with plant-level EROIs ranging from 20 to 81.

• 6. Civilizations and Energy

  ›

  Thomas Homer-Dixon argues in "The Upside of Down" that a falling EROI contributed to the collapse of the Western Roman Empire in the fifth century CE. At the height of the Roman Empire, with a population reaching 60 million, the agrarian base produced roughly 1:12 per hectare for wheat and 1:27 for alfalfa, yielding about 1:2.7 for oxen. That system could sustain a population calculable from a daily requirement of around 2,500 to 3,000 calories per person. But ecological damage - deforestation and soil fertility loss, particularly in southern Spain, southern Italy, Sicily, and especially north Africa - began pulling down agricultural EROI in the 2nd century. By 1084, Rome's population, which had peaked at 1.5 million under the Emperor Trajan, had collapsed to only 15,000. Homer-Dixon sees similar dynamics in the Mayan and Cambodian collapses. Joseph Tainter identifies diminishing EROI returns as a central cause of the collapse of complex societies more broadly, connecting the pattern to peak wood in early civilizations and to high-quality fossil fuel depletion in industrial ones. Newer research from 2024 added a further complication: when EROI is measured at the stage of useful energy rather than final energy, renewable electricity sources outperform fossil fuels even after accounting for intermittency.

• 7. The Measurement Problem

  ›

  Calculating EROI for total energy output is relatively straightforward, particularly when the output is electricity. The harder question is where to draw the line on energy inputs. If steel is used to build a nuclear power plant, should the energy cost of producing that steel be counted? Should the energy used to build the factory that made the steel? What about the roads used to transport the materials, or the energy used to feed the workers? These questions have no clean answers, and they explain why published EROI figures for the same energy source can vary dramatically. In 2016, Hall observed that much of the published research in this field is produced by advocates or individuals connected to the business interests of competing technologies. He noted that government agencies had not yet provided adequate funding for rigorous analysis by more neutral observers. One response to this problem came from a 2010 paper by Murphy and Hall, which proposed an extended boundary protocol intended to produce more realistic and consistent comparisons. Meanwhile, the International Energy Agency has endorsed a different methodology that generates more favorable values for certain technologies, particularly photovoltaic solar - an approach that has remained controversial. Richards and Watt proposed an Energy Yield Ratio specifically for photovoltaic systems as an alternative, using the known design lifetime of a system rather than its actual operational life, allowing multi-component systems with different component lifetimes to be compared on equal terms.

• 8. Energy Cannibalism and Breeding

  ›

  A solar breeder is a photovoltaic manufacturing plant that powers itself using panels installed on its own roof, and the concept points toward one possible answer to what researchers call energy cannibalism. When any energy technology grows rapidly, it must consume energy to build new capacity, and at high enough growth rates it can cannibalize the net output of existing plants. The solar breeder sidesteps this by becoming self-sufficient first, then generating surplus energy for the broader grid. Research on the concept was conducted by the Centre for Photovoltaic Engineering at the University of New South Wales in Australia. A solar module processing plant in Frederick, Maryland was originally planned as such a breeder. In 2009, the Sahara Solar Breeder Project was proposed by the Science Council of Japan as a joint effort between Japan and Algeria, with the stated goal of creating hundreds of gigawatts of capacity within 30 years. On the storage side, a Stanford University team's assessment of ESOEI found that without pumped hydroelectric storage - which carries an ESOEI of 704, against lithium-ion's 32 - pairing wind energy with battery technology alone would not clear the investment threshold, pointing instead toward curtailment as the more rational choice.