🔬 physics

What Do EROIs Measure? Implications for Energy Transition Assessment

This paper argues that only the external formulation of Energy Return on Investment (EROI), which excludes self-consumed energy and properly accounts for upstream embodied energy in multi-source systems, accurately measures the net energy surplus available to society, whereas standard and internal variants misleadingly reflect process efficiency and can significantly alter assessments of energy transition viability.

Original authors: Thomas Norway, Olivier Cavalié

Published 2026-06-02

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Thomas Norway, Olivier Cavalié

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are running a lemonade stand. To make lemonade, you need lemons, sugar, water, and ice. But you also need to run a blender, cool the stand with a fan, and maybe even drive to the market to buy supplies. All of that takes energy.

In the world of energy research, scientists use a tool called EROI (Energy Return on Investment) to measure how "worth it" an energy source is. It's a simple ratio: How much energy do we get out compared to how much energy we had to spend to get it?

If you spend 1 unit of energy to get 10 units of back, that's a great deal (EROI of 10). If you spend 1 unit to get 1.1 units back, that's a bad deal (EROI of 1.1).

This paper argues that for decades, scientists have been calculating this ratio in three different ways, and two of those ways are misleading us about the true health of our energy system.

The Three Ways to Count the "Cost"

The authors say there are three main ways to count the energy you spend (the "investment"):

The "Internal" Way (INT): This counts everything you use, including the energy you burn just to keep your own factory running.
The "Standard" Way (STD): This counts the energy you burn to keep the factory running, but subtracts the energy you actually sell to the customer.
The "External" Way (EXT): This is the method the authors propose. It says: "Don't count the energy we burn to keep our own lights on as an investment cost. That's just a loss, like a leak in a bucket. Only count the energy we had to buy from the outside world to get started."

The "Leaky Bucket" Analogy

Imagine you have a bucket of water (energy) that you are trying to pour into a community fountain.

The Problem: Your bucket has a hole in it (self-consumption). You have to keep pouring water just to keep the hole from draining the whole bucket before it reaches the fountain.
The Old Ways (INT & STD): These methods say, "Look how much water you had to pour into the bucket just to keep it from leaking! That's a huge cost!" They treat the water lost to the hole as a "cost of doing business."
The New Way (EXT): This method says, "Wait a minute. The water that leaked out is gone. It never reached the community. The only thing that matters is: How much water did you have to buy from the well to fill the bucket in the first place?"

The authors argue that the "Internal" and "Standard" ways are like measuring the efficiency of your bucket-making process, not the actual water you delivered to the community. The "External" way measures the actual surplus available to society.

The "Net Energy Cliff"

Why does this distinction matter so much? The paper introduces a concept called the "Net Energy Cliff."

Imagine a hill. At the top, the hill is flat. If you slide down a little bit, you don't fall far. This represents high energy efficiency. But as you slide down toward the bottom, the hill gets steeper and steeper. A tiny step down results in a massive drop.

The authors show that when you use the "Internal" or "Standard" methods, you often make energy sources look like they are sitting on the steep part of the cliff (in danger of collapse). But when you use the "External" method, you realize they are actually still on the flatter, safer part of the hill.

The Analogy: It's like looking at a bank account.

Old Method: "You spent $500 on gas to drive to work, and you only made $550. You're barely breaking even!" (Focuses on the cost of the trip).
New Method: "You spent $500 on gas to drive to work, but you made $550. Your net gain is $50, which is what you can actually spend on dinner." (Focuses on the surplus).

If you miscount the "gas" as a massive investment rather than a simple cost of operation, you might think you're bankrupt when you're actually fine.

The "Passing the Buck" Problem

The paper also tackles a tricky situation where energy systems talk to each other. Imagine a coal mine that uses electricity from a wind farm to run its drills.

The Mistake: If you just add up the coal mine's energy costs and the wind farm's energy costs separately, you might count the electricity twice.
The Fix: The authors created a new formula. They say: "If the coal mine uses electricity, don't count the electricity itself as a cost. Instead, count the hidden energy (the 'embodied energy') that was required to build the wind farm and the power lines that made that electricity possible."

Think of it like a relay race. If Runner A passes a baton to Runner B, you don't count the baton as a new cost for Runner B. You count the energy Runner A used to run their leg of the race. The authors' formula ensures that when we add up the whole team's performance, we don't double-count the effort.

What They Found in Real Life

The authors tested their new "External" method on real data:

US Oil and Gas (1919–2007): When they stopped counting the oil used to pump oil as a "cost" and treated it as a "loss," the efficiency numbers went up significantly.
China's Fossil Fuels (1995–2010): Same thing. By fixing how they counted the energy exchanged between different factories, the systems looked much more efficient than previously thought.

The Bottom Line

The paper concludes that for a long time, we have been using a ruler that is slightly bent. We have been treating the energy we burn to keep our own machines running as a massive "investment," which makes our energy sources look worse than they actually are.

By switching to the External (EXT) method, which only counts the energy we import from the outside world and treats internal consumption as a simple loss, we get a truer picture of how much energy is actually left over to build hospitals, schools, and cities. This doesn't mean our energy problems are solved, but it means we aren't as close to the "cliff" as some previous reports suggested.

Technical Summary: What Do EROIs Measure? Implications for Energy Transition Assessment

Problem Statement
The literature on Energy Return on Investment (EROI) suffers from a lack of methodological consensus, with multiple formulations coexisting that differ primarily in their treatment of self-consumed energy and external direct energy inputs. This diversity creates a critical inconsistency: reported EROI values for the same physical system can vary by a factor of two or more depending on the chosen formulation. Furthermore, aggregated and disaggregated assessments of interconnected systems often yield mathematically inconsistent results. Because the relationship between EROI and the net energy surplus available to society is non-linear (the "net energy cliff"), these methodological distortions can shift the diagnosis of an energy system from gradual decline to imminent crisis, carrying significant normative implications for energy transition policy. The core issue is the conflation of energy products with their production costs, where internal self-consumption is often treated as an investment rather than a production loss, and external direct flows are counted at face value rather than as embodied energy equivalents.

Methodology
The authors adopt a biophysical framework, modeling energy systems as dissipative structures comprising three functional blocks: Direct Energy Production (DEP), Indirect Energy Production (IEP), and Development (DEV). The study proceeds through three main methodological steps:

Benchmarking Formulations: Three established EROI formulations—Internal (INT), Standard (STD), and External (EXT)—are evaluated against theoretical limit cases. Specifically, the authors analyze the "Horn of Plenty" scenario (zero external indirect investment) and systems with increasing net energy availability to determine what physical quantity each metric actually measures.
Consistency Testing in Multi-Source Systems: The authors examine interconnected configurations where the output of one DEP unit serves as the input for another. They derive a consistency criterion requiring that the aggregate EROI of coupled systems must be analytically derivable from individual components without residual error.
Derivation of a Reallocation Formula: Based on the consistency criterion, a generic formula is derived to reallocate indirect (embodied) energy investments when direct energy flows are diverted between interdependent units. This ensures that direct energy inputs entering the denominator of an EROI calculation are converted into their upstream embodied energy equivalents.
Empirical Application: The revised framework is applied to two historical datasets: U.S. oil and gas production (1919–2007) and China's conventional fossil fuel sectors (1995–2010). In both cases, the authors re-calculate EROI using the EXT formulation and the new reallocation rules, comparing the results against the Standard (STD) values reported in the original literature.

Key Contributions

Distinction of Metrics: The paper demonstrates that the INT and STD formulations primarily measure the internal conversion efficiency of a production process, as they converge toward finite asymptotes determined by self-consumption rates regardless of the net energy delivered to society. In contrast, only the External (EXT) formulation correctly measures the net energy surplus available to society, as it treats self-consumption as a production loss and remains sensitive to increases in net output.
Embodied Energy Accounting: The authors establish that in multi-source systems, direct energy flows exchanged between units must not be included at face value in the investment term. Instead, they must be converted into their upstream embodied energy equivalents.
Generic Reallocation Formula: A unique mathematical solution is derived for reallocating indirect investments ( $x$ ) when direct energy is diverted from one unit to another. The formula $x = DIV_{jk} / EXT_j$ ensures that the indirect cost associated with producing a diverted flow is transferred to the consuming unit, preserving aggregation consistency across the system.

Results

Theoretical Findings: The analysis confirms that treating self-consumption as an investment (as in INT and STD) artificially lowers EROI values and decouples the metric from the actual net energy surplus. The EXT formulation is shown to be the only physically consistent metric for assessing societal energy contributions.
Empirical Findings: When applied to U.S. oil and gas and China's fossil fuel sectors, the revised framework yields systematically higher EROI values than standard formulations. In the U.S. case, reclassifying direct energy from an investment to a loss increases the reported EROI. In the China case, applying the reallocation formula to cross-flows between interdependent sectors alters the magnitude of the EROI.
Implications for the "Net Energy Cliff": Because the relationship between EROI and net energy fraction is non-linear, the systematic increase in EROI values provided by the revised framework can place the same physical system on the opposite side of the "net energy cliff" compared to standard assessments. This suggests that previous assessments using STD or INT formulations may have underestimated the net energy surplus available to support socio-economic complexity.

Significance
The paper argues that the choice of EROI formulation is not merely a technical convention but a fundamental determinant of how energy systems are evaluated. By demonstrating that only the EXT formulation, combined with proper embodied energy accounting, provides a physically consistent measure of net energy surplus, the authors provide a robust methodological foundation for energy transition assessments. The revised framework suggests that current policy evaluations based on standard formulations may be overly pessimistic regarding the viability of energy systems to support complex modern societies. The work calls for a shift in the field toward metrics that accurately reflect the net energy delivered to society, thereby offering a more reliable basis for assessing the socio-economic trajectories enabled by the energy transition.