Why decoupling will not happen

The first essay in a five piece series on the economy and resource use

There are many opinions about decoupling and here you have my take. The short version is:

- It is indeed possible to decouple carbon dioxide emissions from economic growth.

- It is more difficult to decouple useful energy from economic growth.

- It is not likely that economic growth can be decoupled from general resource use, because ”the economy” is basically about transformation of natural resources into commodities – regardless of how many layers of services we add.

- Because of the globalisation of value chains, decoupling of national economies are not particularly relevant as a proof of anything.

- How growth of GDP is calculated has a huge impact on the results, but this is rarely recognised. The measure is a lot more subjective and ambiguous than most people understand and the development of GDP over time is based on even more dubious calculations.

- Efficiency is not a recipe for decoupling, nor is circular economy and other green growth fantasies.

Oil pump, Pana, Illinois, image: Gunnar Rundgren

As part of my ongoing book writing project, tentatively called “the weight of money”, I have expanded my analysis of why decoupling is a fantasy. In five posts, I will explain why.

This first article is about what decoupling means and the empirical evidence.

The second article is about how the GDP is calculated and in particular how growth in GDP is calculated and why that has a huge impact on the notion of decoupling.

The third article is about the service paradox of – why the assumed transition to a service economy is not reducing our impact on the environment.

The fourth article is about efficiency and other aspects supposedly making it possible to combine economic growth and lower resource use.

The final article is more directly linking to the main object of the book: despite all abstractions, the economy is basically still about using labour to convert natural resources into commodities to sell.

*

Let’s start with some definitions:

Absolute or relative?

When speaking about decoupling one has to be clear about whether it is an absolute or a relative decoupling that is discussed. An absolute decoupling means that emissions (or whatever is discussed) are decreasing. A relative decoupling means that the emissions don’t grow as fast as the GDP grows. Essentially, an absolute decoupling is necessary to avoid dramatic climate change. ‘

What?

Often, decoupling is specifically about carbon dioxide (CO2) emissions, but sometimes it addresses greenhouse gas emissions (mostly calculated as CO2 equivalents, a very dubious concept), or energy use, pollution or use of “materials”.

Which entity?

It is also essential to clarify which the entity that is subject to research, a production process, a firm, a country, a value chain, the global economy. Boundaries matter – a lot.

Which period?

It is also very important to be clear about which period we are discussing and if that period has some special conditions that needs to be taken into account.

*

There is a flow of reports and articles claiming that absolute decoupling is happening. For example, a report from the Energy&Climate Intelligence Unit (2025)ⁱ demonstrates that since the signing of the Paris Agreement 2015, 43 countries have reached absolute decoupling and 40 countries have reached relative decoupling of carbon dioxide emissions. When you take a critical look in these kinds of reports, you often find weaknesses. In this case, I haven’t really checked it: let’s just assume that they have done the math.

There are no theoretical obstacles for decoupling of emissions to take place. In many countries there have been a development where fossil gas (mostly called “natural gas”, but I reject that green-washing) is replacing coal in power generation or heating^ii. Just by switching from coal to gas in electric power generation, you reduce emissions with 50 percent! ⁱⁱⁱ On a global level around 80 percent of the supplied energy is still of fossil origin, even if coal is declining and gas is increasing. So you can very well increase the use of fossil fuels and still reduce emissions. Another important driver is the increased use of biomass. In climate budgets, carbon from biomass is not counted as emissions (some argue for that it should, but let’s not discuss that now), so when you skip coal and burn wood instead you get zero emissions.

In Sweden, the use of biomass has increased with 350 percent since 1970, and biomass is now the biggest energy source.^iv In the EU as a whole biomass use increased with 200 percent between 1990 and 2010^v and in the US it increased with 240 percent between 1973 and 2023.^vi

We also need to consider prices. In particular when it comes to oil. Oil reached its lowest price, around 12 dollars per barrel, in the end of 1960 Since then oil has risen considerably, albeit with huge fluctuations. The first half of the 20^th century, oil was still relatively expensive. In that period the use of fossil fuels increased with 1.7 percent annually, while the global economy grew by 2.13 percent. Between 1945 and 1973 the fossil fuel use grew with a staggering 4,.48 percent per year while the economy grew with 4.18 percent, even the opposite of relative decoupling (not sure if there is a term for that?).^vii

When oil prices sky-rocketed with the two oil crises in the seventies, people started to economise, cars got smaller, industries reduced waste and farmers stopped using ultra excessive quantities of nitrogen fertilisers (which are just fossil fuels in drag). An even more remarkable case is how use of oil, gas and nitrogen fertilisers shrank dramatically with the collapse of the Soviet Union. In the Soviet Union, energy had been almost free, and waste was rampant.^viii With this in mind, it is not hard to find proof or decoupling of carbon dioxide emissions and economic growth in the period from 1980 and onwards. Still, the overarching global picture is that the only periods of global absolute decoupling is in periods of economic contraction, such as the collapse of the Soviet Union, the Global Financial crisis 2008 and the Covid pandemic.

If one also consider the linkage between useful energy and economic growth we can see that an even stronger correlation. After all, it is the useful energy that drives the economy. For example, if you drive a car, it is the power rotating the wheels that you are after, not the noise, the exhaust or the generation of heat (well, if you live in Sweden, you do like some of that heat half of the year). If you have a car with an internal combustion engine (ICE!) around one third of the energy in the petrol or diesel is converted to that power. In addition, already before that, the oil had to be extracted, refined and transported and pumped, so the actual energy use is even higher. Meanwhile, in an electric car almost 90 percent of the energy is rotating the wheels. And electricity generation itself has gone from mostly coal to hydro, nuclear, sun, wind and gas, all more efficient than coal. Similarly a gas power station has a lot higher energy efficiency.^ix According to the calculations by Carey King, the useful energy and GDP are strongly linked together between 1900 and 2000 in USA, Great Britain and Austria, but not in Japan since 1980^x.

If we, instead of fossil fuels and energy, look into the material footprint, the picture is less clear as there are so many different kinds of materials. However, looking over a period from 1900 to 2020 we can see that total resource use per dollar has decreases, but with variations over time. It went from 3,7 kg of materials per dollar 1900 to 1.5 kg per dollar in 2012. ^{xi xii xiii C}

Clearly a relative decoupling but absolutely not any absolute decoupling. If wee look in more detail we will see that it is almost only biomass use per dollar that has shrunk, while the use of fossil fuels, metals and other materials have fluctuated between 1 and 1.5 kg per dollar, with a peak from WWII until 1970s (see above).^xiv Between 1970 and 2024, metal use increased with 300 percent and gravel, sand and rock increased with 454 percent (mainly as a result of building and construction), while the population increased with “just” 122 percent ^xv

Let’s now turn to which entity we are measuring. Global figures are of course the most relevant as the economy to a large extent is global. But mostly supposed decoupling is demonstrated per country. Already in that case, we have the distinction between territorial or consumption based emissions. If most of the dirty industries are outsourced, it is not hard to show a growing economy and decreasing emissions in the territory. But also the consumption based emissions represent a narrow look. Instead we need to look into where and how the money is generated, because it is the linkage between emissions and the economy that is the topic. If you are an oil producing and exporting country, a huge share of your GDP will be derived from oil exports, but the only emissions of that oil that will be booked on your country is the emissions in the extraction, within country transport and refinery. The combustion of Norwegian oil and gas cause emissions which are ten times as high as Norway’s own emissions. To that should be added the emissions caused by the use of the nitrogen fertilisers that Norway also export.^xvi That Norway is the global champion of adoption of electric vehicles doesn’t make much difference.

As the GDP is measured through the value added in the economy, the most relevant way of assessing any decoupling, is to link them to the emissions in the value chain. A study of 866 products in 28 countries showed that 45 percent of the emissions were caused by suppliers and subcontractors and 23 percent by the controlling company and 32 percent by usage and disposal.^xvii If you are a car company or a maker of mobile phones most of the emissions are caused in those countries where most of the industrial work is done, e.g. in China, while most of the value is captured by controlling company, located in high-income countries.

In a ground-breaking study 2018, Pablo Piňero and colleagues showed that if emission of a value chain is distributed according to the value created in the various stages, countries like Sweden and Germany would cause much higher emissions. In the Swedish case, the emissions based on value addition are 42 percent higher than when calculated based on consumption and 53 percent higher than if you calculate territorial emissions.^xviii

In this article we have only discussed how emissions and resource use are calculated and allocated, assuming that the GDP and GDP-growth are solid scientific facts. But as I will demonstrate in the next article, they are not.

Rreferences

i Energy and Climate Intelligence Unit, 2025, 10 Years Post-Paris: How emissions decoupling has progressed globally, December 2025

ii IEA (2019), The Role of Gas in Today’s Energy Transitions, IEA, Paris https://www.iea.org/reports/the-role-of-gas-in-todays-energy-transitions.

iii EIA 2024, How much carbon dioxide is produced per kilowatthour of U.S. electricity generation?, 11 December 2024 https://www.eia.gov/tools/faqs/faq.php?id=74&t=11.

iv Energimyndigheten 2025, Energiläget i siffror 2024

v Kalt, Gerald 2012, Bioenergy in the context of the EU 2020- and 2050-policy targets: Technology priorities, opportunities and barriers, 12th IAEE European Energy Conference.

vi US EIA 2024, Monthly Energy Review, 23 December, 2024, https://www.eia.gov/totalenergy/data/monthly/.

vii Fressoz, Jean Baptiste och Christoph Bonneuil, the Idea of infinite growth from fossil capitalism to green capitalism, in Borowy, Iris och Matthias Schmelzer 2017, History of he future of economic growth.

viii FAOSTAT

ix Lumley, Graham 2024, Coal vs Natural Gas: Comparing Fossil Fuels, BKV Energy, https://bkvenergy.com/learning-center/coal-vs-natural-gas/.

x King, Carey W 2021, The economic superorganism, Springer

xi Pineault, Éric 2023, A social ecology of capital. Pluto Press.

xii Bithas, Kostas och Panos Kalimeris, Unmaskuing decoupling, Science of the total environment 619-630 (2018) 338-351.

xiii Hickel, Jason och Giorgios Kallis 2020, Is green growth possible, New political economy vol 25, no 4 469-486.

xiv Pineault, Éric 2023, A social ecology of capital. Pluto Press.

xv Global Material Flows Database 2025, International Resource Panel, https://www.resourcepanel.org/global-material-flows-database.

xvi NOU 2023: 25, The transition to low emissions s. 211.

xvii Meinrenken, C.J., Chen, D., Esparza, R.A. et al. Carbon emissions embodied in product value chains and the role of Life Cycle Assessment in curbing them. Sci Rep 10, 6184 (2020). https://www.nature.com/articles/s41598-020-62030-x

xviii Piñero, P., Bruckner, M., Wieland, H., Pongrácz, E., & Giljum, S. (2018). The raw material basis of global value chains: allocating environmental responsibility based on value generation. Economic Systems Research, 31(2), 206–227.