At the centre of contemporary environmental politics sits a question that seems, at first glance, both practical and unavoidable: can economic growth continue without destroying the living world? The evidence of damage is no longer marginal or eccentric. Climate change is accelerating. Biodiversity is declining. Fisheries, forests, soils, rivers and aquifers are being pushed beyond their capacity to recover. Plastics, pesticides, nitrogen, heavy metals and countless synthetic chemicals circulate through bodies and ecosystems. One no longer has to be a radical ecologist to accept that endless material expansion on a finite planet is impossible. That premise has moved from the edge of public thought toward its centre. Only the most technologically enchanted imagine that fusion power, asteroid mining, artificial meat, terraforming and planetary engineering will free us from all earthly limits. Such visions may have billionaire sponsors and glossy renderings, but they remain fantasies of escape.
At the same time, the political religion of economic growth remains remarkably intact. Governments that acknowledge ecological limits still speak as if growth is the condition of all social possibility. Without growth, we are told, employment falters, public budgets collapse, pensions become unaffordable, debts become unmanageable, and politics turns ugly. Growth is presented not only as a way to reduce poverty in poor countries, but as the life force of rich societies too. In wealthy countries, where basic material needs have long been exceeded for much of the population, the argument is rarely that growth is needed to feed the hungry. More often, it is that growth is needed to keep the social machine from stalling.
This produces an apparent contradiction. The ecological story says that material expansion must meet limits. The economic story says that expansion must continue. The concept that promises to reconcile them is “decoupling”. The word sounds technical, almost modest, but it carries enormous political weight. It suggests that economic output can keep rising while resource use and environmental damage stop rising, or even begin to fall. Growth, in this view, can be detached from its dirty material shadow. The economy can become larger while its footprint becomes lighter. We can have more GDP without more ecological harm. Growth can be green.
This idea is attractive because it appears to preserve the basic structure of modern political hope. It tells governments that they need not choose between prosperity and the planet. It tells businesses that innovation will solve the problems created by earlier forms of production. It tells citizens that the familiar promise of rising living standards can survive, merely powered by different energy, cleaner technologies and more efficient systems. The smokestacks can be scrubbed, the cars electrified, the grids decarbonised, the factories optimised, the waste recycled, the forests restored and the economy expanded. The old conflict between growth and limits dissolves into a programme of modernisation.
But this is precisely why decoupling has become such a contested idea. To some, it is the only realistic route through the ecological crisis. To others, it is a comforting illusion, a kind of macroeconomic greenwashing that postpones harder questions about consumption, inequality and power. The public argument has often been conducted as a battle between “green growth” and “degrowth”. Green-growth advocates insist that the evidence already shows decoupling is possible, at least for some pollutants and in some wealthy countries. Degrowth advocates reply that the evidence is partial, that emissions reductions are too slow, that resource use remains coupled to GDP, and that the language of decoupling hides the continued overconsumption of affluent societies.
Paul Krugman, writing in The New York Times, argued that growth can be green and that it is possible to detach economic expansion from environmental harm. His target was not only conservative opponents of environmental policy, but also environmentalists who claim that growth itself is the problem. In his framing, pessimism about decoupling risks becoming an accidental ally of anti-environmental politics: if growth and ecological protection cannot coexist, then those who value growth will reject ecological protection.
Writers at Our World in Data have made similar arguments, often with a particular emphasis on poverty. If the world cannot decouple growth from ecological harm, then global development appears trapped in a cruel dilemma: either poor countries remain poor, or ecological destruction continues. From this perspective, rejecting decoupling can look morally reckless. It can seem to deny the aspirations of billions of people who still need better housing, electricity, sanitation, food security, healthcare, transport and education.
Degrowth thinkers have responded sharply. Timothée Parrique has accused green-growth advocates of selecting convenient evidence and ignoring the scale, speed and breadth of the ecological crisis. Jason Hickel has argued that wealthy economies remain deeply dependent on excessive material and energy use, much of it displaced through global trade. Andrew McAfee, by contrast, has attacked degrowth as unnecessary and even anti-scientific, claiming that rich countries have begun to reduce their pressure on the Earth. These exchanges are often heated because they are not really about one statistic. They are about the shape of the future.
One side fears a politics of ecological austerity, in which limits become a justification for denying development and lowering expectations. The other fears a politics of technological complacency, in which promises of green growth allow wealthy societies to avoid changing the systems that created the crisis. Both fears are serious. Both can be supported by evidence, depending on what one chooses to measure. And that is the first clue that the question “Is decoupling possible?” is badly formed.
The argument about decoupling is not one argument, but several arguments disguised as one.
Before any useful answer can be given, we have to ask what kind of decoupling is being discussed. Relative decoupling means that environmental pressure grows more slowly than the economy. This may sound encouraging, but if the pressure is already beyond sustainable limits, slower growth in damage is still damage. Absolute decoupling means that the economy grows while the environmental pressure actually declines. Even this may not be enough, because decline has to happen at the necessary speed and scale. A country that reduces emissions by 1 per cent a year while needing to reduce them by 7 or 8 per cent a year is technically improving and practically failing.
We must also ask what is being decoupled. “Environmental harm” is not a single thing. Carbon dioxide is not the same as nitrogen runoff. Urban air pollution is not the same as tropical deforestation. Freshwater depletion is not the same as ocean acidification. Species extinction is not the same as plastic waste. A society may improve on one measure while worsening on another. It may clean its city air while importing goods produced through deforestation elsewhere. It may reduce coal use while expanding mining for lithium, copper, nickel and rare earths. It may protect one forest while burning another as “renewable” biomass.
This is where much of the confusion begins. Examples of successful decoupling are often drawn from environmental problems that are relatively tractable. Local air pollution, for example, can be reduced through regulation, filters, fuel standards and the closure of dirty plants. Sewage can be treated. Lead can be removed from petrol. Certain chemicals can be banned. These are real achievements, and they should not be dismissed. But they do not prove that an entire economy can grow indefinitely while its total ecological burden shrinks within planetary limits.
Climate emissions are a more difficult and revealing case. They are deeply connected to energy systems, industry, transport, agriculture and patterns of consumption. Yet even here, some absolute decoupling has occurred. Several affluent countries have increased GDP while reducing territorial carbon emissions. In some cases, consumption-based emissions have also fallen, meaning the decline is not only the result of outsourcing dirty production. Coal has been replaced by gas, wind, solar, nuclear or other lower-carbon sources. Energy efficiency has improved. Climate policy has mattered. Technology has mattered. These are not trivial facts.
But they are not the end of the story. The decline in emissions in wealthy countries has usually been too slow to meet climate targets. The world as a whole has not achieved the necessary absolute decoupling. And climate is only one dimension of the ecological crisis. The danger is that carbon becomes the one problem we know how to count, while the rest of the living world becomes a footnote. A solar panel still requires materials. An electric car still requires roads, batteries, mining, land and electricity. A low-carbon economy can still be a high-extraction economy. A society may decarbonise and still expand its appetite for land, metals, water and biological life.
That is why material use has become central to the decoupling debate. Total resource extraction is an imperfect measure, but it forces attention back to the physical economy. Modern economies have not become weightless. The digital cloud lives in data centres, cables, minerals, cooling systems and electricity grids. The service economy sits on buildings, transport networks, food systems, devices and global supply chains. The paperless office did not abolish forests, mines or landfills. It simply changed the pathways through which material demand appears.
The difficulty is that material flows cannot be reduced to a single moral number. A tonne of sand is not a tonne of fish. A tonne of wheat is not a tonne of uranium. A tonne of sustainably harvested timber is not a tonne of old-growth rainforest. Counting tonnes can obscure as much as it reveals. Yet the aggregate still matters because scale matters. An economy that extracts, moves, processes and discards ever more physical stuff will tend to generate more ecological pressure, even if each unit is produced more efficiently. Efficiency can lower impact per unit, but if the number of units expands fast enough, total impact can still rise.
This is the familiar problem of rebound. A technology becomes more efficient, lowering cost, which can increase use. Cars become more fuel-efficient, but people drive more. Lighting becomes cheaper, and illuminated space expands. Data storage becomes more efficient, and data production explodes. Efficiency is essential, but efficiency inside an expansionary system does not automatically produce restraint. It can become a way of making growth cheaper.
The most interesting question, then, is not whether decoupling can occur in isolated cases. It can. The more important question is whether decoupling can occur across the whole material metabolism of affluent societies, fast enough, without merely shifting burdens elsewhere, while also allowing poorer societies to meet legitimate needs. On that question, the evidence is much less reassuring.
Relative improvement becomes dangerous when it is mistaken for absolute escape.
The word “decoupling” itself may be part of the problem. It suggests that the economy is one thing and the environment another, and that progress consists in separating them. But economies have never been separate from ecosystems. They are organised flows of energy, materials, labour, care, knowledge and waste inside the biosphere. Food, housing, transport, heating, medicine, computing and education all have material conditions. Even finance, that most abstract of economic forms, ultimately commands claims on land, labour, energy and future production.
In this sense, the ambition should not be decoupling in the literal sense. It should be a different kind of coupling. We need to recouple economic activity to ecological reality. The economy should be tied more tightly, not more loosely, to carbon budgets, watershed limits, soil regeneration, biodiversity protection, material caps and human wellbeing. The danger of the decoupling narrative is that it lets GDP float upward as the unquestioned objective while ecological indicators are asked to fall away beneath it. But perhaps GDP is the wrong object to protect. Perhaps the task is not to free growth from nature, but to bind production to the conditions that make life possible.
This shift changes the debate. Instead of asking, “Can GDP keep growing while damage falls?” we should ask, “Which forms of production need to expand, which need to contract, and who gets to decide?” Renewable energy must grow. Public transit may need to grow. Home insulation, ecological restoration, healthcare, education, repair, care work and low-carbon housing may need to grow. Private jets, disposable fashion, oversized vehicles, planned obsolescence, speculative construction, industrial meat and luxury emissions may need to shrink. The economy is not a single balloon that must inflate or deflate as a whole. It is a patterned system. Some parts are starved; others are swollen.
This is where the standard growth debate becomes crude. GDP adds together cancer treatment and cigarette sales, oil spills and cleanup operations, luxury towers and public housing, arms production and school meals. It does not ask whether activity meets needs, whether it damages ecosystems, whether it increases resilience, or whether it concentrates power. A society can raise GDP by doing things that make life worse and lower GDP by doing things that make life better. Everyone knows this, and yet the growth imperative survives because GDP is woven into employment, taxation, debt, pensions, investment and political legitimacy.
Green growth tries to repair this problem by making GDP cleaner. Degrowth tries to challenge the social dependence on GDP growth itself. But perhaps the more precise demand is selective growth and selective contraction under ecological constraints. This is less slogan-like than either green growth or degrowth, but it is closer to the work that must actually be done. We need growth in capabilities, security, repair and ecological knowledge. We need contraction in throughput, waste, extraction and forms of consumption that provide status rather than wellbeing.
The political difficulty is that contraction is never abstract. It always happens somewhere, to someone. A steelworker hears “less material use” differently from a climate academic. A rural community hears “less car dependence” differently from an urban planner. A poor country hears “limits to growth” differently from a rich country whose wealth was built through centuries of extraction and emissions. This is why ecological politics cannot be reduced to planetary accounting. It must also be a politics of justice, transition and power.
Poorer societies need more energy, infrastructure and material goods. Billions of people still require better housing, refrigeration, sanitation, transport, hospitals, schools and communications. For them, some forms of material growth are not excess but liberation. The scandal is that the ecological space needed for that development is being consumed by affluent overuse. The problem is not that humanity as a whole desires too much in the same way. The problem is that some consume far beyond sufficiency while others remain below dignity.
The promise of decoupling is seductive because it seems to avoid this conflict. It implies that rich societies can continue expanding their consumption patterns while poor societies develop, because technology will make the whole arrangement clean. But if total material use must fall, and if some people still need more, then others must use less. This is not a technical detail. It is the political heart of the matter.
There is another reason decoupling is too narrow: it often treats environmental damage as a side effect of production rather than as a feature of provision systems. People need nutrition, not industrial monocultures. They need mobility, not necessarily private cars. They need thermal comfort, not wasteful buildings. They need communication, not disposable devices. They need social participation, not endless commodity turnover. If we ask how to decouple GDP from harm, we begin with the existing economy and try to clean it. If we ask how to provide good lives within ecological limits, we begin with needs and redesign systems around them.
The question is not how to make every existing activity greener, but which activities deserve a future.
Consider transport. A green-growth approach may focus on replacing petrol cars with electric cars. This is necessary, but insufficient. Electric cars reduce tailpipe emissions and can reduce lifecycle emissions, especially with cleaner electricity. But they do not eliminate congestion, road deaths, tyre particles, parking space, sprawl or the material demand of mass car ownership. A provisioning approach asks a different question: how can people access work, school, care, leisure and community with less energy and material throughput? The answer may include electric cars, but also trains, buses, cycling, walking, denser cities, remote services and local planning. The goal is not a greener traffic jam.
Housing offers another example. A decoupling story may celebrate more efficient appliances, heat pumps and lower-carbon construction materials. These matter. But if homes keep getting larger, if cities sprawl outward, if second homes multiply, if speculative real estate leaves existing buildings empty while new ones are built, efficiency will struggle against expansion. The deeper question is how to provide secure, comfortable, affordable housing with fewer materials, less land pressure and lower energy demand. That requires planning, tenure reform, public investment and social norms, not only better technology.
Food is similar. Agricultural emissions can be reduced through better fertiliser management, renewable energy, methane reduction, precision farming and less waste. But diets, land use and power in food systems matter too. Industrial meat, especially beef, places enormous pressure on land, water, feed crops and climate. A society can make farms more efficient while still degrading soils and habitats. Again, the challenge is not merely to decouple the existing food economy from its impacts, but to reshape the system that determines what is produced, how, for whom and at what ecological cost.
The same holds for energy. It is absolutely necessary to build renewable energy at extraordinary speed. But if total energy demand continues to climb without limit, the transition becomes harder, slower and more extractive. A high-energy society powered by renewables is preferable to a high-energy society powered by fossil fuels, but it is not automatically sustainable. The easier path to decarbonisation is not only cleaner supply; it is also intelligent demand reduction, especially among those whose energy use is excessive.
This is where the language of “sacrifice” often misleads. Some reductions would indeed be sacrifices, especially for people already living precariously. But many forms of high-consuming life are not making people happier, healthier or freer. Long commutes, polluted air, debt-driven consumption, insecure work, status competition, throwaway goods and overheated housing markets are not signs of a successful civilisation. A good life within limits is not necessarily a smaller version of consumer society. It may be a different arrangement of time, space, care, security and pleasure.
The idea of “A Good Life for All Within Planetary Boundaries” points in this direction. The relevant question is not whether every country can maximise GDP while reducing every impact. It is whether all people can achieve decent lives while humanity remains within ecological thresholds. At present, no society has fully achieved both. Rich societies tend to meet more social needs but overshoot ecological limits. Poorer societies often remain within some ecological boundaries but fail to meet basic social thresholds. This is not an argument for poverty. It is an argument for redesigning prosperity.
Such redesign will not be delivered by markets alone, though markets can play a role. It requires public investment, regulation, planning, democratic conflict, institutional imagination and cultural change. Carbon prices may help, but they cannot decide what kind of cities to build. Green innovation may help, but it cannot by itself determine who benefits from the transition. Corporate efficiency may help, but corporations are not designed to ask how much is enough. If ecological limits are real, then the economy must become more deliberately governed.
This does not mean replacing all markets with central planning, nor does it mean romanticising poverty or stagnation. It means recognising that the growth imperative is not a law of nature. It is an institutional arrangement. Pension systems, debt systems, employment systems, tax systems and political promises have been built around expectations of growth. If those expectations become ecologically impossible or socially destructive, then the institutions must change. A society dependent on endless expansion is not pragmatic. It is fragile.
The defenders of green growth are right about one thing: technological pessimism can become self-defeating. We should not deny the genuine achievements of clean energy, pollution control, electrification, efficiency and environmental regulation. Solar and wind power have changed the realm of possibility. Batteries are improving. Heat pumps work. Some pollutants have been dramatically reduced. Policy can work. Innovation matters. The future is not a simple extrapolation of the past.
But the critics of green growth are also right about something essential: possibility is not enough. The question is speed, scale, distribution and total impact. A pathway that is technically possible but politically delayed, materially explosive or dependent on continued inequality is not a serious answer. “Can it be done?” must be followed by “for whom, by when, at whose expense, and within which limits?”
The old environmental Kuznets curve suggested that pollution rises as societies industrialise and then declines once they become wealthy enough to care about the environment. There is some truth in this for certain local pollutants. But as a general theory of ecological modernity it is dangerously incomplete. Wealthy societies often clean their immediate surroundings while expanding their distant footprints. They protect local rivers while importing water-intensive crops. They reduce domestic emissions while consuming goods made elsewhere. They preserve landscapes at home while relying on extraction abroad. Riches can buy cleanliness nearby without reducing damage overall.
In the 21st century, the central ecological problem is not only dirty production. It is unequal command over planetary capacity. Some people command vast flows of energy and materials through their purchasing power, investments and infrastructures. Others lack the material basis for a decent life. Decoupling, when framed only as GDP versus environmental impact, can obscure this inequality. It treats the economy as a national or global aggregate, when the real economy is divided by class, region, race, gender and history.
This matters because ecological transition will fail if it is experienced as punishment by those with little power while leaving luxury untouched. Asking ordinary people to accept higher costs, smaller homes or reduced mobility while the wealthy continue to fly privately, own multiple properties and shape policy through lobbying is politically toxic. A serious ecological politics must begin with the high-consuming classes and sectors. The first contraction should occur where excess is greatest and social value is weakest.
None of this means that GDP will instantly disappear from public life. It remains embedded in accounting, comparison and policy. But we should stop treating it as the central moral indicator of progress. A rise in GDP should not be presumed good; a fall should not be presumed bad. The question is what changed in material reality. Are people better housed, better fed, healthier, freer, safer and more educated? Are ecosystems more resilient? Are emissions falling fast enough? Is resource use declining where it must? Are burdens and benefits distributed fairly? These are the questions hidden beneath the abstraction of growth.
In that sense, the debate over decoupling may be a transitional argument. It belongs to a period in which societies are beginning to understand ecological limits but still speak the language of 20th-century economic expansion. The word tries to reassure us that the old dashboard can remain in place if only the engine is redesigned. But the dashboard itself is part of the problem. We need indicators that measure provisioning, resilience, ecological pressure, inequality, health, time, care and democratic capacity. We need to know not only whether the economy is bigger, but whether it is doing the right things with less damage.
The real task is not to decouple growth from harm, but to recouple prosperity to limits.
This recoupling would be practical before it is philosophical. It would mean binding climate policy to material policy, so that decarbonisation does not simply produce new extractive frontiers. It would mean designing cities around access rather than speed, housing around security rather than speculation, food systems around nutrition rather than commodity yield, and energy systems around sufficiency as well as supply. It would mean reducing luxury emissions and wasteful throughput while expanding the material foundations of dignity for those who lack them. It would mean asking, again and again, not whether an activity adds to GDP, but whether it contributes to a viable and decent world.
The challenge is enormous because it cuts against the habits of modern political economy. Growth has been the great solvent of conflict. It allowed societies to promise more to workers without taking too much from owners, to fund welfare without fully confronting wealth, to defer distributional battles into an expanding future. Ecological limits make that bargain harder to sustain. If the pie cannot grow materially forever, then distribution returns to the centre of politics. That is why the decoupling debate is so emotionally charged. It is not only about carbon curves or resource graphs. It is about whether modern societies can remain legitimate without the promise of endless expansion.
There is no single answer that applies everywhere. Poor countries need room to build. Rich countries need to reduce excess. Fossil fuel regions need just transitions. Cities need different solutions from rural areas. Cold countries face different housing and heating needs from warm ones. Mineral extraction for renewable energy must be governed differently from extraction for disposable consumption. Ecological politics must become more specific, not more abstract.
The future will not be saved by one grand theory of growth, nor by one grand theory of degrowth. It will be shaped by thousands of decisions about infrastructure, ownership, technology, labour, land, food, transport, energy and care. Some sectors must expand rapidly. Some must disappear. Some must be transformed beyond recognition. The question is not whether “the economy” grows. The question is what kind of economy is being built, whose needs it serves, and what it asks from the Earth.
Decoupling remains useful if it is treated as a diagnostic tool. It can show whether a given pressure is rising or falling relative to economic output. It can reveal where policy and technology are succeeding. It can puncture lazy claims that every environmental improvement requires recession. But as a political vision, it is too thin. It offers a separation where we need responsibility, an accounting relation where we need transformation.
The living world does not care whether GDP rises or falls. It responds to concentrations of greenhouse gases, the clearing of forests, the draining of wetlands, the acidification of oceans, the poisoning of soils, the fragmentation of habitats, the depletion of aquifers and the pressures placed on species. Human beings, likewise, do not live on GDP. They live through food, shelter, care, energy, friendship, health, meaning, safety and time. The economy is valuable only insofar as it helps provide these conditions without destroying their foundations.
The real environmental question, then, is not whether growth can be made green in the abstract. It is whether societies can learn to pursue good lives within limits, and whether they can do so quickly enough to avoid irreversible damage. That requires more than decoupling. It requires choosing what to grow, what to shrink, what to repair, what to abandon and what to protect. It requires a politics mature enough to distinguish need from excess, innovation from evasion, and prosperity from throughput.
We may yet build economies that are cleaner, fairer and more secure than the ones we have now. But we will not get there by pretending that the economy can float free of the Earth. The task is not to cut the cord between prosperity and nature. There is no such cord to cut. The task is to remember that prosperity has always lived inside nature, and to build institutions that behave accordingly.
