Imagine a house whose smoke alarm has been sounding for years. The inhabitants no longer argue about whether something is burning. They can smell it. Buckets stand ready. The fire brigade has published detailed reports. Yet the water never quite reaches the flames.
This is no longer a story about the alarm. It is about who is standing on the hose — and why that question is asked so rarely.
The ecological emergency is not suffering from a lack of evidence. The years 2015 to 2025 were the eleven warmest in the instrumental record. In 2025, the global mean temperature was about 1.44 °C above the 1850–1900 level. Climate change is only the most visible signal. Six of nine planetary boundaries are now assessed as being outside the safe operating space: climate, biosphere integrity, land-system change, freshwater change, biogeochemical flows and novel entities.
The political gap is just as clear. Current policies still point towards roughly 2.8 °C of warming this century. Even full implementation of existing national pledges would leave the world far outside the Paris target range. The technologies needed for much faster emission cuts already exist.
So the decisive question has changed. It is no longer whether we know enough. It is why well-established knowledge still fails to produce limits commensurate with the danger.
The mechanism we keep missing
There is no single explanation for this failure. Elections reward short horizons. States compete. Infrastructure locks societies into existing paths. Transitions create real distributional conflicts. People deny, defer and defend the familiar. Administrative capacity is often weak.
Reducing all of this to capitalism or corporate power would be analytically lazy.
But one mechanism has an exceptional structural reach. It does not merely sit beside the other barriers. It can finance them, organise them, amplify them and make them politically useful. At the same time, it obstructs its own correction.
I call it the externalisation–power feedback.
Externalised damage creates advantage. Advantage creates power. Power prevents the full attribution of the damage. The advantage therefore persists.
How the loop works
Picture a factory on a river. It saves money by discharging waste downstream, where the damage lands on other people’s fields, wells and children. With the money saved, the factory grows. It buys advertising, funds political campaigns and hires lawyers. When the town finally debates a filter ordinance, those lawyers draft the exemptions. A year later, the river is dirtier, the factory is richer and it is better placed than ever to fight the next ordinance.
Scaled from one town to the planet, the same circle has four steps.
1. The costs are shifted out
An activity generates private or institutional benefits while a substantial share of its ecological and social costs is borne elsewhere: by the public, weaker communities, other states, future generations or non-human life. The displacement may be legal. It can result from missing rules, weak liability, inadequate prices, subsidies or cumulative damage that is difficult to attribute to a single actor. Intent is not decisive. The economic advantage is.
2. Externalised costs become accumulation
Whoever does not carry the full cost can realise higher profits, rents, market share or public revenue. That accumulation creates more than wealth. It buys organisational permanence, expertise, legal capacity, data, communication power and local dependency. The beneficiaries may be private corporations, state-owned companies, financial actors or resource-dependent states. The mechanism is particularly visible in private capital accumulation, but it is not confined to capitalist ownership.
3. Economic power becomes political power
Accumulated resources can be converted into lobbying, campaign finance, business associations, commissioned expertise, litigation, revolving-door careers, advertising, public relations, research funding and apparently independent communication networks. None of these instruments is automatically illegitimate. The problem is the starting asymmetry: actors who profit from externalisation possess above-average resources — generated partly by that externalisation — to shape the rules meant to constrain them.
4. Correction is blocked or weakened
Influence does not need to defeat regulation outright. Delay is often enough. So are exemptions, weak liability, long transition periods, an impossible burden of proof or the transfer of costs to households and the state. A policy can look ambitious while remaining materially inadequate. Harmful activity stays profitable, new resources accumulate and the next round of influence begins.
The externalisation–power feedback loop.
The theoretical point is simple: under certain institutional conditions, ecological harm produces not only profit, but also the power to defend access to that profit. The error finances its own continuation.
Why this barrier is different
Many obstacles to change are serious. Few combine the same properties within one closed circuit.
First, the mechanism is self-financing. The longer costs remain externalised, the larger the financial and organisational means available to defend that externalisation can become.
Second, it activates other barriers. It can magnify fears about jobs and prices, present transitional technologies as final solutions, shift responsibility from structures to individuals and invoke international competition as a reason why no country should move first. These concerns may be real. The mechanism turns them into durable veto positions.
Third, it changes the effective distribution of democratic power. Political rights may be formally equal, but influence is not. A citizen has a vote. A large corporation can employ lobbyists, lawyers, researchers, communication agencies and political networks. Both may speak, but they do not speak with the same volume or endurance.
Fourth, it acts on visibility itself. Money influences which studies are amplified, which proposals are labelled realistic, which voices receive attention and which doubts are repeated until they appear reasonable. This does not mean that every neglected idea is good or that every debate is manipulated. It means that public visibility is not a neutral contest decided by quality alone.
Finally, concentrated wealth can shield the beneficiaries from the damage. The same resources that finance obstruction can buy private adaptation, secure supply, spatial mobility and distance from the places where consequences arrive first. Those with the greatest power to correct the system are often the least immediately exposed to its failure.
A system in which the beneficiaries of harm can buy distance from harm will not reliably correct itself through their own exposure.
The numbers reveal the scale
The International Monetary Fund estimates direct fiscal subsidies for fossil fuels at 725 billion US dollars in 2024. It also calculates 6.7 trillion dollars in implicit subsidies — mainly the underpricing of climate damage, air pollution and other social costs. These are not direct payments and the estimate depends on methodological assumptions. But the structural point is hard to escape: fossil energy remains artificially attractive because major costs do not appear in its market price.
The next link is also documented. More than two billion dollars was spent on climate-related lobbying in the United States between 2000 and 2016. Research on the Waxman–Markey climate bill estimated that corporate lobbying reduced its probability of passage by 13 percentage points, with an expected social cost of 60 billion dollars in 2018 prices.
Corporate funding has also been linked to the production and amplification of polarising climate narratives. Outright denial has increasingly been replaced by subtler delay: someone else should act first; technology will solve it later; the costs are too high; consumers are responsible; meaningful change is impossible.
The strength of the mechanism is difficult to measure precisely because much of it remains opaque. In 2025, only seven per cent of listed energy companies worldwide disclosed their positions on climate regulation, and only 15 per cent reported their memberships in business associations. Among large companies surveyed by the OECD, only a minority published lobbying expenditure.
The evidence is still uneven and heavily concentrated on the United States. No global dataset yet tests the full loop from externalised cost to accumulated power, from power to obstruction and back to renewed externalisation. That is a limitation. It is not a reason to ignore the mechanism; it is a reason to test it properly.
The strongest objection
A serious opponent would say that the argument overstates power and understates complexity.
Billions of people depend on fossil energy, industrial agriculture, global supply chains and infrastructures that cannot be replaced overnight. Governments must weigh energy security, employment, affordability and development. Science can describe risks, but it cannot decide by itself which costs, timelines and distributions are socially legitimate.
This objection is correct as far as it goes. Political delay cannot be reduced to manipulation. Social dependencies are real. Companies can defend existing assets while investing in transformation. States are not merely victims of lobbying; they preserve harmful systems for fiscal, geopolitical and social reasons of their own. Strong regulation can also be badly designed, socially unjust or ineffective.
The feedback thesis therefore does not explain every delay. It does not attribute hidden intent to every actor. And it does not prove that every restrictive policy is good policy.
But complexity does not explain why the speed of correction has remained below the speed of damage for decades, even as the risks have grown and the available technologies have improved.
That is where the feedback matters. It turns real trade-offs into permanent arguments against adequate action. It invokes consumer demand while prices, infrastructure and advertising help produce that demand. It invokes uncertainty while slowing investment in alternatives. It invokes social justice while leaving the underlying distribution of costs untouched.
The mechanism is not the only cause. It occupies an exceptional position because it can finance, organise and defend the conditions under which the other barriers remain effective.
No conspiracy is required
A traffic jam does not require a conspiracy. No driver plans it. Each driver acts more or less rationally, and the aggregate result is paralysis.
The externalisation–power feedback works in the same way. A company protects its assets. An association represents its members. A government protects jobs, tax revenue or energy supply. A bank limits short-term risk. A party avoids unpopular costs. Each decision may be institutionally understandable. Together, they can prevent the correction that all of them ultimately depend on.
That is what makes the mechanism hard. It does not need villains or secret meetings. It rewards behaviour that appears reasonable from each actor’s narrow position, even when the aggregate result destabilises the shared basis of life.
Moral appeals alone cannot break such a structure. The rules must change.
What would break the loop?
If the diagnosis is right, climate and environmental policy cannot focus only on visible emissions and individual behaviour. It must interrupt the conversion of externalised harm into obstruction capacity. That requires action at several points at once.
End externalisation. Ecological damage and risk must be attributed as far as possible to the activities that generate them — through prices, liability, binding budgets and restoration duties.
Limit the conversion of money into political power. Political finance, lobbying, association positions, revolving-door careers and commissioned expertise must be transparent, controllable and restricted where fundamental conflicts of interest arise.
Correct asymmetries of proof and information. Where damage may be severe or irreversible, the burden of proof cannot remain permanently with diffuse victims. Companies and public authorities must disclose cumulative effects, compatibility with binding limits and less damaging alternatives.
Create automatic safeguards. Scientifically established boundary transgressions should trigger predefined legal consequences. Otherwise every protective measure must be fought for again under the same unequal balance of power.
Finance restoration. Revenue from harmful activities must flow into the stabilisation and restoration of shared foundations of life, adaptation and social compensation. This is not retroactive moral punishment. It is present responsibility towards the public, future generations and those already carrying the costs.
Give the biosphere constitutional rank. Its functional integrity cannot remain one ordinary interest among many. It is the material precondition under which economies, legal orders and democratic choice can continue to exist.
The detailed institutional design requires separate work. But the direction is clear. We must regulate not only the harmful material flows, but also the political mechanism through which the harm reproduces itself.
The point we can no longer avoid
We do not know everything. We know enough.
The persistent gap between knowledge and effect can no longer be explained as an information deficit. People repress. States compete. Infrastructure binds decisions. Transformation creates real conflicts. Yet one feedback loop gives these barriers unusual durability: the continuation of the problem finances the power to resist its correction.
A system with this structure does not automatically repair its ecological error. It rewards the error’s continuation.
Effective Earth-system protection therefore cannot stop at better technology, individual virtue or new targets. It must attribute costs, limit asymmetric influence, establish binding safeguards and finance restoration.
Catastrophe is not a path of correction.
It is what remains when correction has been prevented for too long.
—
This essay condenses a working paper. The full version, with complete references and testable hypotheses, is archived at https://doi.org/10.5281/zenodo.21807118 (CC BY 4.0).
Selected sources and further reading
Black, S., et al. (2025). Underpriced and Overused: Fossil Fuel Subsidies Data 2025 Update. IMF Working Paper 2025/270. https://doi.org/10.5089/9798229034715.001
Brand, U., & Wissen, M. (2021). The Imperial Mode of Living. Verso.
Brulle, R. J. (2018). The climate lobby: A sectoral analysis of lobbying spending on climate change in the USA, 2000–2016. Climatic Change, 149, 289–303. https://doi.org/10.1007/s10584-018-2241-z
Colgan, J. D., Green, J. F., & Hale, T. N. (2021). Asset revaluation and the existential politics of climate change. International Organization, 75(2), 586–610. https://doi.org/10.1017/S0020818320000296
Farrell, J. (2016). Corporate funding and ideological polarization about climate change. PNAS, 113(1), 92–97. https://doi.org/10.1073/pnas.1509433112
Gilens, M., & Page, B. I. (2014). Testing theories of American politics: Elites, interest groups, and average citizens. Perspectives on Politics, 12(3), 564–581. https://doi.org/10.1017/S1537592714001595
IPCC. (2022). Climate Change 2022: Mitigation of Climate Change. https://doi.org/10.1017/9781009157926
Lamb, W. F., et al. (2020). Discourses of climate delay. Global Sustainability, 3, e17. https://doi.org/10.1017/sus.2020.13
Meng, K. C., & Rode, A. (2019). The social cost of lobbying over climate policy. Nature Climate Change, 9, 472–476. https://doi.org/10.1038/s41558-019-0489-6
OECD. (2025). Global Corporate Sustainability Report 2025. https://doi.org/10.1787/bc25ce1e-en
Richardson, K., et al. (2023). Earth beyond six of nine planetary boundaries. Science Advances, 9(37), eadh2458. https://doi.org/10.1126/sciadv.adh2458
UNEP. (2025). Emissions Gap Report 2025: Off Target. https://www.unep.org/resources/emissions-gap-report-2025
WMO. (2026). State of the Global Climate 2025. https://wmo.int/publication-series/state-of-global-climate/state-of-global-climate-2025