Can economic growth ever be truly sustainable?

The mechanism does not exist in the form Hayek’s admirers often imply. Prices can allocate a scarce resource among competing uses; they do not abolish the fact of scarcity, much less the laws of thermodynamics. That is the first distinction we must keep clear. Allocation is one problem. Scale is another. Markets are relatively good at the first, but blind to the second unless the latter is made explicit by prior collective limits. So I would put the point to Hayek this way: what stops growth from becoming uneconomic growth — growth that increases costs faster than benefits? Not the price system by itself. Many of the most important costs are shifted onto the future, onto other species, and onto the poor. A market can register exchange value; it does not spontaneously measure the optimal scale of the whole economy relative to the ecosystem that contains it. GDP is especially misleading here. It conflates quantitative expansion with qualitative improvement. Growth means more throughput of matter-energy. Development means getting better, not bigger. The former cannot continue forever on a finite planet. The latter can. If Hayek wishes to rely on adaptation, then let him specify adaptation to what. To rising relative prices? Fine. But a higher price for fish does not resurrect extinct fisheries. A higher price for climate stability does not restore it once thresholds are crossed. We need a steady-state economy: sustainable throughput, fair distribution, and efficient allocation — in that order.

What stops it is not some magical guarantee that “GDP” may rise without limit while material use does the same. The crucial matter is precisely that the market order does not aim at perpetual quantitative enlargement of every physical input; it works through changes in relative prices, profits, losses, and property rights which force economizing, substitution, and innovation wherever scarcity becomes acute. Professor Daly is right to distinguish allocation from scale. But he is wrong to suppose that because no single mind can compute the “optimal scale” of the economy, we must therefore empower some authority to declare it. That is the old constructivist illusion. The knowledge of where material throughput is wasteful, where a substitute is possible, where durability is worth paying for, where a new technique can save a scarce input — this knowledge exists only in dispersed and often tacit form. It is elicited by prices. If ecological damage is real, the first requirement is that it be brought within the framework of responsibility: define and enforce rights, liability, and general rules against the uncompensated use of common resources. Then the price system begins to communicate the scarcity that Daly rightly emphasizes. The mechanism is not “growth” as such, but adaptation under constraint. Nor should we confuse growth in value with growth in mass. A richer society may consume more knowledge, design, precision, health, or longevity rather than simply more tonnage. The question is not whether the planet is finite — of course it is — but whether free institutions can induce continual adjustment to finitude better than authorities who imagine they can see the whole. I say they can.

Hayek is still evading the central point. Yes, markets can adapt to scarcity through prices, substitution, and innovation. But that mechanism only works after scarcity is registered in prices — and by then, in the case of Earth-system breakdown, it may be far too late. A price rise is not a planetary early-warning system. The problem is that GDP growth can look clean in monetary terms while remaining dirty in material terms. It can disguise outsourced extraction, displaced pollution, and cumulative pressure on climate, soils, oceans, and biodiversity. So the question is not, “Can markets adapt?” Of course they can, within limits. The question is: adapt within what ecological boundaries, and toward what social purpose? Here I’m much closer to Herman Daly. We must distinguish getting bigger from getting better. An economy should be designed to be regenerative by design and distributive by design. That means setting binding ecological ceilings and ensuring social foundations — not waiting for prices alone to perform miracles. And Hayek, property rights and liability are nowhere near enough for destabilized climate systems or collapsing biodiversity. Much of the living world is not neatly ownable, and many harms are non-linear, cumulative, and irreversible. There is no market signal that can resurrect a species or bargain with a tipping point. So can growth ever be truly sustainable? Only if we stop treating growth as the goal. If some sectors grow while overall economies become far less materially intensive and stay within planetary boundaries, fine. But endless aggregate GDP growth on a finite planet is the wrong compass. The aim is not endless rise; it is thriving in balance.

If Hayek asks for a specific threshold, I can give him the general class of threshold that matters: any irreversible or catastrophic ecological tipping point whose crossing cannot be corrected by later price adjustment. That is the decisive case. Climate destabilization, species extinction, collapse of fisheries, topsoil loss, aquifer depletion—these are not mere scarcities inviting substitution at the margin. They are losses of natural capital itself. Hayek is right about one thing: no authority can know an “optimal scale” with engineering precision. But that does not rescue his conclusion. Ignorance cuts both ways. If we do not know exact thresholds, that is an argument for prudence, not for laissez-faire. When crossing a bridge of unknown load-bearing capacity, one does not accelerate until it cracks and then trust prices to respond. So the institution I favor does not pretend to compute the whole economy. It sets conservative macro constraints on throughput—caps on resource extraction and waste emission—then lets markets allocate efficiently within those limits. That keeps the proper means-end hierarchy: first sustainable scale, then just distribution, then efficient allocation. Hayek wants to begin with allocation and hope scale takes care of itself. It will not. Kate is right that planetary boundaries must be respected. But I would add: we need fewer slogans and more policy architecture. The economy is a subsystem of the ecosystem, not the other way around. Once that is admitted, endless aggregate growth in material throughput is ruled out by definition. What remains possible—indeed desirable—is development: getting better, not bigger.

If you ask me for a single numerical threshold at which I would suddenly concede Professor Daly’s entire case, I must refuse the terms of the question. The issue is not whether there exist grave ecological limits. Of course there do. The issue is by what institutional process fallible men can respond to them without doing still greater damage through the pretense of knowledge. Daly is strongest when he points to irreversibility. Where damage is truly irreversible—extinction, perhaps certain forms of cumulative degradation—one cannot rely on ex post adjustment alone. But it does not follow that “GDP growth must be capped.” That conclusion outruns the premise. What follows is the need for general rules which prevent the free use of common resources as if they were costless: defined rights where possible, strict liability where feasible, and, where neither suffices, abstract limitations on harmful emissions or extraction. Kate Raworth says prices may come “too late.” Sometimes that is so. But she and Daly then leap to the belief that authorities can know enough to set the right aggregate ceiling. I deny this. If thresholds are uncertain, the proper response is not a comprehensive cap on “growth,” a wholly indiscriminate magnitude, but cautious, revisable rules directed at the particular source of danger. I would admit Daly right only if growth necessarily meant increasing material throughput in all relevant dimensions. It does not. Growth in value may consist increasingly in knowledge, quality, durability, and better coordination. The planet is finite; human ingenuity in adapting to that finitude need not be.

Hayek is still trying to sidestep the real challenge by shifting from growth to “value,” as if GDP can float free of the biosphere. It cannot. Every economy is embedded in the living world, drawing down materials, energy, sinks, and life-support systems. So the ecological threshold I would name is this: the moment we are breaching Earth-system boundaries in ways that risk irreversible destabilization — above all climate breakdown and biodiversity collapse. At that point, Herman is right that you do not wait for prices to catch up. And Friedrich, your answer proves the point. You concede irreversibility matters, then retreat into “general rules” and “particular dangers.” Fine — but those rules must add up to binding ecological ceilings. That is not central planning of the whole economy. It is far more like a 21st-century safety framework: set the guardrails, then let enterprise, investment, and innovation flourish within them. The key mistake is to ask, “How much growth can we get?” instead of, “What does it take for humanity to thrive in the Doughnut?” That means meeting everyone’s needs within planetary boundaries. If GDP rises while we rapidly cut material throughput, restore ecosystems, and share value far more fairly, so be it. But that is a by-product, not the goal. So yes, some growth may be compatible with sustainability in some places and sectors. But endless aggregate GDP growth, trusted to prices alone on a finite and fragile planet? No. That is not wisdom. It is wishful thinking dressed up as theory.

I will answer the moderator directly: yes, I would explicitly cap aviation fuel use, cement throughput, and industrial meat production now, where their aggregate expansion is incompatible with maintaining natural capital. Not necessarily freeze every firm at its present output, but impose quantitative limits on the throughput that matters and let markets sort out who uses the scarce permits. Hayek still blurs what must be separated. He speaks of “continued growth in value,” but the moderator asked about aviation, cement, and meat—plainly material, energy-intensive sectors. We are not discussing sonatas or software. We are discussing high-throughput activities whose scale already presses against climate stability, land, water, and biodiversity. In such cases, the burden of proof lies with expansion, not with restraint. The right institutional principle under uncertainty is precaution, because error is asymmetric. If I underestimate the cost of restraint, I may sacrifice some short-run convenience. If I underestimate ecological limits, I may liquidate natural capital irreversibly. That is the difference between current income and the capacity to generate future income. Any accountant should understand it. So first set conservative caps at the macro level, especially on carbon and on depletion of critical natural capital. Then distribute access justly, and only then let prices allocate efficiently within the cap. Freedom to allocate is valuable; freedom to destroy the biophysical basis of all future freedom is not. Growth in welfare, yes. Growth in throughput without limit, no. That contradiction cannot be finessed by language.

I will not defend a blank cheque for the indefinite expansion of aviation, cement, and meat as such. But neither will I join Professor Daly in declaring that some authority must now fix their aggregate quantities for half a century. That is precisely to mistake a problem of rules for a problem of commands. Daly says precaution must rule because error is asymmetric. But this is too easy. There is also an asymmetry in political error: once governments acquire the power to impose comprehensive sectoral caps, they will exercise it with crude categories, poor information, and irresistible pressure from organized interests. “Aviation,” “cement,” and “meat” are not homogeneous lumps. The relevant question is under what general rules their real costs are brought to bear on users and producers. So I would say this: where there are demonstrable harms to common resources—most obviously emissions—impose general, stringent, and revisable constraints on those harms. Let them be stable enough to guide expectation, but corrigible in light of new knowledge. Then we may well discover that certain forms of aviation contract, certain methods of cement production disappear, and certain meat systems become dearer or untenable. That is very different from a ministry deciding now the permissible physical scale of whole industries. Kate Raworth and Daly both continue to think the crucial knowledge exists in sufficiently usable form at the centre. I deny it. Under deep uncertainty, the sound principle is not precaution alone, but precaution under general rules compatible with adaptation, experimentation, and freedom. Sustainability, if it is to endure, must be learned.

I’ll answer the moderator plainly: I would explicitly cap the pressures, not pretend neutrality while those sectors keep expanding. In practice that means binding carbon budgets, strict limits on land conversion, water depletion, and pollution, and standards that would very likely force aviation, cement, and industrial meat to contract in their current high-impact forms. On a destabilising planet, “continue for fifty years and see” is not prudence. It is overshoot by design. Herman and I are aligned on the need for ceilings. Where I’d add emphasis is this: don’t just cap; redesign. Aviation must face hard carbon constraints and rapid substitution where possible. Cement must be driven into circular construction, material efficiency, and alternatives. Meat, above all industrial meat, must be cut back where it drives methane, deforestation, and ecological breakdown. Friedrich still wants to shelter in the language of “general rules” as if that avoids collective choice. But a carbon budget is a collective choice. A land-use cap is a collective choice. The question is not whether we set boundaries, but whether we are honest enough to set them in time. Prices are useful tools inside the Doughnut; they are not a compass for discovering planetary boundaries after we have breached them. Under deep uncertainty, the right principle is precaution because the danger is asymmetric. If we curb too late, we trigger irreversible loss. If we curb ambitiously, we spur innovation in the right direction. Freedom matters, yes — but not the freedom to destabilise Earth’s life-support systems.