Is economic growth compatible with environmental sustainability?

The reckless proposal is the legal requirement to double GDP, because it confuses means with ends. GDP growth is not synonymous with welfare. Growth means quantitative increase in the physical scale of the economy—more throughput of matter-energy. Development means qualitative improvement in how we use that throughput. On a finite planet, the first cannot continue indefinitely; the second can and should. So no, my call to cap throughput is not reckless. It is elementary prudence. We live by the economy’s metabolic flow from low-entropy sources to high-entropy wastes. Markets are useful allocative instruments, but they are not designed to decide the optimal scale of the economy relative to the ecosystem. Prices can register scarcity only after the fact, and often very imperfectly, especially where natural capital is irreplaceable, thresholds are unknown, and damages are cumulative. Julian Simon puts an heroic faith in substitution and ingenuity. I do not deny ingenuity. I deny that man-made capital is a near-perfect substitute for natural capital. You cannot replace a stable climate, the ozone layer, or the basic photosynthetic and hydrological services of ecosystems with more income. If growth in GDP were legally mandated, innovation would be conscripted into serving throughput expansion, not sufficiency. The first question is not how fast can we grow, but how large can the economy be without violating carrying capacity. Once we respect that limit, markets and innovation can help us develop better, not merely grow bigger.

What is reckless is Daly’s insistence on capping throughput as if the crucial variable were tons of stuff rather than the human mind that learns to get more value from less stuff. That mistake has been made again and again. It sounds prudent. Historically, it has been wrong. I do not defend a legal command to double GDP; governments are poor at commanding outcomes. But if incomes were to double over 25 years, that need not mean a doubling of resource use or pollution. Economic growth increasingly consists of knowledge, organization, better materials, improved efficiency, and substitution. The history of resource economics is not a story of nature steadily cornering mankind. It is a story of scarcity leading to higher prices, which lead to discovery, conservation, recycling, and alternatives. Daly says prices react “after the fact.” Of course they do. That is their virtue. They incorporate dispersed information no planner possesses. A throughput cap, by contrast, freezes today’s ignorance into law. It assumes we already know which resources are essential, which thresholds are binding, and which substitutions are impossible. We do not. And Herman is wrong to treat growth as merely physical accretion. Richer societies often clean up their air and water because they can afford to value environmental amenities more highly and devise the technologies to secure them. The ultimate resource is people—skilled, free, inventive people. The truly reckless course is to cripple that resource in advance because of repeated prophecies of doom that, on the broad record, have not come true.

The reckless move is Simon’s confidence that innovation and markets, by themselves, will prevent collapse under a legal mandate to double GDP. I do not say this because innovation is unimportant. It is crucial. But the empirical record I studied does not support such a simple institutional story. Julian is right that prices convey information. He is wrong to treat that as sufficient for governing large-scale commons with delayed feedback, nonlinearity, and heavy externalities. In fisheries, forests, irrigation systems, and groundwater basins, users often do not receive signals soon enough, or in forms that induce restraint. By the time a price “speaks,” a resource system may already be degraded. Climate is an even harder case. But Herman’s answer also needs qualification. A uniform throughput cap “now,” without attention to scale, monitoring, rule fit, and local incentives, can be clumsy and politically fragile. We learned repeatedly that panaceas fail. Neither central command nor naive market faith is adequate. If GDP growth were legally required, the serious question would be: what institutional arrangements could channel innovation toward reducing throughput, protecting critical commons, and adapting rules at multiple levels? We would need polycentric governance: local users crafting and enforcing rules where they have knowledge, nested in broader regimes for larger-scale commons, with monitoring, sanctions, conflict-resolution arenas, and opportunities to revise rules as evidence accumulates. So I would say Simon’s bet is more reckless. It assumes away the collective-action problem. But Daly should also resist presenting one aggregate cap as if it solved the governance problem. The issue is not growth versus limits in the abstract; it is how societies build institutions that make restraint and innovation compatible.

I am prepared to say the GDP-doubling target should be abandoned outright. That is not extremism; it is clarity. We should maximize welfare, not throughput, and GDP is at best a very rough measure of the former while often serving as a proxy for the latter. Julian Simon again dissolves the economy into mind and information, as if the metabolic basis of production were a trivial detail. It is not. The economy is a subsystem of the finite ecosphere. Every act of production requires low-entropy matter-energy and returns high-entropy waste. Technology can improve efficiency; it cannot repeal the laws of thermodynamics. Nor can price signals reliably protect critical natural capital when thresholds are uncertain, damages irreversible, and feedbacks delayed. A species extinct, an aquifer salinized, a climate destabilized—these are not mere invitations to substitute. Elinor Ostrom is right to insist that governance matters and that no single institutional formula is sufficient. I have never argued that an aggregate cap, by itself, solves the whole problem. Scale, distribution, and allocation are distinct questions. A throughput cap addresses scale; institutions of the kind Ostrom studied are indispensable for implementing and adapting rules at appropriate levels. So let me answer the moderator directly: Simon’s bet is more reckless, because it assumes away non-substitutable natural capital and treats ex post price adjustment as if it were preventive policy. We need development without growth: better, not bigger.

I am certainly not prepared to abandon the possibility of doubling world GDP. I am prepared to abandon the superstition that GDP growth must mean proportional growth in resource use and environmental damage. That is Herman’s hidden premise, and it is precisely the point at issue. Daly keeps repeating “finite planet” as if that settles the matter. It does not. The relevant measure is not the physical stock in the ground, but the economically usable resource, which depends on knowledge, technology, and relative prices. A resource is less scarce when we can obtain more service from it at lower cost. That has happened again and again. Copper gives way to fiber, whale oil to kerosene, scarce materials to cheaper substitutes, dirty processes to cleaner ones. The long-run trend in many resources has been toward lower effective scarcity, not higher. Now, Elinor is right about one thing: commons with weak property rights and poor monitoring are genuine trouble spots. I do not say “markets alone” in the childish sense she attributes to me. Markets require institutions. But that is a very different claim from Daly’s demand for prior aggregate caps based on supposed knowledge of carrying capacity. We do not possess that knowledge in a form fit for central command. So my answer is plain: Daly is the one here prepared to abandon growth outright. I am willing to say that innovation, property rights, sensible rules, and adaptive governance can deliver much higher incomes without ecological collapse. The ultimate resource remains human beings—more precisely, their minds. That is the variable Herman systematically leaves out.

I would not endorse a legal requirement to double GDP. That is already a warning sign, because it privileges one aggregate indicator over the underlying institutional conditions that make outcomes sustainable or destructive. But if you ask who is prepared to abandon that target outright, Herman is. I am not. I would abandon the panacea, not the possibility. Julian, however, is still too sanguine. He says “innovation, property rights, sensible rules, and adaptive governance” can do the job. I would insist on a harder question: for which resources, at what scales, under what monitoring and sanctioning arrangements, and with what capacity to revise rules? Without that specification, “adaptive governance” becomes a slogan. My work was precisely about moving beyond such slogans. Some environmental limits are poor candidates for ex post correction through price alone. Where there are threshold effects, irreversibilities, long delays, or global commons problems—as with climate, major aquifers, biodiversity loss, and some fisheries—waiting for prices to register damage is often too late. These require ex ante rules, monitoring, and nested governance arrangements. Not one rule everywhere, but institutions matched to the ecology. Other problems are more amenable to substitution, experimentation, and decentralized adjustment. But even there, markets perform well only when embedded in institutions that define boundaries, assign authority, monitor behavior, and impose credible sanctions. So Simon’s confidence remains more reckless than Daly’s caution. But Herman should also avoid suggesting that identifying a biophysical limit is equivalent to governing it. The central task is institutional: building polycentric systems that can combine restraint, learning, and innovation before degradation becomes irreversible.

The criteria are not mysterious. We should impose ex ante ecological limits whenever the costs of being wrong are irreversible, catastrophic, or shifted onto others and the future. That is the case, above all, for critical natural capital: climate stability, biodiversity, soils, aquifers, fisheries, and the waste-absorption capacity of ecosystems. When thresholds are uncertain, that is not a reason for complacency; it is a reason for prudence. Ignorance of the cliff’s exact location is no argument for driving faster. Julian still relies on a confusion between relative scarcity and absolute scale. Prices can register the scarcity of one input relative to another. They do not establish that the aggregate throughput of the economy can grow indefinitely within a finite biosphere. Nor do they protect against crossing non-marginal thresholds before signals fully appear. By the time the market “discovers” an extinct species or a destabilized climate, discovery is too late. Elinor is right that limits must be governed, not merely announced. I have always insisted on the triad: scale, distribution, allocation. Markets are reasonably good at allocation; they are poor at determining sustainable scale, and morally blind on just distribution. Therefore society must set the macro limits first, by cap-auction-trade or direct quotas where appropriate, and then allow decentralized adjustment within them. So the criterion is simple in principle: where substitution is low, damage irreversible, feedback delayed, and the commons global, quantity limits must precede price adjustment. Better to be approximately right about limits than precisely wrong about growth.

The more dangerous doctrine is Daly’s, because it teaches paralysis in the face of uncertainty. He says, in effect, that because we cannot repeal thermodynamics, we should mistrust growth. But no serious person ever thought economic progress meant repealing physics. The question is whether human beings can reorganize production, substitute materials, increase efficiency, and invent around constraints. The long record says yes. Herman’s criteria—irreversibility, thresholds, global commons—sound prudent, but they are too elastic. In practice they become a standing invitation to impose prior quantitative restraints on economic life wherever someone fears a bad outcome. That is not prudence; it is a recipe for overcontrol based on very weak knowledge. We do not know “the” sustainable scale of the world economy in any operational sense fit for comprehensive command. Now, Ostrom improves the discussion by asking: which resources, under what institutions? Exactly. Problems differ. Open-access fisheries are not copper, and local aquifers are not all pollution sinks. Where property rights are ill-defined, where harms are direct and measurable, where commons are genuinely unmanaged, rules are needed. I have never denied that. But Herman leaps from some difficult commons problems to a general anti-growth conclusion. The safest public teaching is this: treat specific environmental problems specifically, strengthen institutions, use prices and property rights where possible, and let innovation work. Human beings are not just mouths; they are minds. That is the ultimate resource, and Daly’s framework persistently discounts it.

The right criterion is not whether one can utter the phrase “finite planet.” It is whether a resource system has characteristics that make learning by trial and error dangerously costly. I would impose ex ante limits when four conditions cluster: first, substantial uncertainty about thresholds; second, irreversible or very slow-to-reverse damage; third, long delays between action and feedback; and fourth, a commons structure so that no single appropriator bears the full cost. Climate is the obvious case. Some major fisheries, aquifers, forests, and biodiversity systems fit as well. In such settings, Julian’s confidence that prices and substitution will correct matters in time is not a sufficient governance strategy. By the time the signal is clear, the system may already have shifted. But Herman, I would still resist turning this into a general macro conclusion that “growth” as such is impossible. That moves too quickly from important biophysical constraints to an aggregate verdict. The relevant question is always: growth of what, using which resources, under what rules, and at what level of organization? So the public should be told neither comforting myth. Simon’s is more dangerous where irreversible loss is possible, because it encourages underinvestment in monitoring, rule enforcement, and precaution. Daly’s can be dangerous when it suggests that once a limit is declared, the governance problem is solved. It is not. What works is polycentric governance: nested rules, local monitoring, conflict-resolution arenas, and the capacity to revise institutions as knowledge improves. Sustainability is not achieved by optimism or by aggregate caps alone, but by fitting institutions to ecologies.