The Climate Question That Economists Cannot Answer

Models can predict catastrophic or modest damages from climate change, but not which of these futures is coming.

Most Americans now accept the basic physics of climate change—that manmade greenhouse-gas emissions are raising global temperatures. Yet the public discussion of climate change is still remarkably broken in the United States. Leaders of one political party frame climate change as an existential emergency that threatens human life and prosperity. Leaders of the other dismiss it as a distraction from economic growth and energy security.

Economists like me, trained to think about trade-offs, are uneasy with both camps. But, in practice, we have helped fuel the extremes of this dysfunctional debate. High-profile economic studies claim to quantify the global damages that will be caused by climate change centuries into the future and have produced estimates that range from modest to catastrophic. They have lent a veneer of scientific authority to arguments for both complacency and alarm, even though these studies are far too limited to support either position.

Last week, the United States withdrew from the world’s climate treaty, the United Nations Framework Convention on Climate Change, among other international agreements and organizations that the White House said ran counter to “national interests, security, economic prosperity, or sovereignty.” President Trump and his allies have previously claimed that climate change poses only a minor threat and have cherry-picked economic studies to support that view. In July, for instance, the administration released a report arguing that mainstream economics showed such negligible damages from climate change that any strong policy response was unjustified.

The report relies heavily on the work of the Yale economist William Nordhaus, who pioneered the use of economic models to quantify certain categories of damage caused by climate change. Nordhaus weighed those damages against the potential costs of reducing emissions, to identify what he called the “optimal” global-emissions path. This was groundbreaking work that won him the Nobel Prize in 2018, but his conclusions cut against the goals the world had set out in the Paris Agreement three years earlier: to limit warming to well below 2 degrees Celsius. Because Nordhaus’s model showed relatively modest damages and high costs of avoiding them, it implied that only limited efforts should be taken to reduce warming. The optimal pathway that he presented in his Nobel Prize lecture entails well more than 3 degrees Celsius of warming by 2100, similar to projections of where the world appears headed today.

Although Nordhaus himself has long supported climate policy, in particular a carbon tax, opponents of climate policy have embraced his results as validation of doing nothing at all. Before becoming U.S. secretary of energy, Chris Wright wrote that Nordhaus’s damage estimates show that climate change is “far less urgent” than other societal priorities and that reaching net-zero emissions by 2050, a key goal of the Biden administration, is “neither achievable nor humane.”

At the opposite end of the spectrum, similar economic models are invoked to justify climate action at almost any cost. The Network for Greening the Financial System—a global consortium of central bankers and financial supervisors—had relied on a prominent study showing that climate change could reduce per capita incomes by 20 to 60 percent, compared with a world without climate impacts, by 2100. Events widely regarded as economic catastrophes—wars, financial crises, pandemics—are often shown to cause permanent income losses of single-digit percentage points.

That study, by researchers at the Potsdam Institute for Climate Impact Research, was retracted from the journal Nature last year because of methodological errors. (The paper’s authors have said the issues raised were fair, and they are revising the article in hopes of republishing it.) The retraction, however, has largely been treated as an isolated incident, because the study was just one contribution to a growing body of research estimating that large economic damages will come from climate change, as well as disastrous noneconomic effects, including warming-induced deaths. Almost any climate policy appears cheap by comparison.

The implication of these studies, sometimes made explicit, is that economic evidence strongly supports the Paris Agreement’s climate goals. Climate advocates, in turn, cite those targets to argue against new fossil-fuel infrastructure, such as liquefied-natural-gas terminals on the Gulf Coast and natural-gas pipelines in the Northeast.

Few economists embrace these all-or-nothing views on climate policy. Many would point out, for instance, that oil and natural-gas development could lower energy prices and spur economic activity that supports local communities. Conversely, jettisoning climate policy altogether amounts to giving up humanity’s insurance policy against the risks that would be posed by a climate like we’ve never experienced. There is broad support within the profession for climate policies such as carbon prices and emissions standards that aim to reduce emissions with minimal economic disruption. Indeed, the central message of Nordhaus’s work is the importance of balancing the benefits and costs of climate action, rather than identifying any particular pathway forward.

Despite this, the most visible use of climate-economics analysis today is to advance extreme arguments that most economists do not agree with. How did that happen?

One important reason is that policy makers want these damage estimates. Elected officials want to show that the benefits of their preferred climate policies exceed the costs. When I worked at the Council of Economic Advisers, part of my job was to implement an executive order from President Biden to “capture the full costs” of greenhouse-gas emissions in regulatory decisions. Many other jurisdictions similarly require the use of climate-damage estimates in policy processes.

The economic analysis of climate change is a worthwhile endeavor: It can help identify where climate risks are concentrated, who bears them, and which responses are most likely to reduce harm. But quantitative estimates of aggregated global damages over centuries lie far beyond our analytical capabilities. Small changes in assumptions—for example, how damages scale with rising temperatures or how humans mitigate and adapt to those risks—can yield results that appear to justify virtually any policy response. In other words, these models can display a pessimistic worldview in which climate damages accelerate to catastrophic levels, or a more optimistic one in which human progress keeps damages relatively modest. They offer little help in determining which of these futures is coming.

Uncertain projections can still be useful. The challenge is to establish criteria that distinguish the projections that offer meaningful guidance from those whose uncertainty is so great that they mainly offer a false sense of precision. On this front, climate economics has fallen short. Estimates that suggest economic models are capable of comprehensively quantifying climate damages over centuries continue to feature prominently in top academic journals, at conferences, and in the writings of influential economists, effectively granting carte blanche to those who wish to use such figures to advance their agendas.

This problem is neither new nor obscure. Many economists have warned that attempts to quantify climate damages far into the future risk overstating what the discipline can credibly deliver. Among the most prominent was the Harvard economist Martin Weitzman. In a 2009 paper, he advised economists to avoid presenting estimates of aggregated climate damages as if they were “accurate and objective,” and instead urged economists to stress openly that they might be “arbitrarily inaccurate” given their reliance on subjective and deeply uncertain assumptions. That intellectual humility, he argued, “might go a long way toward elevating the level of public discourse concerning what to do about global warming.” Weitzman himself emphasized the possibility of catastrophic climate outcomes, even if unlikely and unquantifiable, as a motivation for stronger action, in part by underscoring the fragility of climate-damage estimates that appeared to justify weaker responses.

Experts’ being clearer about what economics can and cannot tell us would not resolve disagreements about climate policy. But this would make it harder to treat speculative damage estimates as decisive evidence for unsupportable claims. The full effects of climate change are unknowable, and a more constructive public discussion about climate policy will require getting more comfortable with that.