Climate Economics in Four Easy Pieces
Once upon a time, debates about climate policy were primarily about the science. An inordinate amount of attention was focused on the handful of “climate skeptics” who challenged the scientific understanding of climate change. The influence of the skeptics, however, is rapidly fading; few people were swayed by their arguments, and doubt about the major results of climate science is no longer important in shaping public policy.
As the climate science debate is reaching closure, the climate economics debate is heating up. The controversial issue now is the fear that overly ambitious climate initiatives could hurt the economy. Mainstream economists emphasizing that fear have, in effect, replaced the climate skeptics as the intellectual enablers of inaction.
For example, William Nordhaus, the U.S. economist best known for his work on climate change, pays lip service to scientists’ calls for decisive action. He finds, however, that the “optimal” policy is a very small carbon tax that would reduce greenhouse gas emissions only 25% below “business-as-usual” levels by 2050—that would, in other words, allow emissions to rise well above current levels by mid-century. Richard Tol, a European economist who has written widely on climate change, favors an even smaller carbon tax of just $2 per ton of carbon dioxide. That would amount to all of $0.02 per gallon of gasoline, a microscopic “incentive” for change that consumers would never notice.
There are other voices in the climate economics debate; in particular, the British government’s Stern Review offers a different perspective. Economist Nicholas Stern’s analysis is much less wrong than the traditional Nordhaus-Tol approach, but even Stern has not challenged the conventional view enough.
What will it take to build a better economics of climate change, one that is consistent with the urgency expressed by the latest climate science? The issues that matter are big, non-technical principles, capable of being expressed in bumper-sticker format. Here are the four bumper stickers for a better climate economics:
Our grandchildren’s lives are important
We need to buy insurance for the planet
Climate damages are too valuable to have prices
Some costs are better than others
1. Our Grandchildren's Lives are Important
The most widely debated challenge of climate economics is the valuation of the very long run. For ordinary loans and investments, both the costs today and the resulting future benefits typically occur within a single lifetime. In such cases, it makes sense to think in terms of the same person experiencing and comparing the costs and the benefits.
In the case of climate change, the time spans involved are well beyond those encountered in most areas of economics. The most important consequences of today’s choices will be felt by generations to come, long after all of us making those choices have passed away. As a result, the costs of reducing emissions today and the benefits in the far future will not be experienced by the same people. The economics of climate change is centrally concerned with our relationship to our descendants whom we will never meet. As a bridge to that unknowable future, consider our grandchildren—the last generation most of us will ever know.
Suppose that you want your grandchildren to receive $100 (in today’s dollars, corrected for inflation), 60 years from now. How much would you have to put in a bank account today, to ensure that the $100 will be there 60 years from now? The answer is $55 at 1% interest, or just over $5 at 5%.
In parallel fashion, economists routinely deal with future costs and benefits by “discounting” them, or converting them to “present values”—a process that is simply compound interest in reverse. In the standard jargon, the present value of $100, to be received 60 years from now, is $55 at a 1% discount rate, or about $5 at a 5% discount rate. As this example shows, a higher discount rate implies a smaller present value.
The central problem of climate economics, in a cost-benefit framework, is deciding how much to spend today on preventing future harms. What should we spend to prevent $100 of climate damages 60 years from now? The standard answer is, no more than the present value of that future loss: $55 at a discount rate of 1%, or $5 at 5%. The higher the discount rate, the less it is “worth” spending today on protecting our grandchildren.
The effect of a change in the discount rate becomes much more pronounced as the time period lengthens. Damages of $1 million occurring 200 years from now have a present value of only about $60 at a 5% discount rate, versus more than $130,000 at a 1% discount rate. The choice of the discount rate is all-important to our stance toward the far future: should we spend as much as $130,000, or as little as $60, to avoid one million dollars of climate damages in the early twenty-third century?
For financial transactions within a single lifetime, it makes sense to use market interest rates as the discount rate. Climate change, however, involves public policy decisions with impacts spanning centuries; there is no market in which public resources are traded from one century to the next. The choice of an intergenerational discount rate is a matter of ethics and policy, not a market-determined result.
Economists commonly identify two separate aspects of long-term discounting, each contributing to the discount rate.
One component of the discount rate is based on the assumption of an upward trend in income and wealth. If future generations will be richer than we are, they will need less help from us, and they will get less benefit from an additional dollar of income than we do. So we can discount benefits that will flow to our wealthier descendants, at a rate based on the expected growth of per capita incomes. Among economists, the income-related motive for discounting may be the least controversial part of the picture.
Setting aside changes in per capita income from one generation to the next, there may still be a reason to discount a sum many years in the future. This component of the discount rate, known as “pure time preference,” is the subject of longstanding ethical, philosophical, and economic debate. On the one hand, there are reasons to think that pure time preference is greater than zero: both psychological experiments and common sense suggest that people are impatient, and prefer money now to money later. On the other hand, a pure time preference of zero expresses the equal worth of people of all generations, and the equal importance of reducing climate impacts and other burdens on them (assuming that all generations have equal incomes).
The Stern Review provides an excellent discussion of the debate, explaining Stern’s assumption of pure time preference close to zero and an overall discount rate of 1.4%. This discount rate alone is sufficient to explain Stern’s support for a substantial program of climate protection: at the higher discount rates used in more traditional analyses, the Stern program would look “inefficient,” since the costs would outweigh the present value of the benefits.
2. We Need to Buy Insurance for the Planet
Does climate science predict that things are certain to get worse? Or does it tell us that we are uncertain about what will happen next? Unfortunately, the answer seems to be yes to both questions. For example, the most likely level of sea level rise in this century, according to the latest Intergovernmental Panel on Climate Change reports, is no more than one meter or so—a real threat to low-lying coastal areas and islands that will face increasing storm damages, but survivable, with some adaptation efforts, for most of the world. On the other hand, there is a worst-case risk of an abrupt loss of the Greenland ice sheet, or perhaps of a large portion of the West Antarctic ice sheet. Either one could cause an eventual seven-meter rise in sea level—a catastrophic impact on coastal communities, economic activity, and infrastructure everywhere, and well beyond the range of plausible adaptation efforts in most places.
The evaluation of climate damages thus depends on whether we focus on the most likely outcomes or the credible worst-case risks; the latter, of course, are much larger.
Cost-benefit analysis conventionally rests on average or expected outcomes. But this is not the only way that people make decisions. When faced with uncertain, potentially large risks, people do not normally act on the basis of average outcomes; instead, they typically focus on protection against worst-case scenarios. When you go to the airport, do you leave just enough time for the average traffic delay (so that you would catch your plane, on average, half of the time)? Or do you allow time for some estimate of worst-case traffic jams? Once you get there, of course, you will experience additional delays due to security, which is all about worst cases: your average fellow passenger is not a threat to anyone’s safety.
The very existence of the insurance industry is evidence of the desire to avoid or control worst-case scenarios. It is impossible for an insurance company to pay out in claims as much as its customers pay in premiums; if it did, there would be no money left to pay the costs of running the company, or the profits received by its owners. People who buy insurance are therefore guaranteed to get back less than they, on average, have paid; they (we) are paying for the security that insurance provides in case the worst should happen. This way of thinking does not apply to every decision: in casino games, people make bets based on averages and probabilities, and no one has any insurance against losing the next round. But life is not a casino, and public policy should not be a gamble.
Should climate policy be based on the most likely outcomes, or on the worst-case risks? Should we be investing in climate protection as if we expect sea level rise of one meter, or as if we are buying insurance to be sure of preventing a seven-meters rise?
In fact, the worst-case climate risks are even more unknown than the individual risks of fire and death that motivate insurance purchases. You do not know whether or not you will have a fire next year or die before the year is over, but you have very good information about the likelihood of these tragic events. So does the insurance industry, which is why they are willing to insure you. In contrast, there is no body of statistical information about the probability of Greenland-sized ice sheets collapsing at various temperatures; it’s not an experiment that anyone can perform over and over again.
A recent analysis by Martin Weitzman argues that the probabilities of the worst outcomes are inescapably unknowable—and this deep uncertainty is more important than anything we do know in motivating concern about climate change. There is a technical sense in which the expected value of future climate damages can be infinite because we know so little about the probability of the worst, most damaging possibilities. The practical implication of infinite expected damages is that the most likely outcome is irrelevant; what matters is buying insurance for the planet, i.e., doing our best to understand and prevent the worst-case risks.
3. Climate Damages Are Too Valuable to Have Prices
To decide whether climate protection is worthwhile, in cost-benefit terms, we would need to know the monetary value of everything important that is being protected. Even if we could price everything affected by climate change, the prices would conceal a critical form of international inequity. The emissions that cause climate change have come predominantly from rich countries, while the damages will be felt first and worst in some of the world’s poorest, tropical countries (although no one will be immune from harm for long). There are, however, no meaningful prices for many of the benefits of health and environmental protection. What is the dollar value of a human life saved? How much is it worth to save an endangered species from extinction, or to preserve a unique location or ecosystem? Economists have made up price tags for such priceless values, but the results do not always pass the laugh test.
Is a human life worth $6.1 million, as estimated by the Clinton administration, based on small differences in the wages paid for more and less risky jobs? Or is it worth $3.7 million, as the (second) Bush administration concluded on the basis of questionnaires about people’s willingness to pay for reducing small, hypothetical risks? Are lives of people in rich countries worth much more than those in poor countries, as some economists infamously argued in the IPCC’s 1995 report? Can the value of an endangered species be determined by survey research on how much people would pay to protect it? If, as one study found, the U.S. population as a whole would pay $18 billion to protect the existence of humpback whales, would it be acceptable for someone to pay $36 billion for the right to hunt and kill the entire species?
The only sensible response to such nonsensical questions is that there are many crucially important values that do not have meaningful prices. This is not a new idea: as the 18th-century philosopher Immanuel Kant put it, some things have a price, or relative worth, while other things have a dignity, or inner worth. No price tag does justice to the dignity of human life or the natural world.
Since some of the most important benefits of climate protection are priceless, any monetary value for total benefits will necessarily be incomplete. The corollary is that preventive action may be justified even in the absence of a complete monetary measure of the benefits of doing so.
4. Some Costs Are Better than Others
The language of cost-benefit analysis embodies a clear normative slant: benefits are good, costs are bad. The goal is always to have larger benefits and smaller costs. In some respects, measurement and monetary valuation are easier for costs than for benefits: implementing pollution control measures typically involves changes in such areas as manufacturing, construction, and fuel use, all of which have well-defined prices. Yet conventional economic theory distorts the interpretation of costs in ways that exaggerate the burdens of environmental protection and hide the positive features of some of the “costs.”
For instance, empirical studies of energy use and carbon emissions repeatedly find significant opportunities for emissions reduction at zero or negative net cost—the so-called “no regrets” options.
According to a long-standing tradition in economic theory, however, cost-free energy savings are impossible. The textbook theory of competitive markets assumes that every resource is productively employed in its most valuable use—in other words, that every no-regrets option must already have been taken. As the saying goes, there are no free lunches; there cannot be any $20 bills on the sidewalk because someone would have picked them up already. Any new emissions reduction measures, then, must have positive costs. This leads to greater estimates of climate policy costs than the bottom-up studies that reveal extensive opportunities for costless savings.
In the medium term, we will need to move beyond the no-regrets options; how much will it cost to finish the job of climate protection? Again, there are rival interpretations of the costs based on rival assumptions about the economy. The same economic theory that proclaimed the absence of $20 bills on the sidewalk is responsible for the idea that all costs are bad. Since the free market lets everyone spend their money in whatever way they choose, any new cost must represent a loss: it leaves people with less to spend on whatever purchases they had previously selected to maximize their satisfaction in life. Climate damages are one source of loss, and spending on climate protection is another; both reduce the resources available for the desirable things in life.
But are the two kinds of costs really comparable? Is it really a matter of indifference whether we spend $1 billion on bigger and better levees or lose $1 billion to storm damages? In the real-world economy, money spent on building levees creates jobs and incomes. The construction workers buy groceries, clothing, and so on, indirectly creating other jobs. With more people working, tax revenues increase while unemployment compensation payments decrease.
None of this happens if the levees are not built and the storm damages are allowed to occur. The costs of prevention are good costs, with numerous indirect benefits; the costs of climate damages are bad costs, representing pure physical destruction. One worthwhile goal is to keep total costs as low as possible; another is to have as much as possible of good costs rather than bad costs. Think of it as the cholesterol theory of climate costs.
In the long run, the deep reductions in carbon emissions needed for climate stabilization will require new technologies that have not yet been invented, or at best exist only in small, expensive prototypes. How much will it cost to invent, develop, and implement the low-carbon technologies of the future?
Lacking a rigorous theory of innovation, economists modeling climate change have often assumed that new technologies simply appear, making the economy inexorably more efficient over time. A more realistic view observes that the costs of producing a new product typically decline as industry gains more experience with it, in a pattern called “learning by doing” or the “learning curve” effect. Public investment is often necessary to support the innovation process in its early, expensive stages. Wind power is now relatively cheap and competitive, in suitable locations; this is a direct result of decades of public investment in the United States and Europe, starting when wind turbines were still quite expensive. The costs of climate policy, in the long run, will include doing the same for other promising new technologies, investing public resources in jump-starting a set of slightly different industries than we might have chosen in the absence of climate change. If this is a cost, many communities would be better off with more of it.
A widely publicized, conventional economic analysis recommends inaction on climate change, claiming that the costs currently outweigh the benefits for anything more than the smallest steps toward reducing carbon emissions. Put our “four easy pieces” together, and we have the outline of an economics that complements the science of climate change and endorses active, large-scale climate protection.
How realistic is it to expect that the world will shake off its inertia and act boldly and rapidly enough to make a difference? This may be the last generation that will have a real chance at protecting the earth’s climate. Projections from the latest IPCC reports, the Stern Review, and other sources suggest that it is still possible to save the planet—if we start at once.