Climate change has become the premier environmental issue facing the globe. Carbon dioxide (CO2) emissions continue to grow and accumulate in the atmosphere. The average global temperature in 2014 was the highest recorded over the last century and a half. Most scientists say that climate change is a “very serious problem.” Yet virtually no progress has been made in convincing the general public of its serious nature, nor have significant steps been taken to curb emissions and slow warming. Why has progress been so halting?
The risks of a warming world and potential policies to deal with these risks are the subject of a short book by Gernot Wagner and Martin Weitzman. Wagner is a public policy specialist and lead senior economist at the Environmental Defense Fund and has written widely on energy and climate change. Weitzman is one of the leading economic theorists of our day, having made fundamental contributions to environmental accounting, the relative merits of price and quantity regulation, measurement of species extinction, and in an earlier era the economics of central planning and the Soviet Union.
Their book on climate change is a witty, far-ranging, and literate set of observations, but—unlike many books on climate change—it is always informed by a deep understanding of the complexities of economics and particularly the difficulties of reaching international environmental agreements. While the entire book is worth careful study, its singular contributions are in three areas: the discussions of how nations may “free-ride” on the decisions of others, the ultimate curse on international climate policies; the uncertainties surrounding both climate change and its consequences; and the particular perils of geoengineering to reverse carbon-induced climate change. None of these subjects is well covered in most books on climate change, so I will concentrate on them in this review.
Why has progress in climate change policy been so slow? If you read five books, you will find six theories. Perhaps this is because the public is poorly informed. Or because the science is so hard. Or because industry is putting up such a vicious fight, and policies are being blocked by the oil and coal lobbies. Or because climate deniers have captured the Republican Party. Perhaps because a solution is so expensive. Or because costs must be incurred in the present while benefits accrue far in the future.
Doubtless, each of these plays a part. But the fundamental reason for the lack of progress is the strong incentives for “free-riding” in current international climate agreements. Free-riding occurs when a party receives the benefits of a public good without contributing to the costs. People free-ride when they jump the turnstile on the subway. Nations free-ride in military treaties such as NATO when they enjoy the benefits of the strong US military to protect them while doing little to pay for the common defense.
While free-riding is pervasive in many areas, it is particularly pernicious for global public goods, or global externalities. There are activities for which both the benefits and negative effects are widely and indivisibly dispersed around the globe from the country or countries that were originally responsible for them. In addition to climate change, similar global externalities are ozone depletion on the negative side and discoveries like antibiotics on the positive side.
The benefits from investments to reduce emission of CO2 and other greenhouse gases are a global public good. They require costly investments by individual countries. However, the benefits from the lower emissions are spread widely around the world, and the country undertaking the investments will receive only a tiny fraction of the benefits. Wagner and Weitzman put the point as follows:
Why act, if your actions cost you more than they benefit you personally? Total benefits of your actions may outweigh costs. Yet the benefits get spread across seven billion others, while you incur the full costs. The same logic holds for everybody else. Too few are going to do what is in the common interest. Everyone else free-rides.
Global public goods differ from national or local public goods because there are no mechanisms—either market or governmental—to deal with them effectively. Under international law, countries must consent to joining international agreements, and all agreements are therefore essentially voluntary. If countries act rationally in their own self-interest, they will have a strong incentive to free-ride on the emissions reductions of others. One result of free-riding is the failure of the only significant international climate treaty, the Kyoto Protocol, along with the difficulties of arriving at effective arguments and measures to follow up the treaty.
To illustrate the importance of free-riding, it is useful to compare policies on climate change with those on two other major health and environmental issues, smoking and air pollution. Public policies to curb smoking and air pollution were only slowly enacted in the face of public doubt and industry resistance. Yet major gains were made in both cases.
Take smoking, which kills almost half a million people annually in the US. More than half of men regularly smoked cigarettes in 1965, and most of them thought smoking did not cause cancer. However, as a result of a half-century of education, persuasion, warnings, and personal health concerns, attitudes changed. According to recent studies, close to 90 percent of the population, including smokers, think smoking causes cancer. The fraction of the US adult population that currently smokes has declined to 18 percent. Here was a case where most people learned the facts and worried enough about their own personal health to take steps to stop or never to start smoking. We might call a smoke-free environment a “personal public good.” While not a complete success, the decline in smoking is surely one of the major public health victories of the modern era.
A second example, also highly successful, has been the reduction in air pollution that produces small particles that are harmful to health. Major sources of these harmful particles are chemical reactions of sulfur dioxide, soot, nitrogen oxides, and other compounds. Estimates suggest that perhaps 50,000 early deaths annually in the US are caused by this form of air pollution. National and regional regulations in place largely since 1970 have successfully reduced the worst of the harmful pollutants. For example, from 1970 to 2014, sulfur dioxide emissions declined by 84 percent. It is interesting to note that most of this decline came after the introduction in 1990 of a cap-and-trade program that put a market price on sulfur emissions—this being the first major test of market pricing of emissions.
Regulating sulfur dioxide and other sources of harmful small particles is a “national public good.” It is an example of nations paying the cost of reducing emissions, by accepting higher prices for electricity and other goods produced with coal and other substances that cause harmful emissions. Both costs and benefits are internal to the nation, with small spillovers outside the country. Moreover, the benefits are relatively short-term and occur during the lifetime of those bearing the costs. The examples of reductions in smoking and in sulfur emissions show that well-functioning political systems can overcome public apathy and industrial resistance and take cost-effective steps. The polities of the Unites States, Western Europe, and Japan took those steps.
Free-riding can be prevented for national public goods like air pollution because well-managed national governments can make the cost-benefit calculations that affect their populations and find a reasonable balance of costs and benefits of investments in environmental protection.
For climate change, the national calculus is entirely different. Here, the temptation of free-riding will push policies toward minimal investments in slowing emissions. A nation will look at the benefits and costs of abatement and recognize that most of the benefits will occur outside its borders. If it pursues its national interest, it will invest little or nothing.
An illustrative calculation will show why. Suppose that, as the US government estimates, the total global cost of CO2 emissions (called the “global social cost of carbon” or SCC) is $40 per ton. This is calculated from models that trace the impact on all countries of higher carbon emissions on the climate, in the rise of sea levels, in agricultural production, health, storms, and many other factors. In other words, the $40 is the sum of the different national social costs per ton of carbon. Perhaps the national numbers might be $4 per ton for the US, $7 per ton for China, and $1 per ton for Japan. The $40 estimate is uncertain, but the exact number does not matter for this discussion.
A policy to optimize the global benefits of emissions reductions would require a universal carbon price of $40 per ton. Under this policy, countries might set a carbon tax of $40 per ton on all carbon-emitting activities. This would lead to significant reductions in emissions, perhaps 30 percent on average relative to the emissions that would take place if there were no policy.
Contrast the globally optimal policy I have just described with a free-riding policy. Suppose Japan decides to invest in abatement of CO2 emissions to optimize its own national interest. It would reduce emissions only to the point where the cost would be justified by its national benefits; on this basis, the cost would be no more than its social cost of $1 per ton reduced. It might have a carbon tax of $1 per ton. This would produce far smaller investments than would be justified if Japan had global interests in mind, where it would invest up to a cost of $40 per ton reduced.
If the same logic based on national interest is followed by all countries, then the average level of abatement would be only a tiny fraction of what would be involved in the global policy. Economic models suggest that the average carbon tax based on national interest would be closer to $4 than $40. In other words, when countries look only to their national interest in setting their climate policies, the level of abatement will be close to zero. And this is approximately where we are today.
Once we understand the logic of free-riding, we can understand the reason why existing agreements have failed to slow warming. The first binding treaty was the Kyoto Protocol. It was negotiated in 1997, had minimal impact, and expired in 2012 without a successor. Follow-up agreements are being discussed every year but they get nowhere. The problem is that no country has any incentive to go beyond its low-abatement, free-riding policy. The US and China and Brazil make solemn promises to meet ambitious targets, but the logic of national self-interest up to now has prevented even the most modest targets from being met.
Looking at the different cases of smoking, national air pollution, and climate change, we see that the difficulties posed by international free-riding are a major impediment to reaching effective agreements for deep emissions reductions.
Scientists are increasingly confident that the basic results of climate modeling are accurate. Climate models calculate that past emissions have contributed to warming of almost one degree centigrade over the last century, with rapid continued warming projected over the present century and beyond. On its 2001 report, the Intergovernmental Panel on Climate Change (IPCC) reported that human activity was “likely” to be the source of this warming. The IPCC upgraded this evaluation to “very likely” in its 2007 report and to “extremely likely” in its 2014 report.
The consequences of these temperature changes have been analyzed and widely reported. They involve disruptions to agriculture, water systems, storms, ecosystems, ocean chemistry, and a wide variety of other effects. But even these wide-ranging impacts may understate the potential dangers. Wagner and Weitzman emphasize that the standard analyses of climate change ignore our deep uncertainties about the extent and impacts of changes. They argue that recent developments in earth sciences and other studies suggest that the potential impacts of extreme events—what are known as “tail events” or sometimes more vividly as “black swans”—may dwarf the standard impacts described in the IPCC reports.
Tail events are phenomena that are so surprising, so outside everyday observations, that we are unprepared to deal with them. They are called tail events because they come from the far tail, or most unlikely part, of a probability distribution. If you look at the bell curve shown below, the area (or probability) under the little tails on the far left and right make up a tiny fraction of the total area under the curve—those outcomes are extremely unlikely.
The height of humans has a distribution that closely resembles a bell curve. The average man is seventy inches tall, with a range of variation from this height measured by the standard deviation of four inches. If height follows the bell curve, only 2 percent of men would be over six and one half feet. You would never in a trillion trillion men see one twice as large as the average. However, with “fat-tailed” distributions, outliers that are twice as large as the average, or even more, are possible if infrequent. The impacts of climate change could follow such a fat-tailed distribution, instead of a standard bell curve.1
Wagner and Weitzman emphasize that the tail events associated with climate change would be “profound earth-as-we-know-it-altering changes.” They suggest in their third chapter that a decline of 30 percent in global output is a possible outcome. This decline, particularly if it occurred over a short time span, would be unlike anything seen in modern times. By contrast, global GDP did not fall even during the financial crisis of 2007–2009.
Wagner and Weitzman do not point to a specific scenario that would produce a catastrophic result because we cannot know what it is or how large an event will take place. They emphasize that, for climate change, dealing with possible tail events is the central task of policy. They argue that current policies are leading to a substantial chance (perhaps one in ten) that global temperatures will eventually rise by at least six degrees centigrade. This will, in their words, be “the end of the human adventure on this planet as we now know it.” Policies should above all aim to cut off the possibility, the tail, of catastrophic temperature increases.
There are two ways to slow climate change, and thereby to reduce the likelihood of catastrophic damages. One is the hard slog of reducing emissions. The other is to use geoengineering that attempts to offset the CO2-induced warming.
Wagner and Weitzman provide an illuminating discussion of the dilemmas of geoengineering. Geoengineering here means management of solar radiation—techniques that reflect sunlight back into space and prevent it from warming the earth’s climate. You can think of the process as making the earth “whiter” or more reflective, so that less sunlight is absorbed by the surface of the earth. This cooling effect will offset the warming that comes from the accumulation of CO2 in the atmosphere.
The whitening process is similar to what occurs after large volcanic eruptions. After Mount Pinatubo in the Philippines blasted 20 million tons of sulfur dioxide into the stratosphere in 1991, global temperatures fell by about half a degree centigrade because the particles reflected sunlight away from the earth. Geoengineering can be understood as creating artificial volcanic eruptions, where several artificial Pinatubo-sized eruptions may be needed every year to offset the warming effects of CO2 accumulation.
Many proposals have emerged to whiten the earth. These, along with many of the ethical, political, national security, and environmental dimensions of such projects, are discussed in a recent report by the National Academy of Sciences.2 The standard approach is to deliver sulfur-bearing compounds, presumably specially engineered ones that would act as small mirrors, into the lower stratosphere. A number of techniques have been proposed to do this, such as using naval guns, aircraft, or rockets. Recent studies indicate that such geoengineering can lower global temperatures at very low cost relative to other approaches, such as reducing carbon emissions.
Even though the costs are low and the average impacts on temperature are clear, the dangers are frightening, as is emphasized by Wagner and Weitzman as well as by the National Academy report. The NAS committee concludes that climate “modification strategies are limited primarily by considerations of risk, not by direct costs.” Among the risks are the facts that geoengineering does nothing to reduce the ocean acidification caused by increased CO2; that countries would need to keep a program going virtually forever; that there is a mismatch between the cooling and heating effects; and that there is a high likelihood for redistribution of precipitation in the different regions of the world. An effective program would require virtual unanimity in international governance so as to reduce political frictions among countries.
This can be put intuitively as follows. At present, there have been no large-scale geoengineering experiments on our globe (aside from natural eruptions), so the estimates of its impacts and side effects are based on computer modeling. The major concern is that geoengineering is not really a perfect offset to the greenhouse effect. The little particles or mirrors would reduce incoming solar radiation while the greenhouse effect decreases outgoing radiation. The two effects might lead to zero net warming, but they are very different physically. A useful analogy is turning on your home air conditioner during a heat wave. Your house will be, on average, the same temperature as on a normal day, but some of the rooms might be colder and others warmer, and you definitely will be spending a small fortune on electricity.
Even more frightening is the prospect that climatic engineering could turn into climate warfare—much as cyberspace has at times degenerated into cyberwarfare. This prospect was emphasized by John von Neumann in the 1950s:
The most constructive schemes of climate control would have to be based on insights and techniques that would also lend themselves to forms of climatic warfare as yet unimagined…. Useful and harmful techniques lie everywhere so close together that it is never possible to separate the lions from the lambs.3
So here is the dilemma raised by reading Wagner and Weitzman. They are convincing that the tail events are the ones we should worry about most—that these are the earth-changing risks of continued emissions and warming. We have made no progress in slowing emissions, and concentrations of CO2 continue to build up. Scientists have found a way, in theory, to slow or reverse the progress of the disease, but with some potentially devastating side effects. Should we do some experiments with geoengineering to measure its impacts? Or should we reject that approach as too risky and concentrate entirely on emissions reductions?
If the fat tails of climate change are perilous, and geoengineering is itself a dangerous solution, what remains? Here, Wagner and Weitzman largely follow the standard economists’ prescription: “Stick it to carbon.” We might think that capitalism is the problem because economic growth has led to rising emissions. But, they argue, a modified invisible hand is the only workable solution: “It’s capitalism with all its innovative and entrepreneurial powers that is our only hope of steering clear of the looming climate shock.”
Theirs is definitely not a call for letting markets run free. Rather, markets need to be channeled “by a high enough price on carbon to reflect its true cost to society.” A substantial part of their book dissects the different estimates of the true cost. While the estimates of climate modelers differ widely, the authors point with some approval to the federal government’s estimate of $40 per ton of CO2. But the key point is that the actual price of carbon around the world is close to $1 per ton and therefore nowhere near what any serious climate policy would require.
The low actual price of carbon is an economic reflection of the fact that countries, taken as a whole, have made essentially no effort to reduce carbon emissions. This inconvenient truth can be seen in the trend ratio of carbon emissions relative to global GDP. For the period from 1960 to 2000, global CO2 emissions declined at 1.3 percent per year relative to global GDP. Put differently, global GDP grew at 3.7 percent per year while global CO2 emissions grew at 2.4 percent per year.
However, during the period since 2000, after climate summits at Kyoto, Copenhagen, Durban, and other places, there was no slowing of carbon emissions. Instead, the carbon–GDP ratio declined less rapidly—declining at 0.7 percent per year from 1.3 percent per year over the earlier period (see Figure 1). Diplomacy is cheap, but abatement is expensive.
The reason for the minimal effect, as Wagner and Weitzman emphasize and as I mentioned earlier, is the pervasive tendency for free-riding. International climate treaties contain no incentives for countries to undertake costly abatement. This was in part the reason the United States and Canada withdrew from their emissions commitments under the Kyoto Protocol: when they withdrew, they incurred the wrath of environmentalists but no other consequence.
The major challenge for climate policy is to overcome free-riding. The answer, I would suggest, is to rethink the design of climate treaties. We can look at successful treaties such as the European Union, the World Trade Organization, or military alliances as models for a more promising climate treaty.The essence of these successful treaties is the “club model.” A club is a voluntary group deriving mutual benefits from sharing the costs of producing an activity. Members get the benefits but also pay the dues. The benefits of a successful club are sufficiently large that members will pay dues and adhere to club rules in order to gain them. If we look at successful international clubs, we might see the seeds of an effective international system to deal with climate change.
I recently described a possible Climate Club in the American Economic Review.4 Under the club rules, participating countries would undertake harmonized but costly emissions reductions. For example, they might agree that each country would implement policies that produce a minimum domestic carbon price of $40 per ton of CO2. The easiest way to raise the price is through a carbon tax, but countries might prefer other approaches such as setting quantitative limits on emissions, or hybrid approaches.
A crucial aspect of the club is that countries who are outside the club—and do not share in the burden of emissions reductions—are penalized. Penalties for those outside the club are central to the club mechanism, and penalties are the major difference from all other proposals from Kyoto to the upcoming meeting in Paris. Economic modeling indicates that the most promising penalty is uniform percentage tariffs on the imports of nonparticipants into the club region. A country considering whether to undertake costly abatement would have to weigh those costs against the potentially larger costs of reduced trade with countries in the club.
A central feature of the club is that it creates a strategic situation that is the opposite of today’s free-riding incentives. With a Climate Club, countries acting in their self-interest will choose to enter the club and undertake high levels of emissions reductions because of the penalties for nonparticipation.
Undoubtedly, the concept of a Climate Club is highly idealized and even utopian. But consider the alternatives: reviving the feeble Kyoto Protocol; weak national plans that achieve minimal reductions; geoengineering that runs unknown risks of dangerous side effects; or doing nothing at all and incurring unchecked warming with all its fat-tailed perils. A Climate Club that ensures high prices of carbon emissions around the world, or the equivalent, is an essential step toward an effective policy to slow warming.
Figure 2 shows the difference between a bell curve distribution and one that has thicker tails. ↩
The report on solar radiation management (technically called albedo modification) is National Research Council, Climate Intervention: Reflecting Sunlight to Cool Earth (National Academy Press, 2015). There is a companion report on carbon dioxide removal, but that raises different issues from solar radiation management. ↩
See John von Neumann, “Can We Survive Technology?,” Fortune, June 1955. ↩
See William Nordhaus, “Climate Clubs: Overcoming Free-riding in International Climate Policy,” Presidential Address to the American Economic Association, January 2015, and American Economic Review, Vol. 105, No. 4 (April 2015). ↩