Few respected scientists now doubt that the world is warming, that man-made carbon dioxide (CO2) emissions are largely responsible, and that there will be significant further warming over the coming century. The pace of change and the impact on the climate of specific regions is debated. But it is clear that global warming is a reality, and that severe consequences—ranging from rising sea levels to the spread of tropical diseases—are likely.
Yet despite wide backing for the recent Stern review on the economics of global warming, the case for significant early action to combat it is still contested by opponents, such as Nigel Lawson (see his "Against Kyoto," Prospect online, November 2005), who see themselves as introducing economic realities into an overly emotional debate. The House of Lords economic affairs committee produced a report in July 2005 questioning the severity of the problem. Bjørn Lomborg's Copenhagen Consensus project has argued that offsetting climate change should be a low priority compared with other global challenges, such as Aids. Business lobby groups argue that measures to reduce emissions would harm competitiveness. The Kyoto protocol, in particular, is often attacked as both ineffective and harmful, constraining rich-country emissions but leaving China and India free to use low-cost energy to gain competitive advantage. But these arguments are unconvincing. There is a sound economic case for early action to address climate change. And while Kyoto is an imperfect policy in an imperfect world, its broad approach is right.
(A reminder: the UN framework convention on climate change, signed in 1992, committed the signatories—including the US, China and India—to the principle of action to combat climate change. Within this framework, the 1997 Kyoto protocol committed developed countries, but not developing ones, to cut emissions by, on average, 5.2 per cent below the 1990 level by the end of the first phase in 2010. All developed countries, apart from the US and Australia, are committed by treaty to achieving their targets.)
Economic growth under "business as usual" assumptions—with no action to curb emissions—will drive a massive increase in the stock of greenhouse gases in the atmosphere. In the half million years before the industrial revolution, CO2 concentrations oscillated between about 150 to 280 parts per million (ppm). They have now reached about 380, and without action are likely to reach between 540 and 970 ppm by the end of this century. The precise implications for temperature are unclear, and the emission reductions required to limit concentrations to a safe level are in turn debated. But avoiding major climate change will probably require radical cuts of 60 per cent or more below business as usual—rather than 5 or 10 per cent. Faced with that scale of adjustment, opponents of action put forward a second argument, stressing the huge costs and arguing that rather than accepting the implications for growth and living standards, the world should adapt to climate change.
But the most crucial finding of climate change economics is that the costs of achieving radical emission cuts are not huge. Relative to total GDP, estimates range from minute to very small. The Stern review's best estimate is that the cost of cutting global emissions by 25 per cent below current levels (and about 60 per cent below business-as-usual levels) by 2050 would be just 1 per cent of global GDP. Assuming global growth of 2.5 per cent a year, that implies average global living standards reaching in May 2050 the level they would otherwise reach in January 2050—by when they are likely to be over twice the current level. If the world does take the steps required to offset global warming, the impact on the growth path of 21st-century global GDP will be almost indiscernible.
This conclusion is central to any debate about climate change policy. It is also relatively uncontentious. Even economists who argue against early action use similar estimates: pessimists may suggest a 3 per cent cut in GDP, but almost no serious economist says 10 per cent. This unanimity reflects the following simple but robust logic. Energy costs in rich countries are typically about 4-5.5 per cent of GDP today, but will fall as the energy intensity of GDP declines. Most renewable technologies—wind power, biofuels, solar energy—are currently more expensive than fossil fuels, but not dramatically so, and their costs are falling. If the business-as-usual scenario is that energy costs would require 3 per cent of GDP in 2050, and we then have to pay 50 per cent more for renewable energy, the cost of moving to a low-carbon economy will be 50 per cent of 3 per cent—1.5 per cent of GDP. But it could be less if we achieve more rapid energy efficiency improvements (more fuel-efficient cars) or make lifestyle adjustments (more public transport and cycling). The balance between these three different types of adjustment—renewable energy, energy efficiency and life-style changes—drives the wide range of possible costs, but even the upper end implies only a minimal reduction in future prosperity. The idea that dramatically reducing emissions entails huge economic costs is a delusion propagated by business lobbyists defending vested interests, and by radical environmentalists who want climate change responsibility to mean the rejection of capitalism.
An economic argument against action to offset climate change can still be made, however. For although it would be feasible to move to a low-carbon economy at a cost of, say, 2 per cent of GDP by 2050, it would not be optimal if the adverse consequences of climate change, or the cost of adapting to it, amounted to less than 2 per cent. Modelling the comparative cost of climate change consequences, mitigation and adaptation is therefore vital to the debate.
Making such comparisons requires multiple assumptions, and different economists using different models produce widely divergent results. But the economic and human costs of climate change will almost certainly greatly exceed the cost of mitigation. The Stern review suggests that under business as usual, the long-term adverse consequences for human welfare could be as high as 5 to 20 per cent of global GDP. And two findings common to the more realistic models provide strong arguments for early action.
First, the implications of global warming combine a high probability of significant harm with an outside, but not negligible, chance of catastrophic effects. This reflects deep uncertainties in both the science and economics. The translation of a given rise in greenhouse gas emissions into temperature increase is inherently uncertain, and is made more so by the possibility of self-reinforcing effects. Warming could result in the release of methane currently trapped in frozen tundra; the resulting increase in greenhouse gas concentrations could drive yet more warming. Several such effects mean that the relationship between cumulative man-made emissions and temperature increase is likely to be strongly non-linear.
But the consequences for human welfare of rising temperatures are also likely to be non-linear: a rise of 4 degrees Celsius is likely to be much more than twice as harmful as a rise of 2 degrees. Given people's desire to avoid even a small possibility of disaster (on which the insurance whole industry is based), these uncertainties argue in favour of minimising the danger of temperature rises above the threshold at which self-reinforcing effects become likely. How that threshold is defined is uncertain, but 2 degrees is more likely to be appropriate than 3 degrees or more.
The second argument for early action is that warmer regions will suffer more than colder ones. Up to a certain level, warming could have positive effects in colder regions. A moderate temperature rise in Canada may result in increasing agricultural yields, but even slightly higher temperatures in India would harm both agricultural productivity and human health. Aggregating these effects, some models suggest that the global adverse consequences of modest temperature rises are insufficient to justify even small mitigation costs, but almost all models show that some regions—such as India and much of Africa—would be significant losers from a policy of inaction. Some economists therefore argue that the developed world, rather than acting to reduce emissions, should instead make income transfers to more vulnerable countries to assist adaptation. But there is no certainty that such transfers would occur and little chance they would be effective, given the complexity of possible climate change effects in fragile physical and political environments. If the sea level rises, there is at least a possibility that the cost of improved flood defences for London would be less than the costs of achieving carbon emission reductions. But if creeping desertification in parts of Africa produces famine and movement of people, the idea that aid transfers could be relied on to offset the consequences is naive. Even if we were confident that warming could be limited to below the threshold at which self-reinforcing effects become likely, policy should reflect the responsibility of rich countries for those likely to suffer the consequences of the industrialisation that made the developed world rich.
The argument that the rich world should begin to address climate change out of responsibility to the developing world is rejected by Bjørn Lomborg, organiser of the Copenhagen Consensus project. Lomborg asked eight economists to consider a set of initiatives that the developed world could fund, and to rank them on the basis of their benefit to mankind. The economists placed three variants of climate change mitigation policy (including the Kyoto protocol) towards the bottom of the list, favouring instead projects to combat Aids and malaria, malnutrition and poor sanitation. Lomborg cites this exercise as evidence that Kyoto was flawed.
This interpretation of the Copenhagen findings, however, is deeply misleading, given the specific question posed: "Where should the world invest, say, $50bn extra over the next four years to do the most good?" It is quite possible, in answer to that specific question, to agree that Aids and malaria programmes should have priority while also strongly supporting early action on climate change. Many actions to reduce emissions do not compete for limited resources but actually release them. If people cycle to work or buy more fuel-efficient cars, they will have more resources available for donations to Aids funds. If governments increase taxes on fossil fuels, thereby encouraging people to behave in a more environmentally friendly way, they will not have fewer resources for supporting overseas development, and may have more. If a company, encouraged by a higher tax on electricity but compensated by a lower tax on labour, identified energy efficiency improvements with a positive return, it would be no less able to make corporation tax payments to support the projects that the Copenhagen Consensus favoured.
The question "Should we spend £5bn more on education or on health?" is meaningful, even if some people reply, "Let's do both." But the question "Should we spend £5bn more on health or should we increase the tax on fossil fuels while cutting tax rates on employment?" is meaningless. Only to the extent that climate change initiatives entail budgetary expenditures—for instance, through subsidised research and development—can they be considered alternatives to the other desirable projects the Copenhagen Consensus project considered.
This does not mean that all climate change mitigation is costless. Improving energy efficiency can be, but requiring individuals to buy more expensive renewable energies would require them to sacrifice other forms of consumption. A key economic issue, therefore, is whether it is best for mankind to make sacrifices now or in the future. Several economists argue that although measures to reduce emissions will be appropriate at some point, most of the reduction should be delayed, since future growth will make people richer and better placed to sacrifice marginal consumption in the future.
But the models on which that judgement is reached are debatable in one respect and deficient in another. The optimal balance of emission reductions over time depends on two key factors: the "discount rate" used, and how the costs of mitigation change over time. The discount rate is the rate of return used to compare the present value of a pound spent today with one spent at some point in the future; for instance, an annual discount rate of 10 per cent would mean that £10 spent five years in the future would be worth about £6.21 now. The appropriate rate to use when discussing climate change is hotly debated. If the discount rate applied is high (some economists argue for about 4 per cent real), the cost in 2100 will count for close to nothing in present value terms, and action should be delayed. But pick a discount rate of 1.5 per cent (as William Cline proposed in his input to the Copenhagen project), and early action to offset future dangers becomes rational. The argument for a lower rate is that the trade-off between consumption sacrifice today and harmful consequences later should reflect rational consumer preferences and ethical considerations of intergenerational responsibility, and that using a 4 per cent rate undervalues the welfare of future generations. Say we knew for certain that climate change was going to destroy human life entirely in 150 years, unless we offset this danger by sacrificing 6 per cent of GDP today. Assuming an annual future global growth rate of 2 per cent—which would mean GDP growing almost 20-fold over the next 150 years—a discount rate of 4 per cent would still argue for taking no action now, since the discounted value of 100 per cent of GDP in 150 years' would be only 5.4 per cent of current GDP. The costs of mitigation would exceed the value of the future damage avoided. Economic theory does not provide a definitive answer in this debate. But it does, as Stern shows, suggest that the appropriate rate to apply to very long-term and very large costs and benefits is different, and lower, than that applicable to typical medium-term investment projects. People who place more value on the welfare of future generations can argue for a lower discount rate, and so for earlier action, without diverging from sound economics.
The evolution of mitigation costs over time was also considered in Cline's input to the Copenhagen project. All of the models cited incorporate assumptions on how renewable energy costs compare with fossil fuel costs at different points in time, and therefore how the costs of reducing emissions from a business-as-usual baseline change over time. But, crucially, all also assume that the costs at any time are exogenously (externally) given, uninfluenced by whether emission reductions have already been made. This must be wrong. The costs of renewable energy alternatives at any date are not predetermined facts but strongly influenced by research and development expenditures. The cost of wind-generated electricity has halved since 1990 and is now low enough to compete with fossil fuel-based electricity without subsidy. This would not have occurred without commitments to emission reductions, which created an assured market and provided an initial subsidy for wind power. Looking ahead, the cost of solar photovoltaic electricity in 2050 will be lower the tighter the controls on carbon emissions imposed in the preceding years.
One of the most important arguments for early action on climate change, then, is that it will not only begin to contribute to emission reductions, but will also provide future generations with a capital base and a set of technologies that will make it easier for them, in turn, to achieve further cuts. The models considered by the Copenhagen project fail to allow for this. Vernon Smith, one of the economists involved in the project who most clearly expounded the "leave climate change to later" view, argues that there is a responsibility on earlier generations to "leave subsequent generations a capital stock which has not been diminished by incurring premature abatement costs." But the capital we leave to future generations is defined in qualitative as well as quantitative terms. If early action to offset climate change reduces GDP in 2050 by 2 per cent (the upper end of typical estimates), the capital stock will probably be smaller by the same amount. But the idea that a capital stock of 100 which does not incorporate energy efficiency and renewable energy developments will better equip future generations to cut emissions than a capital stock of 98 which does is absurd. If all Chinese coal power stations were constructed so as to facilitate the subsequent fitting of carbon capture and storage, the ability of future Chinese generations to cut emissions would be increased, not diminished, even if Chinese GDP in 2050 were marginally lower as a result.
Economic analysis and modelling alone cannot define an optimal climate change policy—too many subjective factors (such as judgements about the value we place on avoiding catastrophic consequences, or on our responsibilities to low-income countries and future generations) come into play. But economic analysis can narrow the range of sensible options. It suggests that the long-term aim should be greatly to reduce the risk of going beyond potentially dangerous thresholds of greenhouse gas concentration and temperature increase. It implies that an optimal policy would be one that avoids early emission reductions so radical as seriously to impede economic growth, while establishing a framework for increasing reductions over time, achieving cuts of, say, 60 per cent below business-as-usual levels within 50 years. And it recognises that even modest early reductions can help to stimulate the technologies that make later reductions less costly.
Against these criteria, Kyoto scores well. The impact on growth of the initial commitments is trivial (models typically suggest at most a few tenths of a per cent of GDP foregone by 2010), but the targets are tight enough, if translated into sensible policies, to stimulate changes in behaviour and technology that will make it less costly to achieve more radical reductions in future. And the framework is designed to evolve over time, with new targets established as information on the science and economics becomes available.
Two arguments are made against Kyoto, however. The first, stressed by Lomborg, is that the reductions are insufficient to make an adequate difference. But this argument is flawed. Any policy framework that would guarantee emission cuts sufficient to offset harmful climate change would have to specify global emission levels from now until the end of the century. Many economists (and, one suspects, Lomborg) would rightly attack such a policy as inflexible, removing our ability to evolve policy in the light of new understanding. Conversely, any framework which preserves flexibility will, in its initial commitments, be inadequate. An optimal framework would entail initial commitments that are in themselves inadequate, but that are intended to evolve into adequate commitments later. Kyoto attempts to strike that balance.
The second criticism of Kyoto is that it constrains only developed countries. In the long run, it is obvious that a global framework must cover all countries, through globally agreed emissions caps or carbon taxation, or a mix of both. But it is simply not credible that developing countries such as India, whose per capita emissions are less than a tenth of those of America, are going to make commitments before they see the developed world taking action. Alternatives to Kyoto—such as William Cline's suggestion that the world should agree an optimal carbon tax—are classic examples of the best as the enemy of the good. The most likely route—probably the only route—to a globally agreed framework is one, like Kyoto, in which rich countries lead the way.
If enough developed countries commit to action, moreover, competitiveness concerns become groundless. Even for countries acting alone, competitiveness is only a constraint on adopting a specific subset of climate change mitigation actions. An individual choosing a more fuel-efficient car faces no competitiveness constraint. Nor does a supermarket investing in energy-efficient refrigeration, or any other business operating in the bulk of the economy not traded internationally. Only in internationally traded sectors where energy inputs account for a significant percentage of costs are competitiveness objections to emission caps or carbon taxes sometimes valid, and these sectors' concerns are hugely reduced if developed countries agree a common approach. Even without US membership of Kyoto, a common European approach covers a wide enough area to make effects on competitiveness trivial. The only exceptions are a few highly specialised sectors, in particular aluminium, where energy costs are huge. Arguments for sensitive treatment of sectors with real competitiveness concerns are valid; the assertion that competitiveness makes action on climate change impossible until Kyoto covers developing countries is nonsense.
None of this is to imply that Kyoto is a perfect policy. Nor is it to argue that flexibility, including political flexibility, isn't crucial in deciding the next steps. America's failure to ratify Kyoto, or to pursue emission targets unilaterally, is regrettable. But the reality is that the US, having failed to act, could not now reach its original 2010 targets without incurring big costs. And it is not going to say, "We were wrong, and we will now sign up"; superpowers do not eat humble pie. Once President Bush has left office, flexible ways will need to be found to re-engage the US with climate change policy, and if and when a comprehensive global framework is agreed, a new name may be cosmetically useful. But the basic approach of Kyoto, far from failing the test of economic realism, is precisely what sound economics dictates.
