Energy-Climate Challenge and What To Do About It

THE ENERGY-CLIMATE CHALLENGE AND WHAT TO DO ABOUT IT

byJohn P. Holdren

Energy-Technology Innovation Project
Belfer Center for Science and International Affairs
John F. Kennedy School of Government
Harvard University

July 2000

This paper was prepared, in the format of a memorandum to the newly elected American President in 2001, for an Aspen Institute Environmental Forum (Aspen, Colorado, 8-11 July 2000) on "What will be the most important environmental issues of the next 50 years, and what should be done about them now?" It will appear in a book of such presentations being prepared by the Aspen Institute, and it has also been submitted for publication to the journal Environment. The author is Teresa and John Heinz Professor of Environmental Policy and Director of the Program on Science, Technology, and Public Policy at the Kennedy School of Government, as well as Professor of Environmental Science and Public Policy in the Department of Earth and Planetary Sciences, at Harvard University. He is also Chair of the Energy Panel in President Clinton''s Committee of Advisors on Science and Technology, Chair of the Committee on International Security and Arms Control of the National Academy of Sciences, Chair of the National Research Council Committee on US-India Cooperation on Energy, and member of the Board of Directors of the John D. and Catherine T. MacArthur Foundation. The opinions expressed here are the author''s alone and should not be assumed to be endorsed by any of the organizations with which he is affiliated. Support for the Energy Technology Innovation Project from the Energy Foundation, the Heinz Family Foundation, and the Winslow Foundation is gratefully acknowledged.

HARVARD UNIVERSITY

Cambridge, Massachusetts 02138
john_holdren@harvard.edu21 January 2001

TO: THE PRESIDENT OF THE UNITED STATES
RE: THE ENERGY-CLIMATE CHALLENGE AND WHAT TO DO ABOUT IT

Mr. President:

More than a quarter of a century ago, two immensely important understandings about the contemporary human condition pushed themselves almost simultaneously into public and political consciousness. One of these understandings - reflected in the occasion of the first Earth Day in 1970, the appearance of the report of the MIT-hosted international Study of Critical Environmental Problems that same year, and the convening of the first UN Conference on the Environment in Stockholm in 1972 - was that the environmental conditions and processes increasingly under siege from the expansion of human activities were too important to human well-being to continue to be neglected, as they often had been, in the pell-mell pursuit of increased material prosperity. Ways would need to be found - and could be found - to meet economic aspirations while adequately protecting the environmental underpinnings of well-being. The other understanding, which was thrust on the world by the Arab-OPEC-induced oil-price shocks of the 1970s and the global economic recession that followed, was that a reliable and affordable supply of energy is absolutely critical to maintaining and expanding economic prosperity where such prosperity already exists and to creating it where it does not.

Today, these two understandings have long since become part of conventional knowledge. Essentially everybody recognizes the importance of energy for economic prosperity and the importance, for human well-being, of protecting the environment. But far less widely appreciated is the close connection between these two imperatives - and the immense challenge arising from it - in the form of the central role played by civilization''s principal energy sources in generating the most dangerous and difficult environmental problems facing the planet. Energy supply is the source of most of human exposure to air pollution, most of acid precipitation, much of the toxic contamination of ground water, most of the burden of long-lived radioactive wastes, and most of the anthropogenic alteration of global climate. The constraints imposed by these problems on the composition and expandability of energy supply, moreover, are becoming the most important determinants of energy strategy and, increasingly, of energy''s monetary costs. In short, energy is the most difficult part of the environment problem, and environment is the most difficult part of the energy problem. The core of the challenge of expanding and sustaining economic prosperity is the challenge of limiting, at affordable cost, the environmental impacts of an expanding energy supply.

The most demanding part of this energy/environment/prosperity challenge, moreover, is the challenge posed by anthropogenic climate change. In terms of the current scale of damage to health, property, ecosystems, and quality of life, climate change is not yet comparable to air pollution, or water pollution, or land transformation. But in the long run (and by this I mean a matter of decades, not a matter of centuries ) it will come to be understood as both the most dangerous and the most intractable of the environmental impacts of human activity.

• It is the most dangerous because climate constitutes the envelope within which all other environmental conditions and processes operate, and because the degree of irreversibility of human-induced climate change, once it has been set in motion, is very high. Substantial disruption of the climatic "envelope"places at risk the full array of "service" functions of the environment: the formation and fertilization and retention of soils; the detoxification of pollutants; the provision of fresh water; the distribution of warmth and nutrients by ocean currents; the natural controls on human and plant pathogens and pests; the bounding, within mostly tolerable limits, of extremes of temperature and precipitation and storminess; and much more.1 As discussed below, the evidence that human-induced climate change is on a trajectory to create major damage to these services within the current century is becoming overwhelming. The complacent notion that we are clever enough and rich enough to fully replace these contributions of the natural environment to our well-being with engineered substitutes, moreover, is folly - although, insofar as a substantial degree of disruption of global climate is already inevitable, we shall have to try.
• Climate change is the most intractable of all environmental impacts to address because its primary cause - the increasing emissions, from fossil-fuel combustion, of the heat-trapping ("greenhouse") gas, carbon dioxide - is deeply embedded in the character of civilization''s current energy-supply system, in ways that would be both time consuming and potentially very costly to change. More than 75 percent of the world''s energy supply (and more than 85 percent of that of the United States) currently comes from burning oil, coal, and natural gas.2 The rich countries of the industrialized North achieved their enviable prosperity based on a huge expansion of the use of these versatile and relatively inexpensive fuels, and the "business as usual" energy future would have the developing countries of the South doing the same. Today''s fossil-fuel-dominated world energy system, worth some 10 trillion dollars at replacement cost and characterized by equipment-turnover times of 20 to 50 years, could not be rapidly replaced with non-CO2-emitting alternatives even if these were no costlier than conventional fossil-fuel technologies have been (and, today, the non-CO2 options are considerably costlier); nor is the voluminous carbon-dioxide combustion product (some 3 tonnes of CO2 per tonne of coal or oil) easy to capture with add-on pollution-control equipment for existing engines, furnaces, and power plants. That the impacts of global climate disruption may not become the dominant sources of environmental harm to humans for yet a few more decades cannot be a great consolation, then, given that the time needed to change the energy system enough to avoid this outcome is also on the order of a few decades. It is going to be a very tight race.

The challenge is by no means insurmountable, however. It can be met, but only by a strategy embodying all six of the following components:

• expanded research on the science of climate change, climate-change impacts, enhancement of the uptake of carbon sinks in terrestrial ecosystems and in the oceans, geotechnical engineering to offset the effects of greenhouse gas increases in the atmosphere, and means of adaptation to the degree of climate change that proves unavoidable; and increased investments to exploit the opportunities that this research uncovers;
• increased national and international support for the education, development, social-welfare, and family-planning measures known to be most effective in reducing population growth;
• incentives and other help for firms and consumers to make low- and no-CO2 choices from the menu of energy-supply and energy-end-use-efficiency options available at any given time;
• accelerated research, development, and demonstration of advanced energy-supply and end-use technologies, to steadily expand and improve the menu from which choices are made;
• increased international cooperation to facilitate the application of the results of components (1), (3), and (4) in the developing South as well as throughout the industrialized North;
  
  

• development of a global framework of commitments to long-term restraints on greenhouse-gas emissions designed for sufficiency, equity, and feasibility. While the private sector will clearly need to play an immense role in much of this agenda, the nature of the problem as one involving externalities, common-property resources, public benefits, and binding agreements among states dictates that government policies must play a major role, as well. And the government of the United States - a country with a quarter of the world''s fossil-fuel use and CO2 emissions, the world''s strongest economy, and the world''s most capable scientific and technological establishments - ought to be leading, not following, in this effort that is so crucial to the prospects for a sustainable prosperity for all.

In the remainder of this memo, I elaborate on three important aspects of this argument: what the current state of climate-change science allows one to say about the implications of energy business as usual; the extent of the deflection from business as usual likely to be required to bring the degree of energy-linked disruption of global climate in this century within manageable bounds; and the content and prospects of the six-point strategy summarized above for achieving this deflection.

Business as usual and its climate-change implications

It is illuminating to disaggregate carbon emissions to the atmosphere into four multiplicative factors: the size of the human population, the per capital level of economic activity (measured in purchasing-power-parity-corrected dollars of Gross Domestic Product per person), the energy intensity of economic activity (measured in, say, gigajoules per thousand dollars of GDP), and the carbon-emission intensity of energy supply (measured in, say, kilograms of carbon contained in CO2 emitted to the atmosphere per gigajoule of energy supply). The "business as usual" assumption is not that these factors remain constant, but that their trajectories of change follow recent trends, adjusted for expected patterns of development. In a typical business-as-usual global energy future:3

• World population increases from 6.1 billion in 2000 to 8.5 billion in 2030 and 9.8 billion in 2050, stabilizing by 2100 at about 11 billion. Nearly all of this growth occurs in the developing countries.
• Per capita economic growth is higher in the developing countries than in the industrialized ones, but declines gradually over the century in both. Aggregate economic growth (reflecting the combined effect of growth in population and in per-capita GDP) averages 2.9% per year from 2000 to 2020 and 2.3% per year over the whole century, in real terms. As a result, world economic product (corrected for purchasing power parity) grows from about $38 trillion in 2000 to $87 trillion in 2030, $140 trillion in 2050, and $360 trillion in 2100 (all in 1995 dollars).
• Energy intensity of economic activity is assumed to fall at the long-term historical rate of 1% per year, in industrialized and developing countries alike, for the entire century. With the indicated economic growth, this produces a doubling of world energy use between 2000 and 2040, a tripling by 2070, and a quadrupling by 2100 (by which time the figure is about 1800 exajoules per year, compared to about 450 EJ/yr in the year 2000).
• Carbon intensity of energy supply is assumed to fall at a rate of 0.2% per year, in all countries, for the entire century. With the indicated energy growth, this causes carbon emissions from fossil-fuel combustion to triple over the century, going from a bit over 6 billion tonnes of carbon per year in 2000 to some 20 billion tonnes per year in 2100. Historically, the atmospheric concentration of carbon dioxide rose from a pre-industrial level of about 280 parts per million by volume (ppmv) in 1750 to about 370 ppmv in 2000 - an increase of 32 percent - driven in the first part of this 250-year period mainly by deforestation and in the latter part of the period mainly by fossil-fuel combustion. Under the indicated business-as-usual (BAU) scenario for emissions from fossil fuels (and assuming no further contribution from net deforestation), the atmospheric concentration would be expected to reach 500 ppmv by 2050 and over 700 ppmv by 2100. If this BAU scenario persisted until 2100, moreover, there would be no possibility that the continuing run-up of the atmospheric CO2 concentration thereafter could be stopped below 1100 ppmv (a quadrupling of the pre-industrial value).

Besides the CO2 increase, atmospheric concentrations of a number of other greenhouse gases - mainly methane, nitrous oxide, tropospheric ozone, and halocarbons - have also increased since pre-industrial times. The warming effect of all of these increases together, as of the mid-1990s, was estimated as roughly equal to that of the CO2 increase; but this warming contribution of the non-CO2 greenhouse gases was estimated to be approximately offset by the net cooling effect of increased atmospheric concentrations of particulate matter also caused by human activities.4 Under the indicated BAU scenario, the concentrations of the non-CO2 greenhouse gases increase more slowly over the 21st century than the CO2 concentration does, while concentrations of atmospheric particulates slowly decline. By 2100, the net warming effect of all the greenhouse gases together, less the cooling from particulates, is just slightly larger (about 10 percent) than the warming that would be caused by the increased CO2 alone. Conveniently, then, one can simplify the discussion of future possibilities, without much loss of accuracy, by associating these possibilities with CO2 concentrations alone - leaving out the offsetting complexities of non-CO2 greenhouse gases and particulate matter - and this is often done.5 I shall do so here.

As I write this, the debate about whether the effects of rising CO2 concentrations on global climate are already apparent is essentially over - resolved in the affirmative.6

• In 1995, when the atmospheric CO2 concentration was 360 ppmv, the Intergovernmental Panel on Climate Change (IPCC) wrote in its Second Assessment of the Science of Climate Change that "the balance of evidence suggests a discernible human influence on global climate." This report noted that global mean surface air temperature had by then increased by 0.3 to 0.6ºC since the late 19th century, that this had been accompanied by an increase in global sea level by 10-25 centimeters, and that the patterns of change (in relation to day-night temperature differences, vertical temperature distribution, latitudinal differences, patterns of precipitation, and more) match with quite striking fidelity the patterns predicted, by basic climate science and elaborate computer models alike, to result from the observed increases in greenhouse-gas concentrations, adjusted for the effects of atmospheric particulate matter and the known variability of the sun''s output. These patterns are often described as the "fingerprint" of greenhouse-gas-induced climate change, and no one has postulated a culprit other than greenhouse gases that would have the same fingerprint.7
  
• Since the completion of the IPCC Second Assessment the evidence has only grown stronger. By the end of 1999 it was clear that 15 of the 16 warmest years worldwide, in the 140 years since the global network of thermometer records became adequate to define a global average surface temperature, have occurred since 1980. The seven warmest years in the140-year instrumental record all occurred in the 1990s, notwithstanding the cooling effects of the Mt. Pinatubo volcanic eruption at the beginning of that decade. Based on evidence from ice cores and other paleoclimatological data, it is likely that 1998 was the warmest year in the last thousand, and the last 50 years appear to have been the warmest half century in six thousand years. A National Academy of Sciences report that appeared in January 2000, reviewing modest discrepancies between the surface thermometer records and satellite measurements made over the preceding 20 years, concluded that "the warming trend in global-mean surface temperature observations during the past 20 years is undoubtedly real"; and a comprehensive survey of ocean-temperature measurements published in Science in March 2000 showed widespread warming of the oceans, over the past 40 years, to great depths.
  With rather high confidence, then, one can now say that global warming is being experienced and that greenhouse-gas increases from human activities are its primary cause.8

So, what is to be expected from continuation of business as usual? According to the IPCC Second Assessment, a climate in equilibrium under the influence of twice the pre-industrial CO2 concentration would have warmed by 1.5 to 4.5ºC, with a "best estimate" of 2.5ºC. Because of the large thermal inertia of the oceans, however, the attainment of the equilibrium temperature increase associated with a given CO2 concentration lags by some decades the attainment of that concentration. Thus, although the BAU emissions future described above yields a doubling of the pre-industrial CO2 concentration by 2070, the "best estimate" temperature increase over the pre-industrial value is only about 1.8ºC by 2070, reaching 2.5ºC in 2100.9 These, on the other hand, are global land-and-ocean averages; in general, the increases on land will be higher, and those on land at high latitudes higher still. Sophisticated climate models capable of tracing the time-evolution of these changes typically show mid-continent U.S. temperatures in the range of 2.5 to 4ºC higher than today''s for the middle of the century under business as usual.

The IPCC''s 1995 assessment concluded, for the indicated BAU scenario, that sea level would rise by 2100 to a best estimate of 50 centimeters above today''s value (and would continue to rise for centuries thereafter), and that other characteristics of the warmed climate would be likely to include increases in floods and droughts in some regions,10 increased variability of precipitation in the tropics, and a decrease in the strength of the North Atlantic circulation that warms the southeastern United States and western Europe in winter. (Yes, a warmer climate overall can make it colder in some places at some times!) The assessment found that the expected climate change "is likely to have wide-ranging and mostly adverse effects on human health" (with the increased damage from heat stress, aggravation of the effects of air pollution, and expanded range of tropical diseases more than offsetting the reduced health impacts from cold winters); that northern forests "are likely to undergo irregular and large-scale losses of living trees"; and that agricultural productivity would "increase in some areas and decrease in others, especially the tropics and subtropics" (where malnutrition is already most prevalent).11

The 1995 assessment also emphasized, as all competent reviews of climate-change science do, that many uncertainties about the character, timing, and geographic distribution of the impacts of climate change remain, and that the nonlinear nature of the climate system (in which small causes can have big effects) implies the possibility of surprises that current models cannot capture at all. Such possibilities include increases in the frequency and intensity of destructive storms (which a few models do suggest), larger and more-rapid-than-expected sea-level rise (from, e.g., slumping of the West Antarctic Ice Sheet), and a "runaway" warming effect from release of large quantities of the potent greenhouse gas, methane, immense amounts of which are currently locked in icy methane clathrates beneath permafrost and on the ocean floor. These examples illustrate that "uncertainty" does not necessarily mean, as the public and policy makers sometimes suppose, that when we learn more it will all turn not to be as bad as was feared. It can easily turn out to be worse than the "best estimates", not just better.

While there is room for debate about whether the impacts of doubling the pre-industrial CO2 concentration would be unmanageable, any basis for optimism shrinks when the postulated CO2 level moves to a tripling or a quadrupling.12 A quadrupling of pre-industrial CO2 would yield, under the IPCC''s assumptions, an equilibrium mean global surface temperature increase of 3 to 9ºC, with a "best estimate" of 5ºC. Studies by Princeton''s Geophysical Fluid Dynamics Laboratory - one of the few groups to analyze this case - found equilibrium average temperature increases of 7-10ºC (13-18ºF) for the mid-continental United States after a quadrupling, drops of June-August soil moisture by 40 to 60 percent over most of the country, and a July heat index for the southeastern United States reaching 43ºC (109ºF) compared to the pre-warming value of 30ºC (86ºF).13 The Princeton calculation also showed the North Atlantic thermohaline circulation (which drives the Gulf Stream) shutting down essentially completely under a quadrupling of pre-industrial CO2, accompanied by a rise in sea level at about twice the rate expected for a CO2 doubling.14 While there are of course uncertainties associated with the projections of this particular study - as for all others - it would be folly to suppose that the impacts of the degree of climate disruption that any model will show for a quadrupling of atmospheric CO2 would entail anything other than immense human costs.

How big a departure from business as usual is required?
Stabilizing emissions of carbon dioxide near current levels would not lead to stabilizing the atmospheric CO2 concentration. Constant emissions the mid-1990s rate would lead, instead, to a more or less steady increase of about 1.5 ppmv per year in the concentration, leading to a value of about 520 ppmv by 2100 if the constant emission rate were maintained throughout this century. Stabilizing the atmospheric concentration at any level of possible interest - even at a quadrupling of the pre-industrial level - would require that global emissions drop eventually to a small fraction of the current 6 billion tonnes of contained carbon per year (GtC/yr). It is consistent with ultimate stabilization of the atmospheric concentration, however, that emissions rise for a time - as they are destined to do given the momentum in the current fossil-fuel-dominated energy system - as long as they peak eventually and then fall to levels well below today''s.

If, for example, it is desired to stabilize the atmospheric CO2 concentration at 750 ppmv, the BAU trajectory could be followed until as late as 2035 (when emissions would have reached about 12 GtC/yr); but emissions would then have to level off at not more than 14 GtC/yr by about 2050, fall to about 11 GtC/yr by 2100, and reach 4 GtC/yr by 2200. To stabilize the atmospheric concentration at 550 ppmv - about twice the pre-industrial value - the BAU trajectory could not be followed beyond about 2020, and the concentration would need to peak at not more than about 11 GtC/yr around 2030 and be falling by 2035, reaching 5 GtC/yr by 2100 and 2.5 GtC/yr by 2200. A somewhat different emissions trajectory that would still be compatible with stabilizing the atmospheric CO2 concentration at 550 ppmv would depart from business as usual sooner (essentially immediately), peak lower (at 8-9 GtC/yr) and later (around 2050), and then fall more gradually, becoming coincident with the more sharply peaked 550 ppmv trajectory between 2150 and 2200.

Besides the details of their shapes, emissions trajectories that lead to stabilization of the atmospheric CO2 concentration at various levels can be characterized by the cumulative emissions they entail over the 100 years between 2000 and 2100. Trajectories compatible with stabilization at 550 ppmv would have cumulative emissions in the range of 800 to 900 GtC in this century. Trajectories corresponding to stabilization at 750 ppmv would have 21st century cumulative emissions in the range of 1100 to 1200 GtC. For comparison, cumulative 21st century emissions on the BAU trajectory would be about 1400 GtC.

Under the UN Framework Convention on Climate Change, which was enacted at the "Earth Summit" in Rio in 1992 and subsequently ratified by the United States and more than 170 other nations, the parties agreed to pursue "stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system." There has been no formal or even informal agreement, up until now, on the stabilized concentration that would be considered low enough to meet this criterion. But it is difficult to believe, given the evidence that global climate change is already doing damage, that any level equivalent to more than a doubling of the pre-industrial CO2 concentration could possibly be considered compliant with the Convention. Were it not for concerns about the practicality of meeting a lower target - or, stated another way, concerns that the cost of compliance might exceed the benefits - it seems likely that a level at or below today''s concentration would be chosen. If the target were a "compromise" 450 ppmv (a bit closer to today''s 370 ppmv than to a doubling at 560 ppmv), then cumulative carbon emissions over the 21st century would have to be kept below 600 GtC - a figure 2.5 times smaller than that for business as usual.

The magnitude of the challenge represented by a target this low can be illustrated by considering what would be required to meet it by emissions reductions alone - that is, without reductions below BAU population growth or per-capita economic growth. With the population and per-capita GDP trajectories at their BAU values, a doubling of the century-average rate of decline of energy intensity (energy divided by GDP), from 1.0 to 2.0 percent per year, and a doubling of the rate of decline of carbon intensity (carbon emissions divided by energy). from 0.2 to 0.4 percent per year, would reduce the 21st century emissions to about 700 GtC, establishing a trajectory that could stabilize the atmospheric concentration at about 500 ppmv. If, in this variant, the century-average rate of reduction of carbon intensity were boosted to 0.6 percent per year - three times as fast as business as usual - the result would be a trajectory consistent with stabilization at 450 ppmv.15

Attaining such a trajectory would not be easy, but neither is there reason to consider this impossible. Higher rates of reduction in these intensities than are needed have been achieved in at least some places and some times in the past (although never for as long or as universally as would be required to meet the challenge described here). Between 1973 and 1986, for example, in response to the 1973-74 and 1979 world oil-price shocks, energy intensity in the United States fell at an average of 2.5 percent per year. In France in about the same period (1973-1991), when that country was rapidly nuclearizing its electricity-generation sector, the carbon intensity of energy supply in the French economy fell at an average rate of 2.7 percent per year. Among global scenarios for energy in the 21st century constructed by the Intergovernmental Panel on Climate Change, the joint "Global Energy Futures" study of the World Energy Council and the International Institute for Applied Systems Analysis, and other reputable efforts, there are high-technological-innovation variants with long-term world-average rates of improvement averaging 1.5 to 2.5 percent per year in energy intensity and 0.6 to 1.2 percent per year in carbon intensity.

The six-point action program

A sensible strategy, in my view, would seek to stabilize the atmospheric CO2 concentration below 500 ppmv, while taking additional steps to try to reduce the harm to human well-being that disruptions of climate at even this level of greenhouse-gas increase would tend to cause. In principle, there are just four possible approaches to the overall problem from which the ingredients of such a strategy can be assembled, and they can be enumerated as follows:

• reduce the emissions of greenhouse gases below what they would otherwise be;
• remove from the atmosphere greenhouse gases that have previously been added to it;
• intervene to reduce the effects of greenhouse-gas increases on climatic variables;
• adapt to reduce the human impact of the degree of climate change that cannot be avoided.
  Working up from the bottom of this list:

• It is plain that a considerable amount of adaptation will be needed, inasmuch as climate change and adverse impacts from it are already apparent. Adjustments in agriculture, forestry, fisheries, water storage, flood control, public-health measures, transportation management,16 and protection of coastal settlements, among many other activities, will be required. Some of this is already underway. But a strategy that relies too heavily on adaptation and not enough on avoidance is likely to be very costly, not to mention much less effective in the resource- and infrastructure-poor developing countries than in the industrialized ones.
• Interventions to reduce the effects of greenhouse-gas increases on climatic variables constitute what is often termed "geotechnical engineering". An example would be the insertion of reflecting materials into orbit in order to reduce the sunlight reaching the Earth and thereby offset greenhouse warming. While such ideas are intriguing, they suffer today from insufficient understanding of the intricacies of the planet''s climatic machinery for us to be confident of achieving the desired effects (or, put more pointedly, to be confident of doing more good than harm). We are powerful enough to disrupt the climate, and smart enough to notice we are doing this, but alas not yet remotely competent to fine-tune the complex machinery of climate to our tastes. The possibilities - and the climate system itself - need much more study.
• The best means currently known for removing carbon dioxide from the atmosphere is growing trees. The trick is to increase the inventory of carbon embedded in plant material, of which the most enduring form is wood: just as net deforestation reduces that inventory and adds CO2 to the atmosphere (as has happened on a global-average basis over much of the past few centuries), so does net afforestation increase the inventory and remove CO2 from the atmosphere. Expanding the forested area of the planet is feasible (although not as easy to achieve and to sustain as is sometimes supposed), as is increasing the carbon storage on existing forested land. But given the amount of continuing deforestation in the tropics, we will probably be doing well just to stay even on a global-average basis over the next century. The best imaginable performance at rapidly ending current deforestation and improving other land-management practices that generate greenhouse-gas emissions,17 combined with very aggressive reforestation and afforestation efforts (including widespread, costly restoration of degraded land) might achieve 20 or 25 percent of what is required for a transition, in this century, from the BAU trajectory to a trajectory consistent with stabilizing atmospheric CO2 at 500 ppmv. Far more study and effort than are happening today will be required to achieve even this much.18

The foregoing considerations about the adaptation, geotechnical-engineering, and greenhouse-gas-removal options motivate the first element in my six-point program, namely:

1. expanded research on the science of climate change, climate-change impacts, enhancement of the uptake of carbon sinks in terrestrial ecosystems and in the oceans, geotechnical engineering to offset the effects of greenhouse gas increases in the atmosphere, and means of adaptation to the degree of climate change that proves unavoidable; and increased investments to exploit the opportunities that this research uncovers.19
The same considerations make it plain that, no matter how much ultimately proves to be achievable under these headings, prudence requires pushing forward very aggressively as well on the option of reducing emissions of greenhouse gases below what they would otherwise be. This necessary preoccupation brings me back to the determinants of the most important anthropogenic greenhouse-gas emissions - those of carbon dioxide - in the form of population, GDP per person, energy use per unit of GDP, and carbon emissions per unit of energy. Although the trajectories of these four factors alone are sufficient to specify the trajectory of total carbon emissions, each of the four is influenced in turn by an array of interacting technical, economic, social, and political factors, wherein reside the leverage points for policy.

Let me begin with population, for it would be foolish to ignore it. To show why, it is only necessary to note that the range of plausible world population sizes in 2100 extends at least from 7 billion on the low end to 14 billion on the high end. The difference between these two figures in terms of ease or difficulty of achieving a low-carbon-emission energy future (as well as for a great many other aspects of the human condition) is immense. We should be striving for a result near the low end of these possibilities.

The principal manipulable determinants of human fertility, and hence of population growth, are the prospective parents'' knowledge of reproductive biology, their motivation affecting desired number and spacing of offspring, and the effectiveness and availability of technologies of fertility limitation. Knowledge of reproductive biology is a matter of education - of women even more importantly than of men - which is in turn a matter of development. Motivation about number and spacing of offspring has been shown to depend most directly on the status and education and employment opportunities of women, the survival prospects of offspring, and the availability of a social security system - again, all matters closely related to the process of development itself - as well as government incentives for small families and other factors influencing perceptions about the individual and social costs of large ones. Fertility-limiting technologies (the means of contraception and abortion) are already quite good (although they could always be better); the key factor is access to them on satisfactory terms.

While a few of these fertility-reducing factors have been or could be politically sensitive, nearly all of them are things that most of the world''s people want for their own immediate well-being. That achieving them would also bring a large societal gain in the form of reduced population growth and the benefits of that for addressing the energy/climate challenge (and a great many other resource, environmental, and social problems of the 21st century) means that there can be even less excuse than otherwise for failing to push ahead with the second element in the six-point program:

2. increased national and international support for the education, development, social-welfare, and family-planning measures known to be most effective in reducing population growth.
The GDP-per-person factor in carbon emissions can be dispensed with more quickly, at least in respect to policy leverage for reducing those emissions. Much can be said, of course, about how GDP is influenced by the productivity of labor (which in turn is influenced by health, education, training, organization, technology, and natural resources), but policy is rightly focused on how to increase all this, not on decreasing it as a way to reduce environmental harm. In the long run, GDP per person also depends on the allocation of time between economic and noneconomic activities, influenced in turn by conceptions about the relative importance of economic and noneconomic contributions to well-being; but, as much as some might like to see a re-orientation of human wants away from economic consumption, advocating this explicitly is not likely to become a part of a major political party''s platform for some time to come. (It may happen, nonetheless, that bringing more of the external costs of economic growth into the balance sheets of producers and consumers - as overall economic efficiency requires - will raise the price of growth enough that people will not buy so much of it. But the appropriate policy instruments relate to internalizing the external costs, not to suppressing economic growth per se.)


This brings me to the two more technical determinants of emissions, namely the energy intensity of GDP and the carbon intensity of energy supply. The energy intensity of GDP relates both to "technical efficiency" (a matter of the energy requirement to produce a given good or service) and to the composition of economic output (a matter of the mix, in the economy, of more and less energy-intensive types of goods and services). The carbon intensity of energy supply depends on the characteristics of fossil-fueled energy technologies (specifically, how much carbon they emit per unit of end-use energy they supply to the economy) and the mix of fossil-fueled and non-fossil-fueled energy technologies in the energy system as a whole. The two elements of the six-point program that relate directly to the evolution of these factors in the United States are

3. incentives and other help for firms and consumers to make low- and no-CO2 choices from the menu of energy-supply and energy-end-use-efficiency options available at any given time;
4. accelerated research, development, and demonstration of advanced energy-supply and end-use technologies, to steadily expand and improve the menu from which choices are made.

The range of policy measures that can be considered under item (3) is wide, including: analysis of, and education of firms and individual consumers about, the available options; correction of perverse incentives embedded in existing policies; lowering of bureaucratic barriers to adoption of otherwise desirable options; performance standards (relating, for example, to energy efficiency and to emissions); portfolio standards (relating to the proportion of low- and no-carbon options in the energy-supply mix); preferential financing, tax breaks, and other subsidies for demonstration and widespread deployment of targeted options; overall emissions caps implemented through tradeable permits; and carbon taxes.20

While there is room for innovation and expanded activity on many of these fronts, most economists will argue that the most potent and economically efficient means to encourage low-carbon and no-carbon choices from the menu of available options, as well as to encourage research and development of better choices of these kinds, would be a tax on carbon emissions. They are right. Taxing a widely practiced activity that the society has reason to want to discourage has a long and successful history. Taxing "bads" (such as pollution) is preferable to taxing "goods" (such as income and capital investment) for a variety of reasons, and the revenue stream from taxing the "bads" can be used to reduce the taxes on "goods",21 to reduce the burdens on hard-hit subpopulations (such as coal workers), and to finance research, development, and demonstration of better low-emission technologies. The money does not disappear into a black hole.

Serious advocacy of a carbon tax has been anathema in American political discourse, but it''s far from obvious that the persuasive power of the Presidency would not be enough to sell such a tax to the public. One does not have to leap to the levels of $100 or $200 per tonne of emitted carbon that feature in scare stories about how damaging this approach would be to the fossil-fuel industries; getting our toes wet with a tax of $20 per tonne, as a beginning, would generate a healthy set of incentives for energy firms and energy users to start making more climate-friendly choices, and it would raise about $30 billion per year initially in the United States - of which, say, a tenth could be used to alleviate resulting burdens on the groups hardest hit, a tenth could be used for additional targeted incentives for the adoption of low-carbon energy options from the existing mix, a tenth could be used for more than doubling Federal support for research, development, and demonstration of improved low-carbon options, and 70 percent ($21 billion) would still be left for reducing other taxes.22

As economists like to point out, an effect on the energy marketplace substantially identical to that of a carbon tax can be obtained through the use of an emissions cap implemented through tradeable emissions permits. It is often supposed that this approach would be less problematic politically than a carbon tax, and this may be right. But it is harder to design, harder to calibrate, and harder to implement. For the United States, initially, I believe a carbon tax would be a better bet. Ultimately, however, an international cap-and-trade scheme may be the best negotiable approach for constraining carbon emissions worldwide in an equitable way (about which more below).

Politically easier than either carbon taxes or emissions caps, of course, are the increases in research, development, and demonstration of advanced energy-supply and energy-end-use technologies recommended as part (4) of my six-part program. But even this proved problematic in the last administration. The 1997 report of the President''s Committee of Advisors on Science and Technology (PCAST) on Federal Energy Research and Development for the Challenges of the Twenty First Century concluded that the Federal energy-technology R&D programs then in place were "not commensurate in scope and scale with the energy challenges and opportunities that the twenty-first century will present", taking into account "the contributions to energy R&D that can reasonably be expected to be made by the private sector under market conditions similar to today''s"; and the panel recommended modifications to DOE''s applied energy-technology (fossil, nuclear, renewable, efficiency) R&D programs that would increase funding in these categories from their FY1997 and FY1998 level of $1.3 billion per year to $1.8 billion in FY 1999 and $2.4 billion in FY 2003.23

The administration embodied a considerable fraction of this advice in its FY1999 budget request (which contained a total increment about two-thirds of what PCAST recommended for that year) and the Congress appropriated a considerable fraction of that (about 60 percent of the increment requested by the administration). The net result was an increment about 40 percent as large as PCAST recommended for FY1999. In subsequent budgets, the gap between the PCAST recommendations and what the administration was willing to recommend widened steadily, and Congress continued to appropriate only a fraction of what the administration recommended. This should not have been so hard. As PCAST pointed out, its proposed increases in Federal energy R&D would not only have positioned the country to respond more cost effectively to the need to reduce greenhouse-gas emissions when and if a national decision were made to do this; it also would end up lowering the monetary costs of energy and energy services below what they would otherwise be, increasing the productivity and competitiveness of U.S. manufacturing, reducing U.S. overdependence on oil imports, and reducing emissions of air pollutants directly hazardous to human health and ecosystems, among other benefits. And it would only have restored Federal spending on applied-energy-technology R&D, by FY2003, to its level in the FY1991 and FY1992 Bush Administration budgets (the annual total for which could be raised by an increase of 2.5 cents per gallon in the Federal gasoline tax).

Another PCAST study, this one completed in June 1999, fleshed out the argments for and ingredients of the fifth element of the six-point program I recommend here:

5. increased international cooperation to facilitate the application of the results of components (1), (3), and (4) in the developing South as well as throughout the industrialized North.
The report from that study, entitled Powerful Partnerships: The Federal Role in International Cooperation on Energy Innovation, noted that enhanced U.S. participation in such cooperation would improve the access of U.S. firms to the immense foreign market for energy technologies,24 lower the cost of energy-technology innovation for U.S. domestic application, and help other countries participate effectively in the solution of global energy problems that the United States cannot solve by itself. (The energy technologies that other countries deploy will largely determine not only pace of global climate disruption by fossil-fuel-derived greenhouse gases,25 but also the extent of world dependence on imported oil and the potential for conflict over access to it, the performance of nuclear-energy systems on whose proliferation resistance and safety the whole world depends, and the prospects for trade-enhancing and security-building sustainable economic development in regions where, otherwise, economic deprivation will be a continuing source of conflict.)

This 1999 PCAST study estimated that Federal spending on international cooperation in energy research, development, demonstration, and deployment (ERD3) amounted in FY1997 to about $250 million per year, and it recommended that this figure be increased to about $500 million in the FY2001 budget and to $750 million by FY2005. The increments were for specific initiatives for strengthening the foundations of energy-technology innovation and international cooperation relating to it (including capacity building, energy-sector reform, and mechanisms for demonstration, cost-buy-down, and financing of advanced technologies); for increased cooperation on RD3 of technologies governing the efficiency of energy use in buildings, energy-intensive industries, and small vehicles and buses, as well as of cogeneration of heat and power; and for increased cooperation on RD3 of fossil-fuels-decarbonization and carbon-sequestration technologies, biomass-energy and other renewable-energy technologies, and nuclear fission and fusion. The administration''s FY2001 budget request included an increment of $100 million for these initiatives (as opposed to the $250 million increment proposed by PCAST); at this writing, the fate of this increment in an election-year Congress controlled by the other party is unclear.

But addressing the energy/climate challenge should not be a partisan issue. The values at stake - economic prosperity, environmental quality, international security - are held dear by both parties. The UN Framework Convention on Climate Change(UNFCCC) was signed by a Republican President. And that convention, which unlike the Kyoto Protocol has long since been ratified by the U.S. Senate and is therefore the law of the land, already commits the United States to most of the climate-related actions that climate-change skeptics in the 106th Congress have mistakenly associated with Kyoto and noisily opposed. For example, Article 4 of the UNFCCC commits the parties to "formulate, implement, publish, and regularly update national and, where appropriate regional programmes containing measures to mitigate climate change by addressing anthropogenic emissions by sources and removals by sinks of all greenhouse gases not controlled by the Montreal Protocol, and measures to facilitate adequate adaptation to climate change", as well as to "promote and cooperate in the development, application and diffusion, including transfer, of technologies, practices, and processes that control, reduce, or prevent anthropogenic emissions of greenhouse gases...". This covers a lot of ground - and provides a lot of cover for doing what is required.

Both the 1992 UNFCCC and the unratified 1997 Kyoto Protocol represent early, halting, imperfect steps in the effort to achieve what I list as the sixth element of my six-part program of action on the energy/climate challenge, namely

6. development of a global framework of commitments to long-term restraints on greenhouse-gas emissions designed for sufficiency, equity, and feasibility.

The UNFCCC correctly recognized the asymmetries built into the energy/climate challenge - notably that it has been the industrialized countries who mostly consumed, in the course of their economic development, the capacity of the atmosphere to hold anthropogenic CO2 without entraining intolerable changes in climate; that the industrial countries are far better positioned financially and technologically to undertake early corrective action; and that no approach to planetary emissions limits that closed off the path to development for three quarters of the Earth''s population would be acceptable. Article 3 explicitly affirms, accordingly, that "the developed country Parties should take the lead in combating climate change and the adverse effects thereof." This is only sensible.

The Kyoto negotiation attempted, with insufficient time and insufficient preparation in relation to the complexity of the agenda, both to address a variety of gaps and ambiguities in the UNFCCC''s treatment of the coverage and approach of a global framework for limiting anthropogenic climate change and to agree on an initial set of binding numerical targets and timetables for emissions reductions by the industrialized countries. The biggest shortcoming of the negotiation, in my view, was the degree of preoccupation in the meeting - and in the preparations for it in individual industrialized-country governments - with these numerical targets and timetables, to the near exclusion of addressing the mechanisms (above all, incentives) that might start to move emissions trajectories in the right direction.

The result was a set of targets and timetables for industrialized countries - expressed in terms of percentage reductions from 1990 levels to be achieved in the 2008-2012 time period - that has been assailed both for requiring more than is needed in the short run and for requiring much less than is needed in the long run. It has also been assailed for failing to bind the developing countries. The criticisms about too much and too little have some validity, but the bigger failure is that arguments and agreements about targets and timetables are essentially irrelevant in the absence of mechanisms that might cause them to be achieved.

The idea that the developing countries could have been or should have been included in reductions targets of this character (percentage cuts from 1990 levels) was a nonstarter from the outset - inconsistent with the principles of the UNFCCC and not taken seriously by much of anybody outside the U.S. Senate. When it is time to bring the developing countries into a framework of commitments to reductions (and I believe this will only happen after the industrialized nations have demonstrated a willingness not just to establish targets but to impose mechanisms to make the targets attainable), the formula will need to be based either on carbon intensities (agreement to reduce the ratio of carbon emissions to GDP at a specified rate) or on tradeable emissions permits allocated on a per capita basis. (The idea that the industrialized countries are entitled, permanently, to far higher per capita emissions of greenhouse gases than developing countries are is simply never going to fly...and for good reason.)

A satisfactory global framework for emissions restraints might well employ the two approaches just mentioned in successive stages: commitments based initially on specified annual percentage reductions in the carbon intensity of economic activity, transitioning in the longer term to evolving global emissions caps implemented through tradeable permits. If "sufficiency" in a framework of emissions restraints were defined, as I believe it should be, in terms of getting the world onto a trajectory that would stabilize the atmospheric CO2 concentration at between 450 and 500 ppmv, then the initial commitments for reducing the carbon to GDP ratio might start in the range of 1.5%/yr (not too far above the long-term historical average) and ramp up over a decade or so to the range of 2.5%/yr that would be required, as a century-long worldwide average, to achieve stabilization at the indicated level.

The later phase, employing caps, would be based on the insight that the desired stabilization trajectory cannot have a peak higher than about 10 GtC/yr around 2035 and must be falling thereafter. If one supposes that world population in 2035 could be held to 8 billion persons (somewhat below the 8.8 billion projected for 2035 in our BAU scenario), the per- capita allocation in 2035 would need to be about 1.2 tonnes of carbon per person. This happens to be about 3 times less than what the industrialized nations were averaging at the end of the 1990s, and 3 times more than what the developing countries were averaging then. So, in this strikingly symmetric (and equitable) scheme, the per-capita allowances of the industrialized and developing countries would have converged from opposite sides, after 35 years, to the geometric mean of their current per-capita emissions.26 The emissions cap and the associated per-capita allocations would fall gradually thereafter, tracking the trajectory needed to achieve stabilization at 450-500 ppmv.

Such an approach would certainly be equitable. It is more likely to be sufficient than variants aiming for stabilization at 550 ppmv or above (although worse-than-expected evolution of climate change over time could still show it to be inadequate). And I believe it is feasible, at least from the technical and economic standpoints even if not yet, today, from the political one. But, crucially, it does not need to be politically feasible today, because its most politically problematic ingredient - equal per capita emissions allocations - would not really need to begin being phased in before 2015 or 2020, by which time people''s everyday experience of the impacts of climate change is likely to have stretched considerably the scope of what domestic and international politics will allow.

As for the Kyoto Protocol itself: it is, with all its warts, sufficiently important today as a symbol of the world''s commitment to move forward collectively to address the energy/climate challenge that a serious effort must be made to either salvage or supplant it. The most important ambiguities in it - relating, for example, to the treatment of carbon sinks and to the operation of the Clean Development Mechanism - have been in the process of being ironed out in Conferences of the Parties (to the UNFCCC) subsequent to the Kyoto meeting. As for the "binding" targets and timetables, these might be made acceptable by designing a set of agreed penalties for noncompliance that are more constructive than punitive. (The industrialized countries could agree, for example, to increase their investments in RD3 and international cooperation on low-carbon-emitting energy technologies in proportion to the margin by which they miss their 2008-2012 targets.) If the Kyoto protocol itself proves not to be salvageable in these ways, it will be important to have a new and better agreement that the major emitters in North and South alike are prepared to sign at the same meeting when the final demise of the Kyoto agreement is formally acknowledged.

Conclusion

The energy/climate challenge must be met. And it can be met. There is no shortage of persuasive professional knowledge about why doing so is necessary, nor any shortage of promising proposals about how to proceed. There is not even a good argument that doing this job would be too expensive: the cost of the needed steps almost certainly would be small compared to the cost of the environmental and economic damages averted, as well as small compared to investments the society makes in military forces (where the degree of certainty about the magnitude of the threat - and about the cost-effectiveness of the proposed investments against it - is actually considerably smaller than in the energy/climate case). What has mainly been missing is simply the public understanding and the political conviction that this is a problem to which the nation and the world must give high priority, now. Repairing that deficit is a matter of political leadership, and no one is in as good a position to provide it as the new President of the United States.

* * * * *

ENDNOTES

1 A former member of the Council of Economic Advisors asserted to the contrary, in the early 1990s, that the U.S. economy cannot be very vulnerable to climate change because only 3 percent of our GNP depends on the environment. He apparently was referring to the share of GNP generated by agriculture, forestry, and fishing. But, in a fundamental sense, it all depends on the environment. The economist''s assertion is akin to claiming that human beings shouldn''t worry about heart attacks, insofar as the heart constitutes only 2 percent of the mass of the body.

2 World use of primary energy forms in 1998 amounted to 440 exajoules (1 EJ = 1018 J = 0.95 quadrillion Btu = 22 million tonnes of oil = 33 million tonnes of UN standard coal). Of this total, 35% came from oil, 23% from coal, and 20% from natural gas - a total of 78% from fossil fuels. Nonfossil contributions were principally 13% from biomass fuels (fuelwood, charcoal, crop wastes, dung, and biomass-derived alcohol), 6% from nuclear energy, and 2% from hydropower. Renewable sources other than biomass and hydropower - notably geothermal, wind, and solar energy - contributed altogether less than 0.5 percent. In the United States, the 1998 primary energy supply of 100 EJ was derived 38% from oil, 24% each from coal and natural gas, 8.2% from nuclear energy, 3.8% from biomass, 1.2% from hydropower, and 0.4% from other renewables.
3 The scenario described here closely resembles scenario IS92a from the 1995 Second Assessment of the Intergovernmental Panel on Climate Change, which was a middle-of-the-road scenario in that study and has been widely employed as a reference case in energy/climate discussions since. Scenarios are not predictions; they serve only to illustrate the consequences of the stated assumptions about the evolution of the contributing factors.
4 Atmospheric particulate matter cools the surface of the Earth in some circumstances and warms it in others. The uncertainties associated with these quite complicated phenomena are considerable, and the investigation of the details - including the prevalence of particulate-warming circumstances versus that of particulate-cooling circumstances - is a vigorous focus of current research. A global-average net cooling effect of the magnitude indicated here was the best estimate of the 1995 assessment of the Intergovernmental Panel on Climate Change.
5 This approach tends if anything to underestimate the future warming to be expected, insofar as the most likely deviation from its assumptions is a more rapid decline in particulate concentrations than assumed, as a result of more aggressive programs to control emissions of particulate matter and of its gaseous precursors (mainly oxides of sulfur and nitrogen). The dilemma here is that better control of "conventional" pollutants makes the problem of greenhouse-gas-induced climate change worse!
6 By this I mean the debate among respected analysts. There will always remain a few credentialed skeptics, as there are on such questions as whether cigarette smoking causes lung cancer. This is in the nature of science and the distribution of human characteristics among those who practice it. But the very small and continuously diminishing possibility that the dissidents are right cannot be given much weight in prudent public policy.7 This means the dissidents have two very difficult questions to answer: First, what is the alternative culprit that would account for the observed pattern of effects? Second, how can it be that the measured increase in greenhouse-gas concentrations is not causing the pattern of effects that climate science predicts for it (since, by the dissidents'' postulate, something else is causing this)?

8 It is true, as climate-change skeptics are fond of pointing out, that there is considerable natural variability in global climate (arising, for example, from variations in solar output, cycles in the Earth''s orbital parameters, and "internal" climate-system oscillations involving the interactions of ice, oceans, and atmosphere). But the size of the climate-change "signal" that has emerged over the last few decades from the "noise" of natural variability is large, and, as indicated above and in Note 7, it fits too well the pattern expected from greenhouse-gas-induced warming to be plausibly attributed to a hitherto unsuspected and undetected natural cause.
9 These figures take the realized temperature increase from pre-industrial times to 1990 to be 0.5ºC. The increase projected between 1990 and 2070 under the BAU scenario is 1.3ºC and that between 1990 and 2100 is 2.0ºC.
10 The seeming paradox of having more droughts as well as more floods is explained by the fact that, while a warmed world generates more precipitation with a larger part of it concentrated in extreme events, it also produces greater evaporation and hence more rapid drying out of the soil between precipitation events.
11 The agricultural projections in the IPCC assessment and subsequent ones, such as the recently released draft U.S. National Assessment of Climate-Change Impacts on the United States, are based on analysis of the effects of temperature, moisture, and increased CO2 (which is a plant nutrient) only. They do not account for ecological effects, such as from changes of conditions favorable to plant pests and pathogens. When these ecological factors are taken into account, it seems likely that the net assessment of climate-change effects on agriculture will change from "better in some places, worse in others - balancing out on average" to "more harm than good".
12 The focus of the climate-science community and its literature on the consequences of a doubling, which resulted from an agreement to standardize analyses for the purpose of comparison rather than from any conviction that no more than a doubling is likely, seems to have deflected attention from the full implications of business as usual, which would carry the world past any realistic possibility of avoiding a tripling if it persisted through 2075, and past any realistic possibility of avoiding a quadrupling if it persisted through 2100.
13 The heat index combines temperature and humidity into a single measure of discomfort in hot weather. The figures given are 24-hour, 30-day monthly averages. The best way to make these numbers meaningful is to associate what you know to be the current Washington DC climate in July with the figure 86 and then imagine what 109 would mean.
14 The Princeton group only calculated the sea-level rise from thermal expansion, considering that the then-current models were not capable of a credible calculation of the contribution from melting from the Greenland and Antarctic ice sheets.
15 This could also be achieved with other combinations of energy-intensity and carbon-intensity reductions, as long as the century-average reduction rates for the two ratios add up to 2.6%/yr. Thus, instead of a 2.0%/yr decline in energy intensity and a 0.6%/yr decline in carbon intensity, one could do the same job with an energy intensity decline averaging 1.6%/yr and a carbon intensity decline averaging 1.0%/yr.
16 To pick but one item on this list for brief elaboration, the vulnerability of air travel to increased summer storminess (a likely consequence of warming and the more vigorous hydrologic cycle that goes with it) should be apparent to anyone who attempted much flying in the summer of 2000.
17 Ending deforestation and improving other land-management practices so as to reduce greenhouse-gas emissions from the managed lands are, of course, forms of "emissions control" rather than means of removing greenhouse gases that are already in the atmosphere. But the effect is the same, and it is customary for obvious reasons to treat all of the land-use approaches to carbon management together.
18 It is conceivable that another 10 percent could be achieved by fertilizing the oceans to increase the amount of carbon stored in plant material there. But the biology involved is much less well understood than for terrestrial ecosystems, and the chance of unanticipated and unwanted side effects correspondingly higher.
19 The details of the research programs that need to be pursued under these headings are abundantly spelled out in reports generated over the past few years by the Intergovernmental Panel on Climate Change, the U.S. Global Change Research Program, and the U.S. National Academies complex, among others. 20 The merits and demerits of many of these approaches were treated in the two studies of energy strategy conducted by the President''s Committee of Advisors on Science and Technology in the second Clinton term - Federal Energy Research and Development for the Challenges of the Twenty First Century (November 1997) and Powerful Partnerships: The Federal Role in International Cooperation on Energy Innovation (June 1999) - both available from the Office of Science and Technology Policy in the Executive Office of the President.

21 At least one well regarded economic model, that of Dale Jorgenson of Harvard University, indicates that a revenue-neutral carbon tax,, wherein the revenues were used to reduce income and capital-gains taxes, would lead in the United States not only to a reduction in CO2 emissions below business as usual but to an increase in GDP compared to the reference (no carbon tax) case. (This means that the benefits to the economy from reduced income and capital-gains taxes outweighed the damage to the economy from increased energy costs.)
22 A carbon tax of $20 per tonne would add 5.5 cents per gallon to the price of gasoline (an increase of about 3% on a market price of $1.70 per gallon for regular unleaded); 32 cents per million Btu to the price of natural gas (an increase of about 5% on a market price of $6 per million Btu for residential gas); and $11 per short ton to the price of electric-utility coal (an increase of 28 percent on a market price of $40 per short ton). The increment on the price of natural gas would add 0.2 cents per kilowatt hour to the 3.5 cents per kWh cost of electricity generation using natural gas in a combined-cycle power plant operating at 50% thermal-to-electric efficiency, and the increment on the price of utility coal would add 0.6 cents/kWh to the 6 cents/kWh cost of electricity generation with a conventional pulverized-coal power plant operating at 36% efficiency. But the tenth of the $30 billion/yr total revenue from this tax that would be allocated, by hypothesis, to additional incentives for low-carbon energy choices would be five times larger than the $600 million/yr in such incentives proposed in the Clinton Administration''s initial Climate Change Technologies Initiative.
23 The principal recommended increases were (in descending order) in efficiency, renewable, and nuclear (fusion and fission) technologies; recommended initiatives in the fossil category, while they included increases for work on fuel cells, advanced coal technologies, and carbon capture/sequestration technologies, were largely offset by recommended phase-outs. The bipartisan panel that produced these recommendations contained experts in the full range of energy sources - and from the private sector as well as from the academic and NGO communities - and its conclusions were all unanimous.
24 Most of the multi-hundred-billion dollar per year energy-supply-technology market is outside the United States - and increasingly in developing countries - and this will become even more true as the century wears on.
25 The United States was contributing about a quarter of worldwide fossil-CO2 emissions at the end of the twentieth century, while other industrialized countries contributed about half and the developing countries the remaining quarter. In a BAU energy future, however, the developing countries will become equal to the industrialized ones as emitters of fossil-fuel-derived carbon dioxide by around 2030, and will increasingly dominate global CO2 emissions thereafter.26 This does not mean, of course, that per-capita emissions in all of the industrialized countries would necessarily have fallen this far by 2035, or that per-capita emissions in all of the developing countries would necessarily have risen this far; but, under a cap-and-trade scheme, the industrialized countries that had not gotten this low would need to buy emissions permits from the developing countries that had not gotten this high.

Reprinted with Permission.