7 Global Environmental Sustainability: Protecting the Commons

International Monetary Fund
Published Date:
April 2008
  • ShareShare
Show Summary Details

Sustainable management of global environmental resources—the earth’s climate, ocean fisheries, and biodiversity—is essential to achieving the Millennium Development Goals (MDGs) and, indeed, to ensuring continued economic progress over the next century. Failure to mitigate the impact of greenhouse gas (GHG) emissions on the earth’s climate may lead to disastrous changes in temperature and precipitation and to an increase in extreme weather events. Pollution and overexploitation of marine fisheries can damage or destroy fish populations. Habitat destruction may lead to species extinction. All three global environmental problems, and how the world deals with them, will affect the welfare of the developing world.

The goal of this chapter is to monitor recent progress in dealing with each of the three global environmental problems, with an emphasis on climate change. The chapter begins by describing temperature trends; the relationship between GHG concentrations and climate; and projections, and effects, of future climate change in the absence of any mitigation efforts. The chapter then discusses the sources of and trends in GHG emissions and the opportunities for adapting to changes in climate. Progress in international efforts to develop institutions and policies to deal with climate change is reviewed, and the chapter ends with a review of trends in biodiversity and the health of marine fisheries.1

Climate Change: The Impact of Human Activity on Climate

Deforestation and the burning of fossil fuels produce greenhouse gases that trap incoming solar radiation, leading to a rise in global average surface temperature. Measurements show that the average world temperature has increased since the start of the industrial revolution in the mid-1800s; over the last hundred years, the average temperature has risen 0.74°C.2 Indeed, eleven of the last twelve years rank among the warmest years on record since 1850. Rising sea levels are consistent with warming. Since 1961 global sea levels have risen at an average rate of 1.8 millimeters (mm) a year and since 1993 at an average rate of 3.1 mm a year.3 At the same time snow cover has decreased, and ice fields in the Arctic and Antarctic have shrunk drastically. Average temperatures in the Arctic are rising twice as fast as elsewhere in the world. The polar ice cap as a whole is shrinking: satellites show that the area of permanent ice cover is contracting at a rate of 9 percent each decade. If this melting continues, summers in the Arctic could become nearly ice-free by the end of the century.

More important, scientific research suggests that human activities are contributing to the rise in global temperatures. The concentration of carbon dioxide (CO2) in the atmosphere—the most important GHG—has increased from approximately 277 parts per million volume (ppm) in 1744 to 384 ppm in 2007.4 Models of the determinants of temperature change that take into account the addition of GHGs into the atmosphere from human activities provide much more accurate estimates of historical trends in temperature than do models that ignore these emissions.5

Relationship of GHG Concentrations to Climate Change

The extent of future climate change depends on future GHG emissions and on the relationship between climate and the stock of GHGs in the atmosphere. Table 7.1 shows the likelihood of various changes in mean global surface temperature (relative to levels before the industrial revolution) corresponding to various equilibrium concentrations of GHGs.6 In 2005 the concentration of all GHGs was approximately 375 ppm CO2e (carbon dioxide equivalents).7 Stabilization at 450 ppm CO2e, as advocated by the UN’s Human Development Report, would still carry a risk of an increase in mean surface temperature of at least 3°C.8 Equilibrium GHG concentrations of 650 or 750 ppm CO2e, which are consistent with some of the nonmitigation scenarios in the IPCC (Intergovernmental Panel on Climate Change) Fourth Assessment Report, carry a significant risk of an increase in mean global surface temperature of 5°C.9

Table 7.1Likelihood of various CO2e concentrations exceeding various increases in global mean surface temperaturepercent
Stabilization level (in ppm CO2e)2°C3°C4°C5°C6°C7°C
Source: Stern 2008.
Source: Stern 2008.

A mean increase in global surface temperature of 5°C would result in disastrous consequences: heat waves throughout the world, increases in heavy precipitation in northern latitudes, and drought and decreases in precipitation in most subtropical regions. It would likely lead to the melting of snowpack in the Himalayas and risk the total disappearance of the West Antarctic ice sheet, which could increase the global sea level by six meters. It would also risk “tipping points”—positive feedbacks that would cause atmospheric GHG concentrations and temperature to rise rapidly. These feedbacks include the release of methane from permafrost as warming occurs, the release of carbon from deep oceans as climate change affects deep-sea circulation, and the increased absorption of solar radiation as polar ice caps melt. Any of these effects could lead to truly catastrophic climate changes.10

The Geographic and Temporal Dimensions of Climate Change

How likely are GHG concentrations to reach 650 or 750 ppm, and how fast might this occur? The IPCC in its Fourth Assessment Report estimates the change in the stock of GHGs under various nonmitigation emissions scenarios, together with the corresponding changes in temperature and sea level rise worldwide (table 7.2).11

Table 7.2Changes in mean global temperature and sea level associated with various IPCC scenarios
Temperature change

(°C at 2090–99 relative to 1980–1999)a
Sea level rise

(meters at 2090–99 relative to 1980–99)
CasebBest estimateLikely rangeModel-based range excluding future

rapid dynamic changes in ice flow
Constant year 2000 concentrationsc0.60.3–0.9Not applicable
B1 scenario1.81.1–2.90.18–0.38
A1T scenario2.41.4–3.80.20–0.45
B2 scenario2.41.4–3.80.20–0.43
A1B scenario2.81.7–4.40.21–0.48
A2 scenario3.42.0–5.40.23–0.51
A1FI scenario4.02.4–6.40.26–0.59
Source: Summary for Policy Makers, Fourth Assessment Report, IPCC 2007b.

These estimates are assessed from a hierarchy of models that encompass a simple climate model, several earth system models of intermediate complexity, and a large number of atmosphere-ocean general circulation models (AOGCMs).

The six main scenarios for the projections are described as follows:

  • B1: Convergent world; low population growth; change toward a service and information economy, clean technologies.
  • B2: Regional focus; intermediate population growth; development and technical change; environmental emphasis.
  • A1: Convergent world; population peaks at mid-century; rapid growth and introduction of more efficient technologies that are sourced from either:
  • A1T: Nonfossil energy sources
  • A1B: A balance across all sources
  • A1FI: Fossil-intensive
  • A2: Heterogeneous world; high population growth; slower economic growth and technical change.

Year 2000 constant composition is derived from AOGCMs only.

Source: Summary for Policy Makers, Fourth Assessment Report, IPCC 2007b.

These estimates are assessed from a hierarchy of models that encompass a simple climate model, several earth system models of intermediate complexity, and a large number of atmosphere-ocean general circulation models (AOGCMs).

The six main scenarios for the projections are described as follows:

  • B1: Convergent world; low population growth; change toward a service and information economy, clean technologies.
  • B2: Regional focus; intermediate population growth; development and technical change; environmental emphasis.
  • A1: Convergent world; population peaks at mid-century; rapid growth and introduction of more efficient technologies that are sourced from either:
  • A1T: Nonfossil energy sources
  • A1B: A balance across all sources
  • A1FI: Fossil-intensive
  • A2: Heterogeneous world; high population growth; slower economic growth and technical change.

Year 2000 constant composition is derived from AOGCMs only.

Figure 7.1 shows the geographic distribution of temperature changes for three non-mitigation scenarios: B01, a scenario that results in an increase in mean global temperature of 1.8°C in 2090 (relative to 1980–99 temperatures);12 A1B, a scenario that results in an increase in mean global temperature of 3.3°C in 2090; and A2, which results in an increase in mean global temperature of 3.9°C in 2090. The global distribution of temperature changes (figure 7.1) is roughly the same for all three scenarios: temperature increases are greatest in the northern latitudes, but in scenarios A1B and A2, they rise above 4°C in parts of Latin American and Sub-Saharan Africa, as well as in India and the Middle East.

figure 7.1Projections of surface temperatures for three IPCC scenarios

Source: IPCC 2007a.

Other effects are likely to accompany these temperature changes. Arid and semiarid regions will become drier, while areas in the mid-to-high latitudes will become wetter. Heavy precipitation events are very likely to occur in mid-to-high latitudes, while the likelihood of droughts will increase in areas that are currently dry. Storm surges, cyclones, and hurricanes are also likely to increase in frequency throughout the world. Water supplies are likely to be affected: the melting of glaciers will lead to higher springtime water flows and reduced summertime flows. The majority of the negative effects of climate change are likely to occur in lower latitudes—in the South, rather than the North—implying that developing countries will bear the brunt of these effects.

Climate change is often viewed as a problem for the future, but figure 7.1 suggests otherwise. As the first set of maps indicates, significant temperature changes in Africa and Latin America are likely as early as 2020–29 under the A1B nonmitigation scenario—a scenario of rapid economic and population growth in which the world relies on a combination of fossil fuels and renewable energy sources. More important, avoiding the risk of large temperature changes in 2090–99 requires action now. As the IPCC noted, world GHG emissions would have to decline by 50 to 85 percent of their 2000 levels by 2050 to stabilize concentrations at 450 ppm, depending on the mitigation path chosen.13 World GHG emissions would have to decrease by as much as 30 percent from 2000 levels by 2050 (depending on the mitigation path chosen) to stabilize concentrations at 550 ppm.

The Impacts of Climate Change and opportunities for Adaptation

What impacts would the temperature changes in figure 7.1 have on the economies of developing and developed countries? Table 7.3 describes in qualitative terms some of the likely impacts of climate change on agriculture, forestry and ecosystems, water resources, human health, and human settlements that are expected to occur under the nonmitigation scenarios in figure 7.1. The magnitude of these effects depends on the extent to which countries adapt to them and also on the extent to which mitigation efforts lower GHG emissions. The effects of climate change vary greatly among developing countries—even for countries in the same region. Efforts to adapt to climate change must therefore be tailored to specific country needs.

Table 7.3Possible impacts of climate change in the mid-to-late-21st century
Examples of major projected impacts by sector
Phenomenona and direction of trendLikelihood of future trendsbAgriculture, forestry and ecosystemsWater resourcesHuman healthIndustry, settlement, and society
Over most land areas, warmer and fewer cold days and nights, warmer and more frequent hot days and nightsVirtually certaincIncreased yields in colder environments; decreased yields in warmer environments; increased insect outbreaksEffects on water resources relying on snow melt; effects on some water suppliesReduced human mortality from decreased cold exposureReduced energy demand for heating; increased demand for cooling; declining air quality in cities; reduced disruption to transport from snow, ice; effects on winter tourism
Warm spells/heat waves. Frequency increases over most land areasVery likelyReduced yields in warmer regions from heat stress; increased danger of wildfireIncreased water demand; water quality problems, such as algal bloomsIncreased risk of heat-related mortality, especially for the elderly, chronically sick, very young, and socially isolatedReduction in quality of life for people in warm areas without appropriate housing; impacts on the elderly, very young, and poor
Heavy precipitation events. Frequency increases over most areasVery likelyDamage to crops; soil erosion, inability to cultivate land because of waterlogged soilsAdverse effects on quality of surface and groundwater; contamination of water supply; water scarcity may be relievedIncreased risk of deaths, injuries, and infectious, respiratory and skin diseasesDisruption of settlements, commerce, transport, and societies from flooding; pressures on urban and rural infrastructures; loss of property
Area affected by drought increasesLikelyLand degradation; lower yields, crop damage, and failure; increased livestock deaths; increased risk of wildfireMore widespread water stressIncreased risk of food and water shortages; increased risk of malnutrition; increased risk of water- and food borne diseasesWater shortages for settlements, industry, and societies; reduced hydropower generation potentials; potential for population migration
Intense tropical cyclone activity increasesLikelyDamage to crops; windthrow (uprooting) of trees; damage to coral reefsPower outages causing disruption of public water supplyIncreased risk of deaths, injuries, water- and food- borne diseases; post-traumatic stress disordersDisruption by flood and high winds; withdrawal of private risk insurance coverage in vulnerable areas; potential for population migrations; loss of property
Increased incidence of extreme high sea level (excludes tsunamis)dLikelyeSalinization of irrigation water, estuaries, and freshwater systemsDecreased freshwater availability from saltwater intrusionIncreased risk of deaths and injuries by drowning in floods; migration- related health effectsCosts of coastal protection versus costs of land-use relocation; potential for movement of populations and infrastructure; also see effects of tropical cyclones above
Source: IPCC 2007b.

See Working Group I Fourth Assessment table 3.7 for further details regarding definitions.

Based on projections for 21st century using scenarios in table 7.2.

Warming of the most extreme days and nights each year.

Extreme high sea level depends on average sea level and on regional weather systems. It is defined as the highest 1 percent of hourly values of observed sea level at a station for a given reference period.

In all scenarios, the projected global average sea level in 2100 is higher than in the reference period. The effect of changes in regional weather systems on sea level extremes has not been assessed.

Source: IPCC 2007b.

See Working Group I Fourth Assessment table 3.7 for further details regarding definitions.

Based on projections for 21st century using scenarios in table 7.2.

Warming of the most extreme days and nights each year.

Extreme high sea level depends on average sea level and on regional weather systems. It is defined as the highest 1 percent of hourly values of observed sea level at a station for a given reference period.

In all scenarios, the projected global average sea level in 2100 is higher than in the reference period. The effect of changes in regional weather systems on sea level extremes has not been assessed.

Impact on Agriculture

There is wide recognition that developing countries in general stand to lose more from the effects of climate change on agriculture than developed countries. Although figure 7.1 suggests that temperatures will rise more in northern than in southern latitudes, temperatures in developing countries are already close to thresholds beyond which further increases in temperature will lower productivity. Developing countries are also likely to have fewer opportunities for adaptation. Moreover, the losses in yields that occur in developing countries are likely to affect a larger number of people—especially the poor—because of the greater importance of agriculture in the livelihoods of people in developing countries.14

Cline presents estimates of the impact on agriculture of a 4.4°C increase in mean global temperature and a 2.9 percent mean increase in precipitation occurring during the period 2070–99.15 His estimates combine results from the two main strands of the literature—cross-sectional studies of land values or net revenues (the Ricardian approach) and crop models. The estimates of impacts on yields shown in figure 7.2 incorporate carbon fertilization effects—that is, they allow for the fact that increased carbon in the atmosphere will increase yields by promoting photosynthesis and reducing plant water loss.16 As figure 7.2 clearly shows, the largest agricultural losses will occur in parts of Africa, in South Asia, and in parts of Latin America. In contrast, the United States and Canada, Europe, and China will, in general, benefit under the nonmitigation climate scenario.

figure 7.2Impacts of increases in temperature and precipitation on agricultural yields, 2079–99

Source: Cline 2007.

Why do the estimated impacts differ significantly across countries in Africa and Latin America? The answer in part lies in adaptation: yields on irrigated farmland decrease less than on rain-fed land; in some areas, yields increase. In Africa the value of output per hectare declines less for farmers who can substitute livestock for crops. Two points about adaptation should be noted, however. One is that the Ricardian approach, which allows farmers in different climatic zones to adapt to climate, assumes that prices in the future will remain unchanged. If water shortages increase the price of irrigation, yields may fall more than indicated in figure 7.2. Second, it is the impact of climate change on net revenues that should be measured rather than the impact on yields. Adaptation is costly, and the impact of climate change should be measured as the sum of damages after adaptation plus the costs of adaptation. As Cline notes, output in southwest India falls by approximately 37 percent under the nonmitigation climate scenario, but net revenues fall by 55 percent.17

Impacts on Health

Climate change may affect human health both directly and indirectly. Increased warming in cold climates may reduce cardiovascular and respiratory deaths, but heat waves in both warm and cold climates are likely to increase cardiovascular deaths. Changes in temperature and precipitation also affect diarrheal disease—the second-leading cause of death among children between one and five years. Extreme weather events—hurricanes, floods, and tornadoes—are likely to raise accidental deaths and injuries. Equally important to the poor in developing countries are the indirect effects of climate change on health. As figure 7.2 suggests, climate change, through its impact on agricultural yields, may lower food security and lead to malnutrition. Increased temperatures and precipitation in low latitudes may increase the incidence of malaria and other vector-borne diseases.

The largest impacts of climate change on mortality and morbidity occur through malnutrition, diarrhea, and malaria, and the largest effects geographically are felt in Sub-Saharan Africa, South Asia, and the Middle East. Simply put, the health burden of climate change is borne by the children of the developing world. Table 7.4 shows estimated disability-adjusted life years (DALYs) attributable to climate change in 2000; figure 7.3 shows the distribution of deaths. Climate change in 2000 is associated, worldwide, with 166,000 deaths—77,000 associated with malnutrition, 47,000 with diarrhea, and 27,000 with malaria. The highest number of deaths (per 100,000 persons) occurs in Africa, parts of South Asia (SEAR-D), and the Middle East. The impact of climate change on the United States, Canada, and Europe is negligible, with cardiovascular deaths associated with heat waves cancelling out the benefits of milder winter temperatures.

Table 7.4Estimated DALYs attributed to climate change in 2000, by cause and subregionthousands, unless otherwise indicated
Cause of DALYs
WHO subregionMalnutritionDiarrheaMalariaFloodsAll causesTotal DALYs (per 1 million population)
Source: McMichael and others 2004.

figure 7.3Estimated death rate from climate change in 2000, by WHO subregion

Source: Map created by Center for Sustainability and the Global Environment, University of Wisconsin, using data from McMichael and others 2004.

Note: Change in climate compared to baseline, 1961–90.

The future impacts of climate change are more dramatic than those in 2000. In 2030, assuming that GHG emissions are stabilized at 750 ppm by 2210, the risk of malnutrition is predicted to be 11 percent higher in Latin America than it was in 1990, and 17 percent higher in South Asia (SEAR-D). The risk of diarrhea is predicted to be 6 percent higher in Sub-Saharan Africa and 7 percent higher in South Asia (SEAR-D) than in 1990. It should be emphasized that these increased risks apply to large exposed populations.

These calculations assume little adaptation to climate change—for example, a program that eliminated the anopheles mosquito from Sub-Saharan Africa, or the development of an effective malaria vaccine, would of course reduce malaria risks. A program to improve food security in the region would reduce deaths caused by malnutrition.

Sea Level Rise

Although the mean increases in sea level rise associated with the IPCC nonmitigation scenarios are modest—ranging from 0.2 to 0.5 meters during this century (see table 7.2)—these estimates exclude future rapid dynamic changes in ice flow. Velicogna and Wahr have measured variations in the Antarctic ice sheet during 2002–05.18 Their results indicate that the mass of the West Antarctic ice sheet decreased significantly, at a rate several times greater than assumed by the IPCC in its Third Assessment Report. Climate change could possibly cause the West Antarctic ice sheet to slide into the ocean, which would raise average sea level by approximately five to six meters, even if the ice sheet did not melt.

Measuring the vulnerability of developing countries to rising sea levels—given the current location of settlements—provides a useful starting point for measuring the benefits of adaptation. Dasgupta and others estimate the impact of various possible increases in sea level on 84 coastal developing countries.19 Using Geographic Information System techniques, they estimate the fraction of land area, agricultural land, wetlands, urban land area, population, and GDP that would be affected by increases in sea level of one to five meters. These calculations pertain to current land uses and assume no adaptation.

Figure 7.4 shows the share of various classes of land area, population, and GDP affected by sea level rise, by World Bank region. The impacts of sea level rise are greatest—in virtually all dimensions—in East Asia and the Pacific, followed by the Middle East and North Africa. Effects, however, vary significantly among countries within each region. Table 7.5 shows the ten countries most affected by an increase in sea level of one meter for four dimensions of vulnerability. With no adaptation Vietnam would lose 10 percent of its GDP; the Arab Republic of Egypt, over 6 percent. Egypt would stand to lose 13 percent of its agricultural land (not shown), and Vietnam 28 percent of its wetlands. Twelve percent of the Bahamas would be submerged. As table 7.5 indicates, Vietnam ranks among the top five countries most affected by a one meter rise in sea level; the Bahamas, Egypt, and Suriname also rank among the countries most vulnerable to sea level rise.

figure 7.4Vulnerability to sea level rise

Table 7.5Ten countries most affected by a one meter rise in sea levelpercentage affected
RankPopulationGDPUrban areasWetlands




2Egypt, Arab Rep. of





Egypt, Arab Rep. of

French Guiana








Egypt, Arab Rep. of

The Bahamas

6French Guiana

The Bahamas





United Arab Emirates


8United Arab Emirates

French Guiana



9The Bahamas






The Bahamas

Taiwan, China


Extreme Weather Events

Although regional forecasts of climate change are uncertain, it is likely that weather variability will increase, and with it, extreme weather events. To the extent that future events follow historical patterns, the damages from past weather events—such as droughts, heat waves, and floods—provide an additional index of vulnerability to climate change. Buys and others have compiled a country vulnerability index based on droughts, heat waves, floods, wildfires, and wind storms that occurred between 1960 and 2002.20 The index gives persons killed in these events a weight of 1,000, persons rendered homeless a weight of 10, and persons affected by each event a weight of 1. This sum is divided by population for 1980 (the midpoint of the period) to develop an index of population impact relative to population size.

Table 7.6 presents the index of vulnerability to extreme weather events, showing the 10 most vulnerable countries in each World Bank region. Again, the differences across countries are striking: in per capita terms, Bangladesh is affected more than three times as much as India by extreme weather events—on a par with Ethiopia. Countries in East Asia are—in per capita terms—affected much less than countries in South Asia or Africa, although total damages are high.

Table 7.6Weather damage index (WDI), by country and region
Sub-Saharan AfricaWDIEast Asia & PacificWDILatin America & the CaribbeanWDIMiddle East & North AfricaWDISouth AsiaWDI
Ethiopia1809Tonga698Honduras819Iran, Islamic Rep. of183Bangladesh1940
Mozambique1134Samoa589Antigua Barbados387Jordan32.9India566
Sudan999Laos PDR573Belize385Tunisia29.3Sri Lanka318
Djibouti586Solomon Islands416Haiti254Yemen, Rep. of27.5Pakistan172
Botswana536Philippines392Nicaragua242Syrian Arab Rep.18.4Maldives151
Somalia497Vanuatu340Venezuela, R. B. de215Algeria17.6Nepal84.4
Mauritania433Fiji310St. Lucia212Oman14.5Afghanistan73.5
Malawi411Vietnam235Dominican Republic191Morocco13.3Bhutan64.5

Adaptation to Climate Change

Nobel laureate Tom Schelling has argued that the best way for developing countries to adapt to climate change is to develop.21 In many ways this prescription is correct. Preventing the health impacts of climate change means making progress toward reducing malnutrition, eliminating diarrhea as a leading cause of death among children under five years of age, and eradicating malaria. Achieving MDGs 1, 4, and 6 would constitute effective adaptation to the most adverse health effects of climate change. Development also would reduce the impacts of climate change by helping developing countries to diversify their economies. Agricultural economies are more vulnerable to the effects of climate change than economies where employment is concentrated primarily in manufacturing and services. The yield impacts pictured in figure 7.2 would be less serious in a world in which a smaller share of employment and GDP in developing countries depended on agriculture than is currently the case.22

Economic growth would also reduce the damages associated with extreme weather events.23 Yohe and Tol explain variation across countries in the fraction of the population affected by extreme weather events between 1990 and 2000.24 They find that the fraction of the population affected by natural disasters decreases with increases in per capita income (elasticity =–1); increases with increases in income inequality (elasticity = 2.2), and increases with increases in population density (elasticity = 0.24).

In addition to pursuing economic growth, developing countries will also have to adapt to climate change. People in developing countries are already adapting to annual variations in temperature and precipitation, as well as to droughts, floods, and cyclones. In agriculture adaptation to temperature is reflected in crop choice. In Africa, for example, farmers select sorghum and maize-millet in cooler regions; maize-beans, maize-groundnut, and maize in moderately warm regions; and cow-pea, cowpea-sorghum, and millet-groundnut in hot regions. As precipitation increases or decreases, farmers shift toward water-loving or drought-tolerant crops.25 In the Indian state of Orissa, champeswar rice—a flood-resistant strain—is grown to provide insurance against agricultural losses. Farmers in the Mekong Delta build dikes to control flood waters.26 And community microinsurance schemes have been implemented in India’s Andra Pradesh to provide insurance against natural disasters.27

Climate change means that the need for this sort of adaptation will become greater. Much adaptation to climate change is a private good. But government actions to strengthen private adaptation to climate change will be needed in four areas: to provide those inputs to adaptation that are public goods—information about climate impacts, early warning systems for heat waves and floods, and construction of defensive public infrastructure; to take climate impacts into account in designing roads, bridges, dams, and other public infrastructure that may be affected by climate; to correct market failures that may impede adaptation; and to provide social safety nets that will sustain the poor through natural disasters.

Information about expected precipitation or early warnings about floods and heat waves can help people adjust to adverse weather conditions. In Mali the national meteorological service distributes information about precipitation and soil moisture through a network of farmers’ organizations and local governments. This information is transmitted throughout the growing season to allow farmers to adjust production practices. Obtaining information about weather risks depends on having enough monitoring stations and an adequate budget for collecting meteorological data, which can be facilitated by donor contributions and through transfer of technology for predicting weather events.28

Defensive infrastructure includes sea walls to protect against storm surges and irrigation systems that store monsoon rains. The Stern Review reports that expenditures of $3.15 billion on flood control in China between 1960 and 2000 avoided losses of $12 billion, while flood control projects in Rio de Janeiro yielded an internal rate of return of over 50 percent. Climate-proofing of roads, dams, and other infrastructure that may be affected by climate changes can also yield high returns. In dam construction in Bangladesh and South Africa, benefit-cost analyses have determined that it pays to increase the size of reservoirs to accommodate increased water runoff.29 Studies by the World Bank and Asian Development Bank have helped to identify cost-effective measures to climate-proof infrastructure in small island states.30

Governments can also help promote efficient market responses to climate risks. These include promoting insurance markets and making sure that credit is available, especially to the poor, to finance private adaptation. In high-income countries, one-third of losses associated with natural disasters are insured, compared with only 3 percent of losses in developing countries.31 Governments can promote weather insurance when private markets fail. The development of weather-index insurance to reduce farmers’ vulnerability to weather shocks is another example of the use of insurance markets to reduce climate risk (box 7.1).

In addition, governments can build institutions to help with disaster relief and create social programs to cushion households from income shocks. The Maharashtra Employment Guarantee Scheme, which was developed in the 1970s to help households cope with crop losses and other negative income shocks, is an excellent example of this, as are the employment creation programs adopted in Indonesia in 1997.32

BOX 7.1Weather-index insurance

One of the biggest problems faced by farmers in developing countries is dealing with weather shocks and adverse weather conditions, a problem that will only be exacerbated by climate change. The problem is especially acute for small farmers who are the most vulnerable to the increased frequency and magnitude of droughts, cyclones, and floods. Public programs to deal with weather risk include crop insurance, which reimburses farmers for yield losses and, more recently, weather-index insurance (WII). Weather-index insurance differs from traditional crop insurance because it pays farmers based on realizations of an index that is highly correlated with farm-level yields and can be used as a proxy for production losses. The index is based on the objective measurement of weather variables, such as the deficit of precipitation at a weather station or the trajectory and wind speed of a tropical cyclone. Weather-index insurance has several advantages over traditional crop insurance: adverse selection and information asymmetries are reduced since both the insurer and the insured can observe the same weather index; farmers cannot influence the results of the index (as opposed to the yield in their fields), and index-based payouts reduce administrative costs since a field-based loss assessment is not required. The success of WII depends on the availability of sufficient meteorological stations, which may be a problem, especially in Sub-Saharan Africa.

Weather index insurance has recently been researched or introduced in pilot projects in Ethiopia, India, Kenya, Malawi, Mexico, Morocco, Nicaragua, Peru, Thailand, Tunisia, and Ukraine. The introduction in India of rainfall insurance by BASIX and ICICI Lombard in 2003 was the first index insurance initiative launched at the farmer level in the developing world, and this insurance is now expanding in the Indian private and public insurance sectors.a A World Bank initiative in Malawi has succeeded in reaching small-scale farmers of maize and groundnuts. Policies sold to farmers are based on a rainfall index calibrated to the rainfall needs of the crop. The Malawi WII has been bundled with credit to allow farmers to repay input loans in the face of severe drought.

The extent to which these activities will be undertaken depends on institutional capacities in developing countries and on the availability of funding. Determining what should be done requires planning. The heterogeneity in climate impacts described above and highlighted in box 7.2 suggests the need for impact studies and benefit-cost analyses of specific adaptation strategies at the country level. Even though some of the most severe climate impacts may not occur until the second half of the century, developing countries are already vulnerable to variations in temperature and precipitation and extreme weather events. Projects that cushion these shocks are likely to have positive net benefits, although further studies are required.

Many studies are already under way. The development of National Adaptation Programmes of Action (NAPAs) by the United Nations Framework Convention on Climate Change (UNFCCC) is an attempt to help developing countries cope with the adverse effects of climate change. Each NAPA takes into account existing coping strategies at the grassroots level and builds upon them to identify priority activities. Currently 46 countries are preparing (or have prepared) NAPAs, with financial assistance from the UNFCCC’s Least-Developed Countries Fund (UNFCCC 2008). Multilateral development banks are also sponsoring studies: adaptation strategies are currently being prepared by the World Bank for each World Bank region. The Asian Development Bank, following its case studies of adaptation options in Micronesia and the Cook Islands has, with the World Bank, initiated a study of climate change impacts in four Asian coastal cities (Bangkok, Ho Chi Minh City, Kolkata, Manila).33 This study is tied to the Southeast Asia “mini-Stern” review, one of several regional climate impact studies currently in progress.

BOX 7.2Adaptation to climate change

The heterogeneity in climate change impacts across countries suggests that country-level studies are required to measure climate effects and to assess the benefits and costs of various adaptation measures in each individual country.

The three figures below depict the distribution of temperature and precipitation impacts in agriculture, the percent of population affected by a three meter rise in sea level, and the distribution of flood risk damages across countries.

The distribution of agricultural productivity losses, which assume no carbon fertilization effect, suggests that 20 developing countries would suffer yield losses of 30 percent or more. Adaptive agriculture programs should be examined in countries facing large agricultural productivity losses, such as India, Mexico, Senegal, and Sudan. Broader micro-insurance coverage for the poor should also be part of these programs.

The distribution of losses from sea level rise is highly skewed. Countries facing huge losses from rising sea levels, such as Egypt, Suriname, and Vietnam, will need to examine the net benefits of adaptive infrastructure and urbanization programs.

The distribution of flood risks (shown on a per capita basis) is also highly skewed. Programs combining adaptive infrastructure and micro-insurance should be the focus for countries facing high risk of flood disaster, such as Bangladesh, Benin, Cambodia, Honduras, Jamaica, and Mozambique.

Distribution of climate change impacts in developing countries

Source: Wheeler 2007, based on Cline 2007, Dasgupta and others 2007, and Buys and others 2007.

Note: The term “country rank” refers to the ranking of countries by size of damages, going from the country with the largest damages (first country) to the country with the smallest damages. In the top panel, the country with the largest agricultural losses (number 1) suffers yield declines of about 60 percent, while the country ranked 20th loses about 30 percent of its output.

Resources, beyond traditional development assistance, are available to help finance adaptation. The UNFCCC Special Climate Change Fund (SCCF) was established in 2001 to finance projects relating to adaptation, technology transfer, and capacity building. The UNFCCC Adaptation Fund, established in December 2007, will provide funds for adaptation by taxing emission reduction credits generated under the Clean Development Mechanism (box 7.3). These funds are small, however; currently the SCCF is approximately $60 million and the Adaptation Fund, $45 million.

Emission Trends and Progress toward Mitigation

Although differences of opinion exist about stabilization targets and means of achieving them, there is broad agreement that GHG emissions must be reduced over the coming decades to avoid serious alternation of the earth’s climate. GHG emissions have continued to increase since 1990, although the rate of increase in emissions has slowed for some sectors.

GHG Sources and Distribution

Figure 7.5 and table 7.7 show the breakdown of world GHG emissions in 2000 by sector.34 Approximately 65 percent of GHG emissions come from energy consumption and industrial processes, 18 percent from land use change (deforestation), and the remaining 17 percent from agriculture and waste. Deforestation and fossil fuel consumption primarily produce CO2, while agriculture and waste are the main source of methane and nitrous oxide emissions.35 When GHG emissions from energy are broken down by sector, over one-third of energy emissions are from power generation, approximately 22 percent from industry, and 22 percent from transportation.36

figure 7.5World GHG emissions, by sector, 2000

Table 7.7GHG emissions, by sector and region, 2000

metric tons of CO2e

EnergyIndustrial processesAgricultureLand use change and forestryWasteTotal
East Asia & Pacific4,0094281,4023,5362399,613
South Asia1,206655501451512,117
Middle East & North Africa868497822421,059
Europe & Central Asia3,504101354861464,190
Latin America & the Caribbean1,361821,0092,3571344,943
Sub-Saharan Africa553232941,379592,307
High-Income countries15,4816222,0439359118,830
Source: WRI.Note: The figures in parentheses are percentages of total emissions.
Source: WRI.Note: The figures in parentheses are percentages of total emissions.

The source of GHGs by sector varies widely across countries and regions (table 7.7). For the very poorest countries, most GHG emissions come from agriculture and changes in land use. Indeed, for the International Development Association (IDA) countries, only 29 percent of GHG emissions come from energy use.37 The ranking of the world’s largest emitters of CO2 depends on whether emissions from land use change are counted in the total. When they are not, the top 10 emitters account for 73 percent of CO2 emissions, and China and India are the only developing countries in the top 10. When emissions from land use change are included, the top 10 emitters account for two-thirds of CO2 emissions, and three developing countries—Brazil, Indonesia, and Malaysia—join China and India in the list of top 10 emitters.38

The rank of emitters based on per capita emissions is quite different. In 2004 world emissions per capita were 4.5 tons of CO2 per person from the burning of fossil fuel. The average emissions were 13.3 tons per person in high-income countries, 4.0 in middle-income countries, and only 0.9 tons per person in low-income countries. The map in figure 7.6 illustrates the striking disparity in per capita CO2 emissions between developing and developed countries, even when land use change is included as a source of emissions.

figure 7.6Per capita GHG emissions in 2000, including emissions from land use change

Source: Map created by Vinny Burgoo ( using CAIT 4.0 database of WRI.

How have emissions changed over time, and how are they likely to change if no steps are taken to reduce GHGs? Figures 7.7 and 7.8 show historic CO2 emissions from fossil fuel combustion and project them into the future under the IPCC A1FI scenario, which assumes high reliance on fossil fuels and rapid economic and population growth (see table 7.2). Emissions are broken down between those countries that agreed to limit GHG emissions under the UNFCCC—labeled Annex I countries—and the developing world (non-Annex I countries). Table 7.8 shows complementary information for emissions of all GHGs in 2000, broken down by Annex I and non-Annex I countries.

figure 7.7Annual CO2 emissions under the A1FI scenario, 1965–2035

figure 7.8Cumulative atmospheric CO2 under the A1FI scenario, 1965–2035

Table 7.8Comparison of GHG emissions for Annex I and non-Annex I countries
CategoryMeasurementAnnex INon-Annex I
GHG emissions in 2000: CO2, CH4, N2O, PFCs, HFCs, SF6 (including land use change)Percent of total emissions42.058.0
Tons of CO2e per person13.94.9
Cumulative CO2 emissions, 1950–2000 (including land use change)Percent of total emissions52.547.5
Tons of CO2 per person457103
Carbon intensity of electricity productionGrams of CO2/kilowatt hour436679
CO2 intensity of economy (excluding land use change)Tons of CO2/ million

Source: WRI 2007.
Source: WRI 2007.

Carbon emissions by both high-income (Annex I) and developing countries have continued to increase and are predicted to increase—by over 60 percent by 2035 from 2004 levels under the A1FI scenario. Moreover, developing countries’ CO2 emissions from fossil fuel will soon equal those of high-income countries (figure 7.7). Indeed, by 2035 developing countries will equal high-income countries in their contribution to the stock of CO2 in the atmosphere if the world follows the A1FI trajectory (figure 7.8). If all sources of GHGs are included, non-Annex I countries already emit more GHGs than Annex I countries (table 7.8). This does not imply that the total emissions of developing countries should immediately be reduced, but it does indicate that their magnitude cannot be ignored.

Understanding Sources of Change in CO2 Emissions from Fossil Fuel

To better understand sources of growth in CO2 emissions, it is useful to decompose the change in CO2 emissions into three components: the change in CO2 per unit of GDP (carbon intensity of output); the change in per capita income; and the change in population.39 For emissions to decline as population, per capita incomes, or both rise, the CO2 intensity of output must decrease. A recent World Bank study decomposes the change in fossil fuel emissions for the 70 largest emitters of CO2 from fossil fuel over the period 1994–2004 to see which countries were able to offset some of the growth in emissions that results from income (GDP) growth by reducing the carbon intensity of output.40

For the 70 countries as a whole, which accounted for about 95 percent of global CO2 emissions from fossil fuel in 2004, CO2 emissions from fossil fuel increased by approximately 5,000 million metric tons between 1994 and 2004. This change can be decomposed into a per capita GDP effect equal to 5,735 metric tons, a population effect of 2,665 tons, and a carbon intensity effect of –3,400 tons. This implies that the largest factor behind CO2 growth was the growth in per capita incomes. The effect of population growth was about half as large. Improvements in carbon intensity, however, offset 40 percent (–3,400/8,400) of the growth in CO2 from growth in population and per capita incomes.

How did reductions in the carbon intensity of output vary across countries? Figure 7.9 groups countries according to the percentage of increase in CO2 emissions from GDP growth (growth in GDP per capita plus growth in population) that was offset by a decline in the carbon intensity of output. In 15 countries, shown in the right bar of the figure, the percentage decline in the carbon intensity of output was greater than the percentage increase in GDP, implying that more than 100 percent of the increase in CO2 emissions due to GDP growth was offset. In these countries, which include Denmark, Germany, the Russian Federation, and Sweden, and some other countries that were part of the former Soviet Union, CO2 emissions actually declined between 1994 and 2004. Of the world’s 10 top emitters of CO2, only 2 countries (Germany and the Russian Federation) were in this group. For 36 countries (reflected in the middle bar of figure 7.9) the carbon intensity of output declined, but the percentage decrease in carbon intensity was smaller than the percentage increase in GDP, implying offsetting between 0 and 100 percent. For the remaining 19 countries, the carbon intensity of output actually increased, implying no offsetting.

figure 7.9Change in annual CO2 emissions by carbon intensity class, 1994–2004

Source: Derived from Bacon and Bhattacharya 2007.

Note: Countries are categorized according to whether changes in carbon intensity of output offset growth in GDP. Offsetting greater than 100 percent indicates that reductions in carbon intensity of output more than offset emissions from GDP growth, resulting in a decline in total emissions; offsetting between 0 and 100 percent indicates that some fraction of emissions from GDP growth was offset by reductions in carbon intensity; and “no offsetting” indicates that carbon intensity of output increased, adding to the increase in emissions from GDP growth.

What figure 7.9 makes clear is that the declines in CO2 emissions by the countries in the right bar of the figure were swamped by the increases in emissions of the other two groups. Countries in the middle group, in spite of reductions in the carbon intensity of their output, increased carbon emissions substantially—by nearly 4 billion tons a year in the aggregate. Countries whose carbon intensity increased caused world emissions to rise by 1.24 billion tons a year. In contrast, countries in the right bar caused annual emissions to drop by only 200 million tons a year.

Although the carbon intensity of GDP fell for 51 out of the 70 largest emitters of CO2 between 1994 and 2004, it must fall even faster if world carbon emissions are to decrease. For developing countries, carbon per unit of GDP must decrease even if their total carbon emissions are allowed to increase. Suppose, for example, that the carbon emissions of developing countries are allowed to double over the next 20 years, implying an annual growth rate in emissions of 3.5 percent. For carbon emissions to grow at a rate of 3.5 percent a year when GDP is growing at a rate of 10 percent a year—growth rates that India and China have recently experienced—carbon per dollar of GDP must fall at a rate of 6.5 percent a year.

Balancing Economic Growth and Reductions in Carbon Intensity

How can the carbon intensity of GDP be reduced as countries continue to grow? This must occur by reducing either the energy intensity of GDP (the energy used per unit of output), the fossil fuel intensity of energy (the fossil fuel used per unit of energy), or the carbon intensity of fossil fuel (the amount of carbon in a unit of fossil fuel), or by some combination of the three. Between 1994 and 2004 the reduction in the carbon intensity of GDP came almost entirely from reductions in the energy intensity of GDP. The carbon intensity of fossil fuel decreased slightly, reflecting a shift from coal to natural gas, but this reduction was offset by an increase in the fossil fuel intensity of energy.

Improving energy efficiency.

Figure 7.10 shows the energy intensity of GDP for World Bank regions. Eastern Europe and Central Asia had the highest energy intensity in 2004, largely because of the continued use of old, inefficient production equipment across various industries, dilapidated heating systems in cities and towns, high transmission and distribution losses, and inefficient stocks of household appliances. In China the widespread use of inefficient, coal-based power plants and small boilers for heating has offset an increasing trend in efficiency in other sectors. In both China and India a large proportion of small- and medium-scale industries continues to use old and inefficient technologies that contribute to high energy-intensity levels. Even though Sub-Saharan Africa’s energy use is small on a global scale (it used only 4 percent of global energy supply in 2004), as the industrial sector in the region develops, the adoption of new technologies will be needed if energy intensity is to improve.

figure 7.10Energy intensity by region, 2004

How great is the technical scope for improving energy efficiency in developing countries? The International Energy Agency has recently completed a global analysis of energy efficiency in manufacturing.41 The study found that manufacturing accountsfor about a third of world energy consumption, and three industries—chemicals and petrochemicals, iron and steel, and nonmetallic minerals—account for over half of manufacturing energy use and over 70 percent of CO2 emissions from manufacturing. Table 7.9 compares the energy efficiency of three production processes in various countries with each other and with best-available technology. There is clear variation in energy efficiency across countries: China, the world’s largest producer of cement, is less efficient than India or Japan. However, the energy efficiency of cement production could be increased even in Europe and Japan. Similar gains in efficiency could be realized in steel and ammonia production. Overall, the IEA study estimates that between 18 and 26 percent of world industrial energy use could be reduced by using best-practice technologies. This would reduce CO2 emissions by between 1.9 billion and 3.2 billion tons a year.

Table 7.9Comparison of industrial energy efficiency across countries
Energy consumption per unit produced (100 = most efficient country)
United States120145105
Best-available technology759060
Source: Watson and others 2007.n.a. = Not available.
Source: Watson and others 2007.n.a. = Not available.

Improving energy efficiency in power generation will also reduce energy CO2 intensities, especially in countries such as India and China that depend on coal for power generation. The average thermal efficiency (the amount of power produced per unit of heat input) of power plants in India and China is between 29 and 30 percent, compared with 36 percent in developed countries. Supercritical plants can achieve efficiencies up to 45 percent. In China, installed capacity is expected to double—from 500 to 1,000 gigawatts (GW) between 2007 and 2015. India is expected to add 100 GW of capacity over the same period.42 Installing thermal power plants with an efficiency of 38 percent in China would reduce carbon emissions at a typical plant by 22 percent. Emission reductions of up to 92 percent could be achieved by building supercritical plants with carbon capture and storage.43

The adoption of some low-carbon options will clearly require external financing and a positive shadow price for carbon. There are, however, some “no regrets,” win-win options for improving energy efficiency that would pay for themselves in fuel savings if subsidies to energy consumption and production were removed.44 These include reducing losses in the transmission and distribution of electricity, some improvements in power plant efficiency, insulation of buildings and improvements in appliance and vehicle efficiency. In many developing countries, demand-side incentives to improve energy efficiency are weak because electricity is not priced to recover the costs of generation. Failure to reform the electricity sector may also hamper access to financing more efficient power plants.

Low-carbon investments that would not pay for themselves in fuel savings or ancillary benefits could be financed by selling the emission reduction credits in a world in which long-term commitments to reduce CO2 emissions establish a price path for carbon. This is now possible under the Kyoto Protocol’s Clean Development Mechanism (box 7.3); however, because the Kyoto Protocol ends in 2012, the Clean Development Mechanism does not currently provide long-term financing opportunities. With donor support, international financial institutions are attempting to fill this market void. Currently, the World Bank manages nine carbon funds totaling more than $2.5 billion. The International Finance Corporation and European Bank for Reconstruction and Development manage three additional carbon funds. These funds support more fuel-efficient thermal power generation as well as renewable energy sources.45

Reducing the carbon intensity of energy use.

The carbon intensity of energy used by the top 70 emitters of CO2 did not improve over the 1994–2004 period—although the carbon intensity of fossil fuel decreased slightly, the share of fossil fuel in energy increased. Substituting renewable energy sources for fossil fuels does, however, represent another means of reducing the carbon intensity of GDP. Although many sources of renewable energy may not be cost-effective at current energy prices, the potential for tapping these sources exists in many developing countries. And, given a functioning carbon market, these sources would likely be exploited eventually.

A recent World Bank study has estimated the potential for developing five sources of renewable energy—solar power, wind power, hydro power, geothermal energy, and biofuels—in developing and developed countries. 46 In each case potential energy supply is expressed as a fraction of current energy consumption. Table 7.10 shows the availability of renewable energy sources, relative to current consumption, for developing countries by World Bank region. The opportunities for renewable energy are greatest in Sub-Saharan Africa and parts of Latin America. Of the top 35 countries in the world with the most solar energy potential, 17 are in Sub-Saharan Africa and 7 in Latin America. Of the top 35 countries in the world with the most biofuel potential, 25 are in Sub-Saharan Africa. Note that table 7.10 measures the technical potential for developing renewable energy sources. For such development to be economically feasible, the world would have to make a significant commitment to GHG reduction. In the case of biofuels, the implications of their development on land use and food security must also be considered.47

Table 7.10Availability of renewable resources relative to current consumption, by World Bank regionannual renewable energy potential in years of current energy consumption
Sub-Saharan AfricaEast Asia and PacificLatin America and the CaribbeanSouth Asia
Central African Republic90.9Papua New Guinea12.6Uruguay31.7Pakistan1.9
Mauritania86.2Solomon Islands9.3Argentina27.5Sri Lanka1.2
Chad77.3Lao PDR8.8Guyana19.3Bangladesh1.1
Congo, Rep. of43.6Vanuatu3.3Brazil6.4
Congo, Dem. Rep.24.7Vietnam0.7Belize3.8
Mozambique23.4Thailand0.6Venezuela. R. B. de2.6
Burkina Faso15.9Panama1.9
Madagascar14.6Costa Rica1.8
Sierra Leone10.1

Reducing deforestation.

Land use change currently accounts for 18 percent of GHG emissions. As figure 7.11 shows, CO2 emissions from changes in land use increased more or less steadily from 1850 until 2000. Since the early 20th century emissions from land use change in developing countries have dominated emissions from Annex I countries. In recent years two countries—Brazil and Indonesia—have produced over half of all world emissions from land use change. In Brazil, forests in the Amazon have been cleared to make way for pasture and cropland. The ultimate drivers of deforestation in Brazil are the demand for beef, soybeans, and lumber. Deforestation in Indonesia has been driven by the demand for timber and pulp and for land for palm oil plantations.48 In both countries, deforestation has been undertaken by large corporate interests as well as by small-holders. Although the data pictured in figure 7.11 stop in 2000, annual hectares deforested in Indonesia were approximately the same between 2000 and 2005 as between 1990 and 2000.49 In Brazil hectares deforested actually increased from 2.7 million annually between 1990 and 2000 to 3.1 million annually between 2000 and 2005.

figure 7.11CO2 emissions from land use change, 1850–2000

Source: CDIAC 2008.

As many have observed, the continued conversion of the world’s forests for agriculture would not be economical if there were a well-functioning carbon market.50 The present value of a hectare of crop- or pastureland in the Brazilian Amazon is worth between $100 and $200.51 Clearing a hectare of dense rainforest could release 500 tons of CO2. At a carbon price of even $10 per ton of CO2, an asset worth $5,000 is being destroyed for a land use that is one-twentieth as valuable.

Currently the carbon market does not extend to avoided deforestation. The Clean Development Mechanism allows parties to the Kyoto Protocol to purchase emission reduction credits from projects in developing countries that reduce CO2 emissions (box 7.3). These reductions must be additional to what would have occurred under business as usual and can include reforestation projects. The mechanism, however, does not allow developing countries to create emission reduction credits from avoided deforestation.

A new carbon credit program is currently under negotiation within the UNFCCC that would compensate countries with carbon credits for avoided deforestation (see box 7.3). This complements donor efforts to fund avoided deforestation, including the World Bank’s Forest Carbon Partnership Facility, which will help developing countries improve their estimates of forest carbon stocks and fund pilot projects to reduce deforestation, and the Bank’s BioCarbon Funds.

Progress on Institutions and Policies to Deal with Climate Change

Because the abatement of greenhouse gases is a global public good, policies to reduce GHGs require international coordination. Beginning with the formation of the IPCC in 1988 and continuing with the establishment of the UNFCCC in 1992, the nations of the world have taken steps to address the effects of human actions on the earth’s climate. Progress in establishing a link between human actions and climate change—and drawing public attention to this fact—is the first step in formulating effective public policies. The successful regulation of ozone-depleting substances under the Montreal Protocol would never have occurred had scientists not demonstrated that 40 percent of the stratospheric ozone layer had disappeared between 1957 and 1984 and linked pictures of the hole in the ozone layer to emissions of chlorofluorocarbons (CFCs) and other ozone-depleting substances.

BOX 7.3Sources of carbon finance under the UNFCCC

Under the Clean Development Mechanism (CDM), parties to the Kyoto Protocol can meet their obligations to reduce GHG emissions by purchasing emission reduction credits (ERCs) from projects in developing countries. An ERC is generated if a project reduces its carbon emissions below what would have occurred without the CDM. The credits must be certified by the UNFCCC before they can be used to meet obligations under the Kyoto Protocol.

The CDM market is growing rapidly. As of January 2008, the UNFCCC had registered 901 projects with ERCs totaling 1.15 billion tons of CO2 equivalent. Most of these projects have originated in Asia or Latin America, with fewer than 3 percent originating in Africa. If projects are weighted by the number of ERCs delivered, China was the largest seller of ERCs in transactions that occurred between January 2005 and September 2006, accounting for 61 percent of credits sold.a Most transactions have involved energy and manufacturing projects. Under the Marrakesh accords, land use projects are limited to afforestatation and reforestation projects.

A new carbon credit program is under negotiation within the UNFCCC—Reducing Emissions in Deforestation and Forest Degradation (REDD)—that would compensate countries with carbon credits for their efforts in reducing CO2 emissions through forest conservation and by controlling forest degradation. A recent study by the Woods Hole Research Centerb develops a conceptual framework of the costs to tropical countries of implementing REDD programs. It estimates that, in the Brazilian Amazon, approximately 90 percent of the opportunity costs of maintaining existing forest could be compensated for $3 per ton of carbon (approximately $1 per ton of CO2). Under the program, forest families would double their incomes, fire-related damages would be avoided, and carbon emissions would be reduced over 30 years by 6.3 billion tons, equivalent to 23 billion tons of CO2.

a.Lecocq and Ambrosi 2007.b.Woods Hole Research Center 2007.

International policies to deal with climate change have been organized under the United Nations Framework Convention on Climate Change, which was signed in Rio de Janeiro in 1992, went into force in 1994, and has been ratified by 190 countries. The UNFCCC created an international frame work for climate change policy consisting of four elements: a long-term goal of stabilizing GHG concentrations in the atmosphere at a level that would prevent dangerous interference with the climate system; a short-term goal for developed (Annex I) countries to stabilize their emissions at 1990 levels by 2000; a principle of “common but differentiated responsibilities,” suggesting that developing countries should not be expected to undertake the same obligations as developed countries; and opportunities for realizing more cost-effective reductions in GHG emissions through joint implementation.52 Under joint implementation developed countries were allowed to invest in emission-reducing projects in developing countries to meet their 2000 emission reduction goals. Although only a few Annex I countries had met their emissions goals by 2000, the Rio accords established important principles that continue to be reflected in policy discussions.

The Kyoto Protocol

The Kyoto Protocol, which came into force in February 2005, committed most industrial countries and some transition economies (referred to as the Annex B countries) to specific GHG emissions targets. Over 2008–12, the total emissions of Annex B countries are to be reduced 5 percent below 1990 levels. Countries can either reduce GHG emissions or enhance the amount of carbon captured in “carbon sinks” (by sequestering GHG from the atmosphere) such as reforestation programs. The protocol also allows countries to buy emission rights from other Annex B countries whose emissions are below their limits and to assist non–Annex B countries to implement projects that reduce GHG emissions through the Clean Development Mechanism (see box 7.3).

The Kyoto Protocol represents a major attempt by the international community to come to grips with climate change. By signaling the intention of many countries to reduce GHG emissions, it may encourage investors to adopt more efficient, low-carbon technologies. Through provisions for carbon trading and the Clean Development Mechanism, the protocol helps to establish the principle that emissions reductions should be achieved in a cost-effective manner. It is also equitable, in the sense that it imposes no restrictions on the emissions of developing countries, which on a per capita basis have contributed less to the existing stock of greenhouse gases than developed countries.

The Kyoto Protocol has nevertheless been subject to many criticisms. For one thing, it does not limit the emissions of three of the world’s five largest emitters of greenhouse gases—the United States, China, and India. The United States did not ratify the treaty, and the Kyoto Protocol does not extend to developing countries. It is too early to judge compliance (obligations to curtail are legally binding only for the 2008–12 period), but transition economies have more than satisfied their Kyoto targets because of a major decline in economic activity after 1990, while most developed-country parties to the Protocol have thus far not met their targets (figure 7.12).53

figure 7.12Kyoto targets and changes in GHG emissions for Annex B countries

Source: UNFCCC 2008, EEA.

Beyond Kyoto

International agreements to deal with climate change in the future will have to deal with several issues. Progress toward global environmental sustainability will depend on how agreements measure up against the following criteria.54 First, an agreement must achieve a desirable environmental outcome. This could be stated as an emissions (or concentration) target or as a temperature goal. Second, the agreement should be efficient—it should achieve the environmental outcome at least cost, both in the timing of actions and in minimizing the costs of abatement across countries. Third, the obligations and results of the policies should be viewed as equitable, both across countries and, given the long-term nature of climate change, across generations. Fourth, the policies should be flexible—they should be able to accommodate changes in information about climate science. And, finally, the agreement should encourage wide participation and compliance among countries.

Whatever form an international agreement takes, it will have to provide incentives to reduce GHG emissions and an institution to collect and verify information on GHG emissions so that progress toward mitigation goals can be monitored.55 The verification and publicizing of GHG emissions at the country level is necessary for the enforcement of an international agreement, will signal the willingness of countries to participate in the agreement, and will provide the means for stakeholders to put pressure on major emitters.56 To provide an incentive to reduce GHG emissions, emissions must be priced, whether through a carbon tax, a permit market, or some combination of the two. The agreement will also have to make some provisions for the accelerated development of clean technologies, including clean energy technologies, carbon capture and storage, and geoengineering, and it will need to finance the diffusion of these technologies in developing countries. Finally, the agreement will need to support developing-country adaptation to the impacts of unavoidable climate change.

Recent Trends in Biodiversity and Marine Fisheries

The global commons includes the animal and plant species that inhabit the planet, as well as the earth’s climate. Protecting the diversity of animal and plant life is important for both economic and noneconomic reasons: humans attach a value to the existence of diversity per se, quite apart from the role that biological organisms play in the production of goods and services. At the same time, continued diversity of animal and plant species is important to the world’s economy, and especially to the lives of the poor in developing countries. Ocean fisheries, in particular, constitute an important source of food, and of livelihoods, for developing countries.


One of the targets of MDG 7 is to reduce biodiversity loss. The two indicators associated with this target are the proportion of terrestrial and marine areas that are protected, and the proportion of the earth’s species that are threatened with extinction. Figure 7.13 depicts trends in protected areas from 1990 to 2005. As the figure shows, the percent of area protected in all World Bank regions has increased since 1990. On average, approximately 15 of territorial area is protected worldwide, although the percent protected varies considerably across countries and regions.

figure 7.13Proportion of terrestrial and marine area protected

Measuring the health of a wide variety of animal and plant species is inherently more difficult than measuring the fraction of land that is protected. Data on birds and mammals come from sightings of individual members of populations in a limited numbers of locations. The size of fish populations is often inferred from the ratio of harvests to effort (such as number of boat days) and is likewise subject to error, especially for individual species. Obtaining reliable data on the proportion of individual species threatened with extinction (indicator 7.6) is therefore difficult. The World Wildlife Fund (WWF) summarizes changes in populations of vertebrate species in its Living Planet Index (LPI)—the index measures trends in the planet’s biodiversity by tracking over 3,600 populations of 1,313 vertebrate species.57 Separate indexes are computed for terrestrial, marine, and freshwater organisms using data from a variety of sources. Indexes are also computed for different biogeographic regions of the world.

The LPI (figure 7.14) decreased from a value of 1.0 in 1970 to 0.71 in 2003, suggesting a downward trend in vertebrate populations overall. Each of the three component indexes also declined by approximately 30 percent. These aggregate trends, however, mask important regional changes in biodiversity (figure 7.15). The decline in the terrestrial index reflects a slight increase in the population of temperate species, but a 55 percent decrease in the populations of tropical species. The rapid decline in the terrestrial index in tropical regions reflects the conversion of natural habitat to cropland or pasture. The most rapid conversion over the past 20 years occurred in the forests of Southeast Asia and in South America.

figure 7.14Living Planet Index, 1970–2003

Source: WWF 2006.

figure 7.15Living Planet Indexes for terrestrial, marine, and freshwater organisms

Source: WWF 2006.

The marine subindex declined overall by 27 percent between 1970 and 2003, but trends in the four ocean basins varied greatly. Monitored populations in the Atlantic-Arctic oceans actually increased, while populations in the Pacific were at approximately the same levels in 2003 as in 1970. In contrast, marine populations in the Indian Ocean declined by 55 percent, while populations in the Southern Ocean decreased by 30 percent. The relative stability of populations in the Pacific, the world’s largest commercial fishery, masks declines in economically important species such as cod and tuna as a result of overfishing.

The freshwater index shows that species populations in this group declined by 30 percent between 1970 and 2003. This represents a stable trend in bird populations, but a 50 percent decline in fish species, attributable to habitat destruction, overfishing, and pollution. The damming of rivers for industrial and domestic use is likely responsible for much of the habitat destruction. The alteration of natural river flows alters the migration and dispersal of fish. More than 70 percent of large river systems (measured by catchment area) in virtually all biomes have been disrupted, primarily for irrigation.58

Marine Fisheries

The health of marine fisheries is especially important to developing countries. Fish provide 2.6 billion people with over 20 percent of their protein intake. 59 Two-thirds of world fisheries production comes from marine and freshwater fish capture; the remainder comes from aquaculture. Developing countries are among the top 10 countries in fish capture: together China, Peru, Chile, Indonesia, and India accounted for 45 percent of inland and marine fish catches in 2004.60 While the number of fishers has been declining in most high-income countries, it has increased in China, Peru, and Indonesia since 1990.

Indicator 7.3 of MDG 7 calls for monitoring the proportion of fish populations within safe biological limits. The Food and Agriculture Organization (FAO) has monitored the world’s marine stocks since 1974. As figure 7.16 reveals, about half of all stocks are fully exploited, implying that production is close to maximum sustained yield. The share of fish stocks that are moderately exploited or underexploited has fallen, from 40 percent in 1974 to 25 percent in 2006, while the share of overexploited fish populations has increased, from 10 percent in 1974 to 25 percent. The increase in the number of overexploited stocks occurred primarily during the 1970s and 1980s, however, and the share overexploited has stabilized since 1990.

figure 7.16Global trends in the world’s marine stocks since 1974

Source: FAO 2007.

The data in figure 7.16 are consistent with trends in capture fisheries production. Production from marine and inland fisheries increased rapidly from 1950 until 1970, grew more slowly from 1970 until 1990, and has stabilized since then.61 Since the world’s fishing fleet has also been approximately stable between 1990 and 2004, the stable catch is consistent with fish populations that are, in the aggregate, stable.

This does not, however, mean that there is no cause for concern. The most commercially successful species are all fully exploited or overexploited. Examples of the latter include the blue whiting in the Northeast Atlantic and the Chilean jack mackerel and some anchoveta stocks in the Southeast Pacific. The percent of stocks that are overexploited varies by area. The areas with the highest proportion (46–60 percent) of overexploited species are the Southeast Atlantic, the Southeast Pacific, the Northeast Atlantic, and the high seas. The FAO suggests that deep water species in the high seas are at particular risk of exploitation because of their slow growth rates and late age at first maturity.62


Following are the key points summarizing the findings of this chapter:

  • The world has been warming since the industrial revolution as a result of human emissions of greenhouse gases. This effect has accelerated in the second half of the 20th century and especially since 1990. If past trends in emissions continue, the world could experience mean global temperature increases of 2 to 6 degrees centigrade by the end of the century.
  • These temperature increases, and accompanying changes in precipitation, sea level rise, and extreme weather events will not be evenly distributed across countries. Temperatures will rise more in northern latitudes than in subtropical regions. But temperature increases in subtropical regions will reach levels where agricultural productivity is likely to decline. Heat waves will be more likely in both southern and northern latitudes. Dry areas are likely to become drier and wet areas wetter.
  • Poor countries will suffer the most from, and are able to adapt the least to, the effects of climate change. These include impacts on agriculture and human health and effects caused by rising sea levels and extreme weather events. However, vulnerability to climate impacts varies widely among developing countries, suggesting that adaptation planning must be country-specific.
  • For developing countries, the best way to adapt to climate change is to promote inclusive development. This will help to reduce vulnerability to climate impacts through economic diversification and by providing the poor with the resources they need to adapt. Achieving MDGs 1, 4, and 6 would constitute effective adaptation to the health effects of climate change.
  • Although much adaptation is a private good, governments have a role to play in fostering adaptation: they can help provide information, including weather forecasts; they can facilitate infrastructure investments; they can promote efficient market responses to climate change—such as weather-index and flood insurance; and they can build institutions to help with disaster relief and set up social programs to cushion households from income shocks.
  • Preventing dangerous changes in climate will necessarily involve mitigation of GHGs. This includes CO2 from fossil fuel use, but also mitigation of CO2 from deforestation and reduction of methane and nitrous oxide from agriculture. Better data are needed on GHG emissions from land use and agriculture, as these sources currently account for one-third of GHG emissions.
  • CO2 emissions from fossil fuel can be reduced by reducing the energy intensity of output and the carbon intensity of energy. Studies of the technical feasibility of improving energy efficiency indicate considerable scope for improving energy efficiency and for replacing fossil fuels with renewable energy sources.
  • The use of technologies that are more energy efficient and the tapping of renewable energy sources will depend in part on the world’s making a commitment to reduce GHG emissions. If carbon is priced, then the reductions in carbon emissions in developing countries could be sold on the carbon market to finance low-carbon technologies. This approach, however, would require a long-term commitment since low-carbon capital investments will yield carbon reductions over a long horizon. International financial institutions may be able to bridge the gap between current schemes (such as the Clean Development Mechanism) and those that will replace them in the longer term.
  • Carbon finance can also help reduce emissions from deforestation. Using carbon finance to protect forests will require the development of institutions to monitor and protect forests at the national level, as well as funding from developed countries, through a carbon market or other forms of assistance.
  • The world has made progress in dealing with climate change in the past 20 years, most notably by establishing the IPCC and UNFCCC. The UNFCCC has established important principles in dealing with climate change: that the world should stabilize GHG concentrations in the atmosphere at a level that would prevent dangerous interference with the climate system; that this goal should be achieved through “common but differentiated responsibilities,” suggesting that developing countries should not be expected to undertake the same obligations as industrialized countries; and that reductions in GHG emissions should be achieved in a cost-effective manner.
  • The formulation of an international architecture to deal with climate change is an ongoing process. Despite its limitations, the Kyoto Protocol has helped establish a foundation for global collective action to build on. Future agreements will be judged according to their ability to limit GHG emissions significantly, to do so in a cost-effective and equitable manner, and to ensure widespread compliance.
1.In this respect the chapter has a very different objective from the UNDP’s 2007 Human Development Report, which discusses the impacts of climate change and the costs of mitigation and argues for specific emissions targets, and the Stern Review (Stern and others 2006), which presents a detailed analysis of the economics of climate change. The International Monetary Fund (2008) in chapter 4 of the April 2008 World Economic Outlook discusses the macroeconomic implications of climate change, including the costs of GHG mitigation and their fiscal implications.
4.CDIAC 2008; NOAA 2008. Other GHGs include methane (CH4), nitrous oxide (N2O), fluorocarbons (perfluorocarbon (PFC), hydro-fluorocarbon (HFC), and sulphur hexafluoride (SF6). The concentrations of these GHGs in the atmosphere are described in terms of CO2 equivalents. Equivalent CO2 (CO2e) is the concentration of CO2 that would cause the same level of radiative forcing as a given type and concentration of greenhouse gas.
5.IPCC 2007b, figure SPM.4.
6.A change of 2°C relative to pre-industrial times represents a change of 1.5°C relative to 1980–99 levels.
7.IPCC 2007b, figure SPM 6. The CO2 equivalent concentration of 375 ppm reflects the effects of aerosols as well as long-lived GHGs.
10.Weitzman (2007) has recently emphasized this point.
12.Or 2.3°C compared with pre-industrial levels.
13.2007b, table SPM.6.
16.The carbon fertilization effect raises yields approximately 15 percent for crops such as rice, wheat, and soybeans.
17.Cline 2007, table 6.2. It should be noted that Cline’s analysis ignores the effects of yield changes on agricultural prices.
19.Dasgupta and others 2007. The impacts of sea level rise on GDP are calculated by attributing current GDP to each square mile of a country’s surface area. The percent of GDP affected by sea level rise is calculated based on this attribution and ignores the macroeconomic impacts of sea level rise on GDP.
20.Buys and others 2007. The data for the index come from the Emergency Disasters Database (EM-DAT) at the Center for Research on the Epidemiology of Disasters, Université Catholique de Louvain.
22.In 2004, 55 percent of the labor force in South Asia, and 58 percent of the labor force in Sub-Saharan Africa and East Asia and the Pacific were employed in agriculture (World Bank 2007a). On average, the share of agriculture in GDP in 2004 was 22 percent in low-income countries and only 2 percent in high-income countries.
34.Information on CO2 emissions from land use, based on Houghton (2003), have been published for 1850–2000. Data on non-CO2 gases are available from the U.S. Environmental Protection Agency for 1990, 1995, and 2000 (WRI 2007).
35.In terms of CO2 equivalents (see note 4), carbon dioxide accounts for approximately 78 percent, methane for 14 percent, and nitrous oxide for 7 percent of GHG emissions. Fluorocarbons and sulphur hexafluoride account for the remaining GHG emissions.
36.There are various possible sectoral breakdowns, depending on whether or not electricity is attributed to end users (such as agriculture and the residential sector).
39.Formally, emissions = carbon intensity effect + per capita GDP effect + population effect. The carbon intensity effect = emissions * [rate of growth carbon intensity/rate of growth in emissions]. Other effects are defined similarly.
40.Note that CO2 intensity is calculated using PPP (purchasing power parity) GDP; however, using GDP at market exchange rates makes little difference in the rates of change computed in the study (Bacon and Bhattacharya 2007).
45.For more on external financing of climate change mitigation and adaptation in developing countries, see chapter 3.
53.The penalties for noncompliance are not likely to change behavior. Countries that fail to meet their targets in 2008–12 must make up for this shortfall in the subsequent commitment period, plus a 30 percent penalty. A country liable for the penalty could fail to ratify the extension or insist on raising its emissions limit as a condition of participation.
54.Aldy and others 2003; Aldy and Stavins 2007.
56.Currently the UNFCCC reviews the GHG inventories of Annex I countries. For a description of the review process, see (
61.The yearly variation in production between 1990 and 2004 is due almost entirely to variation in production from the Peruvian anchovy fishery and is associated with El Niño.

    ADB (Asian Development Bank). 2005. “Climate Proofing: A Risk-Based Approach to Adaptation.”Asian Development BankManila.

    AldyJoseph E. and Robert N.Stavinseds.2007. Architectures for Agreement: Addressing Global Climate Change in the Post-Kyoto World.Cambridge, U.K.: Cambridge University Press.

    • Search Google Scholar
    • Export Citation

    BaconRobert W. and SomaBhattacharya.2007. “Growth and CO2 Emissions: How Do Different Countries Fare?”Environment Department Paper Climate Change Series No. 113World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    BarrettScott.2005. “The Theory of International Environmental Agreements.”In Handbook of Environmental Economics;vol. 314571516ed.K. G.Mäler andJ.Vincent.Amsterdam: Elsevier.

    • Search Google Scholar

    BettencourtSofiaRichardCroadPaulFreemanJohnHayRogerJonesPeterKingPadmaLalAlanMearnsGeoffMillerIdahPswarayi-RiddihoughAlfSimpsonNakibaeTeuataboUlric Trotz andMaarten VanAalst.2006. “Not If, But When: Adapting to natural hazards in the Pacific Islands Region.”Policy NoteWorld BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    BuysPietUweDeichmannCraigMeisnerThao TonThat andDavidWheeler.2007. “Country Stakes in Climate Negotiations: Two Dimensions of Vulnerability.”Policy Research Working Paper No. 4300World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    CapoorKaran andPhilippeAmbrosi.2007. “State and Trends of the Carbon Market 2007.”World Bank, with IETA (International Emissions Trading Association) Washington, DC.

    • Search Google Scholar

    CDIAC (Carbon Dioxide Information Analysis Center). 2008.

    ChomitzKennethM.2002. “Baseline, Leakage, and Measurement Issues: How Do Forestry and Energy Projects Compare?”Climate Policy2 (1): 3549.

    • Search Google Scholar

    ChomitzKennethM.PietBuysGiacomo DeLucaTimothy S.Thomas andSheilaWertz-Kanounnikoff.2007. “At Loggerheads? Agricultural Expansion, Poverty Reduction, and Environment in the Tropical Forests.”World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    ClineWilliamR.2007. “Global Warming and Agriculture: Impact Estimates by Country.”Peterson InstituteWashington, DC.

    Commission of the European Communities. 2005. “Winning the Battle Against Global Climate Change.”Communication from the Commission to the Council the European Parliament the European Economic and Social Committee and the Committee of the Regions COM(2005)35European Union, Brussels.

    • Search Google Scholar

    DasguptaSusmitaBenoitLaplanteCraigMeisnerDavidWheeler andDavid JianpingYan.2007. “The Impact of Sea Level Rise on Developing Countries: A Comparative Analysis.”Policy Research Working Paper No. 4136World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    DinarArielRobertMendelsohnRobertEvensonJyotiParikhApurvaSanghiKaviKumarJamesMcKinsey andStephenLonergan.1998. “Measuring the Impact of Climate Change on Indian Agriculture.”Technical Paper No. 402World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    FAO (Food and Agriculture Organization of the United Nations). 2007. The State of World Fisheries and Aquaculture 2006.Rome: FAO.

    FAO (Food and Agriculture Organization of the United Nations). 2008. “Change in Extent of Forest and Other Wooded Land, 1900–2005.”Table in Global Forest Resources Assessment.

    • Search Google Scholar
    • Export Citation

    HoughtonR. A.2003. “Revised Estimates of the Annual Net Flux of Carbon to the Atmosphere from Changes in Land Use and Land Management 1850–2000.”Tellus55B:37890.

    • Search Google Scholar
    • Export Citation

    IEA (International Energy Agency). 2007. World Energy Outlook 2007.Paris: IEA.

    IMF (International Monetary Fund). 2008 (April). World Economic Outlook.”Washington, DC: IMF.

    IPCC (Intergovernmental Panel on Climate Change). 2007a. “Climate Change 2007: The Physical Science Basis.”Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University PressCambridge, U.K., and New York.

    • Search Google Scholar
    • Export Citation

    IPCC (Intergovernmental Panel on Climate Change). 2007b. “Climate Change 2007: Synthesis Report.”Summary for PolicymakersFourth Assessment Report of the Intergovernmental Panel on Climate Change.Cambridge University PressCambridge, United Kingdom and New York, NY, USA.

    • Search Google Scholar
    • Export Citation

    KahnMatthew E.2005. “The Death Toll from Natural Disasters: The Role of Income, Geography, and Institutions.”Review of Economics and Statistics87 (2): 27184.

    • Search Google Scholar
    • Export Citation

    KurukulasuriyaPradeep andRobert O.Mendelsohn.2007. “Crop Selection: Adapting to Climate Change in Africa.”Policy Research Working Paper No. 4307World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    KurukulasuriyaPradeepRobertMendelsohnRashidHassanJamesBenhinTemesgenDeressaMbayeDiopHelmy MohamedEidK. YerfiFosuGlwadysGbetibouoSumanJainAliMahamadouRennethManoJaneKabubo-MariaraSamiaEl-MarsafawyErnestMoluaSamihaOudaMathieuOuedraogoIsidorSéneDavidMaddisonS. NiggolSeo andArielDinar.2006. “Will African Agriculture Survive Climate Change?”World Bank Economic Review20 (3): 36788.

    • Search Google Scholar

    LecocqFranck andPhilippeAmbrosi.2007. “The Clean Development Mechanism: History, Status, and Prospects.”Review of Environmental Economics and Policy1 (1): 13451.

    • Search Google Scholar
    • Export Citation

    LopezAlan.2006. “Global and Regional Burden of Disease and Risk Factors, 2001: Systematic Analysis of Population Health Data.”The Lancet367:174757.

    • Search Google Scholar

    ManuamornOrnsaran Pomme.2005. “Scaling-Up Micro Insurance. The Case of Weather Insurance for Smallholders in India,”World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    McMichaelAnthony J.DiarmidCampbellLendrumSariKovatsSallyEdwardsPaulWilkinsonTheresaWilsonRobertNichollsSimonHalesFrankTanserDavid LeSueurMichaelSchlesingerand NatashaAndronova.2004. “Global Climate Change.”In Comparative Quantification of Health Risks: Global and Regional Burden of Disease Due to Selected Major Risk Factorsed.M. J.Ezzati et al.15431649. Geneva, Switzerland: World Health Organization.

    • Search Google Scholar
    • Export Citation

    MurdochJames C. and ToddSandler.1997. “Voluntary Cutbacks and Pre-Treaty Behavior: The Helsinki Protocol and Sulfur Emissions.”Public Finance Quarterly25 (2): 13962.

    • Search Google Scholar

    NOAA (National Oceanic and Atmospheric Administration). 2008. “Trends in Atmospheric Carbon Dioxide.”Earth System Research Laboratory, Global Monitoring Division.

    • Search Google Scholar
    • Export Citation

    ParryM.C.RosenzweigA.IglesiasM.Livermore andG.Fischer.2004. “Effects of Climate Change on Global Food Production under SRES Emissions and Socio-Economic Scenarios.”Global Environmental Change11 (3): 13.

    • Search Google Scholar

    SchellingThomas.1992. “Some Economics of Global Warming.”American Economic Review82 (1): 114.

    SternNicholas2008. Richard T. Ely LectureAmerican Economic Association.

    SternN.S.PetersV.BakhshiA.BowenC.CameronS.CatovskyD.CraneS.CruickshankS.DietzN.EdmonsonS.-L.GarbettL.HamidG.HoffmanD.IngramB.JonesN.PatmoreH.RadcliffeR.SathiyarajahM.StockC.TaylorT.VernonH.Wanjieand D.Zenghelis.2006. “Stern Review on the Economics of Climate Change.”HM Treasury, London. report.cfm

    • Search Google Scholar
    • Export Citation

    SunsteinCarl R.2007. Worst-Case Scenarios.Cambridge, MA: Harvard University Press.

    • Search Google Scholar

    SuryahadiAsepSudarnoSumarto andLantPritchett.2003. “Evolution of Poverty during the Crisis in Indonesia,”Asian Economic Journal17 (3): 22141.

    • Search Google Scholar
    • Export Citation

    UNDP (United Nations Development Programme). 2007. Human Development Report 2007/2008—Fighting Climate Change: Human Solidarity in a Divided World.New York: UNDP.

    • Search Google Scholar
    • Export Citation

    U.S. EPA (U.S. Environmental Protection Agency). 1999. The Benefits and Costs of the Clean Air Act: 1990–2010.Washington, DC: Office of Policy Analysis, U.S. EPA.

    • Search Google Scholar

    VelicognaIsabella andJohnWahr.2006. “Measurements of Time-Variable Gravity Show Mass Loss in Antarctica.”Science311 (5768): 175456.

    • Search Google Scholar
    • Export Citation

    WatsonJimGordonMacKerronDavidOckwell andTaoWang.2007. “Technology and Carbon Mitigation in Developing Countries: Are Cleaner Coal Technologies a Viable Option?”Occasional Paper 2007/16Human Development Report OfficeUNDP, New York.

    • Search Google Scholar
    • Export Citation

    WeitzmanMartin L.2007. “Role of Uncertainty in the Economics of Catastrophic Climate Change.”Working Paper No. 07-11AEI Center for Regulatory and Market Studies (formerly AEI-Brookings Joint Center) Washington, DC.

    • Search Google Scholar
    • Export Citation

    WheelerDavid.2007. “Greenhouse Emissions and Climate Change: Implications for Developing Countries and Public Policy.”Mimeo. Center for Global DevelopmentWashington, DC.

    • Search Google Scholar
    • Export Citation

    WheelerDavid andKevinUmmel2007. “Another Inconvenient Truth: A Carbon-Intensive South Faces Environmental Disaster, No Matter What the North Does.”CGD Working Paper 134Center for Global DevelopmentWashington, DC.

    • Search Google Scholar
    • Export Citation

    Woods Hole Research Center.2007. “Reducing Emissions from Deforestation and Forest Degradation (REDD): The Costs and Benefits of Reducing Carbon Emissions from Deforestation and Forest Degradation in the Brazilian Amazon.”A Report for the United Nations Framework Convention on Climate Change (UNFCCC) Conference of the Parties (COP), Thirteenth SessionBali, IndonesiaDecember3142007.

    • Search Google Scholar
    • Export Citation

    World Bank. 2006. Clean Energy and Development: Towards an Investment Framework. April232006.

    World Bank. 2007a. World Development Indicators 2007.Washington, DC: World Bank.

    World Bank. 2007b. “IDA and Climate Change: Making Climate Action Work for Development.”World BankWashington, DC.

    World Bank. 2007c. “Global Economic Prospects 2007: Managing the Next Wave of Globalization.”World BankWashington, DC.

    • Search Google Scholar

    World Bank. 2007d. World Development Report 2008: Agriculture for Development.Washington, DC: World Bank.

    WWF (World Wildlife Fund). 2006. Living Planet Report 2006.Geneva, Switzerland: WWF International; London: Institute of Zoology; Oakland, CA: Global Footprint Network.

    • Search Google Scholar
    • Export Citation

    WRI (World Resources Institute). 2007. “Climate Analysis Indicators Tool (CAIT) database 5.0.”World Resources InstituteWashington, DC.

    • Search Google Scholar
    • Export Citation

    YoheG.W. andR.S.J.Tol2002. “Indicators for Social and Economic Coping Capacity–Moving Towards a Working Definition of Adaptive Capacity.”Global Environmental Change12 (1) 2540.

    • Search Google Scholar
    • Export Citation

    Other Resources Citing This Publication