Some developing countries have reached the limits of land expansion. India's population has more than doubled since 1950, but cropland has only expanded by 15 percent or so. Crop yields have increased, and multiple cropping has increased to enable more than a doubling of agricultural product.
For a number of developing countries, this historical process has not yet been completed, however. In parts of Southeast Asia, most of sub-Saharan Africa, and parts of Latin America, notably Brazil (and Colombia), cropland and pastureland expansion continues.
Land Use and Carbon Sinks
Cropland and pastureland constitute significant carbon sinks even though they do not store much carbon in vegetation. The carbon in the soil for cropland and pasture is significant and can actually be higher than for certain woodland types and for semiarid savannah-type lands. Significant carbon is stored in animal stocks as well.
Historically, cropland and pastureland expansion in the temperate zone countries has tended to be "sink-reducing" (expansion against forests) or "sink-neutral" (expansion on prairie lands). In countries still in the expansion process, it is probably on balance sink-reducing, but there is quite considerable expansion that is either sink-neutral or sink-expanding. In addition, most of the expansion on savannah-type land has probably also been sink-expanding. Population change, technology, and government policies affect land use patterns. Typically, as population grows (with constant technology of production) relative to land resources, cropland expands at the expense of other land uses. As land best suited to cropland (and pasture)
is settled (i.e., the frontier is closed), various land-saving options are employed. Improved varietal technology (high-yielding varieties) reduces the pressure on resources. The combination of changing economic conditions and new technology brings cropland expansion to a halt in developed countries. Cropland expansion in farms in the United States stopped around 1920, and cropland area has declined in recent years. (Farm production has tripled since 1920.) This is also true for pastureland. The same situation holds in Europe generally.
Large areas of savannah-type land exist in sub-Saharan Africa and in the local Cerrado-Llianos region in Brazil and Colombia (with some in Bolivia and Paraguay). Agricultural research programs in these countries have sought to achieve efficient land use expansion and have been successful in facilitating land expansion in the Cerrado regions in Brazil. This expansion has also been fueled by subsidized credit, which has fueled expansion in the Amazon, where it is sink-reducing. On balance, the agricultural research systems in Brazil and Africa have facilitated expansion on sink-neutral or sink-expanding areas. It is not clear that any policies can materially change some of the land use patterns that will occur in much of Africa over the next few decades. Populations are growing at rapid rates, and few countries have effective family planning. To the extent that improved agricultural technology can be developed, it will alter the ultimate course of expansion of cropped areas. Industrial development and nonfarm employment opportunities for workers will, as well. Much of this expansion will be sink-neutral, however, because savannah lands are not large sinks. The most severe problems will be associated with desertification and the management of shorter fallow systems on savannah soils.
In developed and developing countries alike, however, even if oil prices do not rise appreciably over the next two decades, continued technological improvements are likely to bring some biomass energy options into the competitive range. No major breakthroughs are necessary (although some may be achieved). Continued support for well-established plant breeding and agronomic research programs is required to bring this biomass energy option.
Agricultural Greenhouse Gas Mitigation
For purposes of assessing the relative impacts of U.S. emission controls and controls in other countries, the range of emission reduction (million tons of carbon) from a 10 percent reduction in rice production or ruminant production in different regions is used in this analysis.1 The United States is a minor contributor of CH4 from rice paddies and contributes virtually
nothing from work animals. It is an important source of CH4 from other ruminant animals (as are other industrialized countries).
The mechanisms by which reductions in CH4 emissions in the United States and in other countries can be achieved include the following options:
1. Elimination of existing subsidies that stimulate more of the activity than dictated by market equilibrium conditions.
2. Taxation of the CH4-emitting activity.
3. Quantitative regulation of activities (i.e., through quotas on production and trade) or regulations regarding burning and waste management.
4. Buyouts, either through purchase of assets (e.g., rice paddy land, dairy cows, or pastureland) or payments to induce alternative activities (e.g., paying rice farmers not to produce rice, but allowing them to produce an alternative crop on the land).
These mechanisms vary in effectiveness and cost per unit of CH4 reduction achieved. The lowest-cost option is generally option 1 because subsidies induce inefficient resource use, and their elimination is actually an economic gain. However, they exist because interest groups have used political power to put them in place. Thus their elimination has political implications.
Option 2 is costly from a consumer's standpoint because it induces inefficient resource use due to market distortions. A producer will produce less of a taxed good than an untaxed good. Therefore from a consumer's standpoint, too little of the taxed good is produced. Calculation of these inefficiency costs is complex and requires estimates of supply and demand responses to a tax.
Option 3 also causes inefficiency losses, and this option, too, requires complex cost calculations.
Option 4 has the seeming merit of being directly calculable. For example, a government agency might pay a rice farmer $100 per acre not to produce paddy rice but allow him to produce an alternative crop on the land. One could then compute the CH4 emission from an acre of paddy land and arrive at a cost per ton of CH4 mitigated. Alternatively, a government agency might purchase rice paddy land (for $5000 per acre) and leave it idle (or reforest it). Then the investment could be amortized at alternative interest rates, and a cost per ton of CH4 reduced can be obtained. The difficulty with this mechanism is that other rice farms might respond to this action by producing more paddy rice.
There are some options associated with farming practices, particularly minimum tillage options, but the scope for extensive adoption of these practices is limited because most farmers are aware of them and have tested them. Where they have been found effective, they have already been adopted.
In addition, as with production practices generally, use of these practices tends to depend on pricesespecially the price of energy.
It should be further noted that some options have consequences that may be undesirable. Substitution of tractors for work animals may eliminate CH4 emissions, but constitutes an increase in fossil fuel use and CO2 emissions (and some increase in chemical fertilizer use). Swampland drainage has consequences for wildlife and species diversity (Matthews and Fung, 1987).
Options for Rice Paddies
Almost all of the approximately 1 million hectares (ha) of production in the United States is irrigated paddy rice. All water regimes, except upland, produce paddy rice (i.e., rice grown under standing-water conditions). There are no practical options to grow alternative crops in deep- and medium-water rainfed regimes. Perhaps half of the shallow-water-rainfed and irrigated regimes could be shifted to alternative crops but at some cost. Upland rice is not an alternative crop to paddy rice. It is produced under conditions quite different from paddy riceusually on semiarid land. Even if land is no longer being used for paddy rice, it is highly unlikely to be planted with upland rice.
The United States has a little less than 1 percent of the world's paddy rice land but produces about 1.3 percent of the world's paddy rice. It accounts for approximately 20 percent of world rice exports. Approximately 90 percent of the world's paddy rice production is in Asia, with China, India, Indonesia, Bangladesh, Thailand, Vietnam, and Japan being the leading producers (International Rice Research Institute, 1988).
Several rice-importing countries intervene in rice markets with high tariffs to protect domestic rice producers. The ratio of domestic prices to world prices was over 7 in Japan in 1985. South Korea, Taiwan, and most European economies also protect domestic producers (with ratios of domestic to world prices greater than 2 (World Bank, 1987)). Elimination of this protection in these countries would result in lower domestic prices, decreased domestic production, more imports, and increased domestic consumption. World rice prices would rise in response, and this would reduce consumption marginally in other countries. The net effect of the removal of such protection on global CH4 emissions would probably be quite small.
Exporting countries, such as the United States, cannot maintain domestic prices at levels above world prices except at high costs to subsidize exports. Consequently, the United States has a relatively modest program of subsidies to rice producers. A price support program is in place, with support prices roughly 30 percent above export prices (Gardner, 1987). Elimination
of these subsidy programs would reduce U.S. production and exports, but would be partially offset by increases in production in other countries.
Taxation of paddy rice in the United States could be undertaken, but would be difficult politically because subsidies are now in place. It should be noted, however, that the other major rice-exporting country, Thailand, has used a rice tax (called the rice premium) for years to reduce rice exports, realize a higher export price, and increase government revenues. If such a tax were imposed on U.S. rice farmers (presumably after subsidies had been eliminated) and the supply response elasticity to the tax were significantly negative (Gardner, 1987), a reduction in U.S. production and CH4 emissions could be achieved. If domestic prices did not change (i.e., were determined by world prices), domestic consumption would not change, so the full effect would be felt in reduced exports. Because the United States is a leading exporter, this could have an impact on world rice prices. The cost of this tax would be loss of the export revenue (10 percent of the value of production) minus the value of other crops that could be produced on land formerly devoted to paddy rice (0.95 × 10 percent of the value of production) (Gardner, 1987). This assumes a demand elasticity of -0.5 and a supply elasticity of 0.5 (see Barker and Herdt, 1985). Thus a 10 percent tax on U.S. paddy production would reduce emissions by 200,000 t C as CH4/yr (computed as 2 t C/ha). The cost would be 0.005 × 6 Mt × $300/t, or $9 million, giving a cost per ton of carbon of $45 (the cost per ton CH4 is $16).
A quota system has been suggested as a way to reduce production and has been used in a number of countries for other products. This could also be applied to paddy rice producers (Johnson, 1990). Licenses might be required to sell paddy rice; these could be traded and the total available licenses to farmers reduced by 10 percent (or some other level). This option would have the same costs as the tax option, provided the licenses were negotiable.
The buyout options for paddy rice in the United States are actually the simplest to analyze. A government agency would have two alternatives:
1. Purchase rice paddy land directly from farmers and convert it to idle land or to some other use (e.g., it could plant trees on the land, although some rice paddy land is probably poorly suited to tree production);
2. Offer a payment to buy the land out of paddy production for 1 or more years and allow farmers to produce an alternative crop.
The option of buying paddy land out of production is the more costly of the two, but it could complement other policies. Paddy land would probably cost $7,000 to $10,000/ha. The annualized interest costs at 3 percent would be $210 to $300, and approximately 2 t CH4 (carbon equivalent) would be mitigated. At 6 and 10 percent interest rates the costs would be
$420 to $600, and $700 to $1000. The cost per ton of carbon mitigated would range from $150 to $500.
The option of yearly or multiyear arrangements to pay farmers for not producing paddy rice but allow them to produce alternative crops would depend on the suitability of the land for alternative crops. For land where the substitution could be made easily, payments could be modest (e.g., $100/ha). For land where drainage and other modifications would be required, these payments could rise to the full buyout option costs. Water pricing policies have also affected paddy rice production in California. Farmers have access to water at rates far below the real value of water. If water were priced at market rates, much of the California rice production would be uneconomical. It would be a wise policy for all parties to devise a compensation scheme to enable more efficient water pricing, and this would reduce rice production and CH4 emissions at little or no cost.
Other countries would incur similar costs if they were to attempt to reduce rice production. It would be difficult to manage the tax options in countries that do not export (where the tax can be levied on the exported goods). Taxation leading to rice price increases could also have severe implications for large low-income populations, where rice is often the staple food. It would also be difficult to manage a coupon or quota system in countries where rice is consumed by the households producing it (most developing countries).
Thus the realistic options in developing countries are the buyout options, and these, if pursued in substantial degree, will have the consequence of raising rice prices, which will induce more conversion of nonrice areas to rice, partially offsetting the effects of the reductions. The options for increasing upland nonpaddy rice are quite limited, in part because little technological progress has been made in upland rice production.
Options for Ruminant Products
A shift from ruminant products (dairy products, beef, and mutton) to cereal-based products would reduce CH4 emissions. The United States is a major producer and consumer of ruminant products, and several mitigation options are open to it. In fact, a number of these options have been pursued, although not to mitigate CH4 emissions. The United States, Canada, and the European Economic Community (EEC) countries have been intervening in dairy product markets for many years to achieve prices to producers (and consumers) that are higher than equilibrium market prices. This has been undertaken via import controls in EEC countries and via support prices and regulated trade in milk markets in the United States and in Europe as well (Barichello, 1984).
When prices are supported above market equilibrium levels, consumers
wish to consume less and producers wish to produce more than equilibrium levels. This results in the accumulation of surpluses unless supply control measures are taken. Such surpluses have been a common phenomenon in the United States, Canada, and Europe, and in general, these countries have probably produced as much dairy products as would have been the case in an unregulated market even though consumers have consumed less (World Bank, 1987).
Quantitative supply control programs could be employed to reduce both production and consumption further, however. Taxes on dairy products or on meats, except for the normal sales taxes that affect all foods, would also discourage consumption. Political factors, however, produce subsidies and price support systems, not taxes, in the United States and European economies (Johnson, 1990).
Livestock commodities other than dairy products have experienced less government program intervention because of the high costs of surplus storage.
The elimination of existing (costly) dairy support and feed grain support programs (which indirectly support meat production) in the United States (and the EEC) would lead to increased consumption of ruminant products and probably to increased ruminant production and CH4 emissions. A tax on ruminant consumption and production would probably be politically unacceptable. It could achieve CH4 mitigation, however. The cost of a tax using standard measurement techniques (see Gardner, 1987) would depend on supply and demand responses to the tax.
Estimates of demand responses (Gardner, 1987) range from -0.4 to -0.7 (i.e., a 10 percent tax would reduce consumption by 4 to 7 percent). Few estimates of supply responses are available, but it is reasonable to postulate a relatively high long-run supply response. Thus a 10 percent tax could reduce consumption by roughly 5 percent (based on medium demand estimates) (Gardner, 1987).
Quota systems have been used to control dairy production in Canada but have generally been costly to monitor and administer. If effective, they have the same efficiency costs as the ruminant tax option but different income options (Barichello, 1984).
Buyout options to reduce dairy product supply have also been used in the United States. These options have generally been ineffective because the compliance of nonparticipants cannot be ensured.
Options for Work Animals
Approximately 300 million of the 970 million cattle and buffalo worldwide are used primarily as work animals. Most south Asian and sub-Saharan African farms are not yet mechanized. It is generally thought that
existing subsidy programs, particularly credit subsidies, induce the substitution of machines (tractors) for animals and thus encourage ''overmechanization."
Such subsidies could futher reduce the world's work animal stock and thus CH4 emissions; however, fossil fuel use could increase, reducing the greenhouse gas benefit from the subsidies. Thus a trade-off between reduced CH4 and increased fossil fuel use must be addressed. On balance, it is probably not wise to encourage more mechanization in developing countries (Binswanger, 1986).
Options for Biomass Burning
Biomass burningto clear land for agricultural production or to carry out general farm management activitiescontributes to CH4 emissions. Burning in sugarcane fields is a low-cost way to reduce trash and facilitate harvest and processing. Rice straw and other plant residues are sometimes burned to lower the cost of plowing and land preparation even though burning reduces the amount of organic matter in soil.
Thus, although some biomass burning by farmers may constitute poor management, most burning by farmers is done for cost considerations. Regulation of this burning (e.g., banning some of the burning in specific situations) is probably not costly and in some cases may actually bring about managerial improvement. Thus selective judicious controls on biomass burning may be cost-effective in CH4 (and N2O) mitigation.
Options for Biogas from Animal Waste
Confined animal production systems require special waste management practices and offer some potential for the production of CH4 biogas. Most cities and counties in developed countries regulate waste management largely for pollution reasons. Waste from animals is used widely as a fuel in many developing countries and as organic fertilizer in most countries. Biogas projects have been implemented in many countries but have not attained widespread use.
Further judicious regulations and technological improvements in biogas production will achieve some mitigation of CH4 (and replacement of fossil fuel), but these are not likely to be large effects.
Options for Fertilizers
Several studies of fertilizer demand estimate that a 10 percent increase in price (from a tax) would decrease use by roughly 5 percent (Gardner, 1987). The efficiency cost of such a tax would be only 0.0025 percent of the total fertilizer value if farmers were in equilibrium and the long-run supply of
fertilizer were perfectly elastic. If such a tax were applied on all nitrogen fertilizer in the United States, it would have an efficiency cost of $25 million and would reduce N2O emission by 50,000 t N/yr at a cost of $500/t N.
The cost-effectiveness of the policy measures described above is summarized in Table 25.4. To determine the cost-effectiveness in terms of CO2 equivalence, the U.S. information in Table 25.4 is used, and CH4 and N2O emission reductions are weighted by the global warming potential factor (21 for CH4 and 190 for N2O (per discussion of global warming potential in Chapter 19). Therefore
(3 × 106 t C as CH4)(16 CH4/12 C)(21 CO2/1 CH4) = 84 × 106 t CO2 eq. ($100/t C as CH4)(12 C/16 CH4)(1 CH4/21 CO2) = $3.6/t CO2 eq.
(4.5 × 106 t C as CH4)(16 CH4/12 C)(21 CO2/1 CH4) = 126 × 106 tCO2 eq. ($150/t C as CH4)(12 C/16 CH4)(1 CH4/21 CO2) = $5.4/t CO2 eq.
(0.05 × 106 t N as N2O)(44 N2O/28 N)(290 CO2/N2O) = 23 × 106 t CO2 eq. ($500/t N as N2O)(28 N/44 N2O)(1 N2O/290 CO2) = $1.1/t CO2 eq.
1. Throughout this report, tons (t) are metric; 1 Mt = 1 megaton = 1 million tons.
Barichello, R. R. 1984. Analyzing an agricultural marketing quota. Discussion Paper 454. Economic Growth Center, Yale University, New Haven, Conn.
Barker, R., and R. Herdt. 1985. The Rice Economy of Asia. Washington, D.C.: Resources for the Future.
Binswanger, H. 1986. Agricultural Mechanization: A Comparative Historical Perspective. The World Bank Research Observer 1. Washington, D.C.: World Bank.
Gardner, B. 1987. The Economics of Agricultural Policies. New York: Macmillan.
Hayami, Y., and V. W. Ruttan. 1985. Agricultural Development, An International Perspective. Baltimore: John Hopkins Press.
International Rice Research Institute. 1988. World Rice Statistics. Los Ranes, Laguna, Philippines: International Rice Research Institute.
Johnson, D. G. 1990. World Agriculture in Disarray. Chicago: University of Chicago Press.
Matthews, E., and I. Fung. 1987. Methane emissions from natural wetlands: Global distribution, area and environmental characteristics of sources. Global Biogeochemical Cycles 1:61–86.
World Bank. 1987. World Development Report. Washington, D.C.: World Bank.