Scenarios of Future U.S. Agricultural Production and Technology and Their Environmental Costs

There is mounting concern about the long-term impacts of U.S. agriculture on the nation's environment. The concern centers mainly on the reduction of soil productivity and water quality because of soil erosion, the pollution of ground and surface waters by pesticides and fertilizers, and the loss of wildlife habitat through the clearing of woodlands and the draining of wetlands for crop production.

It is useful to think of these environmental consequences of agriculture as the outcome of five sequentially linked factors: (1) the quantity of production; (2) the technologies and management practices used in production; (3) the effects of these technologies and practices on the physical productivity of land and water (such as the number of kilos of corn produced per hectare per year), on the quantity and quality of plant and animal habitat, on concentrations of wastes (such as sediment and agricultural chemicals) in water runoff from land and in groundwater; (4) the effects of these wastes on the quantity and quality of the receiving media—land, water, and the atmosphere—and on human health; and (5) the effects of the third and fourth factors on the social value of land, water, and atmospheric resources.

The consequences of the third and fourth factors are measured in physical units, such as a hectare of lost habitat on a unit increase in a measure of water turbidity. The consequences of the fifth factor are measured in dollars—that is, the dollar value of the physical damages resulting from the third and fourth factors.

For environmental policy, it is the social value of the damages that counts. There is an implicit philosophical assumption here—not shared by all—that policy legitimately deals only with damages to resources that humans value. Thus an increase in the turbidity of water that has no direct or indirect value to humans, either in the present or in the future, is not a reason for policy action to clean up the water.

Many losses of social value are not reflected in market transactions. Economists have developed survey and other techniques to estimate social values, but comprehensive estimates—for example, of the social value of annual losses of wildlife habitat when farmers drain wetlands—are not available. In recognition of this, much environmental policy uses physical measures to set standards of maximum acceptable environmental damage. For example, a concentration of nitrate-nitrogen in groundwater in excess of 10 parts per million is considered a threat to humans consuming the water and particularly to babies in whom ingestion may cause methemoglobinemia ("blue baby syndrome").

The focus in this article is primarily on physical measures of environmental damage due to agricultural production. In this connection it is important to note that the social value of damage can change even if the physical measures of damage do not. Because of rising population and income in the United States, demand for the nation's environmental resources increases over time. Consequently, the social value of a given amount of damage would tend to rise. For example, with the demand for recreational uses of water increasing, the social cost of a given amount of turbidity in water would tend to rise. Constancy in physical measures of environmental damages thus would likely mask increasing losses of social value wherever demand for environmental resources is increasing.

Scenarios of future production and technology

Scenarios are useful for organizing thinking about what the future may hold. They are not predictions, but sketches of certain key features of the future situation based on plausible extensions of existing trends. The following scenarios project the state of U.S. agricultural production and technology and the environmental consequences of this production and technology for the year 2010.

Future U.S. agricultural production will reflect domestic and export demands for food and fiber. Population growth is the main driver of domestic demand. Population projections by the United Nations indicate about a 13 percent increase in the U.S. population between 1990 and 2010. Increases in income can also drive demand for food, but because most Americans now are adequately nourished, these increases will add little to demand.

Exports reflect both the growth of demand in foreign markets and the ability of American farmers to compete in those markets. Foreign demand for food reflects population growth outside the United States and per-capita income growth in the less developed countries (LDCs). According to United Nations projections, 95 percent of global population growth between 1990 and 2010 will be in the LDCs. Because millions of people in these countries are ill-nourished, increases in their incomes would stimulate increases in demand for food.

The scant evidence available suggests that the expansion of agricultural production over the last forty years has resulted in significant environmental harm. Sediment from agricultural land has caused damage to the quality of the nation's surface waters, although the quality of these waters may have improved due to federal and state policies to reduce industrial and municipal sources of water pollution.

The expansion of production also appears to have caused serious losses of wildlife habitat. Farmers moved away from diversified crop-animal production systems toward more homogeneous crop production systems and removed fences and hedgerows to accommodate larger farm machinery. By reducing vegetative diversity on the farm, these changes had unfavorable impacts on wildlife habitat.

From the mid-1950s to the 1980s, farmers also drained some 12 million to 15 million acres of wetland, primarily in the Mississippi delta and in the northern plains states and Minnesota. Wetlands provide rich habitat for a wide range of wildlife, but especially for migratory waterfowl. Studies show that the number of these birds declined from some 44 million in 1972 to about 28 million in 1985. The loss of wetlands could not be the sole reason for the decline, but it likely was a major contributor.

It is fair to say that when the American people contemplate the environmental consequences of agriculture they think not about damages to the supplies of land and water, to the quality of water, or to natural habitats but about threats to human health from fertilizers and pesticides. The U.S. Environmental Protection Agency's nationwide survey of pesticides and nitrates in wells containing water for human consumption suggests that this may be a misreading of the relative importance of the various kinds of environmental damage. The survey revealed that only about 1 percent of the wells had pesticide concentrations high enough to pose a threat to human health.

It also revealed that about 2 percent of the nation's people (some 4.5 million) drink well water in which nitrate concentrations exceed the U.S. Public Health Service's standard of 10 parts per million. These findings are for numbers of people exposed to contaminated water; they say nothing about how many people may actually have been injured by drinking the water. While the numbers appear small in relation to the total U.S. population, their significance must be carefully assessed. Are damages, or potential damages, to human health to be weighed on the same dollar scale as other environmental damages—for example, an X-million-dollar-per-year loss in recreational uses of water because of sediment damage? Economists, philosophers, and others argue about the question. If, however, the costs of pesticide and nitrate poisonings are valued by the costs of treatment and time lost from work due to illness and death, they likely would not be large in relation to the costs of sediment damage and habitat loss.

If the sequential linkages between agricultural production and the social costs of the resulting environmental damage remain the same as they are at present, then the roughly 34 percent baseline increase in production between 1987/1991 and 2010 would impose a proportional increase in environmental costs. The linkages are not likely to remain the same, however. Changes in agricultural technology and in environmental policies will alter the physical connection between production and damage, and increases in demand for environmental resources will tend to increase the social costs per unit of damage measured in physical units.

The concern of American people about environmental protection almost surely will move the country's agricultural technologies in a less environmentally damaging direction. Research already is under way to find weed and insect control systems less dependent on synthetic inorganic pesticides, although success in making these alternative systems economically attractive to farmers still is in doubt. Should the effort fail, the American people may nonetheless place constraints on pesticide use, accepting the higher costs of food as the price for improvements in the quality of environmental resources.

Trends in fertilizer prices and in technologies of fertilizer use suggest that the amounts of fertilizers used per acre will slowly increase over the next twenty years. The possible effects on human health of the resulting increase of nitrate concentrations in groundwater do not appear seriously threatening, but would bear watching. However, more nitrogen and phosphorus carried by runoff into lakes, reservoirs, and estuaries would stimulate the growth and subsequent decay of aquatic plants, with consequent damage to the quality of these waters. Given continuing increases in the demand for recreational uses of water, the social cost of this damage likely will rise faster than the physical measure of the damage.

For several decades American farmers have been moving toward greater use of tillage techniques that reduce soil erosion by keeping more crop residues on the soil surface. Studies have shown that on sloping land these techniques can reduce erosion by 50 to 90 percent. Both economics and government policy likely will encourage even wider adoption of these techniques, with consequent reductions in soil erosion. However, less soil erosion would not directly translate into less damage from sediments, at least not in the short run. Much previously eroded soil is stored in the many nooks and crannies scattered around the landscape. Consequently, when erosion on farmers' fields is reduced, the water runoff from the fields is free to pick up this stored soil and carry it downstream. Thus deliveries of sediment to places where it causes damage might not be reduced. Even if the deliveries are reduced, the rising demand for recreational uses of water could nonetheless increase the social cost of damage from sediment.

The small increase in cropland in the baseline scenario may imply more habitat loss, particularly if the increase comes from land now in forest or in wetlands. Whether losses in fact are incurred will depend in large part on federal and state policies. The 1985 and 1990 farm bills contained provisions to withhold government program benefits from farmers who violate policies to protect wetlands. And the Bush administration has adopted a policy of no net loss of wetlands. The effectiveness of these various policies has been questioned, but they clearly are aimed in the right direction. As with the case of sediment damages, however, the continuing rise in demand for wetland services could increase the social cost of habitat loss even if the loss in terms of acres were reduced.

In the high export demand scenario, production of grains and soybeans is 95 mmt (22 percent) above the baseline scenario. This scenario clearly would imply greater environmental costs than the baseline scenario unless, because of higher costs, major breakthroughs occurred in pest management, fertilizer use practices, tillage techniques to reduce soil erosion, and measures to improve the productivity of remaining habitat. Absent such breakthroughs, either environmental costs would be greater in the high export demand scenario than in the baseline scenario—probably proportionately more than the difference between production in the two scenarios—or government policies for controlling the environmental costs would push up the economic costs. Either way, the social costs of the high export demand scenario would be sharply higher than those of the baseline scenario.

In the low export demand scenario, production of grains and soybeans is 62 mmt (15 percent) below the baseline scenario. The trends in pest management and in tillage techniques discussed above suggest that the low export demand scenario would imply reduced physical damage to environmental resources by 2010. Although the rising demand for these resources likely would increase the social cost of damage per unit of physical damage, the low export demand scenario by and large does not appear to present a serious threat to the nation's environmental resources.

On balance, the American people are right to be concerned about the implications for the environment of future demands for the nation's agricultural output. The United States is not likely to face a crisis of economic or environmental costs in meeting these demands. But alertness to emerging signs of pressures that would increase these costs, and readiness to move to control these pressures, clearly are called for. Hysteria in misreading the signs of pressure will only make things worse. Coolness in assessment and prudence in action should be our bywords.