Agricultural impacts of global warming: discussion.

It is indeed a pleasure to discuss three excellent papers on the potential effects of atmospheric greenhouse gas accumulation on the agricultural sector around the world. Our discussion proceeds in three parts, beginning with a synthesis of conclusions from this session on important ingredients

Climate Change Effects on Agriculture

Most of the extant analyses of climate change have sought to predict the effects on a variety of phenomena of doubling the level of greenhouse gases in the earth's atmosphere. Such a doubling could occur in the next fifty to one hundred years if measures are not implemented to curb future greenhouse gas emissions. To predict the potential effects of such greenhouse gas accumulation on agricultural production and prices requires extremely complex modeling with the following ingredients:

1. Effects of greenhouse gases on climate itself, including ambient temperature, precipitation, ENSO cycles, and climatic events such as cyclones and hurricanes. Such effects on the climate system may result in a complicated mosaic of changes in climatic conditions with a wide geographical variance and strong localized effects driven primarily by precipitation patterns.

2. Physical responses of agricultural systems to greenhouse gases and climate change. Such responses include the so-called C[O.sub.2]-fertilization effect and potential changes in pest problems from warmer and more humid conditions.

3. Behavioral responses to climate change for given technologies and institutions. Such adaptations may include not only farmer shifts in cropping patterns, planting dates, tillage practices, irrigation techniques, and other management approaches, but also regional adaptations in, for example, water delivery systems.

4. Technological change, which may include both exogenous improvements in agricultural productivity and changes in technology induced by climatic changes.

5. Equilibrium effects in international agricultural markets.

No analysis, thus far, has successfully incorporated all of these ingredients. However, the evidence so well surveyed by this session's participants suggests some general conclusions:

1. Estimated costs of greenhouse gas accumulation in the agricultural sector depend crucially upon the global circulation model that one uses, with such costs varying by large orders of magnitude across different models (Reilly, table 1).

2. In terms of agricultural impacts, developing countries are potentially much larger losers than developed countries such as the United States (Kaiser and Crosson, Reilly).

3. Farmer adaptation to climate change (with alternative cultivation practices that are now available) may significantly reduce agriculture-related costs of climate change, particularly in the United States (Kaiser and Crosson, Reilly).

4. In large part because the agricultural sector has a relatively small share of the overall economy, particularly in developed countries like the United States, agricultural impacts of greenhouse gas accumulation are likely to be small when measured against overall economic activity (Kaiser and Crosson, Reilly, p. 3).

5. The economic benefits of the C[O.sub.2]-fertilization effect (which enhances agricultural production) may be important in offsetting other agriculture-related costs of global warming (Reilly). However, these benefits may be exaggerated by a failure to account for yield costs of open field conditions, including increases in levels and variability of temperature and pest outbreaks that may attend greenhouse gas accumulation (Kaiser and Crosson).

6. Finally, induced innovation in response to changes in climatic conditions may potentially mitigate adverse effects of global warming on agriculture, but estimates of such impacts must await empirical implementation of models such as the one developed here by John Antle.

Problems and Opportunities in Climate Change Analysis

In reading this session's three papers, one is struck by the difficulty of coming up with believable estimates of greenhouse gas impacts on agriculture. Beginning with the very first ingredient in such an estimation exercise, there is tremendous uncertainty and little confidence about the specific climatic effects of greenhouse gas accumulation in the atmosphere. Beyond impacts on average global temperatures, climate change models are hampered by poor resolution (and attendant inability to predict local climatic conditions), an inability to predict rates and magnitudes of climatic change, and little reliability in predicting precipitation, cloud cover, episodic climatic events, and other climatic variables that are important in the agricultural context (Antle). Building on weak climate models with the many other necessary ingredients to a plausible agriculture sector model of climate change effects, potential specification errors make matters worse. Such difficulties may argue for a research agenda that focuses at least as much on the search for qualitative insights into climate change effects as on quantitative estimation.

In searching for qualitative insights, the papers in this session identify two striking features of the climate change problem: (a) the tremendous uncertainty about the likely effects of greenhouse gas accumulation, and (b) a dynamic process of gradual change. Both aspects of the problem merit research attention.

To a large extent, policy measures to reduce or restrain growth in greenhouse gas emissions are motivated by desires for insurance against very adverse outcomes. Such concern motivates thinking about probability distributions for possible outcomes, and of the likelihood of particularly bad outcomes, at least as much as about possible average steady-state conditions that may result from greenhouse gas accumulation. Conceptual models that incorporate uncertainty and dynamics are necessary to think about probabilistic outcomes and may also help us to understand the policy implications of several important phenomena:

1. Learning about climate change over time. Can we wait until we learn more about climate change before we adopt costly greenhouse gas reduction strategies?

2. Innovations in greenhouse gas abatement technologies over time. Can we wait with costly emission controls until we have cheaper abatement opportunities? Can we pursue strategies that lead to cheaper abatement opportunities in the future?

3. Opportunities for "closed loop" decision making that responds to improved knowledge and technology, and permits assessment of desirable policy strategies to pursue today in view of opportunities for future policy responses to threats of climate change. Again, do we need to act today or should we wait until we learn more? Can we undertake cheap measures that enhance our ability to respond to bad outcomes?

4. Risk, risk attitudes, and the desire for "insurance." For given risk attitudes and costs of different levels of climate change insurance, how much of this insurance should society purchase?

Of course, these issues take us beyond the agricultural sector and thus permit us to raise a general question that is motivated by casual empiricism but that may merit more rigorous investigation. Even if economic activity were not much affected by climate change, social welfare could be. Residents of Tucson, for example, may be willing to pay quite a bit to avoid increased summertime temperatures, just as residents of Seattle may be willing to pay quite a bit to avoid increased cloud cover. Similarly, residents in the Northeast may be willing to pay quite a bit to obtain warmer winters. Even with endogenous location choices, the papers in this session suggest that such costs and benefits of climate change may swamp agricultural impacts.

Policy Issues

The international debate on future actions to reduce greenhouse gases and vulnerability to climate change is continuing to evolve. In March of 1995, the first Conference of the Parties to the Framework Convention on Climate Change took place in Berlin, Germany. In light of continuing discussions on the appropriate timing and types of future actions to be taken, the policy conclusions of this session's papers merit emphasis and some elaboration.

Research

Kaiser and Crosson rightly emphasize the importance of supporting research that may aid in agricultural adaptation to climate change. However, agriculture may not only be a victim or beneficiary of climate change, it may also contribute to greenhouse gas reductions. For example, the Clinton administration is exploring the possibility of bioenergy as an important new source of electricity production in the next century, supplanting the burning of some fossil fuels, which generates greenhouse gases. Research, development, and demonstration projects can potentially lead to new bioenergy-generating technologies and feedstock crop systems that make biomass power generation economically viable.

With respect to research on climate change itself, the United States and other international research programs are heavily weighted toward understanding the climate system, with an aim toward improving future predictions. Such programs are arguably weakest in modeling feedbacks from human activities, including effects of trends toward greater urbanization and deforestation on local and regional climates. The dramatic rate of loss of the rain forest in Latin America and elsewhere is altering regional hydrologic systems and ultimately influencing regional climate systems. Economic incentives and culturally motivated practices are in large part driving changes in land use. Understanding incentives and responses by individuals, companies, and governments in developing countries will strengthen the human behavioral component of feedbacks to the climate system.

The value of such research endeavors will, to a great extent, hinge on the quality of data. Antle rightly points out the importance of collecting agricultural economic data that are compatible with spatially referenced scientific data. One trend to note is the curtailment of natural resource survey data efforts as research budgets have grown smaller over the past few years. Yet, as every one of this session's papers attests, the complicated modeling requirements of climate change research demand improved environmental and economic data.

Free Markets and Free Trade

Kaiser and Crosson point out that U.S. farm policy merits reform to increase farmers' flexibility in responding to climatic changes without financial penalties that government programs may potentially give to such responses. For example, farmers can adapt to warmer and/or drier growing conditions by planting alternative crops, adopting soil moisture-conserving tillage and rotation practices, cultivating less intensively or less frequently in some areas, and entering or exiting from crop production in areas made more or less viable for crop production due to climate change (see Lewandrowski and Brazee, as cited in Kaiser and Crosson). Current U.S. deficiency payment programs may inhibit some of these adaptations by providing farmers with incentives to continue to plant program crops to preserve base acreage in program commodities and obtain deficiency payments on base acreage planted to program crops. Price support programs may also inhibit climate change adaptation. In fact, in recent years, price support payments have been turned into classic deficiency payments, with all of the same economic costs, by so-called marketing loan provisions. Today, farmers are paid any positive difference between the support price for any program commodity and the international market price. For years, economists have criticized these commodity programs as impeding efficiency in agricultural markets by interfering with market signals of social values. Impediments to climate change adaptation add another cost to current program structures, and thus add more fuel to the case for farm policy liberalization in the coming farm bill.

Kaiser and Crosson also emphasize the importance of free trade in moderating climate change effects, particularly the effects on food consumption patterns. International commodity markets smooth the price effects of production and consumption shocks, so changes in the patterns of food consumption induced by climate change thus are tempered by open trade. Efficient adaptation in crop production patterns is also promoted by free trade. Free trade is now moving forward through the GATT, NAFTA, and other historic trading agreements.

Implementation of Climate Change Mitigation Policy

Economics delivers central insights to the construction of efficient mechanisms for achievement of target greenhouse gas emission reductions. One manifestation of such insights is the joint implementation approach to greenhouse gas reductions initiated by several countries, including the United States. Under joint implementation, the least costly projects to reduce greenhouse gas emissions or enhance carbon sinks can be pursued jointly across countries by, in essence, giving private agents and governments opportunities to meet emission reduction targets anywhere and in cooperation with any others around the world. Forestry projects are likely to be an important outcome of joint implementation in agriculture. Planting trees in reforestation or afforestation projects enhances the absorptive capacity of the biosphere and leads to carbon dioxide reductions in the atmosphere. Forestry projects can also provide benefits beyond climate change, including benefits to the local environment and to biodiversity.

In summary, this session suggests directions for future climate change research on both theoretical and empirical fronts. Foremost among such research directions are joint modeling of uncertainty and dynamics, empirical work on rates of climate change and induced innovation, and modeling climate change feedbacks from human activities and behavior. Such research should be supported by improved data and can contribute to a better informed policy debate on the timing and nature of actions to curb greenhouse gas emissions. While we are learning more about the economics and geophysics of climate change, policy makers should continue to seek ways to minimize social costs of climate change and climate change mitigation, a process to which the economics profession has much to contribute. Prominent examples of efficiency-enhancing policy measures are the promotion of free trade, the liberalization of farm policy, and joint implementation of greenhouse gas reduction objectives under the Framework Convention on Climate Change.

Robert Innes is senior economist with the Council of Economic Advisers, and professor in the Department of Agricultural and Resource Economics at the University of Arizona. Sally Kane is senior economist with the Council of Economic Advisers, and is on leave from the National Oceanic and Atmospheric Administration (Department of Commerce).