Researchers:
- Tom Hertel
- Jev Steinbuks
We are working to better understand how uncertainty in climate impacts, mitigation policies, and technology affect the optimal path of global land allocations between food, biofuel, forest products and ecosystem services. The project involves the development and use of a multi-sector, partial equilibrium model of global land use, elements of which will eventually be merged with the general equilibrium models currently under development in this project.
This project is motivated by the fact that agriculture, forestry and associated land use change together account for one quarter of GHG emissions today. Recent global economic analyses undertaken at Purdue suggests that these same sectors could account for up to 50% of economically efficient abatement at modest carbon prices ($28/ton CO2eq). Yet research efforts devoted to understanding the potential for such mitigation, as well as the associated limitations remain poorly developed, relative to those focusing on fossil fuel combustion in transportation, industry and power generation. There are several features of land-based climate analysis that make this area particularly challenging. First of all, in addition to offering considerable abatement potential, agriculture and forestry are also two of the sectors which are most exposed to climate change impacts and these are expected to negatively affect productivity in many parts of the world, especially over the longer run as temperatures continue to rise. We seek to inform decisions about the optimal trajectory for land-based GHG policies, even as the impacts of these policies depend critically on their implementation.
The second feature that makes land-based climate policy decisions difficult is the irreversible nature of many such decisions. Deforestation of old-growth forests and the associated loss of biodiversity are decisions which are difficult, if not impossible to reverse. Such irreversibility is particularly problematic in light of the tremendous uncertainty about climate impacts on these sectors, as well as their potential for GHG mitigation. In short, we must make decisions with long run ramifications today, even though we have a poor idea of their consequences in the future. Our core research activity to date has been to build, for the first time, an economic model of global land use, which is capable of meeting this research goal. In the first stage of the project we developed a long-run, forward-looking, dynamic partial equilibrium model in the absence of uncertainty. The model encompasses two natural resources: land and fossil fuels, along with eight sectors, each producing intermediate and final goods and services. The societal objective function being maximized places value on services derived from food, energy, timber, and ecosystem services. Emissions of GHGs are central to the problem at hand, and are currently treated as a time-varying constraint on the flow of GHGs. The model we develop is at global scale, in the spirit of the DICE model, and offers an elaborate treatment of the different sectors as well as a 200 year time horizon. Implementation of the empirical model required assembly of a wide range of data and model parameters, including inputs and outputs for each of the sectors, elasticities of substitution in production, and parameters governing the long run evolution of consumer demands for food, fuel, forest products and ecosystem services. These data and parameters have been obtained from a variety of sources, including the Global Trade Analysis Project (GTAP) Integrated Global Land Use and Global Forestry databases, the U.N. Food and Agriculture Organization, U.S. Department of Agriculture, U.S. Energy Information Administration, in addition to several secondary sources.
A great deal of time was invested in developing a meaningful 200 year baseline: 2000-2200 which accu- rately reflects developments in global land use over the 10 years that have already transpired, while also incorporating projections of population, income and demand growth from a variety of international agencies. After finalizing the baseline, we consider three counterfactual scenarios, each relating to one of the important sources of uncertainty facing decision makers today (higher growth in energy prices, lower growth in agricul- tural productivity, and global GHG emissions regulations). Key findings are summarized in [38]. Specifically, we show that, while some deforestation is optimal in the near term, the desirability of further deforestation is eliminated by mid-century under the baseline scenario. While the adverse productivity shocks from climate change have a modest effect on global land use, when combined with high growth in energy prices, they lead to significant deforestation and higher GHG emissions than under the baseline. Imposition of a GHG emissions constraint further heightens the competition for land, as fertilizer use declines and land-based mitigation strategies expand. However, the effectiveness of such a pre-announced constraint is completely diluted by the inter-temporal substitution of deforestation from the constrained future to the unconstrained present.