U.S. C[O.sub.2] Mitigation in a Global Context: Welfare, Trade and Land Use.

Author:Sands, Ronald D.

    In this paper we describe carbon dioxide (C[O.sub.2]) reference and mitigation scenarios specified by the Energy Modeling Forum study (EMF-24) "U.S. Technology Transitions under Alternative Climate Policies," using the Future Agricultural Resources Model (FARM), a global computable general equilibrium (CGE) model that simulates world energy and agricultural systems in five-year time steps starting in year 2004. EMF-24 guidance specifies 42 scenarios designed to cover a range of technology and policy options for greenhouse gas mitigation in the United States through year 2050.

    Using the GTAP 7 data set as a benchmark social accounting matrix in 2004, we partition the world into 15 regions with the United States as a single region. Taking advantage of a balanced approach to agriculture and energy in a global economic framework, we address the following questions using the FARM model. How does a C[O.sub.2] mitigation strategy in the U.S. affect economic welfare? How does a C[O.sub.2] mitigation strategy in the U.S. affect land use in the U.S.? Are there significant spillover effects through international trade in economic welfare or land use?

    Technology assumptions in all scenarios apply to all 15 world regions. Cap-and-trade mitigation scenarios apply to the United States, the European Union, and other member countries of the Organization for Economic Cooperation and Development (OECD), but not to developing countries. This allows a calculation of carbon leakage through international trade. Other C[O.sub.2] mitigation scenarios, applied only to the United States, include a renewable portfolio standard (RPS) for electricity generation, a clean electricity standard (CES), and transportation regulations to improve fuel efficiency. For each mitigation scenario and world region, change in consumer welfare is computed as equivalent variation relative to the corresponding reference scenario.

    Bio-electricity is the primary link between energy and agricultural systems in these FARM scenarios. Bio-electricity is present in all reference scenarios as combustion of solid biomass, or co-firing solid biomass with coal, to generate electricity. Bio-electricity grows rapidly in the U.S. in most mitigation scenarios, with the quantity of bio-electricity depending on the C[O.sub.2] price, the price of certificates in a renewable portfolio standard, the rate of improvement of biomass crop yield over time, and the availability of other mitigation technologies.

    The presence or absence of C[O.sub.2] capture and storage (CCS) distinguishes many of the technology scenarios. If available, CCS can be used with any electricity generating technology that emits C[O.sub.2], provided the C[O.sub.2] price is high enough to cover the cost of CCS. Bio-electricity combined with CCS is a special case, with electricity and net carbon sequestration as joint products. (1)

    FARM is one of ten models participating in EMF-24 and each model has its own strengths and limitations. The primary strengths of FARM within EMF-24 are global coverage and a balanced approach to energy and agricultural systems within a CGE model. This allows calculation of carbon leakage, changes in welfare as equivalent variation, and changes in land use. The EMF-24 study is unique in its analysis of a wide range of alternative U.S. climate policies across varied technology futures. A particular strength of EMF-24 is an analysis of the tradeoff between emissions reductions and welfare cost for various climate policies, especially policies not on the efficient frontier.

    Carbon leakage has been addressed in a theoretical context by Hoel (1991) and in a recent multi-model study organized by EMF (Bohringer, Balistreri and Rutherford, 2012). Other studies have addressed the greenhouse gas implications of biofuel targets (e.g. Timilsina and Mevel, 2013; Beckman, Jones and Sands, 2011). However, the studies most closely related to EMF-24 are a concurrent study with global greenhouse gas concentration targets (EMF-27) and previous EMF studies (e.g. Clarke and Weyant, 2009).

    We provide a description of EMF-24 scenarios in Section 2 of this paper, including specific assumptions used in FARM implementation of scenarios. In Section 3 we describe the economic framework of the FARM model, including benchmark data and functional form of production and demand systems. Selected output on C[O.sub.2] prices and emissions across EMF-24 scenarios is provided in Section 4. In Section 5 we use welfare calculations from the cap-and-trade scenarios to display the tradeoff between emissions reductions and welfare cost. Emissions leakage to countries outside OECD and the European Union, and welfare impacts on other countries, are the topics of Section 6. The main topic of Section 7 is land use change in response to C[O.sub.2] mitigation scenarios.


    Out of 42 scenarios in the EMF-24 study, eight are reference scenarios with varying assumptions across five major technology groups: end-use technology, C[O.sub.2] capture and storage, nuclear electricity generation, wind and solar power, and bioenergy. All of the remaining scenarios are mitigation scenarios. Eighteen mitigation scenarios are economy-wide cap-and-trade with varying degrees of stringency using either US01 or US02 as the corresponding reference scenarios (Table 1). Another six mitigation scenarios are cap-and-trade using the six other reference scenarios. The remaining mitigation scenarios are targeted to transportation and electricity generation. Scenarios US13 and US14 apply optimistic assumptions to all technologies ("all good"); scenarios US23 and US24 use pessimistic assumptions ("all bad").

    Table 2 provides specifics of the way technologies are represented in the FARM model. In the case of end-use energy technologies, the difference between low and high efficiency is somewhat less than EMF-24 guidance. (2) In the cases of wind, solar, and bio-electricity, specifics of technical change over time were left as modeler's choice. Table 3 provides FARM specifics for EMF-24 mitigation policies.


    The computational framework for all scenarios is the Future Agricultural Resources Model. The first version of FARM was constructed in the early 1990s by Roy Darwin of the Economic Research Service, and was used for analysis of climate impacts on global agriculture. Construction of a new version of FARM began in 2010 with model development driven by requirements of two international model-comparison activities: the Stanford Energy Modeling Forum and the Agricultural Model Inter-comparison and Improvement Project (AgMIP). This required a capability to simulate global energy and agricultural systems through at least 2050, with scenarios that vary across technology availability and policy environment. Bioenergy provides an interface between agricultural and energy systems in some scenarios, especially greenhouse gas mitigation scenarios. (3)

    New tools and data have become available since the first version of FARM was constructed, most notably global social accounting matrices provided by the Global Trade Analysis Project (GTAP) at Purdue University (Hertel, 1997), and tools for using GTAP data in the GAMS (4) programming language (Rutherford, 2010). Therefore, development of the new FARM model did not start from scratch: the starting point is software provided by Rutherford. This software provides a comparative-static global CGE model fully compatible with GTAP 7 social accounts with bilateral trade between world regions.

    The FARM model has been extended in many ways beyond the model in Rutherford...

To continue reading