learn more about working at cbo and check out the agency’s career opportunities
working paper
June 18, 2010
read complete document (pdf, 282 kb)
Kevin Perese
Abstract
The risk of significant climate change caused by greenhouse gas emissions is currently one of the largest environmental and economic issues facing policymakers in the United States and around the world. Carbon dioxide (CO2) is one of the most prevalent greenhouse gases released into the atmosphere, so some policymakers have placed their focus on reducing these emissions. Economists generally agree that efficient regulation of CO2 emissions involves placing a price on them. An input–output (IO) model of the U.S. economy provides a framework that can be used to estimate detailed commodity price effects in response to a placing price on the emission of CO2 into the atmosphere.
This paper provides a general overview of IO models and a specific application of an IO model to estimate the effect of a $20 tax per metric ton of CO2 emissions. In comparison with previous work by other analysts using an IO model for this type of analysis, the model presented here uses more recent (though less detailed) data, holds the price of most imported commodities fixed while subjecting imported petroleum, natural gas, and coal to the tax, and makes adjustments for the noncombusted uses of fossil fuels.
Results from the model, which can only be interpreted as the first-order effects of the policy, imply that in response to a $20 tax on CO2 emissions, energy commodities such as natural gas, electricity, and gasoline will experience price increases of approximately 10%, but the vast majority of commodities will experience much smaller price increases of approximately 1%. The distribution of the policy effects across sectors of the economy are based on the relative price increases and the mix of commodities consumed in each sector. Based on the estimated price increases and the mix of commodity consumption observed in the 2006 input–output tables, consumers would bear approximately 70% of the aggregate policy effects; federal, state, and local governments would bear about 12% of the aggregate policy effects; and private fixed investment costs would be approximately 8% higher.
Highlights
The risk of significant climate change caused by greenhouse gas emissions is one of the largest environmental and economic issues currently facing policymakers in the United States and around the world. The release of carbon dioxide (CO2) into the atmosphere as a by-product of burning fossil fuels for energy accounted for approximately 80% of all greenhouse gas emissions in 2006. As scientific consensus solidifies the link between CO2 emissions from the combustion of fossil fuels and the atmospheric changes these emissions produce, economists have simultaneously been investigating the most economically efficient approaches for reducing CO2 emissions to avert or reduce the consequences of significant global climate change. In response to these scientific developments and economic findings, legislators in the United States have been considering an array of policy options to reduce CO2 emissions.
Economists generally agree that efficient regulation of CO2 emissions involves placing a price on them. Pricing CO2 emissions can be done either directly through a tax on emissions or indirectly by creating a cap-and-trade program. A cap-and-trade program would create emission allowances as a new commodity that would give entities covered by the program the right to emit a metric ton of carbon into the atmosphere in exchange for an allowance they own or purchase.
A carbon tax and a carbon cap-and-trade system are both market-based approaches to reducing carbon emissions. Each would produce two effects: The number of carbon emissions would be reduced, and the cost of emitting each metric ton of carbon would increase. Levying a tax on carbon emissions would directly increase the price of carbon emissions, which would result in market forces reducing carbon emissions. Conversely, a cap-and-trade policy would explicitly limit the number of emissions allowed, which would result in market forces driving up the price of carbon emissions. A carbon tax operates directly on the price of emitting carbon as the policy lever; under a cap-and-trade system, the quantity of emissions serves as the policy lever.
An input-output model (or IO model) of the U.S. economy provides a framework that can be used to estimate detailed commodity price effects in response to a carbon policy. An IO model is constructed from a large database of intermediate transactions in the production of all goods and services as well as the distribution of all the final goods produced in an economy. At the heart of the input-output model is the intermediate transaction matrix, which describes the mix of production inputs required for every commodity output in an economy. These data can be used to estimate how a price on carbon emissions (through either a direct tax or a cap-and-trade policy) would filter through to every good and service produced and sold in the economy. Such a model is capable of capturing not only the direct effects based on the carbon intensity of inputs used in production, but also the sum of all the indirect effects based on the carbon intensity of all the secondary, tertiary, and higher-order inputs to production (that is, the inputs to the inputs to the inputs, and so on).
The IO model described here is based largely on previous models used to analyze carbon policies (most notably Fullerton, 1996, and Metcalf, 1999). Similar to these models, the one presented here makes two important assumptions. First, the model assumes that labor and capital markets are perfectly competitive and that a price on carbon is passed on to consumers in the form of higher prices for carbon intensive energy sources and for commodities that rely heavily on these energy sources in their production process. The second assumption, inherent to most IO models, is that production functions are fixed, which precludes any factor substitution in response to higher (or lower) input prices. Because of that assumption, the results from these models can only be interpreted as the short-run, first-order effects of a carbon pricing policy. Firms will, however, respond to the carbon policy and will seek lower-priced alternative inputs to their production processes. To the extent that production substitution is able to lower the initial cost of the policy, the estimated effects presented here are likely to be upper bounds beyond the short term.
There are however, some important differences between the model presented here and those used by other researchers to analyze the impact of various carbon policies. Relative to previous models, the model described in this paper:
- Uses more recent but less detailed data;
- Is commodity-based rather than industry-based;
- Holds prices fixed for most imported commodities; and
- Makes adjustments for the non-combustive uses of carbon.
This paper provides both a general background on input-output modeling and a description of a specific application of input-output modeling to analyze the effects of a carbon policy on commodity prices. Section 2 of the paper provides an overview of input-output models, and Section 3 provides an overview of the Make-Use framework in which modern input-output data are collected. Section 4 provides details on the conversion of Make-Use tables to a square input-output matrix. For those who are well versed in input-output modeling techniques and issues, Sections 5, 6, and 7 cover the specific application of an IO model to analyze a policy that places a price on carbon emissions, accounts for international imports, and provides some model results from the analysis.