« links for 2009-08-29 | Main | Hoover's "Pro-Labor Stance" Did Not Help to Cause the Great Depression »

Saturday, August 29, 2009

"When Carbon is Priced, Who Ultimately Pays?"

Who pays the costs of a cap and trade system for carbon emissions? This analysis of who pays based upon the legislation passed by the House in June suggests that "the distributional consequences have been addressed fairly effectively," and that "Strong opposition to the legislation will probably be based more on ideological grounds than distributional concerns":

When carbon is priced, who ultimately pays?, by Corbett Grainger and Charles Kolstad, Vox EU: Climate change legislation is winding its way through the US Congress. In June (2009), the House passed complex legislation involving, among other things, a cap and trade system for carbon emissions in the US. The legislation passed by the slimmest of majorities – 219 to 212 with 3 not voting. Although most Republicans opposed the legislation, a significant number of Democrats, primarily those from coal-rich or rural districts, also voted no. Politics is local, even for global warming.

The Senate will consider similar legislation in September, and its passage is likely to be even more difficult. Several issues are at the core of the opposition:

  1. The competitive effect on industry. Will higher costs cause US industry to lose market share to foreign producers? In order to win over skeptics, legislators mandated import tariffs on goods from countries without climate legislation (such as China). President Obama has subsequently condemned such mandatory trade barriers.
  2. The burden of increased energy prices on lower-income individuals. The poor often drive older cars, commute further to work, and in other respects consume goods and services differently than wealthier members of society. Furthermore, to the extent that energy is a necessity, one would expect price increases to hit the poor particularly hard. In other words, carbon regulation may be regressive.
  3. The burden of higher prices for consumers in states heavily reliant on coal-based electricity. Coal has the largest carbon emissions per unit of heat energy of any of the fossil fuels (by a wide margin) and thus will be hit hard by any legislation.

Measuring the burden of a carbon price

It is important to distinguish between a tax’s statutory incidence (i.e., who writes the check to the government) and its ultimate burden (who ends up paying the cost of the tax). A cap-and-trade system is not a tax in the normal sense, but it does induce a tax-like carbon price change. Raising the price of carbon increases the costs of doing business, and those costs may be passed on to consumers, workers, or owners.

The Congressional Budget Office estimates that the House legislation will result in an initial price of carbon on the order of $15 per ton of CO2 (CBO 2009a).1 Because CO2 is mostly O2, simple chemistry tells us that this amounts to a price on carbon of about $55 per ton.

When the price of carbon increases, the direct effect is that fuels that consumers and industry purchase will be more expensive. There is also an indirect effect, in the sense that industries that use fossil fuels will see costs rise and will pass some or all of those increased costs on through increased prices of goods and services. The simplest example is electricity. Higher fuel prices raise the costs of generating electricity, which may in turn lead to increased electricity prices. A subtler example is construction, which would face both higher fuel prices and higher raw materials prices, due to the carbon price faced by producers and transporters of those raw materials.

It turns out that tracing the effect of an increased price of carbon through the US economy is not so hard. The US national accounts track how much industries buy from other industries in order to produce. It is straightforward to calculate how a CO2 price would affect the prices of different products in the US economy, assuming that industries are able to pass on all of their extra carbon costs to their consumers. Disaggregating the US economy into nearly 500 different industrial sectors and using a fairly recent characterisation of nationwide consumption patterns, we have estimated the increase in costs associated with a $15/ton price of CO2 (Grainger and Kolstad 2009).

This is probably an overestimate of the effect of a carbon price on product prices for two basic reasons:

  • Some industries will simply not be able to pass on the entire extra cost due to competitive pressures.
  • People and firms typically substitute away from goods that see price increases.

This may involve simply doing without or switching to other goods. Predicting how much of the carbon price can be passed on or how carbon users will change behaviour is not easy.

The effect of a carbon price on industry

The first political issue we address is how badly specific industries would be affected by a price on carbon. Of the approximately 500 industries considered, only five are estimated to see a cost increase in excess of 5%. Table 1 shows the top ten industries facing expected cost increases. Although some sectors do see significant cost increases, what is striking is how few industries see costs increase more than a few percent.

Table 1. Ten highest impact sectors from $15 per ton carbon price

How regressive is a carbon price?

A second political issue in the debate over climate regulation is how regressive a carbon price may be. The US Consumer Expenditure Survey tracks consumption decisions by households throughout the income distribution. Coupling that with data on interindustry transactions allows us to trace a $15 price on CO2 through industries to consumers. As before, this will be an overestimate of the effect on consumers, since we assume that industry can pass on the full carbon price to consumers and consumers do not change their behaviour in response to an elevated carbon price.

In order to understand how regressive or progressive a price on carbon may be, it is important to estimate the fraction of household income that may be absorbed by the carbon price. It turns out that there are several measures of income that may be relevant. One is simply conventional income. Another is lifetime income, based on current expenditures, smoothing out the fact that we tend to earn less early in life and more later but spend based on our expected lifetime income. Two more variations relate to household size. It is cheaper to keep one household of two people than two households of one person, so we can adjust income to “equivalent” income, reflecting the economies of scale in operating a household.

Figure 1 shows the carbon price burden as a fraction of income for different US income groups.

Figure 1. Burden of $15/ton tax on CO2 equivalent, as a fraction of income


As can be seen from the figure, a price on carbon is regressive in that lower income households spend a larger fraction of their income on payments to carbon. Although the effect differs depending on which measure of income is used, the qualitative story is the same. For very low income (less than $35,000 per year), the burden as a fraction of income is highly sensitive to income, increasing substantially as income drops. For greater incomes, the burden is somewhat regressive though not remarkably so, dropping from 1.5-2.5% for $35,000 incomes to 1-2% for the highest income bracket.

In fact, the House climate legislation contains specific provisions to offset the regressive aspects of the cap and trade program. The Congressional Budget Office has analysed these components of the bill and has concluded that the lowest quintile would actually see a gain of 0.2% whereas the highest quintile would see a cost of 0.1% (CBO 2009b). Virtually all of the regressivity has been neutralised, though regional differences may still persist, an issue addressed by Hassett, Mathur, and Metcalf (2009).


The politics of climate change regulation is as much tied up with the distribution of costs as their overall magnitude. Our analysis of the burden of the legislation passed by the House in June suggests that the distributional consequences have been addressed fairly effectively. Strong opposition to the legislation will probably be based more on ideological grounds than distributional concerns.


Congressional Budget Office (2009a) CBO Analysis 5 June 2009.
Congressional Budget Office (2009b), “The Estimated Costs to Households From the Cap-and-Trade Provisions of H.R. 2454,” (Washington, DC, 19 June 2009)
EPA (2009) “EPA Analysis of the American Clean Energy and Security Act of 2009 HR 2454 in the 111th Congress,” 23 June 2009.
Grainger, Corbett and Charles Kolstad (2009), “Who Pays a Price on Carbon?”, NBER Working Paper 15239.
Hassett, Kevin, Aparna Mathur and Gilbert Metcalf (2009), “The Incidence of a US Carbon Tax: A Lifetime and Regional Analysis,” Energy Journal 30.2 (2009): 155-177.

1 The CBO (2009a) estimates a price of $15 per ton of CO2 equivalent in 2011, rising to about $26 per ton in 2019. The EPA (2009) estimates slightly lower prices – $13 in 2015 and $16 in 2020.

    Posted by on Saturday, August 29, 2009 at 12:36 AM in Economics, Environment | Permalink  Comments (23)


    Feed You can follow this conversation by subscribing to the comment feed for this post.