Tuesday, August 18, 2015

Carbon Tax: Practicalities of Cutting a Deal

The key practical questions about a carbon tax include what should be taxed and how much is should be taxed. The what is fairly clear; the how much is fuzzier. But if advocates of a carbon tax could agree on the size and shape of such a tax, they could offer some interesting incentives for political wheeling and dealing. Donald Marron, Eric Toder, and Lydia Austin write about "Taxing Carbon: What, Why, and How" in a June 2015 exposition for the Tax Policy Center (which is a joint venture of Brookings and the Urban Institute).

Discussions of climate change typically focus on carbon emissions, but there are other greenhouse gases (and non-gases, like soot) that also affect can trap heat in the atmosphere. Here's a list of the major greenhouse gases, and how much heat they trap relative to carbon. It's common in discussions of this subject to refer to all taxes on greenhouse gases as a "carbon tax," and to express the emissions of other gases in terms of "carbon dioxide equivalents."




Marron, Toder, and Austin write:

"[P]olicymakers must address the fact that greenhouse gases differ in their chemical and atmospheric properties. Methane, for example, traps more heat, gram-for-gram, than carbon dioxide does, but it has a shorter atmospheric lifetime. A cost-effective tax should reflect such differences, raising the tax rate for gases that are more potent and lowering it for gases that stay in the atmosphere for less time. Analysts have developed measures known as global warming potentials to make such comparisons. According to the potentials the EPA uses, methane is 21 times more potent than carbon dioxide over a century, and nitrous oxide is 310 times as potent (table 1). By those measures, a $10 per  ton tax on carbon dioxide would imply a $210 per ton tax on methane and a $3,100 per ton tax on nitrous oxide."

But those numbers are about the ratio of taxes on greenhouse gases relative to each other. What should the actual taxes themselves be? Economists argue that the price placed on greenhouse gas emissions should be set according to the damage caused by those emissions--in effect, consumption that leads to carbon emissions should pay the price for harm caused. But estimating the social cost of carbon emissions is very difficult, and estimates are all over the map. Marron, Toder, and Austin: 
Estimates of the marginal social cost of carbon thus vary widely. In developing a cost to inform US climate policy, an interagency working group commissioned 150,000 simulations from three leading models, all using the same 3 percent real discount rate. The resulting estimates fell mostly in the -$10 to $50 per ton range (in today’s dollars), with a few lower and some significantly higher. The central tendency was a cost of $27 per ton in 2015 and rising in the future. An update increased that figure to about $42 per ton in 2015, with estimates again ranging from slightly below zero to more than $100. These wide ranges, and the underlying uncertainty about long-term economic and geophysical  responses to rising greenhouse gas concentrations, have left some analysts pessimistic about the ability of such modeling efforts to identify an appropriate price for carbon. ... 
Along with all the uncertainties of how these emissions affect climate, and the placing an economic value on these changes, another problem is that the benefits of limiting climate change are international, but the costs of a US carbon tax are national. Of course, the ultimate hope is for a coordinated international effort, but with the world moving to more carbon-intensive energy sources, there's no assurance this will happen. Marron, Toder, and Austin point out:

"[A] coordinated international response should focus on worldwide emissions and impacts. If a nation considers unilateral action, however, it must decide whether to focus on domestic costs and benefits or to consider other nations as well. The difference is large. Greenstone, Kopits, and Wolverton estimate that the United States bears only 7 to 10 percent of the worldwide marginal social costs of carbon. If each new metric ton of carbon dioxide emissions imposes $40 in worldwide damages, only $3 to $4 would fall on the United States. They argue that the United States ought to use the global measure when evaluating regulatory policies, but this view is not universal. Indeed, policymakers take a US-only view when evaluating other energy and environmental policies that
have international spillovers."

In the face of such uncertainties, a possible approach is to set a relatively low carbon tax that would slowly rise over time. The rising level over time would encourage actions to reduce emissions of greenhouse gases over time. And there could be an official process, perhaps every 5-10 years, for evaluating if the tax was at roughly the correct level. With some level of a carbon tax in mind, the question becomes one of practical politics. One of my takeaways from the Marron, Toder, and Austin essay is that advocates of a carbon tax have some arguments they could make that might intrigue the undecided middle ground. Here are several such arguments:

1) A carbon tax would reduce the use of fossil fuels, and thus would reduce a number of conventional pollutants. The gains or "co-benefits" from reducing conventional pollutants are substantial--indeed, by some estimates the co-benefits of a moderate carbon tax might make the tax worthwhile even if reducing carbon emissions brought no other gains. Marron, Toder, and Austin put it this way:

 Climate change is not the only harm associated with burning fossil fuels. Power plants, factories, vehicles, and other sources also emit air pollutants that directly harm human health, including fine particulate matter, sulfur dioxide, and nitrogen oxides. Vehicle use also imposes other external costs, including congestion, road damage, and accidents. ... As a result, a carbon tax would generate “co-benefits”—improvements in human health and well-being unrelated to climate concerns. The magnitude of those co-benefits depends on several factors, including the prevalence and value of potential health improvements (e.g., reduced asthma, bronchitis, heart attacks) and the scope of benefits included (e.g., just air pollution from fossil fuels or also congestion and accidents that result from driving). In a comprehensive analysis including both air pollution and vehicle externalities, Parry, Veung, and Heine estimate that the co-benefits of a carbon tax in the United States would be about $35 per ton. In a narrower analysis of the co-benefits from its proposed regulations on power plants, the EPA estimates that the co-benefits of reduced air pollution are at least as large as potential climate benefits. These estimates thus suggest that, in the absence of new policies addressing those harms, a substantial carbon tax would improve US well-being even if we give no weight to climate
change.
2) The revenues from a carbon tax could allow offsetting tax cuts in other area. Taxes discourage certain behaviors, and it's surely better to discourage carbon emissions than to discourage work and saving. Marron, Toder, and Austin write:

The main objective of a carbon tax is to reduce environmental damage by encouraging producers and consumers to cut back on activities that release greenhouse gases. This is its first dividend. A carbon tax can also generate a second dividend: an improvement in economic efficiency by using the resulting revenue to reduce distortionary taxes, such as those on income or payroll. ... For legislative purposes, the most important estimates are those of the Congressional scoring agencies, the Joint Committee on Taxation, and the Congressional Budget Office. In late 2013, they estimated the revenue effects of a tax on most greenhouse emissions starting at $25 per ton and increasing 2 percent faster than inflation. Scaling those estimates to CBO’s latest budget projections, they imply net revenue of about $90 billion in its first complete year and about $1.2 trillion over its first decade.

3)  A meaningful carbon tax would mean that there was less need for a range of other government rules and regulations. For example, other rules about vehicle efficiency or conservation would be less necessary. Subsidies for renewable energy sources could be scaled back, because they could benefit from not paying the carbon tax instead. One last time, here are Marron, Toder, and Austin:

In the absence of a broad, substantial price on carbon, policymakers have attempted to reduce carbon emissions through a mix of narrower policies. The Environmental Protection Agency is developing emissions standards for new and existing power plants, the Department of Transportation has expanded vehicle fuel economy standards, and the Department of Energy has expanded appliance energy efficiency standards. Tax subsidies and renewable fuel standards favor renewable and low-carbon fuels, such as wind, solar, biomass, geothermal, and nuclear, and biodiesel and electric vehicles. A sufficiently high and broad carbon tax would reduce the benefit of these policies. If policymakers contemplate such a tax, it would be appropriate to reassess these policies to see whether their benefits justify their costs.

Some of those concerned about climate change won't like the policy tradeoffs I'm suggesting here. But lasting political coalitions need broad support.