My anti-anti-carbon tax argument [corrected]

UPDATE 12 Noon ET: I made what is, for someone who's worked as a professional fact checker and copy editor, the rather terrible error of not looking at the byline of what I was criticizing. Pethokoukis had linked to the post, but it was written by Benjamin Zycher. Corrected below.

James Pethokoukis Benjamin Zycher has an anti-carbon tax response to Greg Mankiw’s recent pro-carbon tax column. I think Pethokoukis’ Zycher's argument is erroneous for several reasons.

Pethokoukis Zycher writes:

Economics first.  The US emits about 17 % of worldwide emissions of greenhouse gases (GHG).  If we apply the MAGICC climate model developed at the National Center for Atmospheric Research, and assume the IPCC mid-range emissions path for the 21st century, an immediate cut in US emissions by half would reduce global temperatures a century by now by 0.1 degrees.  As the standard deviation is 0.11 degrees, that reduction could not be measured reliably.

Me: Any analysis that limits the effects of a U.S. carbon tax to U.S. emissions is unreliable. A U.S. carbon tax could and should be applied to imports from non-carbon tax countries as well. It should aim, as the book The Carbon Crunch advocates, at applying to carbon production, not consumption. A U.S. carbon tax would also, besides pricing out some U.S. carbon emissions, tend to spur low- or no-carbon technological developments that also could be exported and applied internationally. Pethokoukis’s Zycher's focus only on the obvious, direct effects of a U.S. carbon tax is the kind of static analysis that supply-siders used to decry.

Pethokoukis Zycher writes:

Science next.  Assume that the IPCC-class climate models are correct, despite the reality that they can predict neither the past nor the present. Assume any set of predicted effects, however apocalyptic.  It remains the case that the proposed tax is inconsistent with the underlying climate science.  Why?  Because the effect of increasing GHG concentrations is logarithmic: The marginal impact of emissions declines as concentrations increase.  As the developing world consumes more fossil fuels—the inexorable human pursuit of higher living standards is universal—GHG concentrations will rise, which means that (changes in) US emissions will matter less and less, unless one assumes a large increase in the vulnerable capital stock without adaptation by markets. 

Me: The effect of increasing GHG concentrations is logarithmic, in that each doubling of emissions is expected to produce roughly the same temperature increase. This does not mean that going forward there is less and less reason to worry about greenhouse gases. Emissions may well rise fast enough to outpace the logarithmic effect and produce an accelerating warming (and delayed temperature effects of carbon already in the atmosphere may also be significant). But in any case, what we ultimately care about is the impact of the temperature increases. Does Pethokoukis Zycher feel confident that, say, the difference between a 3.5 degree warming and a 2 degree warming is insignificant (because we avoided a doubling of the first 2 degrees)?

Pethokoukis Zycher also asserts a carbon tax should decline over time (to reflect the alleged declining impact of warming), rather than rise as proponents often suggest. But note that a gradual ramp-up of a tax would reduce shocks to the economy and offer incentives to reduce carbon emissions (and develop new technologies) in advance of expected increases in the tax.

Pethokoukis Zycher writes:

Moreover, the tax would be inconsistent with the principles of efficient taxation under democratic institutions.  Taxes are prices for public services, and the pursuit of efficiency in public spending means that tax prices should reflect the valuations that individuals place on those services.  Proposals for carbon taxes offer no plausible link between the economic burdens that the tax would impose and the differing benefits of public spending across individuals.  Because the tax would be hidden in the prices of goods and services, it would obscure rather than clarify the cost of government, a “fiscal illusion” effect inconsistent with the pursuit of fiscal discipline narrowly and efficiency in the size of government generally.

Me: I’m not sure what kind of tax really makes “a plausible link between the economic burdens that the tax would impose and the differing benefits of public spending across individuals.” Certainly the income tax doesn’t do this; one taxpayer might pay much more than another, in absolute or proportional terms, without necessarily benefitting more from whatever the government does with the money. As for being “hidden in the price of goods and services,” surely people will know that increased prices at the gas pump and elsewhere are related to any carbon tax that’s been imposed by the federal government; if not, that would suggest the tax really isn’t much of a burden, and in any case some form of disclosure could be added (e.g., your gas receipt saying x dollars or percent of this bill resulted from the carbon tax).

Pethokoukis Zycher writes:

The carbon revenues will engender a large rent-seeking tug-of-war exercise the result of which is difficult to predict; but it is far, far, far from obvious that improved “economic efficiency” would be prominent among them.

Me: Yes, there will be fights over government funding, just as there are now. If a carbon tax replaces (as it should) some portion of income taxes, payroll taxes or other taxes, this effect could be limited. In the present situation, fossil fuel companies have done an excellent job of promoting their own interests over broader public interests, so a carbon tax will have the benefit of reducing that effect. The question is not whether a carbon tax will be perfectly designed and implemented, but whether it would be an improvement over current public policy. I don’t find in Pethokoukis’s Zycher's argument a compelling reason to think it would not.