Let us now recall the blessed memory of Godzilla, King of the Monsters. I know him as Zilly, as we have grown close over the years and the 30-plus movies that bear his name. Anyway, at the end of that original timeless classic of the silver screen, an “oxygen destroyer” reduced Zilly to a skeleton at the bottom of Tokyo Bay, as Raymond Burr and other immortals looked on.

An oxygen destroyer sounds vaguely similar to the greenhouse gas (GHG) climate change monster now purportedly wreaking havoc, for which proposition there is virtually no evidence, and the reverse is more likely to be true, at least in the immediate term. Moreover, even the Intergovernmental Panel on Climate Change in its Fifth Assessment Report is deeply dubious (see Table 12.4 and attendant discussion) about the various horror stories popularized as looming effects of increasing atmospheric concentrations of greenhouse gases.

But never mind. The point to be observed here is that former skeleton Zilly has returned repeatedly to do battle with such beasts as Biollante, Mothra, Ghidorah, Gigan, King Kong, and a long list of other threats to civilized life. (Sadly for ardent fans of Zilly, Harvey Weinstein seems to have arrived too late to the monster mash to take his place on this particular red carpet; oh, what I would give to see Zilly’s contract negotiation scene in Harvey’s office. But I digress.)

Consider now the ever-evolving case for a “carbon” tax, that is, a tax on GHG emissions. Note that carbon dioxide is not “carbon,” to which point I return briefly below. Just as Zilly returned from the deep time and again to confront an increasingly terrifying parade of grotesqueries, the policy problems that the carbon tax promises to solve have expanded as well. Originally, it was a textbook Pigouvian externality tax designed (ostensibly) to reduce GHG emissions to socially efficient levels. Revenues per se decidedly were not the goal, and many proposals for a Pigouvian carbon tax incorporated reductions in other taxes so as to achieve approximate revenue neutrality. This rationale implies that the tax would be set at the marginal “uninternalized” social cost of GHG emissions; that government has perverse incentives (and poor information) with respect to determining the efficient tax is a topic for another day.

That secondary goal of revenue neutrality through tax offsets led to a very different objective for a carbon tax: Why not use those revenues to fund a reduction in other distortionary taxes to increase aggregate economic performance? In many economic models—including prominent work from my AEI colleague Aparna Mathur and Adele Morris of Brookings—it is reductions in taxes on capital that yield the greatest benefit in terms of improved investment and growth. This rationale implies a very different carbon tax per ton of emissions, one chosen to optimize a complex mix of higher energy costs, carbon-tax revenues, and reduced capital taxation implemented in pursuit of higher permanent economic growth.

Alas, revenue neutrality is so yesterday. Mathur and Morris now argue that a carbon tax ought to be used to fund an increase in the earned income tax credit (EITC), thus serving to “directly [help] working families,” “fill the deficit hole,” and “get Republicans out of the corner they have painted themselves into on climate change.” That last is particularly amusing: In 2009–10, the Democrats—while controlling the House of Representatives, 60 votes in the Senate, and the presidency—failed to pass climate change legislation, suggesting that the climate change political “corner” looks rather different than that apparently perceived by Mathur and Morris. Beware economists pretending to be politicians.

Competition of ideas: Read Mathur and Morris’s take: How to Improve Tax Reform: A Carbon Tax and Expanded Benefits for Working Families

In any event, this different policy goal obviously implies yet a different carbon tax per ton, one that would yield sufficient revenue to satisfy the first two of those objectives. Given the magnitude of prospective federal budget deficits in the absence of serious reforms of entitlement programs, this third carbon tax, in principle, would maximize revenues (or the present value of the revenue stream) over some time horizon. Note that Mathur and Morris assume implicitly that there would not be a stampede of interests demanding a share of the loot; only working families and deficit reduction would be the funding goals. Seriously?

The specific arguments offered by Mathur and Morris can be summarized as follows:

• The new tax bill (then being negotiated) “will add over a trillion dollars to the deficit over a ten-year window,” a problem that “can be solved with a carbon tax paired with an expansion of the EITC.”
• A carbon tax of $25 per metric ton, “rising at 5% per year over inflation” would “[reduce] US CO2 emissions . . . over 50% by 2040 relative to a business-as-usual emissions projection,” while “benefiting the environment” by “reduc[ing] harmful air pollutants like sulfur dioxide, mercury, particulate matter, and nitrogen oxides.”
• “Any carbon tax large enough to fill the GOP’s deficit hole would be large enough to deliver on the US commitment to the Paris climate agreement without a single additional regulatory measure.”
• “About 11 to 19% of the carbon revenues would keep the poorest 20 to 40% of low-income families whole on average . . . [leaving] at least 80% of revenues to cover the reduction in other taxes.”

Where to begin? Since Mathur and Morris do not pretend that their per-ton carbon tax has anything to do with the purported marginal social cost of GHG emissions, it is not quite clear why they need a “carbon” tax at all. Why not a tax on, say, okra, or fat-free ice cream, or argyle socks, or any of the other myriad monstrosities confronting modern mankind? Actually, it is clear: A carbon tax is where the big money (revenue) is. (Willie Sutton would be proud.) So in the rigorous analytic world of Mathur-Morris public finance, efficient taxation is driven no more by considerations of excess burden or deadweight losses (e.g., the Ramsey rule) or by allocating the costs of government outlays in accordance with varying demands for public spending. Instead, an efficient tax is one that fills a “deficit hole” that seems to exist independent of the spending decisions made (or not made) by Congress. As an aside, why is it “the GOP’s deficit hole?” Would a Democratic Congress spend less? Why is it not the “Beltway’s deficit hole?”

But never mind. Mathur and Morris might respond that the real key to filling the deficit hole—entitlement reform—is unavailable politically, in particular with a president (Mr. Trump) who campaigned against it and whose political coalition includes large numbers of voters who oppose it. So more revenue—lots of it—is the only game in town. Fair enough. But if Mathur and Morris are going to use political reality as a constraint driving their policy proposal, then we must ask what that reality says about their use of the carbon tax revenues to expand the EITC and to replace the revenues lost to a reduction in capital taxation.

Note again their claim that “about 11 to 19% of the carbon revenues would keep the poorest 20 to 40% of low-income families whole on average . . . [leaving] at least 80% of revenues to cover the reduction in other taxes.” What about the other 60–80% of low-income families? Will they not also demand to made whole? And the families neither low- nor high-income, that is, the vast middle class: How happy will they be to bear ever-higher energy costs while most or all of the revenues are used to subsidize others? Will their representatives in Congress not respond to their complaints? Would the recent cut in capital taxation—controversial enough all on its own—have passed Congress if it had been tied to an increase in individual taxes, whether on incomes or “carbon?” The question answers itself.

More generally, the implicit Mathur-Morris assumption that a new tax yielding massive new revenues somehow would not create a life-or-death tug-of-war over (new) spending simply is not credible. In other words, the real problem with the Mathur-Morris analytic framework is the implicit assumption—so very prevalent in academic public finance—that the magnitude and allocation of public spending are exogenous with respect to the taxes imposed by Congress. That cannot possibly be correct; indeed, it is difficult to believe that a carbon tax would emerge from the congressional bargaining process without an explicit quid pro quo in the form of expanded spending for groups harmed on net by the carbon tax and/or for groups viewed politically as the marginal (or “median”) voters. Mathur and Morris seem actually to believe that the majority coalition in Congress enacting a carbon tax will be willing to take the heat for higher energy costs without using the revenues to create some sort of offsetting political benefit. Why then have both Democratic and Republican Congresses refused to enact such a tax, a cap-and-trade system, or any other statutory constraints on the emissions of GHG, that is, a substantial increase in energy costs?

Note that their proposed carbon tax would begin at $25 per metric ton of CO2 (equivalent) and then would rise “at 5% per year over inflation,” apparently permanently. So the tax initially would add about 22 cents per gallon to the retail price of gasoline. (Consumption of a gallon of 10 percent ethanol-gasoline blend emits about 18.9 pounds of CO2.) Average household gasoline consumption is about 1,120 gallons per year. If we assume a national average gasoline price of $2.50 per gallon and a demand elasticity of 0.3 (in absolute value), household consumption would decline to about 1,090 gallons. Accordingly, the carbon tax paid by the average household (I assume perfectly elastic supply over the relevant range) would be about $240 per year, which is an underestimate of the economic cost of the gasoline component of the carbon tax imposed on households because the reduced gasoline consumption is a cost in terms of what economists call lost “consumer surplus.” In any event, $240 is about a quarter of the average household tax cut just enacted.

Since the tax rises at a real rate of 5 percent annually, it would be 28 cents per gallon after five years, 36 cents after 10 years, and 58 cents after 20 years. These figures shunt aside the increases in the prices of a vast array of goods and services engendered by the Mathur-Morris carbon tax—the tax means automatically that the private sector shrinks while the government sector grows—but even the narrow gasoline component after 20 years would represent, annually, more than half of the recent income tax cut for households. Mathur and Morris might respond that the carbon tax is merely an offset for reductions in other (corporate) taxes, but as discussed above the assumption that the carbon tax would emerge from Congress without massive new spending is not to be taken seriously.

Let us turn now to the not-very-rigorous analysis of climate and environment policy used as a partial justification for the Mathur-Morris proposal. Their claim that the tax would “[reduce] US CO2 emissions . . . over 50% by 2040 relative to a business-as-usual emissions projection” sounds impressive—if one assumes that increasing GHG emissions are a serious problem, a proposition vastly less obvious than commonly asserted. But Mathur and Morris seem curiously uninterested in the future temperature effect of that reduction in US GHG emissions. After all, is that not the central goal of GHG policy? Put aside the fact that there are many “business-as-usual” emissions scenarios, not all of which are very plausible. If we choose one “low” emissions path and one “high” one and apply to them the EPA climate model under several assumptions (in particular, a climate sensitivity of 4.5 degrees for a doubling of GHG concentrations) that exaggerate the future temperature effect of that GHG reduction, we get an average temperature effect in 2100 of 0.07 degrees.

That is smaller than the standard deviation (about 0.11 degrees) of the surface (land-ocean) temperature record. More centrally for policy analysis: How much is that trivial temperature effect—effectively zero—worth? Note that the political cost of the Obama climate action plan—a reduction in US GHG emissions of 17 percent—was perceived to be sufficiently high that it was not even considered by the 2009–10 Democratic Congress. What does that tell us about the politics of a 50 percent reduction, putting aside the differences in timing and other details?

With respect to the ancillary reduction in “harmful air pollutants like sulfur dioxide, mercury, particulate matter, and nitrogen oxides,” Mathur and Morris are well-trained economists and clearly understand that such reductions are not free. Accordingly, emissions or levels of pollutants (or ambient air quality) can be too low or too high in a benefit-cost sense. Because the EPA, upon determining that a given effluent endangers the public health and safety, is required to promulgate primary and secondary national ambient standards that “protect the public health” [with] “an adequate margin of safety,” we have such standards and emissions limits for all the pollutants noted by Mathur and Morris and for many others. Are Mathur and Morris arguing that the current system of limiting air pollutants fails to satisfy the requirements of the law? Are they arguing that “protect[ion of] the public health” [with] “an adequate margin of safety” is too lax a standard? Are they assuming that any reduction in effluents by definition is efficient? (If so, why are they not living on a pristine desert island?) How much thought have Mathur and Morris given this issue?

The ice beneath their feet is no thicker when they assert that their proposed carbon tax “would be large enough to deliver on the US commitment to the Paris climate agreement without a single additional regulatory measure.” Ask not about the utter silliness of the Paris climate agreement; ask instead the central question that any economist should address, again the one ignored by Mathur and Morris: What are the respective temperature effects in 2100? Assuming the entire Paris agreement is implemented immediately and that every party adheres to it strictly: 0.17 degrees. For the US: 0.015 degrees. Add another 0.01 degrees if you believe that the Obama pseudo-agreement with China is meaningful. (It is not.)

The phrase “carbon tax,” however solidly embedded in the public discourse, is a misnomer in that carbon dioxide is not “carbon” and it is not a pollutant. By far the most important GHG in terms of the radiative properties of the troposphere is water vapor; why does no one call it a “pollutant?” Obviously, it is because ocean evaporation is a natural process as are volcanic eruptions, the emissions from which of fluorine, sulfur, mercury, and ash are pollutants by any definition. However cumbersome, the term “GHG tax” would be more accurate and more consistent with rigorous thinking.

Like Zilly, who confronted many threats over the years, so does the Mathur-Morris carbon tax supposedly solve a number of problems at once. That alone is a sound reason to be skeptical. Unlike Zilly, in the beginning a bestial horror show that consumed Japanese cities, the carbon tax instead destroys gobs of other people’s money with no environmental benefits whatever and with a notional reduction in budget deficits that is almost certain not to result. Mathur and Morris should rethink their analysis.

Benjamin Zycher is a resident scholar at the American Enterprise Institute.