Government can use three different policies to combat carbon emissions and climate change:

• Make carbon-free energy cheaper with subsidies and government-funded research and development.
• Make carbon-intensive energy more expensive using such mechanisms as a carbon tax.
• Mandate a certain level of energy production from "clean" sources and "cap" carbon emissions using cap-and-trade mechanisms.

A carbon tax is the preferred choice for many economists because it would force emitters to contend with the full cost of producing a ton of carbon, instead of just the emitter's private cost. Burning carbon has an external cost because it produces a greenhouse gas (GHG) that accumulates in the atmosphere and risks unwanted climate change: higher global temperatures, greater climate variability, and possible increases in sea levels. Analysts and policymakers refer to this external cost as the social cost of carbon (SCC).

In economics, a carbon tax would be a Pigouvian tax on a negative externality. The tax would convey proper price signals to consumers and producers (e.g., the price level corresponds to the social cost of production that includes the damage from GHG emissions), stimulate R&D on clean technologies, and avoid the use of such traditional policies as command-and-control regulation, subsidies, and other inherently inefficient schemes. It could also motivate actors to adopt cost-effective emissions mitigation, perhaps by using different energy technologies or avoiding emissions altogether. For these reasons, economists believe prices should reflect scarcity and the social value of a good or service.

A related argument offered by economists for a carbon tax is that it represents a market-oriented solution that corrects for market inefficiency or failure. It places the different energy sources on a level playing field, making low-carbon technologies and energy efficiency more attractive. Because non-clean energy currently receives an implicit subsidy by avoiding the social cost, a tax would undo or at least mitigate that subsidy. But, as discussed below, if the SCC is so speculative that we have little idea of its optimal value, there is the risk that an "excessively high" tax on carbon would result in the opposite problem: an inefficient subsidy to clean energy.

Blackboard economics / The takeaway from Davis's article is that, though economists like to work in an idealized world with stylized facts—one in which a Pigouvian tax simply resolves the carbon emissions problem—policymakers must consider the context and details of the politics and measurement problems with the SCC. Because a tax would lower GHG emissions, the tendency is to think that we could then devote less effort toward adaptation to climate change, say, in agriculture or water-management planning. As discussed later, adaptation may be more effective and economical than reducing GHG emissions.

A carbon tax could have negative net benefits (assuming a reasonable discount rate): costs are incurred now but benefits from such reductions are in the distant future and highly uncertain.

Under specific conditions, a quantity-based measure such as a cap-and-trade mechanism would be preferred to a carbon tax. For example, in comparison with a tax, tradable emissions permits reduce the uncertainty in attaining a specified level of carbon in the atmosphere. This is because the responsible governmental agency controls the number of emissions permits that it issues.

Implementation issues relate to the size of the tax and how soon it should be imposed. For example, how large would a carbon tax have to be to reduce U.S. carbon emissions by 30%? There is much uncertainty over how economic actors would respond to such a tax and, thus, on the amount of carbon emissions reduced.

Even with a carbon tax, climate advocates would surely also want more stringent (inefficient) regulations on energy production and use. They would never agree to a carbon tax unless it was extremely high. Even then, the world would likely fall short of the 2°C limit on global warming advocated by climate activists; meanwhile, such a high tax rate would face strong opposition from the political right and center. We know from the past that when governmental action leads to higher energy prices, often there is a public outcry. We also know that environmentalists are generally skeptical of market-based approaches; their preferences are for subsidies for clean energy technologies and caps on emissions. The trouble with subsidies is that they reflect heavy-handed regulation and require policymakers with imperfect information to choose specific technologies for preferential treatment.

Proposals for a carbon tax almost always specify how the resulting revenues should be allocated. Proposed uses include reducing the national debt and cutting taxes, funding R&D of clean energy technologies, and distributing lump-sum rebates to households. One argument for the last option is that it would compensate low-income households for the regressive outcome from a carbon tax. It would also supposedly garner wider political support than the other options. Some have argued that if a carbon tax can help eliminate or reduce distortionary taxes (e.g., income and payroll taxes)—which is a big if—society would be better off when government revenues derive from Pigouvian-type taxes (i.e., taxes on a negative externality) rather than from other taxes.

Some advocates of a carbon tax argue that it would be easier to undo a tax—if that became necessary—than other policy approaches. But we know from experiences with other taxes that beneficiaries (e.g., clean energy producers) will exert strong political opposition to the abolition of a tax.

Setting the tax / In theory, a carbon tax should equal the SCC. But there are a number of problems with turning that simple theory into reality.

For one thing, estimates of the SCC are wildly speculative and vary within a wide range. Some analysts argue that climate change will be moderate, will occur in the distant future, and will have only a small effect on the economies of most countries. This would imply that the SCC is small, perhaps only around $10 per ton of carbon dioxide emissions. Others argue that without an immediate and stringent GHG abatement policy, there is a reasonable chance of substantial temperature increases that could have a catastrophic economic effect. If so, the SCC is large, perhaps as high as $200 per ton of carbon dioxide. The SCC is extremely sensitive to parameters that are subjective—"garbage in, garbage out." One can come up with an SCC that best advances one's agenda; for example, the SCC is especially sensitive to the discount rate because most climate change scenarios predict major damages only after several decades have passed.

Another problem is the United States would experience only a small part of the global harm from climate change, even though a ton of U.S. emissions produces the same warming as a foreign ton. As a result, Americans might only accept a low carbon tax, which would have only a small effect on change in global temperature.

Also, it is questionable whether a carbon tax would have even a detectable effect on climate change. If the demand for "dirty" energy is highly inelastic, people may be willing to pay the tax and continue emitting carbon. This undermines the position of those who believe that even if a carbon tax fails a benefit–cost test, it can still act as a form of "insurance" by preventing "worst case" scenarios (or "fat tails," in the literature) of climate change. The overall effect is the government collecting huge tax revenues while showing little effect on global temperatures.

Yet another problem is that, while assuming a zero value for the SCC may be wrong, it may be better (i.e., reduce welfare losses by less) than to assume some grossly high level for a carbon tax. Also, estimates of the SCC require forecasts of climate change to relate, for example, economic welfare losses to global temperature change. As of today, those forecasts are highly speculative, diminishing their use for setting the tax. One final problem is that a carbon tax would do little to reduce carbon emissions outside the United States, where the climate-change battle will be won or lost. While it has long been noted that the United States was the largest (and now second-largest, after China) producer of GHG, over 85% of carbon emissions originate outside the United States.

Policy implications / Given the problems with the Pigouvian tax on GHG, we should give more consideration to adaptive strategies, which can evolve over time in response to new information. The best scientific evidence shows that warming of the Earth's atmosphere will occur gradually, allowing ample time for adaptive measures to mitigate the effects of climate change. As we learn from basic economics and other contexts, people adapt when change implies a need to reoptimize.

Adaptation can involve adjusting planting dates and crop locations, building and strengthening seawalls, and urban and rural flood management. Adaptation does not require international cooperation, which, as discussed below, is highly difficult to achieve. Arguably, this is a more effective, less costly, and more practical strategy than futilely trying to limit global emissions in order to achieve a stringent temperature-change target.

A U.S. carbon tax by itself would accomplish little in reducing global temperature. Deep decarbonization to achieve stringent temperature-change limits (e.g., no more than 2° C) requires collective action between countries. One country, even one as big as the United States, cannot do it alone with a carbon tax. But it is hard to reach an effective agreement on limiting emissions even if it is in the world's common interest because of free-riding or countries understating their willingness to pay the costs for climate-change mitigation. (The rise of populism has also aggravated the difficulty for international cooperation.) Internal costs of these policies for individual countries could be significant, but benefits are unclear, distant, and broadly spread across all countries. In effect, countries see potentially high costs of actions that primarily benefit others.

Lastly, a carbon tax would meet with strong political opposition: the left wants more draconian measures like the Green New Deal while the right would oppose any tax at all. Although a favorite of many economists, we are unlikely to see an effective U.S. carbon tax unless the political and economic dynamics radically change.