Written by Frank Ackerman
February 23, 2019
Fourth in a series on climate policy.
We need a price on carbon emissions. This opinion, virtually unanimous among economists, is also shared by a growing number of advocates and policymakers. But unanimity disappears in the debate over how to price carbon: there is continuing controversy about the merits of taxes vs. cap-and-trade systems for pricing emissions, and about the role for complementary, non-price policies.
At the risk of spoiling the suspense, this blog post reaches two main conclusions: First, under either a carbon tax or a cap-and-trade system, the price level matters more than the mechanism used to reach that price. Second, under either approach, a reasonably high price is necessary but not sufficient for climate policy; other measures are needed to complement price incentives.
Why taxes and cap-and-trade systems are similar
A carbon tax raises the cost of fossil fuels directly, by taxing their carbon emissions from combustion. This is most easily done upstream, i.e. taxing the oil or gas well, coal mine, or fuel importer, who presumably passes the tax on to end users. There are only hundreds of upstream fuel producers and importers to keep track of, compared to millions of end users.
A cap-and-trade system accomplishes the same thing indirectly, by setting a cap on total allowable emissions, and issuing that many annual allowances. Companies that want to sell or use fossil fuels are required to hold allowances equal to their emissions. If the cap is low enough to make allowances a scarce resource, then the market will establish a price on allowances – in effect, a price on greenhouse gas emissions. Again, it is easier to apply allowance requirements, and thus induce carbon trading, at the upstream level rather than on millions of end users.
If the price of emissions is, for example, $50 per ton of carbon dioxide, then any firm that can reduce emissions for less than $50 a ton will do so – under either a tax or cap-and-trade system. Cutting emissions reduces tax payments, under a carbon tax; it reduces the need to buy allowances under a cap-and-trade system. The price, not the mechanism, is what matters for this incentive effect.
A review of the economics literature on carbon taxes vs. cap-and-trade systems found a number of other points of similarity. Either system can be configured to achieve a desired distribution of the burden on households and industries, e.g. via free allocation of some allowances, or partial exemption from taxes. Money raised from either taxes or allowance auctions could be wholly or partially refunded to households. Either approach can be manipulated to reduce effects on international competitiveness.
And problems raised with offsets – along the lines of credits given too casually for tree-planting – are not unique to cap and trade. A carbon tax could emerge from Congress riddled with obscure loopholes, which could be as damaging to the integrity of carbon pricing as any of the poorly written offset provisions of existing cap-and-trade systems. More positively speaking, either approach to carbon pricing can be carried out either with or without offsets and tax exemptions.
Why taxes and cap-and-trade systems are different
Compared to the numerous similarities between the two approaches, the list of differences is a shorter one. A carbon tax is easier and cheaper to administer. In theory, a carbon tax provides certainty about the price of emissions, while a cap-and-trade system provides certainty about the quantity of emissions (in practice, these certainties can be undone by too-frequent tinkering with tax rates or emissions caps).
Cap-and-trade systems have been more widely used in practice. The European Union’s Emissions Trading System (EU ETS) is the world’s largest carbon market. Others include the linked carbon market of California and several Canadian provinces, and the Regional Greenhouse Gas Initiative (RGGI) among states in the Northeast.
Numerous critics have pointed to potential flaws in cap-and-trade, such as overly generous, poorly monitored offsets. Many recent cap-and-trade systems, introduced in a conservative era, began with caps so high and prices so low that they have little effect (leaving them open to the criticism that the administrative costs are not justified by the skimpy results). The price must be high enough, and the cap must be low enough, to alter the behavior of major emitters.
The same applies, of course, to a carbon tax. Starting with a trivial level of carbon tax, in order to calm opponents of the measure, runs the risk of “proving” that a carbon price has no effect. The correct starting price under either system is the highest price that is politically acceptable; there is no hope of “getting the prices right” due to the uncertain and potentially disastrous scope of climate damages.
Perhaps the most salient difference between taxes and cap-and-trade is political rather than economic: in an era when people like to chant “no new taxes”, the prospects for any initiative seem worse if it involves a new tax. This could explain why there is so much more experience to date with cap-and-trade systems.
Beyond price incentives
Some carbon emitters, for instance in electricity generation, have multiple choices among alternative technologies. In such cases, price incentives alone are powerful, and producers can respond incrementally, retiring and replacing individual plants when appropriate. Other sectors face barriers that an individual firm cannot usually overcome on its own. Electric vehicles are not practical without an extensive recharging and repair infrastructure, which is just beginning to exist in a few parts of the country. In this case, no reasonable level of carbon price can, by itself, bring an adequate nationwide electric vehicle infrastructure into existence. Policies that build and promote electric vehicle infrastructure are valuable complements to a carbon price: they create a combined incentive to move away from gasoline.
Yet another reason for combining non-price climate policies with a carbon price is that purely price-based decision-making can be exhausting. People could calculate for themselves the fuel saved by buying a more fuel-efficient car and subtract that from the sticker price of the vehicle, but it is not an easy calculation. Federal and state fuel economy standards make the process simpler, by setting a floor underneath vehicle fuel efficiency.
When buying a major appliance, it is possible in theory to read the energy efficiency sticker on the carton, calculate your average annual use of the appliance, convert it to dollars saved per year, and see if that savings justifies purchase of a more efficient appliance. But who does all that arithmetic? Even I don’t want to do that calculation, and I have a PhD in economics and enjoy playing with numbers. My guess is that virtually no one does the calculation consistently and correctly. On the other hand, federal and state appliance efficiency standards have often set minimum levels of required efficiency, which increase over time. It’s much more fun to buy something off the shelf that meets those standards, instead of settling in for an extended data-crunching session any time you need a new fridge, air conditioner, washing machine…
In short, the carbon price is what matters, not the mechanism used to adopt that price. And whatever the price, non-price climate policies are needed as well – both to build things that no one company can do on its own, and to make energy-efficient choices accessible to all, without heroic feats of calculation.