Cap-and-Trade vs. Emission Tax – Design Issues
August 20th, 2007 <-- by Joseph Aldy -->The cost and environmental efficacy of either a cap-and-trade program or an emission tax will depend on several important design issues: the point of regulation; the coverage of emission sources; “complementary” programs; and possible hybrid policies that integrate elements of both approaches.
Point of regulation. It would be infeasible to regulate all greenhouse gas emissions at the very point where they enter the atmosphere – monitors will not be placed on every car and truck tailpipe, every home that heats with natural gas or heating oil, as well as every smokestack in the industrial and electricity sectors. One approach – often referred to as upstream regulation – would impose the emission tax or the requirement to hold permits on energy suppliers, such as at coal mines, natural gas wellheads, petroleum product refineries and importers. The carbon content of all fossil fuels that enter the U.S. energy system would be covered by this approach. This approach would be administratively simple and straightforward because it accounts for almost all U.S. CO2 emissions (more than 98 percent) by focusing on a relatively small number of firms; incorporates existing monitoring and measurement of fuel supplies; and takes advantage of the fundamental molecular properties of fossil fuels that allow for precise measurement of emissions based on the physical amounts of these fuels.
A midstream approach could focus instead on major sources of emissions, such as electric utilities and facilities with large industrial boilers, in a similar fashion as the SO2 trading program. The question then becomes how to integrate other emission sources. In a cap-and-trade program, emissions targets are met at least cost when all sources face the same marginal cost for their abatement. It may be difficult to cover other sources, such as in the transportation sector, in a similar midstream or downstream manner. The point of regulation can influence the extent of coverage of the domestic policy, which raises the costs and lowers the environmental benefits of an emission tax or cap-and-trade program.
Coverage. The broader the coverage of a domestic policy, the lower the costs and the greater the environmental benefits. Some proposals would exclude some sources of greenhouse gas emissions by focusing on just one sector, such as power generation. This raises the costs of the program – with utilities facing a cost for emitting carbon while other sources in the economy facing no cost, the economy foregoes low-cost abatement opportunities. It can also reduce the environmental benefits, as consumers and firms respond to the effect of the program on the utility sector – households install natural gas water heaters (uncovered) in place of electric water heaters (effectively covered), and manufacturing facilities decide to generate their own power instead of buying power from covered utilities. All of these actions would cause emissions leakage – emissions abatement in the covered sector is partially offset by increases in emissions outside of this sector.
The coverage of a tax or cap-and-trade program can also be influenced by “carve-outs.” A domestic policy could exempt some firms based on their size (e.g., those emitting less than a specific threshold) or if they are intensive users of energy. Such exemptions, in either cap-and-trade or a tax, reduce the cost-effectiveness of the policy.
“Complementary” programs. Some have advocated for performance standards – such as on appliances, buildings, fuel economy, and even power plants – to “complement” a cap-and-trade program. These standards do not complement cap-and-trade or emission tax approaches – they increase the total costs to the U.S. economy of a domestic climate change policy. These standards increase costs because they cause marginal costs to differ across sources. There is no environmental need to implement such standards with a cap-and-trade program – the cap limits the emissions, so the only effect of standards is to raise costs. Standards with an emission tax may lower tax payments but increase regulatory compliance costs, and thus increase the costs of goods and services to consumers. Well-designed complementary programs should instead target climate-friendly R&D through support for more basic research and incentives for the private sector to increase their investment in R&D.
Hybrid policies. Some proposals would integrate elements of both cap-and-trade and an emission tax. For example, a cap-and-trade program with a safety valve would allocate emission allowances to covered firms just like a traditional cap-and-trade system. Firms could trade allowances among themselves, or tap the safety valve – the government would sell an unlimited number of allowances at a predetermined price. This effectively acts as a price ceiling in the allowance market – at low allowance prices, the system is more like a normal cap-and-trade, but if costs are unexpectedly high and the safety valve is triggered, then it is more like an emission tax.
August 21st, 2007 at 5:47 pm
Do you have sources that support your argument against performance standards?
August 21st, 2007 at 9:04 pm
This is certainly a useful series!
August 21st, 2007 at 10:58 pm
Response to Query on Sources about Performance Standards
Performance standards raise costs relative to market-based approaches such as cap-and-trade and emission taxes because standards are not designed to equate marginal costs across all emission sources. Under a tax, all sources face the same marginal cost for another ton of emissions – the per ton tax. Under a cap-and-trade, all sources face the same marginal cost for another ton of emissions – the cost of a one-ton allowance. Performance standards do not equate marginal costs across sources; they equate some measure of effort (such as grams per mile or pounds per megawatt-hour). When firms in the economy have heterogeneous costs of abatement, then making them all do the same thing is more costly than allowing a government-created allowance market to work to achieve the same environmental outcome.
Suppose that there are two utilities: The Power Station and Robert Plant. They both emit five tons of a pollutant and they both generate the same amount of electricity. They differ, however, in how much it costs to abate their emissions. At The Power Station, it costs $1 to abate one ton of emissions, $2 for the second ton, and so on. At Robert Plant, it costs $3 to abate one ton, $6 for the second ton and so on. The government decides that pollution should be reduced four tons (from 10 to 6) to benefit the public health. A regulator could decide to assign these two utilities the same reduction of two tons each since they emit the same amount and generate the same amount of power. This would cost the economy $1+$2 (The Power Station) + $3+$6 (Robert Plant) = $12. If instead the regulator decides to set the cap equal to six tons – the amount these two utilities are allowed to emit in total – and allocates allowances equal to six tons, then they will buy and sell allowances and find the most economical way to reduce their emissions. After trading, The Power Station should hold two allowances, and abate three tons of emissions. The cost of their last ton of abatement is $3 and their total cost is $6 (1+2+3). Robert Plant will hold four allowances and abate one ton of emissions at a marginal and total cost of $3. The total cost to the economy of cap-and-trade is $9, or 25 percent less than the performance standard in this example.
Here are some sources on this point (I’m sorry, I only know of URLs for some of these papers):
For a good review of the economic explanations for why standards cost the economy more than emission taxes and cap-and-trade as well as an assessment of the early performance of the SO2 cap-and-trade market (including estimates that cap-and-trade costs half as much as performance standards), you can refer to:
Robert N. Stavins. “What Can We Learn from the Grand Policy Experiment? Positive and Normative Lessons from SO2 Allowance Trading.” Journal of Economic Perspectives, Volume 12, Number 3, pages 69-88, Summer 1998. http://ksghome.harvard.edu/~rstavins/Papers/What%20Can%20We%20Learn%20from%20the%20Grand%20Policy%20Experiment….pdf
Richard Schmalensee, Paul L. Joskow, A. Denny Ellerman, Juan Pablo Montero, and Elizabeth M. Bailey. “An Interim Evaluation of Sulfur Dioxide Emissions Trading.” Journal of Economic Perspectives, Volume 12, Number 3, pages 53-68, Summer 1998.
Curtis Carlson, Dallas Burtraw, Maureen Cropper, and Karen L. Palmer. “SO2 Control by Electric Utilities: What are the Gains from Trade?” Journal of Political Economy Vol. 108, No. 6, pages 1292-1326, December 2000.
http://www.rff.org/Documents/RFF-DP-98-44-REV.pdf
For fuel economy standards with relevant comparisons to gasoline taxes and carbon taxes, you can refer to:
Crandall, Robert 1992. “Policy Watch: Corporate Average Fuel Economy Standards.” Journal of Economic Perspectives 6(2): 171-180.
Kleit, Andrew N. “Impacts of Long-Range Increases in the Corporate Average Fuel Economy (CAFE) Standard,” Economic Inquiry 42:2 (April 2004) 279-294. http://aei.brookings.org/admin/pdffiles/phpQQ.pdf
Jacobsen, Mark. “Evaluating U.S. Fuel Economy Standards in a Model with Producer and Household Heterogeneity” Working Paper, Stanford University, 2007.
http://www.stanford.edu/~mrj/Jacobsen_job_market_paper.pdf
August 22nd, 2007 at 2:57 pm
First question: it is cheaper to include as many sectors as possible in an emissions trading program, but from a long-term view, does it ever make sense to focus on a sector where there are long-term factors, such as longer timeline for R&D, or perhaps the length of time the building is standing?
Second, I find one point confusing — what is the mechanism?
Presumably some company will find it cheaper to convince people to shift to compact fluorescent bulbs and more efficient appliances — is it the power sector that can produce limited GHG? Or do the manufacturers somehow pay for the use of the product as well? Who pays for emissions of new buildings with poor construction?
It is cheaper already, without any GHG tax or cap, for people to buy low-energy bulbs. Ditto for high fuel economy (though smaller) cars. Yet most people opt for expensive bulbs and cars. Wouldn’t standards to some extent reinforce the principle of opting for cheaper solutions?
It seems to me that standards are simpler whenever there are a large number of decision-makers. Or will incandescent bulb manufacturers go out of business because they can’t afford the permits?
Thanks for this series.
September 4th, 2007 at 4:47 pm
The first question raises an important point in evaluating the design of a domestic climate change policy. Because the lifetimes of capital varies across sectors, it may be valuable to complement a trading program with an R&D program focused on reducing the emissions intensity of long-lived capital. In this sense, targeting R&D efforts on reducing emissions in the utility sector or increasing energy efficiency in buildings may be valuable. It is important to recognize, however, that a broader cap-and-trade or tax policy will still be preferred to a narrow, sector-oriented approach. The broader policy can lower total costs, which allows for more ambitious emissions abatement to be achieved at lower economic cost.
The mechanism to promote emissions abatement is effectively the market mechanism modified by the cap-and-trade or tax policy. Either of these policies will increase the cost of emissions-intensive goods and services. Firms and consumers will respond to these higher prices for these goods and services and find ways to economize in their use of them. For example, an upstream cap-and-trade or CO2 tax will increase the cost of consuming fossil-based energy by consumers. This increases the incentive for consumers to buy more efficient light bulbs and more efficient automobiles. Just as prices in the general economy serve as the mechanism for effectively allocating goods and services among consumers and firms, the price of fossil fuel-based goods and services with the embedded carbon price will also allocate the effort implicit in lowering emissions. Firms that can easily respond to a carbon price – say utilities that can dispatch more natural gas or wind power in lieu of coal-based electricity – will adjust quickly to the carbon price. Others, with higher costs of adjustment, may not respond as significantly.
The benefit of this mechanism is that it allows firms and consumers to decide how best to respond to the carbon constraint and emissions price (under cap-and-trade). Standards do not provide this flexibility, and in doing so, increase the costs of reducing emissions. Mandating that all consumers drive super-efficient cars, even if this were the “cheaper solution”, would not necessarily make them better off. A large family, or a building contractor, or a delivery driver, may all need less fuel-efficient, larger vehicles, than say a single 20-something. Ideally, we would like to implement policies that ensure that individuals and firms are as well off as possible given that we meet our emissions abatement goals. Making someone drive a car that they would otherwise not want to drive imposes a cost. Providing the flexibility of a market-based approach, like a tax or cap-and-trade program, will give these individuals and firms the freedom to meet the environmental goal at lower cost than a standard.
Finally, with a large number of decision-makers, a market for tradable permits could be fairly deep and liquid. Most visions of a cap-and-trade program would not require the manufacturer of light bulbs hold permits, or the manufacturers of most other goods. Instead, the program would require either the suppliers of fossil fuels to hold permits to cover the CO2 content of those fuels or require major emitters of greenhouse gases (such as power plants and industrial boilers) to hold permits.
September 6th, 2007 at 9:18 pm
Thanks again, but I don’t understand several points, and I (sort of) teach the subject.
Let’s just look at one: me trying to make decisions about remodeling my house, in the absence of any regulations on insulation, lighting, and smoke alarms.
I as the person who will be buying more expensive electricity (21% coal, 40% natural gas in my state), as well as gas for heating, will figure out the optimum amount of insulation, and optimum lighting fixtures, ditto for smoke alarms to protect my property, to minimize my overall costs. There is no need for regulations on these, or the type of windows. Is this what you intend? FYI, this didn’t work for me last time.
I keep reading that regulation is needed when market forces are too nuanced, for example, compact fluorescents, with close to zero market share, are already cheaper than incandescents. But you disagree with this? Along the same lines, the increasing price of oil, damaging to the third world and not all that good for our economy, was supposed to lead to demand decreases.
It seems to me that a combination of regulation and cap-and-trade would make my decision-making process easier!
Thanks for your help.
September 11th, 2007 at 9:43 am
Hi Joseph,
First let me join others in thanking you for the very useful and informative posts. Too often it seems the blogosphere tends to focus on the tribal aspects of the science of climate change rather than the pros/cons of various policy solutions (i.e. standard vs tax vs emissions trading).
In your posts you mention some of the trade-offs between these solutions, for example cost certainty with taxes, and emissions certainty with cap-and-trade. I’m wondering what your thoughts are on some of the other issues that come to mind. In particular, how do think these three approaches compare when you include transaction costs?
October 26th, 2007 at 8:54 pm
Hi Joseph,
As with the others, thanks again for the insightful series and thoughtful approach. I think what I take away from this upstream regulation makes more sense from a regulatory perspective, and that it’s the completeness of the coverage, much more than the cap-and-trade or tax design is what will make a climate change policy work efficiently.
My question for you is that given that in an ideal world we can debate about whether cost-certainty or emission-certainty is preferable, but of course we don’t set policy in an ideal world, we set policy in a highly convoluted political process with all sorts of agendas having political pull. So, given the political reality of trying to get something like this through congress esp. with something close to 100% coverage, do you think that there is a political advantage to either a tax or a cap-and-trade policy?
December 22nd, 2007 at 3:31 pm
i agree that things need to be done on a global scale to be effective
what happened to the proposal for a natural capital depletion tax? by the way, america is historically anti-tax (thoreau; alexis de toqueville). why did the democrats propose a tax amendment to the energy bill when they could have proposed a tax credit or shifted subsidy, the oil, gas, and coal lobbies being the largest in washington?
grant