The February 2007 report from the Intergovernmental Panel on Climate Change declared with 90 percent certainty that human activity was responsible for global warming. Some thirty-five developed nations have agreed to mandatory emissions targets set by the Kyoto Protocol. Yet Washington, which experts say is the linchpin for more significant international action, has yet to impose caps on greenhouse gas emissions. Most agree that U.S. regulation is now only a matter of time, but there is disagreement over how to curb carbon emissions.
William A. Pizer, senior fellow at Resources for the Future, and Kenneth P. Green, resident scholar at the American Enterprise Institute, debate how the United States should regulate greenhouse gas emissions.
June 29, 2007
Kenneth P. Green
Mr. Pizer raises an important, if subjective question: which carbon control system is more likely to be well-designed. I suspect this is where we will have to agree to disagree. Mr. Pizer thinks an emission-trading regime would be more likely to embody the set of features we both agree upon, while I believe that a revenue-neutral carbon tax is more likely to do so. To me, the answer to this question lies in the nature of the incentives created along the way, and the institutional transparency that tends to limit overt corruption.
In an emission-trading regime revenue would flow largely from consumers to for-profit, private-sector entities that claim reduced emissions, and to the many traders and auditors that would be involved in the process. One has to presume the profit incentive offered would lead participants to push the limits of legality at all stages in the process, from the assessment of historic emissions; to efforts to claim routine equipment upgrades as credit-worthy actions; to lobbying against an auction; and, in some cases, to outright false-credit generation. There is also potential for profiteering at the consumer’s expense. If one pays a business for not doing something, they’ll gladly take the money and save the labor. If we pay mildly-profitable companies more in carbon credits than they can make via product sales, a goodly percentage will take the credits and reduce productive output, curtailing consumer choice. Most importantly, under an emission-trading regime, every entity that profits has an incentive to not only perpetuate the scheme, but to push for more restrictive carbon caps, while groups paying out will lobby for the right to buy cheap (unverifiable) credits abroad. Finally, most companies would have incentive to avoid revealing how much of their price fluctuations are caused by the carbon-permit market, both to please regulators, and to preserve secrecy about their profit margins.
While a carbon-tax regime has some latitude for cheating (understating emissions) and economic distortion (tax shifting), the incentives for most other forms of cheating, such as lobbying for tighter caps; seeking credit for business-as-usual investments, creating false sequestration credits; reducing useful output; and so on are negative. Very few companies will see a potential profit in an increase in the carbon tax, or be willing to spend money lobbying for it. The normal tax aversion of the public would serve as a useful check on the stringency of a carbon tax. And the fixed value of a carbon tax could easily be displayed on the gas pump, on an electricity bill, or on a natural gas bill. The rebate could be easily seen on a tax return.
For these reasons, I believe a carbon tax regime would be superior to an emission-trading regime for greenhouse gas emission control.
June 28, 2007
William A. Pizer
Mr. Green asks two questions. First, does SO2 and lead trading demonstrate emission trading is a viable option for greenhouse gas emissions? Many of the concerns he raises have nothing to do with whether a trading system will work. Proxy indicators, in particular, are recommended by the IPCC (PDF) and would presumably also be used in a tax approach. Among the other issues, the idea of too many diverse sources is interesting. Usually, more trading and greater diversity of traders would be good. But even by the numbers, there is not much difference between the SO2 program (3,456 electric generating units in 2005) and an upstream CO2 system (around 2,000 units). Regarding technological solutions, he ignores fuel switching and conservation—exactly the same results that would arise under a tax.
The second question is whether a safety valve can fix the volatility issue. Mr. Green discusses the recent run-up in SO2 prices and the failure of an apparent safety valve mechanism. But the recent run-up is not volatility—it reflects the tightening of the program under the new Clean Air Interstate Rule, with firms reducing emissions now to bank for the future. Recent SO2 prices (PDF) of around $1500 are extremely close to future compliance costs predicted by economic analyses (PDF). While Mr. Green suggests there is currently a safety valve, that is not true. The mechanism he refers to is a special reserve, offering up to fifty thousand allowances at $1500 (adjusted for inflation). The relevant proposals for a CO2 safety valve would not be limited to 0.5 percent of total allowances and would cap volatility. Discussions of the European Union program are irrelevant: They do not have a safety valve. Interestingly, there was no discussion of the effectiveness of true safety valve mechanisms in the trade and renewable energy applications where they have been used with considerable success.
In the end, much of our debate is really about whether a trading program or tax is more likely to be well-designed. We agree on the basic features that are important—coverage, price stability, raising revenue to cut other taxes, minimizing rent seeking. There are plenty of practical examples of how those features can be addressed in either approach. The question is whether they will be. Mr. Green claims that emission-trading schemes have been plagued by corruption and subversion. While I am not sure his claim is true, it is certainly true that taxes have been subject to such manipulations on a grand scale. And, while such forces in a tradable permit system lead to redistribution, in a tax system they lead to distortions—a much worse outcome.
June 27, 2007
Kenneth P. Green
Mr. Pizer makes many claims for the superiority of emission-trading regimes over a revenue-neutral carbon tax. I’ll examine two here: First, the claim that the successes of lead and sulfur trading demonstrate the utility of carbon emission trading regimes. Second, that price volatility can be averted with the proper use of “safety valves.”
Do SO2 and lead trading demonstrate emission trading is a viable option for greenhouse gases? A bit of drilling down suggests not. Sulfur and lead trading were local issues, and were pollutants of relatively short duration in the environment (before being rained out). There were far less entities that had to trade (as compared with greenhouse gas trading), and the chemicals in question were easily measured at the point of emission. Initially, SO2 trading was only applied to a single sector: Only 110 coal-fired power plants were included in the system, subsequently expanded to 445 plants. In addition, there were readily available technological options to reduce emissions. Carbon trading features none of these: There are no off-the-shelf technologies that can reduce the carbon content of fuel; there would be many thousands of diverse trading entities across multiple sectors of the economy (not simply coal power generation); emissions can only be monitored by proxy indicators; and the pollutants themselves are of long to extremely-long duration in the environment.
Can safety valves fix the volatility issue? Mr. Pizer points to the success of the SO2 trading program to suggest carbon trading would work well to control carbon and avoid price volatility. But as my colleagues and I point out in a recent AEI Environmental Policy Outlook, “There has been significant volatility in emission permit prices, ranging from a low of $66 per ton in 1997 to $860 per ton in 2006, as the overall emissions cap has been tightened, with the price moving up and down as much as 43 percent in a year. Over the last three years, SO2 permit prices have risen 80 percent a year, despite the EPA's authority to auction additional permits as a "safety valve" to smooth out this severe price volatility.” Carbon trading has fared no better in initial runs, as economist William Nordhaus points out: “We have preliminary indications that European trading prices for CO2 are highly volatile, fluctuating in a band and [changing] +/- 50 percent over the last year.”
It is true, in theory, that a perfectly designed carbon trading system can match the efficiency of a carbon tax. However, in practice, emission trading systems have been plagued by corruption and subversion that make such a perfect scheme highly unlikely.
June 26, 2007
William A. Pizer
It is useful to first highlight where Mr. Green and I agree—bans and regulations are highly inefficient. We also agree on the desirable goals for a domestic policy: an economy-wide incentive, transparency, strong institutional frameworks, minimizing rent-seeking and transfers, government revenue to cut other taxes, avoiding unnecessary price volatility, and (possibly) harmonizing policy internationally.
However, in contrast to Mr. Green, I see tax and cap-and-trade equivalent on many of these goals, and cap-and-trade exceeding taxes on several. Both policies can be imposed upstream in the fossil-fuel supply chain, covering virtually all U.S. emissions. Both involve a transparent price signal. While Mr. Green suggests trading systems are unenforceable and the institutions untested, I would argue they are as enforceable and tested as taxes. Both systems require identical measuring and reporting of associated emissions at some point in the fossil-fuel supply chain. And emissions trading institutions have proven themselves in everything from the 1980s lead phasedown in gasoline to the 1990s acid rain program, with a minimum of administration (fifty EPA [Environmental Protection Agency] staff manage the current acid rain program covering thousands of sources).
It is ironic that he would suggest, after the American Job Creation Act of 2004, that taxes are less likely to incite rent seeking than tradable permits. Indeed, rent seeking in a carbon tax program creates distortions in behavior, as well as redistributing rents—making the problem worse. While we both favor raising government revenue to cut other taxes, it is ridiculous to imagine that raising $50 billion of revenue through a carbon tax or permit auction to finance such a cut is not a redistribution from people paying for carbon emissions to those currently paying the other tax that will be cut. Indeed, perhaps such a swap is—at least initially—a less fair redistribution than giving some free permits to those industries and regions bearing more of the burden under a carbon tax or tradable permit program.
Avoiding price volatility is important and easily done in a trading program with a safety valve—exactly the same mechanism applied to policies as diverse as trade (tariff rate quotas) to performance standards (alternative compliance payments). Given its wide application, it is hard to argue it is a particularly complicated or nontransparent approach.
All of this goes to highlight exactly why I prefer a tradable permit system. It can do everything a tax does—and more. Most importantly, it addresses rent seeking head on rather than pretending it does not exist under a tax and ending up with a debate over exemptions and special interest deals.
June 25, 2007
Kenneth P. Green
Policymakers wishing to restrain greenhouse gas emissions have a broad range of potential instruments available. They can criminalize emitting activities, they can regulate emissions via technology requirements or emission standards, or they can put a price on the activity via a tax or trading scheme.
There is widespread agreement among economists and public policy analysts that activity bans and regulations are highly inefficient approaches to managing environmental externalities, particularly those such as climate change, where polluting activities span nearly all aspects of human life; cross all jurisdictional borders; have high levels of uncertainty with regard to costs and benefit delivery; and impose asymmetric costs and benefits. Thus, emission pricing—through taxation or the establishment of a pseudo-market that trades in emission permits—has been widely favored (by analysts) over regulatory approaches for several decades. Both taxes and emission-trading (also called cap-and-trade) impose a price on emissions, but the two systems are very different.
Taxing greenhouse gas emissions accomplishes several desirable goals in one stroke: It creates an economy-wide incentive to reduce greenhouse gas emissions; it is largely transparent; it operates within preexisting institutional frameworks adept at fraud prevention; it minimizes the potential for rent seeking; it does not lead to wealth transfer between regions with different forms of economic activity; it produces revenue that can be used to reduce other taxes in order to offset the economic harm of higher energy prices; it is predictable, adjustable, and can thereby avoid price volatility; it shifts some revenue generation from production to consumption; and it can be harmonized internationally if desired.
Emission trading systems by contrast (particularly international systems) are virtually unenforceable; create massive incentives for fraudulent claims of prior emission estimates and emission reductions; lead to massive wealth redistribution; require new untested institutions that have performed poorly in pilot testing (the European Trading System is a prime example); create incentives for rent seeking; generate no revenue to offset the economic harm of higher energy prices; require complicated “safety valves” to prevent massive energy price volatility; and are largely nontransparent.
We recently estimated that a tax of fifteen dollars per ton of CO2-emitted levied on the carbon content of fossil fuels would produce an 11 percent reduction in greenhouse gas emissions while raising the costs of crude oil and natural gas modestly. The majority of the price increase would affect coal prices and hence coal-based electricity. This is entirely appropriate as that, to paraphrase bank robber Willy Sutton, is where the emissions are.
For these reasons, I believe that a revenue-neutral carbon tax is a superior policy alternative to emission trading, regulation, or activity bans if our goal is the cost-efficient reduction of greenhouse gas emissions.
June 23, 2007
William A. Pizer
For most of the past thirty years, economists have strongly advocated market mechanisms (PDF)—either taxes or tradable permits—to encourage emissions reductions because they effectively reduce emissions at the lowest cost to society. The alternative—regulation that tells businesses how much to emit or what technology to use—risks requiring expensive options in some sectors while cheaper options in other sectors remain untouched. Because climate change is easily the most expensive environmental problem we have ever tried to tackle, doing so at the lowest cost is critical.
Therefore, the most important feature of a U.S. program is that it should be a single market-based policy—either a tax or a tradable permit system—and it should cover as much of total U.S. emissions as possible, across regions, sectors, and various greenhouse gases.
Between taxes and tradable permits, I believe a tradable permit system is a better way to go for two overarching reasons. First, a tradable permit system can be designed to mimic all of the key features of a tax and can do more; and second, this flexibility to do more—specifically to easily provide compensation to various businesses and individuals through a free allocation—takes away pressure to exclude some sectors.
Taxes are typically advocated because they fix the price rather than leaving it to fluctuate in response to volatility in permit demand, and they raise government revenue that can be used to cut other taxes. Yet, a tradable permit system—through use of a safety valve and permit auctions—can match these two features. A safety valve (PDF) would limit the permit price by having the government supply unlimited “extra” permits at a specified price. If the emission cap is low enough, the safety valve can fix the price with certainty. Similarly, the government can sell off permits and raise money to cut other taxes.
But a permit system can do more. While a safety valve can fix the price like a tax, the safety valve can also be removed as more emission certainty is desired. And, all the permits need not be auctioned—some can be given away or used to finance related technology investments. This latter flexibility is the key to why I favor permit trading.
In the end, regulation is not entirely about efficiency—it is also about the distribution of costs (PDF). Market-based climate policies, in particular, have very transparent costs in the form of higher energy prices. This creates obvious and disproportionate burden in some sectors and regions of the country, and consequent political efforts to seek redress. In a permit system, this is easily addressed by a free permit allocation. No such flexibility exists in a tax system—and the logical result is to begin excluding those sectors that face the greatest burdens. Unfortunately, this undermines the primary argument for market-based policies in general—that everyone faces a transparent price and therefore seeks out the cheapest options wherever they exist.