For a complete listing of all On the Margin articles, please visit: www.aei.org/onthemargin/.
Some proposed cap-and-trade systems for carbon dioxide emissions would allocate the emission permits free of charge to firms that have engaged in past emissions rather than selling those permits at auction. Although supporters of free permit allocation contend that it would protect consumers from energy price increases, economic theory and evidence show that it would actually provide windfall gains to stockholders without restraining energy prices. Free allocation would also deprive the government of revenue that could be used to finance reductions in marginal income or payroll tax rates to offset the economic disincentives arising from cap and trade.
Any policy to control carbon dioxide emissions should take the form of a carbon tax or a cap-and-trade system with all permits sold at auction. Much or all of the revenue raised by the carbon tax or cap-and-trade auction should be devoted to income or payroll tax cuts that reduce marginal tax rates or to deficit reduction.
As discussed below, a cap-and-trade system with free permit allocation is economically equivalent to adopting a carbon tax and using the resulting revenue to make transfer payments to firms that have emitted in the past. Like those transfer payments, the allocation of free permits enriches the firms' stockholders, not their customers. Unlike marginal tax rate cuts, free permit allocation provides no incentives for investment or other economic activity and therefore does not promote economic efficiency. Accordingly, economists across the political spectrum have condemned free permit allocation as a pernicious combination of inefficiency and inequity.
In the end, the support for free permit allocation rests on an artifact of labeling. "Cap and trade with free permit allocation" and "carbon tax with all revenue used to make transfer payments to stockholders" are two labels for the same underlying policy. When that policy is presented under the second label, it draws no support. Proposals for explicit carbon taxes do not call for using the revenue to make transfer payments to stockholders, any more than the revenues raised by excise taxes on other items are used to make transfer payments to the stockholders of firms that produce those items. Support for the policy emerges only when it is presented under the first label. Of course, labeling a policy "free permit allocation" rather than "transfer payments" does not make it any more sensible. The disguised transfer payments embedded in the free allocation of cap-and-trade permits should be rejected just as emphatically as explicit transfer payments have been rejected under proposed carbon taxes and other excise taxes. To fully understand why free permit allocation is misguided, it is necessary to look behind the labels. I begin by analyzing a carbon tax and drawing out its relationship to cap and trade.
Carbon Tax Efficiently Reduces Emissions
In a world with no carbon tax, no cap-and-trade system, and no regulations, any carbon emissions that provide a net benefit to the consumers and producers who participate in the associated economic activity will occur. Unfortunately, some of those emissions are actually undesirable because of their adverse environmental effects, which the consumers and producers do not take into account. For example, suppose the adverse environmental effects impose a cost of $20 per ton of carbon dioxide. Then, any emissions that provide less than a $20-per-ton net benefit to consumers and producers should be avoided.
Stated differently, any reduction in emissions (relative to the no-tax world) that can be achieved at a cost of less than $20 per ton should be undertaken. The costs of reducing emissions include production costs incurred by firms, the loss of benefit for consumers who forgo carbon-intensive products, and the costs of developing new technologies that allow emissions to be avoided while minimizing the other costs. Of course, it would be impossible for the government to identify which emissions can be reduced at a cost of less than $20 per ton and to impose regulations mandating those particular reductions. Fortunately, a carbon tax makes those regulations unnecessary. Under a $20 carbon tax, all emission reductions that can be achieved at a cost of less than $20 per ton become economically attractive and are therefore undertaken by producers and consumers in the pursuit of their economic self-interest.
The carbon tax is therefore a market-based mechanism to reduce carbon emissions. Of course, the decision to reduce emissions is not market-based; rather, it is the correction of a market failure. But the manner in which the reductions are achieved is market-based because private parties choose the reductions based on the tax- inclusive prices and the costs they face. Emissions reduction is therefore less costly under the carbon tax than it would be under a regulatory approach, such as the one that the Environmental Protection Agency is moving toward implementing under the Clean Air Act.
Like any other tax, the carbon tax imposes economic burdens by changing prices, wages, and other incomes. Suppose that, in the no- tax world, the production of an item costs $100 (and therefore sells at that price) and requires one ton of emissions. The imposition of the $20 carbon tax increases the price to $120 if the production costs and the carbon emission remain unchanged. In that case, consumers bear the burden of the tax.
Consumers bear less than the full burden of the tax, however, if the tax reduces producers' incomes. That could happen for the following reason. When the price of the item rises, consumers are likely to buy a smaller quantity of it; indeed, that is part of how the tax reduces carbon emissions. At the lower production level, the industry producing the item hires fewer of any specialized workers required to produce the item, which drives down their wages. For example, if the price of coal-powered electricity rises and consumers buy less of it, there is less demand for coal miners and their wages fall. Specialized producers other than workers may be similarly affected. If the plant and equipment used in the industry is specialized (not readily transferable to other industries), the tax will reduce the stockholder profits generated by the plant and equipment.
In the above example, suppose that specialized producers' wages fall by $2. Then, the other production costs fall from $100 to $98 and the item sells at a tax-inclusive price of $118. In that case, consumers bear $18 of the tax burden and the specialized producers bear the other $2.
In general, supply and demand determine how the burden is divided. Statistical analysis indicates that a carbon tax will largely be manifested in higher prices for carbon-intensive items. A careful and widely cited study estimates that 85 percent of the burden will fall on consumers.
A carbon tax also raises revenue. For example, if five billion tons of emissions occur with a $20 per ton tax, revenue is $100 billion. As discussed further below, the revenue can finance efficiency-promoting reductions in marginal tax rates. Before discussing the use of revenue, however, I explain the equivalence between cap and trade with auction and a carbon tax.
Cap and Trade With Auction Equals a Carbon Tax
A few bills pending in Congress would institute a carbon tax, including H.R. 594, introduced by House Ways and Means Committee member Fortney Pete Stark, D-Calif., on January 15, and H.R. 1337, introduced by Rep. John B. Larson, D-Conn., on March 5. But cap-and- trade proposals have received more attention. In accordance with his campaign promises, President Obama has proposed to institute a cap-and-trade system in 2012 and to auction all the permits. The plan would require greenhouse gas emissions to be reduced 14 percent below 2005 levels by 2020 and 83 percent below 2005 levels by 2050. At least two bills pending in Congress, H.R. 1666, introduced by Ways and Means Committee member Lloyd Doggett, D-Texas, on March 23, and H.R. 1862, introduced by Rep. Chris Van Hollen, D- Md., on April 1, call for cap and trade with all permits auctioned.
Suppose the carbon tax described above is replaced by a cap-and- trade system with a cap of five billion tons (the emissions level assumed to occur under the $20 tax) and with all permits auctioned by the government. Then, the market-clearing price for permits is $20 per ton. This must be true because at a $20 permit price, cap and trade gives producers and consumers the same incentives as the $20 tax. As before, each ton of emissions still requires a $20 outlay, although the outlay is now a permit purchase rather than a tax payment. As with the $20 carbon tax, a cap-and-trade system with a $20 permit price causes all emission reductions that can be achieved at a cost below $20 per ton, and only those, to be undertaken; the aggregate level of emissions is therefore unchanged. So, if five billion tons of emissions occur with a $20 carbon tax, five billion tons also occur under cap and trade with a $20 permit price. Working backward reveals that if the cap is set at five billion tons, $20 is the equilibrium permit price consistent with that total.
As explained above, the two systems give rise, not only to the same aggregate volume of emissions, but to exactly the same set of emissions. The emission reductions are achieved in the same cost-efficient way as before. Cap and trade with auction, no less than a carbon tax, is a market-based mechanism.
Moreover, cap and trade affects producers and consumers in the same way as the carbon tax because the obligation to buy permits has the same effect as the tax in pushing up prices and pushing down wages. As stated before, the burden is largely on consumers, with some burden on specialized producers. Finally, the government collects the same $100 billion revenue as before because it auctions five billion tons of permits at $20 per ton.
At the level of generality considered here, cap and trade with auction is identical to a carbon tax, a point that is stressed in leading public finance textbooks. The equivalence can be seen at a glance by realizing that it makes no difference whether the $20 payment to the government is labeled as a permit purchase or as a tax. In practice, the two policies actually differ regarding administration and their response to various cost fluctuations; as discussed later in this article, those differences generally make the carbon tax the superior policy. For the present, however, they can be treated as equivalent.
As discussed below, cap and trade with free permit allocation is equivalent to these policies in many respects but has fundamentally different revenue effects. Before examining free allocation, it is appropriate to consider the use of revenue and its efficiency implications.
Using Revenue to Reduce Marginal Tax Rates
Both the carbon tax and cap and trade with auction produce a revenue stream for the government. From the standpoint of economic efficiency, there is a strong case for using a large portion of the revenue to reduce to distortionary taxes--that is, to reduce marginal tax rates of taxes, such as the payroll tax and individual and corporate income taxes, that impose work and saving disincentives. Alternatively, using the revenue for deficit reduction would have similarly beneficial economic effects because it would tend to increase capital accumulation and would allow future reductions in distortionary taxes.
In earlier years, some analysts assumed that a carbon tax or cap-and-trade system did not itself create inefficient work and saving disincentives. Using the tax revenue or auction proceeds to reduce distortionary taxes would then produce a double dividend: It would both improve the environment and reduce work and saving disincentives.
Later work pointed out that this analysis was flawed. A carbon tax or cap-and-trade system actually interacts with distortionary taxes in a way that amplifies work disincentives. Because households work to be able to buy goods and services, a tax on goods and services (including carbon-intensive goods and services) is a tax on work. A carbon tax or cap and trade does not, as is sometimes claimed, tax pollution rather than work; instead, it taxes pollution in addition to work. Even if the revenues are used to reduce distortionary taxes, it is hard to achieve a net reduction in disincentives and reap the double dividend.
If the revenue or auction proceeds are not used to reduce distortionary taxes, however, the carbon tax or cap and trade necessarily worsens disincentives. Rather than offering an opportunity to raise revenue that can be used to reduce disincentives, a carbon tax or cap and trade with auction imposes an increase in disincentives that can be counteracted only if the associated revenue is used to cut distortionary taxes. As the Congressional Budget Office explains:
"Selling emission allowances could raise sizable revenues that lawmakers could use [to] lower the cap's total cost to the economy . . . the government could use the revenues to reduce existing taxes . . . on labor, capital, or personal income . . . Research indicates that a CO2 cap would worsen the negative effects of those taxes: The higher prices caused by the cap would lower real (inflation-adjusted) wages and real returns on capital, indirectly raising marginal tax rates on those sources of income. Using the allowance value to reduce existing taxes could help mitigate that adverse effect of the cap."
Of course, reducing disincentives is not the only goal of tax policy and it is also necessary to consider distributional issues. Some of the revenue may be used to make transfer payments to vulnerable groups who will be harmed by the energy price increases. Because the burden of a carbon tax or cap and trade is widely distributed across consumers, however, it should be possible to adopt reductions in distortionary taxes that also benefit consumers at large. Such a policy would provide both consumer compensation and block a harmful increase in disincentives.
Obama's proposal stresses consumer compensation rather than reduction of disincentives. His budget outline earmarks $120 billion of auction receipts in fiscal 2012 through 2019 for clean-energy research and earmarks $526 billion to permanently extend the refundable Making Work Pay tax credit, which is scheduled to expire at the end of 2010. The extension of the credit would increase disposable income for nearly all American workers but would provide little or no net reduction in marginal tax rates.
All of this analysis hinges on there being tax revenue or auction proceeds available to the government. The revenue implications change dramatically when the permits are given away.
With Free Permits, Stockholders Get the Revenue
Free allocation has been the most common method of distributing permits under cap-and-trade systems. Most of the permits have been freely allocated under the trading program for sulfur dioxide (acid rain) in the United States and the European carbon cap-and-trade system. Although Obama supports 100 percent permit auction, there is strong support in Congress for giving away at least some of the permits. Recent remarks by the president and his science adviser suggest that the administration may be willing to give ground on the issue. The 650-page discussion draft released by Rep. Henry A. Waxman, D-Calif., and Rep. Edward J. Markey, D-Mass., on March 31 ducks the question by ignoring permit allocation. The House Energy and Commerce Committee held hearings on this draft on April 21 through 24, at which a number of witnesses called for free permit allocation.
I now consider a cap-and-trade system in which the permits are allocated to firms based on their past emissions or some other past characteristic. I initially assume that none of the firms are price-regulated utilities, but I discuss that complication later in the article.
If the cap is set at five billion tons, the emission level that prevailed under the other two policies, the market-clearing price at which firms trade the permits must be $20 per ton, the tax rate that was set under the first policy and the auction price that prevailed under the second policy. The fact that the permits' original holders receive them for free doesn't change their market value. Because the usefulness of the permits has not changed, buyers are willing to pay the same amount as before and sellers are still able to charge the same amount as before, causing the price to remain unchanged at $20.
To confirm this conclusion, note that a $20-per-ton permit price leaves incentives unchanged from the two prior cases and therefore results in the same emissions reductions. At that price, emitting an additional ton continues to impose a $20 marginal cost. The cost is readily seen for a firm that has no allocated permit to cover the emission because it must pay $20 to buy a permit from someone else. But the cost is equally real, if slightly less visible, for a firm that was allocated more permits than it needs. If that firm emits an additional ton, it must use a permit that it otherwise would have sold to another firm for $20. Emitting an additional ton still imposes a $20 cost; economists refer to it as an opportunity cost because it takes the form of a forgone opportunity.
As before, cap and trade causes any emission reductions that can be done at a cost below $20 per ton, and only those reductions, to occur. Because the same emissions occur as before, the aggregate volume of emissions is unchanged and the reduction in emissions continues to be achieved in the cost-effective manner. Even with free permit allocation, cap and trade remains a market-based mechanism.
The burden of the cap-and-trade system is divided among consumers and producers in the same manner as before because costs (including opportunity costs) are affected the same as before. If production of an item requires a ton of carbon emissions, the firm must be compensated by $20 for the opportunity cost of the permit required for the emissions. Again, even if the firm has a freely allocated permit available for use, using the permit in production is worthwhile for the firm only if it receives a price increase that compensates it for the $20 it forgoes by not selling the permit. So the price must rise by the same amount as it did in the other two cases, namely $20 minus any reductions in payments to specialized producers. Consumers therefore continue to bear most of the burden.
It may seem counterintuitive that the firm recovers from its customers the values of the permits it uses when it never actually paid for them. Why does the firm seek and obtain this windfall? In fact, the firm already has a windfall as soon as it receives the free permits because it can sell them to other firms. A self-interested firm will not forgo the sale to pursue another course of action, such as using the permits in production, unless that course of action provides an equal or larger windfall.
The CBO notes that this conclusion is supported by both economic logic and real-world experience:
"Regardless of how the allowances were distributed, most of the cost of meeting a cap on CO2 emissions would be borne by consumers . . . A common misconception is that freely distributing emission allowances to producers would prevent consumer prices from rising as a result of the cap. Although producers would not bear out-of-pocket costs for allowances that they were given, using those allowances would create an 'opportunity cost' for them because it would mean forgoing the income they could earn by selling the allowances. Producers would pass that opportunity costs on to their customers in the same way that they would pass along actual expenses. That result was borne out in the cap-and-trade programs for sulfur dioxide in the United States and for CO2 in Europe, where consumer prices rose even though producers were given allowances for free."
Several other authors also note that freely allocated permits result in higher consumer prices in the same manner as auctioned permits or carbon taxes.
In each of the two preceding cases, carbon tax and cap and trade with auction, the government collected $100 billion of revenue. Cap and trade with free allocation imposes the same burdens on consumers and producers as those policies. Because the government collects no revenue, however, it has $100 billion less in resources than in the other two cases. In contrast, the firms that were allocated the free permits have $100 billion more in resources.
Consider a firm that is allocated 10 free permits, emits 4 tons, and sells the 6 remaining permits for $120. The firm is in the same situation as if it had been subjected to the $20 carbon tax or auction requirement and had also been given a $200 transfer payment. Its production decisions and costs are the same under both policies and it charges the same prices. Under cap and trade with free permits, the firm uses four permits and receives $120 from selling the other six permits; under a carbon tax or cap and trade with auction, it pays $80 for its four tons of emissions and receives the $200 transfer payment, for an identical $120 net gain.
The same logic holds for any volume of emissions and permit allocations. Cap and trade with free allocation is equivalent to a carbon tax with transfer payments to firms. Of course, only people, not artificial entities such as firms, can possess or spend wealth; a transfer payment to a firm enriches the people who are its residual claimants, normally its common stockholders.
To be sure, part of the transfer payment will be recouped by federal, state, and local governments through individual and corporate income taxes on the firm's profits. References in this article to the transfer payment to the firm or its stockholders refer to the transfer payment net of those taxes.
The CBO has incorporated the economic effects described above into its budgetary accounting. It generally records the market value of permits, whether auctioned or given away, as government revenue and then records the value of any permits that are given away as spending. So, if permits worth $100 are given to a firm, the CBO treats the transaction as a permit sale that raised $100 revenue, followed by a $100 transfer payment to the firm, which precisely captures the underlying economic reality.
With this information, we can evaluate the policy implications of free permit allocation.
Inefficiency and Inequity
Free permit allocation has devastating implications for economic efficiency. Because prices rise in the same manner as they did under the carbon tax and cap and trade with auction, work disincentives are amplified in the same manner as under those policies. With free allocation, however, there is no revenue available to reduce distortionary taxes because the revenue has effectively been given away to stockholders. An increase in economic disincentives is therefore unavoidable, unless free permit allocation itself reduces disincentives.
Unfortunately, free permit allocation does no such thing because it does not lower marginal tax rates. One might initially think that it would have effects similar to a corporate income tax rate reduction because stockholders are the direct beneficiaries in each case. In reality, however, free permit allocation and corporate tax rate reduction have completely different incentive effects.
Because the corporate income tax penalizes investment by taxing the income from investments, lowering the tax rate reduces investment disincentives. With a 35 percent effective corporate income tax rate, for example, an investment must yield a 6.15 percent annual before- tax return to be economically viable, assuming that stockholders demand a 4 percent annual after-tax return. Lowering the tax rate to 30 percent lowers the threshold before-tax return to 5.71 percent if there is no change in the required after-tax return, clearing the way for those investments with before-tax returns between 5.71 percent and 6.15 percent to go forward.
In contrast, free permit allocation does not change the viability of any investments and therefore provides no incentives. The firm receives the same volume of free permits regardless of the level of its investments and other economic activity. Any investment that failed to meet the threshold return before the free permit allocation still fails to do so after the allocation.
The efficiency losses from free permit allocation alone may not doom the policy. After all, we are willing to impose distortionary taxes to finance transfer payments to the poor and those in need because we view the beneficial effects on equity or the income distribution as outweighing the resulting economic disincentives. That logic clearly cannot justify free permit allocation, however, because stockholders are largely wealthy.
In July 2007, then-CBO-Director Peter Orszag succinctly summarized the matter:
"Because giving allowances to energy producers would disproportionately benefit higher-income households and would preclude the possibility of using the allowance value to reduce taxes on capital and labor, such a strategy would appear to rate low from both a distributional and an efficiency perspective."
A few months earlier, the CBO provided a broader summary of the issue:
"Selling the allowances and using the proceeds either to cut taxes on earnings from labor or capital or to decrease the budget deficit would strengthen the economy . . . Because most or all of the cost of the cap would ultimately be borne by consumers, giving away nearly all of the allowances to affected energy producers would . . . transfer income from energy consumers--among whom lower-income households would bear disproportionately large burdens--to shareholders of energy companies, who are disproportionately higher-income households."
A wide range of other analysts have noted the inefficiency and inequity of free permit allocation:
- Gilbert E. Metcalf (Tufts University): "Policymakers have used the free allocation of permits for cap-and-trade programs. This practice comes at considerable distributional and efficiency costs. From an efficiency point of view, there always exists [a reduction in distortionary taxes] that is welfare-enhancing relative to a lump-sum return of the revenue. From a distributional perspective, free permits provide windfall profits to permit recipients. These windfalls show up as increases in equity value of the firms receiving permits. Since equity holdings tend to be concentrated in the upper part of the income distribution, this windfall transfer is quite regressive."
- Robert Greenstein (Center on Budget and Policy Priorities): "Giving away a substantial fraction of emission allowances to existing energy producers would do almost nothing to compensate low- and moderate-income families for their losses. A very large percentage of the benefits of such a giveaway would go to shareholders of the energy companies, most of whom have high incomes."
- Ian W.H. Parry (Resources for the Future), Hilary Sigman (Rutgers University), Margaret Walls (Resources for the Future), and Roberton C. Williams III (University of Texas): "Freely allocated tradable emission permits may actually hurt the poor the most, as they transfer income to shareholders via scarcity rents created at the expense of higher prices. On the other hand, emissions taxes (or auctioned emission permits) offer the opportunity to offset regressive effects, if revenues are recycled to finance progressive changes to the tax system."
- N. Gregory Mankiw (Harvard University): "Economists recognize that a cap-and-trade system [with free permit allocation] is equivalent to a tax on carbon emissions with the tax revenue rebated to existing carbon emitters, such as energy companies. That is, Cap-and-trade = Carbon tax + Corporate welfare. If the public understood this theorem, the carbon tax alternative, with revenues rebated to households through lower payroll or income taxes, would attract a lot more interest."
- James Barrett (Redefining Progress): "Free permit distribution would represent the largest windfall distribution of wealth in this country's history. Households, businesses, and industrial energy consumers will transfer their wealth to the owners of energy producing companies, already among the richest corporations in the world . . . When the government collects revenues from permit sales, it can put those revenues to economically productive uses, such as reducing taxes on labor and capital gains that diminish the incentive to work or to invest."
The breadth of the economic consensus against free permit allocation was demonstrated in March 2009, when the Southern Alliance for Clean Energy unveiled a cap-and-trade economist statement signed by 600 economists across the political spectrum (including myself). Calling for immediate 100 percent auction if cap and trade is adopted, the statement noted that free allocation would "do little or nothing to protect families and businesses from higher energy costs," would "represent a significant and undeserved windfall to utilities and other greenhouse gas producers," and would "deny the government the necessary resources to reduce the economic costs of combating climate change, and will thus generate needlessly higher costs of achieving any reduction target."
The case against free permit allocation appears unassailable, whether one emphasizes efficiency or equity. Supporters of free allocation often reply, however, that the argument ignores the fact that some of the firms receiving free permits would be price-regulated electric utilities.
The analysis has to be modified for those utilities because their prices are set by regulators rather than by the supply and demand factors considered above. Regulators would presumably prohibit those firms from raising their rates to cover the opportunity cost of the permits because there are no associated out-of-pocket costs.
Interestingly, the existence of price-regulated utilities actually strengthens the case against free allocation. The existence of price-regulated utilities causes the emissions reduction to be achieved in a cost-ineffective manner without reducing the aggregate consumer burden or the aggregate windfall for stockholders. I consider a simple model of utility regulation, in which regulators require utilities to produce in a cost-minimizing manner and then set prices equal to the average cost of production (including a required rate of return on stockholders' capital).
Under this assumption, cap and trade has the same impact on production technology in the regulated sector as it has on technology in the rest of the economy. The utility will be required by the regulator to make any changes to its production process that can reduce carbon emissions at a cost lower than the market price of permits because such changes are cost minimizing in the presence of cap and trade. Recall that profit-maximizing firms in the rest of the economy also make those changes.
Regulation does, however, change the impact of cap and trade on the utility's prices. Under average-cost pricing, the consumers, in addition to bearing the costs of making the technology changes, pay only the cost of any permit purchases by the utility; if the utility sells permits, consumers receive the sales proceeds. Unlike in the rest of the economy, consumers do not pay the opportunity cost of the allocated permits. Average-cost pricing also guarantees that the utility's stockholders reap no gain from the free permits. Looking at regulated utilities in isolation, price regulation appears to lessen consumer burdens and stockholder windfalls.
But, the economywide effects are quite different. Because the regulated utilities' consumers face lower prices, they buy more electricity, causing the utilities to reduce emissions by less than they would have done in the absence of free allocation. The cap-and-trade system rigidly requires, however, a specified volume of national emission reductions; if one sector makes fewer reductions, the rest of the economy must make more to meet the fixed cap. To force deeper reductions throughout the unregulated economy, the market price of permits must rise in equilibrium.
The higher permit price imposes larger burdens on the unregulated firms' consumers, who pay the opportunity cost of allocated permits. As a result, the aggregate burden on consumers is roughly unchanged. The higher permit price also implies a roughly unchanged aggregate windfall to stockholders, matching the unchanged aggregate consumer burden. Although stockholders of the regulated utilities receive no windfall at all, all other stockholders in the economy receive larger windfalls from the free permit allocation because each permit has a higher market value. As the CBO notes, "In regulated electricity industries, distributing the permits at no cost might mitigate or prevent price increases in those markets but only at the cost of requiring even larger price increases in other markets. Ultimately, consumers will, in one way or another, bear costs roughly equal to the value of the permits."
In the end, the existence of price-regulated utilities rearranges, rather than reduces, consumer burdens and stockholder windfalls. In the process, however, cap and trade's status as a market-based mechanism for reducing carbon emissions is impaired. The emissions reduction is no longer achieved in a cost-effective manner because there is too small of a reduction in the output of regulated utilities and too much reduction in emissions in the rest of the economy. Metcalf and Sergey Paltsev, John Reilly, Henry Jacoby, and Jennifer H. Holak, all of MIT note that if regulators do not allow utilities to charge consumers the market value of free permits, "consumers will face no incentive to reduce electricity consumption, thereby forcing more of the abatement elsewhere at higher cost."
As a side note, if this article's advice is rejected and permits are freely allocated, Congress should require regulators to charge consumers the opportunity cost of the allocated permits. The electricity produced by regulated firms would then be priced on a marginal-cost basis and emission reductions would be made efficiently throughout the economy. The resulting windfall for stockholders of regulated utilities would be no different than the windfalls enjoyed by all other stockholders.
The Tyranny of Labels
The support for free permit allocation is an artifact of labeling. If a carbon tax were imposed, no one would suggest using the tax revenue to make transfer payments to stockholders. No previous excise taxes have been used to finance transfer payments to firms and their stockholders; no tobacco-tax revenues are earmarked for transfer payments to Philip Morris. When the carbon tax is relabeled as cap and trade, however, such transfer payments, relabeled as free permit allocation, are viewed as conceivable or even natural.
Readers who encounter an argument for free permit allocation should translate it into the corresponding argument for using carbon tax revenue to make transfer payments to stockholders. Unless the translated version of the argument makes sense, the original argument should be rejected. It is useful to apply this technique to some commonly heard arguments:
- "Giving away permits is better than letting the politicians spend the auction proceeds." Translation: "It is better to use carbon tax revenue for transfer payments to stockholders than to let the politicians spend it." But, making transfer payments to stockholders is itself a type of spending adopted by politicians, a type that is inequitable and offers no efficiency gains. The right way to keep the politicians from spending the revenue is to use it to cut distortionary taxes.
- "Giving away permits keeps the money in the communities in which the burdens of cap and trade are felt." Translation: "Using carbon tax revenue for transfer payments to stockholders keeps the money in the communities in which consumers bear the burdens of the carbon tax." It clearly does no such thing because firms' stockholders and consumers generally do not live in the same community, sometimes not even in the same country.
- "Giving away the permits allows firms to spend money on clean-energy research." Translation: "Transfer payments to firms allow them to spend money on clean-energy research." If the government wishes to support research, it can provide research grants that require the money to be spent on that purpose. A large volume of government transfer payments and private gifts are made in the United States each year; the fraction of these payments spent by recipients on clean-energy research is essentially zero. When the National Institutes of Health seeks to promote medical research, it does not make transfer payments and hope that some of the recipients will spend some of the payments on medical research; instead, it provides research grants with contractual obligations to engage in research. Furthermore, by making carbon emissions more expensive, a carbon tax or cap-and-trade system already provides incentives to invent technology that minimizes such emissions.
- "Firms must receive free permits to compensate for the burden of cap and trade; they cannot afford 100 percent auction." Translation: "Stockholders must receive the revenue from a carbon tax to compensate them for its burden; they cannot afford to pay a carbon tax if the revenues are not transferred to stockholders." A. Lans Bovenberg and Lawrence H. Goulder, estimating that no more than 15 percent of the tax burden would fall on stockholders, conclude that 100 percent free allocation would "generate substantial windfalls to firms" because the gains to firms would be "many times larger than the income losses otherwise generated by the policy." That most of the burden is passed on to consumers is part of the explanation for why excise taxes are not normally accompanied by transfer payments to firms. A deeper explanation is that stockholders, as residual claimants, are routinely exposed to many risks affecting the operation of their industries, including the risk of tax policy changes, and should not be insulated from the impact of such changes.
The fact that transfer payments to stockholders are not considered under an explicit carbon tax while free permit allocation gains instant support under cap and trade is an advantage of the carbon tax, as several authors have noted. The simplest way to head off free permit allocation may be to adopt a carbon tax rather than cap and trade.
A carbon tax has other advantages over cap and trade, once we move beyond the simple framework presented earlier, in which they were identical. In particular, a carbon tax is likely to be better than cap and trade regarding administration, response to fluctuations in the cost of reducing emissions, and allocation of emissions reductions across different years. It is possible, however, to design cap and trade in a way that largely replicates those advantages, by imposing price floors and ceilings and allowing banking and borrowing of permits across years.
Any carbon control policy adopted in the United States should take the form of a carbon tax or a cap-and-trade system with full auction of permits from the outset of the program. A large part of the resulting tax revenue or auction proceeds should be used to reduce marginal payroll or income tax rates or to reduce the deficit. Such a policy would reduce disincentives to work and save while also compensating the public for the burden of higher energy prices. In contrast, free allocation of permits to firms is equivalent to making unproductive transfer payments to high-income stockholders. Don't give away the cap-and-trade permits!
Alan D. Viard is a resident scholar at AEI.
Ian W.H. Parry of Resources for the Future recently suggested $20 per ton of carbon dioxide as the initial value of a carbon tax, "Raise $100 Billion From a $20 CO2 Tax," Tax Notes, Apr. 13, 2009, p. 243, Doc 2009-5800 [PDF], 2009 TNT 69-16.
On April 2, 2007, the U.S. Supreme Court ruled 5-4 that greenhouse gases are "air pollutants" within the meaning of 42 U.S.C. section 7602(g), Massachusetts v. Environmental Protection Agency, 549 U.S. 497 (2007). The EPA recently published in the Federal Register a proposed finding that greenhouse gases are reasonably anticipated to endanger public health and welfare within the meaning of 42 U.S.C. section 7521(a)(1), 74 Fed. Reg. 18,886 (Apr. 24, 2009). The proposed finding, which is open for public comment through June 23, 2009, does not itself impose any regulations, but is a potential prelude to future regulation.
If the producing firm can avoid the emission at a cost of $15, the firm incurs that cost rather than pay the $20 tax. Continuing to assume that the other production costs remain at $100, the item then sells for $115 and consumers bear the cost of avoiding the emission.
The industry will also hire fewer workers with general skills, but their wages are likely to be unaffected.
A. Lans Bovenberg and Lawrence H. Goulder, "Neutralizing the Adverse Industry Impacts of CO Abatement Policies: What Does it Cost?" in Behavioral and Distributional Effects of Environmental Policy, eds. Carlo Carraco and Gilbert E. Metcalf (Chicago: University of Chicago Press, 2001), pp. 45-85.
These numbers line up with the rough numbers used by Parry, supra note 1.
Office of Management and Budget, A New Era of Responsibility: Renewing America's Promise, Feb. 2009, p. 21.
See Harvey S. Rosen and Ted Gayer, Public Finance, 8th edition (New York: McGraw-Hill/Irwin, 2008), p. 90; and Jonathan Gruber, Public Finance and Public Policy, 2d edition (New York: Worth Publishers, 2007), p. 143.
Gilbert E. Metcalf, "Environmental Taxation: What Have We Learned in this Decade?" in Tax Policy Lessons from the 2000s, ed. Alan D. Viard (Washington, D.C.: American Enterprise Institute, 2009), pp. 7-34, at pp. 13-17, surveys the literature on the interaction of environmental taxes and distortionary taxes. Also, see the references cited by Kevin A. Hassett, Aparna Mathur, and Metcalf, "The Consumer Burden of a Cap-and-Trade System with Freely Allocated Permits," American Enterprise Institute Working Paper No. 144, Dec. 23, 2008, p. 6, n.4. Further, see the estimates in Bovenberg and Goulder, supra note 5, pp. 68-69, and the studies they cite.
"Trade-Offs in Allocating Allowances for CO2 Emissions," CBO Economic and Budget Issue Brief, Apr. 25, 2007, p. 4, available at http://www.cbo.gov/ftpdocs/89xx/doc8946/04-25-Cap_Trade.pdf.
New Era, supra note 7, p. 115. The budget outline states that the auction may produce additional receipts and that any such receipts "will be used to compensate the public."
The credit, as set forth in section 36A, reduces marginal tax rates on labor earnings by 6.2 percentage points for workers with earnings below $6,452 ($12,903 for couples), many of whom already face negative marginal tax rates. On the other hand, the credit's income-based phaseout increases effective marginal income tax rates by 2 percentage points for workers with modified adjusted gross income between $75,000 and $95,000 ($150,000 and $190,000 for couples). For most workers, the credit is simply a $400 transfer payment ($800 for couples) that does not change marginal tax rates.
Rosen and Gayer, supra note 8, p. 97; Gruber, supra note 8, p. 153.
Government Accountability Office, "International Climate Change Programs: Lessons Learned from the European Union's Trading Scheme and the Kyoto Protocol's Clean Development Mechanism," Nov. 2008, p. 4; and Joann M. Weiner, "Taxing or Trading Carbon," Tax Notes, Nov. 5, 2007, p. 570, Doc 2007-24362 [PDF], 2007 TNT 215-17.
At a March 12, 2009, Business Roundtable meeting, President Obama reiterated his preference for 100 percent auction, but he noted the need to work with businesses that support free permit allocation and said that if cap and trade "is so onerous that people can't meet it, then it defeats the purpose--and politically we can't get it done anyway," available at http://www.whitehouse.gov/the_press_office/Remarks-by-the-President-at-the-Business-Roundtable/. John P. Holdren, director of the White House Office of Science and Technology Policy, stated that whether to start with 100 percent auction or "get there over a period of time" was "being discussed" and a White House spokesperson stated that "the administration will be flexible" about 100 percent auction, quoted by Juliet Eilperin, "Science Chief Discusses Climate Strategy," The Washington Post, Apr. 9, 2009.
The draft is available at http://energycommerce.house.gov/Press_111/20090331/acesa_discussiondraft.pdf.
The fact that the same emissions reductions occur whether the permits are auctioned or given away is an illustration of the Coase Theorem, which says that an efficient solution to an externality can be attained under any assignment of property rights. See Rosen and Gayer, supra note 8, pp. 79-81; and Gruber, supra note 8, pp. 130-131.
CBO Brief, "Trade-Offs," supra note 10, pp. 1, 5.
Hassett, Mathur, and Metcalf, supra note 9, p. 5 (free permits "will increase prices by the same amount as if the permits were auctioned. This scenario has played out in existing cap-and-trade systems and is beyond dispute in the economics profession."); Bovenberg and Goulder, supra note 5, p. 58; Metcalf, supra note 9, p. 25; Ian W.H. Parry, Hilary Sigman, Margaret Walls, and Roberton C. Williams III, "The Incidence of Pollution Control Policies," National Bureau of Economic Research Working Paper No. 11438, June 2005, p. 7; and Robert Greenstein, executive director of Center on Budget and Policy Priorities, Testimony before Senate Finance Committee, April 24, 2008, p. 6, available at http://www.cbpp.org/files/4-24-08climate-testimony.pdf.
If the firm is in financial distress, its residual claimants may be its preferred stockholders or even its bondholders and other creditors.
CBO, "Cost Estimate, S. 2191, America's Climate Security Act of 2007," Apr. 10, 2008, p. 7, available at http://www.cbo.gov/ftpdocs/91xx/doc9120/s2191.pdf.
Orszag letter to Rep. Jeff Bingaman, D-N.M., July 9, 2007, pp. 3-4, available at http://www.cbo.gov/ftpdocs/82xx/doc8286/07-09-BingamanLetter.pdf.
CBO Brief, "Trade-Offs," supra note 10, p. 2.
Metcalf, supra note 9, p. 25.
Greenstein, supra note 19, p. 6.
"Incidence," supra note 19, pp. 31- 32.
"The Fundamental Theorem of Carbon Taxation," available at http://gregmankiw.blogspot.com/2007/08/fundamental-theorem-of-carbon-taxation.html.
"The True Cost of Free Pollution Permits: A Redefining Progress Issue Brief," Feb. 2008, p. 2, available at http://www.rprogress.org/publications/2008/True%20Cost%20Issue%20Brief%2002-08.pdf.
See http://www.cleanenergy.org/images/position_ statements/SACE_EconStatement_FullList.pdf.
Electric utilities are still regulated in most states. Hassett, Mathur, and Metcalf, supra note 9, p. 4, list 12 states that, along with the District of Columbia, have deregulated electric utilities.
Testimony of then-CBO-Director Peter Orszag before the Senate Committee on Energy and Natural Resources, May 20, 2008, p. 13, available at http://www.cbo.gov/ftpdocs/92xx/doc9276/05-20-Cap_Trade_Testimony.pdf.
This rearrangement of consumer burdens does not appear to be equitable. Hassett, Mathur, and Metcalf, supra note 9, p. 13, note that electricity consumers in deregulated states have already faced steeper price increases than those in regulated states, a disparity that would be aggravated by free allocation to regulated utilities.
"Analysis of U.S. Greenhouse Gas Tax Proposals," National Bureau of Economic Research Working Paper No. 13980, Apr. 2008, p. 4.
Congress could implement such a result by specifying that regulated utilities would receive no free permits if regulators do not allow the opportunity cost to be charged to consumers. Section 168(f)(2) similarly denies accelerated depreciation to public utilities if regulators do not allow normalization accounting.
If firms face financial constraints, providing a transfer payment could spur some additional research. But transfer payments are an inefficient way to achieve that goal relative to loans.
Bovenberg and Goulder, supra note 5, pp. 48 n.3, 70-77.
Kenneth P. Green, Steven F. Hayward, and Kevin A. Hassett, "Climate Change: Caps vs. Taxes," American Enterprise Institute Environmental Policy Outlook, June 2007, p. 7; "Analysis," supra note 33, p. 5; and Parry, supra note 1, p. 245.
See Green, Hayward, and Hassett, supra note 37, pp. 5-7; Metcalf, supra note 9, pp. 22-26; Gruber, supra note 8, p. 162; and Parry, supra note 1, pp. 245- 246.