Printer Friendly

Environmental Taxation in an Optimal Tax Framework.


An extended and descriptive optimal tax framework is used to identify the objectives of environmental taxation, analyze the portfolio problem, and explore the trade-offs of policy design. Optimal tax criteria include efficiency, equity, administration, compliance, and revenue. The portfolio problem means that the policy maker must compromise to accommodate trade-offs of competing policy objectives. The regulator may attempt to increase efficiency, maintain a fair distributional impact, minimize the costs of administration and compliance, and implement an efficient use of revenue. The best opportunity to enhance policy effectiveness lies in improving the trade-offs between policy objectives. (JEL Q28)


Environmental taxes [1] enhance environmental quality and raise revenue. The revenue may be used to reduce rates on distorting taxes or fund cleanup programs. However, by reducing the tax base through pollution abatement, environmental taxes decrease revenue yields and create both efficiency costs and additional policy concerns. Because of these and other trade-offs, governing authorities cannot achieve each optimal tax criterion (policy objective) as proposed by Aim [1996]: the minimization of efficiency losses, equitable impacts, cost minimization for both tax administrator and recipient, and a stable revenue yield.

Optimal tax research [2] has analyzed institutional aspects of income and commodity taxes but not charges that are intended to discourage environmentally harmful activity or fund cleanup programs. In the design of environmental taxes, many important trade-offs exist beyond the standard efficiency/equity and environmental quality/revenue trade-offs. By exploring trade-offs and multiple policy objectives, it is shown that the regulator must compromise to accommodate competing policy goals. The best opportunity to enhance policy effectiveness lies in improving the trade-offs between policy objectives.

To develop this concept and to bridge the gap between the environmental tax literature and the optimal tax literature, this paper is organized as follows. The second section discusses the perceptions of environmental taxation and second-best policy. The third section provides an analysis, in an extended and descriptive optimal tax framework, of the objectives and conceptual trade-offs of policy design and public choice implications.

Perceptions of Environmental Taxation and Second-Best Policy

Many economists propose a greater use of environmental taxation as a percentage of public revenue. [3] Repetto et al. [1992], calculate that in the U.S., a greater use of pay-by-the-bag household collection charges, rush-hour tolls, and carbon taxes could raise $100 billion annually. Muller and Hoerner [1994] argue that a $70 per ton state-level carbon tax in the U.S. could increase economic efficiency and raise more than $70 billion annually while reducing energy use and carbon dioxide emissions.

Goulder [1995, 2000] and Bovenberg [1998] question the efficiency consequences of new environmental taxes. In particular, the benefit of pollution abatement, which depends on marginal damage, may not exceed net policy cost. Bovenberg and de Mooij [1994] and Bovenberg and Goulder [1996] reveal that the efficiency costs of environmental taxation are higher than previously thought. A policy of revenue substitution (for the purpose of political feasibility, advocated by Pearce [1991], Von Weizsacker and Jesinghaus [1992], and Hamond et al. [1997], among others) would create a tax interaction (TI) effect that would partially offset the benefit of pollution abatement.

However, the problem with papers that isolate efficiency costs, including those that advocate environmental taxation, is the absence of a comprehensive analysis of all policy objectives. An analysis of efficiency impacts may not consider that political feasibility may depend more on distributional consequences. Even the inclusion of equity does not provide a complete assessment because the potential for net policy benefits also depends on the costs of administration and compliance and the use of tax revenue. To enhance environmental quality, achieve a fair distribution of benefits and costs, and reveal the conceptual trade-offs of policy design, the regulator requires an analysis of all policy objectives. An encompassing framework with all objectives is developed in this paper.

In a first-best world without preexisting tax distortions, the optimal marginal tax rate equals marginal environmental damage (MED). Identifiable abatement and damage functions provide accurate information for the administrator while firms exhibit perfect compliance. In this setting, a Pigouvian tax is imposed at such a rate that no additional net benefits are possible with further pollution reduction. The Pigouvian tax links directly to the pollution source, internalizes the negative externality at the optimal level (where the marginal benefit of pollution reduction equals marginal abatement cost), and does not discourage the production of economic output or related economic decisions [Baumol and Oates, 1988].

However, for three reasons, it is difficult to formulate first-best tax rates. First, because of unpredictable meteorological conditions and imprecise valuation techniques, the determination of the economic value of marginal damage, in a damage function approach, is an imprecise procedure. The damage function approach estimates emissions, models dispersion, quantifies impacts, applies an economic valuation, and determines whether the externality is already internalized by policy [McKenzie, 1994].

Second, to provide revenue for cleanup programs, U.S. governing authorities tax final output and intermediate inputs but do not provide incentive for pollution abatement with tax policy [Fullerton, 1996]. Because these environmental taxes do not focus on the polluting activity, the taxes represent nonoptimal solutions. Third, preexisting taxes that distort behavior, present in all industrial economies, affect the efficiency consequences of new environmental levies.

Goulder et al. [1997] model the impact of preexisting taxes in a second-best world and conclude that the gross efficiency cost (net of environmental benefits) of environmental taxes is much higher than in a first-best model. In particular, the optimal second-best environmental tax rate lies below the Pigouvian rate. Bovenberg and van der Ploeg [1994] show that:

[t.sup.*] = MED/ [micro],

where [t.sup.*] is the optimal tax and [micro] is the marginal cost of public funds or any efficiency cost of taxation. The first-best environmental tax, [t.sup.*], equals MED, but in the world of the second-best, a positive value for [micro] implies that it is costly for government to provide environmental quality. Therefore, the higher the value of [micro], the lower the level of environmental quality, and a lower tax rate [t.sup.*].

Optimal Tax Criteria, Policy Trade-Offs, and the Portfolio Problem

An analysis of environmental taxation in an extended and descriptive optimal tax framework differs from the model proposed by Alm [1996] in an important way: the environment must be considered a major component of the efficiency criterion. The optimal tax criteria (efficiency, equity, administration, compliance, and revenue) include subcategories and provide a framework to understand the important objectives and tradeoffs of environmental taxation.

While no environmental tax achieves each policy objective, a tax might be designed to discourage pollution, burden those responsible for pollution damage, and minimize administrative and compliance costs [Fullerton, 1996]. Other objectives might include the minimization of efficiency losses, the avoidance of information and measurement problems, policy flexibility during changing environmental and economic conditions, stable tax rates so businesses can plan payment strategies, the minimization of collection points, and the establishment of an effective enforcement mechanism.

Because of policy trade-offs, regulators will not achieve each objective. With sales taxes, for instance, the regulator attempts to minimize risk (that is, revenue instability) and maximize yield (that is, elasticity) through the choice of sales tax structure [Murray, 1997]. With environmental taxes, an increase in efficiency from pollution abatement may come at the expense of both a stable revenue stream and the minimization of administrative, compliance, and distribution costs. The implementation of a stable tax rate may sacrifice any flexibility in rate adjustment, which could respond to higher levels of pollution abatement. As will be explained, other trade-offs exist between revenue yield and adequacy, environmental quality and administrative cost, enforcement and compliance, and efficiency and revenue functions. Balancing the trade-offs of each objective becomes a portfolio choice for the regulator.

Table 1 provides a comprehensive list of the optimal tax criteria and the important subcategories not previously addressed in a single framework. The description of each criterion allows for a deeper understanding of the objectives and trade-offs of policy design. The extended and descriptive optimal tax framework also elucidates the complexity of policy implementation and the possibility of identifying tax reforms that could achieve more of certain objectives without compromising others. In particular, to increase efficiency while focusing on the tax burden, higher tax rates could provide the incentive for abatement. Instead of using earmarked funds, environmental cleanup programs could be funded with revenue from the general budget. As opposed to focusing solely on efficiency or the combination of efficiency and equity, the analysis points toward a comprehensive assessment of policy implications using all of the tax criteria in Table 1.


Efficiency encompasses all aggregate welfare impacts from environmental taxation, including any environmental effects [Goulder, 1994]. Theoretically, an environmental tax reduces excess burden over some range of implementation [4] by improving resource allocation and discouraging corrupt production processes. If the tax is linked to the polluting activity and is set at a marginal rate that exceeds the marginal abatement cost, it will provide the incentive for pollution reduction per unit of output. Incentive policy improves on command and control regulation because in response to incentive policy, polluters will implement a least-cost method to control waste emissions by choosing between new abatement technology, alternative inputs, or paying the tax [Baumol and Oates, 1988; Carlin, 1992].

With environmental taxation as a revenue-raising instrument, the governing authority must determine how to adjust its relative contribution, as one of a number of taxes, to raise public revenue. Recent research on this efficiency consideration reveals that relative to a first-best (optimal) world, the presence of preexisting taxes significantly raises the cost of environmental policy, particularly when environmental tax revenue is used to finance lower rates on other taxes [Oates, 1995].

In a second-best setting, environmental taxes may discourage related economic activity by distorting choices at the labor-leisure margin, the intermediate resource margin, the intertemporal margin (choice between consuming today or tomorrow by saving and investing), or the margin of consumer product choice [Goulder, 1994]. The first two choices are components of the factor market while the last two exist in the domain of the product market. Recent research addresses the labor/leisure trade-off. [5]

Since the early 1990s, many economists have addressed the potential of a corrective environmental tax achieving a double dividend [Goulder, 1995] by reducing pollution and lowering distortions in the tax system. Terkia [1984], Oates [1995] and others emphasize that to the extent that RR reduces deadweight loss, future distortions from preexisting levies can be avoided. This RR benefit implies that the gross cost of environmental taxation is lower in a second-best economy with preexisting taxes than in a first-best setting. Given the uncertain benefits and costs of environmental taxation, the potential of achieving the RR benefit creates the appealing possibility of no-cost policy reform.

However, Bovenberg and de Mooij [1994], Parry [1995], Bovenberg and Goulder [1996], Bovenberg [1998], and Goulder [2000], among others, recognize the importance of the interaction between environmental policies and nonenvironmental distortions. Pollution taxes, in particular, are implicit levies on resource inputs such as labor and capital. According to these economists, by increasing the cost of production, environmental taxes exacerbate, in a general equilibrium framework, the distortions from preexisting taxes. This TI effect leads to higher output prices and lower real after-tax returns to resource inputs. In fact, the larger the tax rates on preexisting levies, the greater the TI effect and the higher the gross costs from new environmental tax policy. Contrary to the RR impact, the TI effect implies higher gross costs in a second-best economy than in a first-best setting.

The research by Bovenberg and Goulder [1996] and Parry [1995] shows that in plausible circumstances, the TI effect overrides the RR impact: RR reduces factor market distortions, but such recycling does not reverse the relative change in consumer prices and the associated distortion in consumption choices. As a result, environmental taxes and RR distort the household choice between work and leisure, and employment falls. The second dividend of reducing labor-market distortions fails because regulators cannot introduce pollution taxes in a costless manner. This result implies that the optimal environmental tax lies below the levy that internalizes the negative externality (and the target level of emissions is greater than the level that internalizes the externality).

If environmental taxation creates efficiency losses under specific circumstances, such as low levels of marginal damage and high efficiency costs from TI, the relevant policy question concerns preference. If society values a clean environment and allocates resources to this end, then which policy among all environmental policy alternatives will create the lowest efficiency loss when marginal damage is low and the highest efficiency gain when marginal damage is high?

Using analytical and numerical general equilibrium models to calculate efficiency impacts, Goulder et al. [1999] conclude that revenue-raising environmental policies such as taxation and auctioned quotas lead to lower gross efficiency costs than their nonrevenue-raising counterparts. With TI and RR effects, nonrevenue-raising policy instruments, such as grandfathered quotas or permits, do not generate RR benefits to offset any TI costs. Therefore, if the environmental benefits from pollution reduction are below some threshold, any pollution abatement with nonrevenue-raising policy is efficiency reducing.

Papers published on the efficiency consequences of environmental taxation, such as Goulder et al. [1999], seem to conclude that for political necessity or otherwise, policy implementation will most likely occur through tax reform, with the revenues from environmental taxation financing reductions in income or other taxes. If such reform produces a nonenvironmental dividend, environmental taxation can be justified, even if the magnitude of environmental benefits is unknown.

The gross costs of environmental taxation depend not only on the costs attributable to the policy itself, but also the RR benefits (avoided costs) from reducing other tax rates. For net efficiency gains when the second dividend does not hold, the benefits of the first dividend of environmental quality must be quantifiable and enough to offset gross efficiency costs. With a revenue-raising policy such as environmental taxation, Goulder et al. [1997] calculate small efficiency gains when the benefit of pollution abatement equals $100 per ton of emission reduction.

In a general equilibrium framework, additional research must consider the case when environmental taxation replaces direct control (technology and performance standards) and administrative and compliance costs decrease. An important assumption in the above research is that the introduction of new environmental policy exacerbates preexisting tax distortions. However, avoided costs besides those captured by the RR effect could include the money saved when incentive policy replaces direct control. Many analytical and empirical studies, including Baumol and Oates [1988] and Carlin [1992], conclude that direct control normally entails higher policy costs than incentive regulation. When environmental taxes replace direct control, the magnitude of gross policy costs could decrease. Given a level of environmental benefits, this possibility could impact the value of the second dividend.


Fairness in environmental taxation means balancing policy costs and benefits across all affected parties. While the polluter-pays principle aligns itself with common sense, all taxes are ultimately borne by households. When environmental tax policy distributes benefits or allows those who benefit to pay an appropriate percentage of the cost, undue burden does not occur. However, environmental taxes that impact decision making, particularly those on energy consumption, may create an undue burden on lower income households. In the absence of redistribution, regulation that burdens a particular income group, region, or industry limits policy effectiveness. Baumol and Oates [1988, p. 236] suggest that:

"...measures to improve environmental quality may indeed have a very uneven pattern of incidence, particularly during the period of adjustment to a new composition of output and employment... the evidence suggests that we can typically expect a...regressive pattern of distribution of the benefits and costs from environmental programs."

Given the political importance of distribution analysis, economists have studied this issue, as it applies to environmental taxation, relatively little [Repetto et al., 1992]. Research on the topic normally focuses on the impact of environmental taxation on different income groups in the household sector [Organization for Economic Cooperation and Development (OECD), 1997]. Concerning distributional implications, the policy maker must consider five relevant issues: costs and benefits, direct and indirect effects, formal and effective incidence, the baseline for comparison, and dimensions of distribution.

The regulator must first consider the potential benefits and costs of policy design. In particular, what changes in the environment is the tax expected to induce? What will be the accompanying costs and benefits of policy implementation? Short-run distributional changes, which result from relevant costs and benefits, occur during the period of policy adjustment, but long-term shifts in distribution reflect changes in economic activity. Since environmental taxation targets pollution reduction, benefits in the form of cleaner air, water, or land may accrue to all residents of a polluted area. However, pollution tends to be more serious in poorer, industrial neighborhoods. Therefore, environmental tax policy may generate benefits that are more critical to the poor. However, if regulation stifles development by reducing output and employment over time, falling incomes could offset the value of environmental benefits [Baumol and Oates, 1988].

Although the physical quantity of pollution reduction may benefit industrial areas, the poor may not value these improvements more than others. While regulation creates short-term health benefits, a cleaner environment through higher land values and rent could force low-income households to pay the bulk of policy cost. This result implies a redistribution of wealth. Even though price changes could take years, higher rents and living costs in the long term could reduce net policy benefits. In contrast, households with sufficient resources may achieve a desired level of environmental quality, given improvement costs. For example, they may insulate their homes from air pollution. From the point of view of these residents, regulation imposes additional costs [Baumol and Oates, 1988].

The policy maker must also identify the indirect and direct distributional effects. Indirect effects include the change in consumer prices for production inputs, such as gasoline and fertilizer. Direct effects include household purchases of final taxed products [OECD, 1997]. The indirect effects of environmental taxation on inputs have been studied less than the direct burden on final products. One reason for this is that the data requirements for the calculation of indirect effects, including information on both firms and individuals, are imposing. In addition, the complexity of the variables (input substitution elasticities and changes in relative prices) impact distributional calculations [OECD, 1993].

If firms pass on abatement costs to consumers through higher prices, environmental tax policy will distort consumer choice. Measuring the distribution of industrial abatement costs across income classes, Robison [1985] found environmental policy to be regressive when calculated as a percentage of consumption. The magnitude of the burden depends on the relative price of corrupt consumption relative to noncorrupt consumption, the percentage of lower-income consumers who pay for abatement costs, and how changes in the optimal consumption mix impact household utility. The distributional incidence from these factors cannot be determined a priori [OECD, 1993].

After the determination of direct and indirect effects, the governing authority must estimate the formal incidence, or those responsible for the initial tax payment, and the effective incidence, or the ultimate recipient of financial burden. Congress can assign legal liability in terms of who pays a tax but cannot legislate the ultimate tax burden. A tax on one product, for example, may ultimately impact other prices. Additionally, particular products that are not taxed may end up with higher prices, and anyone purchasing them would bear the burden [Kotlikoff and Summers, 1987].

In a revealing analysis of the distributional impacts of environmental taxes on intermediate inputs, Fullerton [1996] explains that Congress cannot target the burden of individual environmental taxes on particular industries. Since each industry buys intermediate inputs from all other industries, production costs depend, in part, on the gross-of-tax cost of purchasing inputs. Since environmental taxes increase the cost of production by raising the price of inputs, the prices of many goods that are not taxed directly by environmental taxes but are produced in industries that employ taxed intermediate inputs are higher [Fullerton, 1996]. Policy reform that would provide incentive for pollution abatement could focus policy burden on industries thought responsible for environmental damage.

The degree to which initial tax burden is borne by business or households depends on the elasticities of supply and demand for taxed products. If supply is relatively elastic but demand is relatively inelastic, the consumer will bear a higher burden. In contrast, if supply is relatively inelastic and demand is relatively elastic, the supplier will bear relative burden. If the tax burdens the seller, formal incidence may target capital, labor, resource owners, and other factors of production [OECD, 1995b].

Effective incidence depends, in part, on the international context in which the environmental tax is introduced. If the tax is implemented at the federal level and other countries adopt similar regulation, it is more likely that owners rather than consumers will bear a greater burden. This issue may also be considered in a time perspective. In the short run, consumers are likely to bear the burden through higher prices. However, in the long term, a greater tax incidence may be borne by the owners of the means of production [OECD, 1997].

Policy authorities must consider whether tax payment and the value of redistribution exceed the economic value of pollution damage. If they are excessive, environmental taxation may create an undue burden and may limit the possibility of net policy benefits. This question is empirical in nature, arises in the presence of any environmental regulation, and depends on a baseline comparison. The impact must be measured relative to a reference situation, such as equivalent command-and-control policy or no regulation.

Governing authority must also address the different dimensions of distribution, including the effects on income groups, businesses, households, industries, and regions [OECD, 1995b]. Compared with the attention given to carbon and energy taxes, relatively little research exists on the distributional consequences of nonenergy environmental taxes. Nonenergy environmental taxes raise little revenue and are tied to less-significant components of household spending. As a result, the impact of nonenergy environmental taxes on household spending could be less regressive [OECD, 1997].


A policy goal of environmental taxation is to minimize administrative cost. To date, little information exists on how administrative cost differs with alternative fiscal policy instruments [Alm, 1996]. In the U.S., multiple resources are allocated to tax collection, but a lack of coordination exists between different levels of government [Graetz et al., 1989]. While various public authorities may oversee environmental policy, national, state, and local governments are not in competition with one another for jurisdiction [OECD, 1995b]. Most likely, administrative cost corresponds in large, discrete levels with the scale of collections and exhibits economies of scale [Fullerton, 1996]. However, detailed aspects of collection cost technology are unknown [Alm, 1996]. Five regulatory issues impact the scope of administration: pollution identification, the ability to administer, oversee, and adapt policy, revenue collection mechanisms, enforcement, and the problem of pollution source and destination.

First, the identification of specific forms of pollution in air, water, and land allows the estimation of environmental damage. The damage function, which calculates the causal relationship between pollution and certain damages suffered by society, may form a basis for environmental policy. Damages may take the form of diminished enjoyment of the outdoors, an increase in the prevalence of mortality and disease, more frequent replacement and maintenance of equipment, and many other less-identified losses. A decrease in pollution levels will bring about a corresponding reduction in pollution damage. The benefit of lower levels of damage equals the difference in damage before and after reduction [Oak Ridge National Laboratory and Resources for the Future, 1994].

Second, administrative costs and policy effectiveness depend on the level of government control, the ministry responsible for tax administration [OECD, 1995b]. An effective environmental tax will encourage pollution abatement but will also be flexible as damage valuation techniques improve and abatement technology is implemented. Even though businesses want stable marginal tax rates so they can plan tax payments, regulators must prepare to adjust tax rates as environmental and economic conditions change. Environmental taxes in the U.S. incur administrative costs but achieve neither targeted incentives nor burdens [Fullerton, 1996].

Third, the minimization of collection points and the recognition of the extent of coverage limits administrative duty [Muller and Hoerner, 1994]. Depending on the tax base, authorities must gather emissions, disposal, and resource/product-use data. Instead of designing a new system, if taxes can be incorporated into an existing regulatory mechanism, cost savings occur and compliance increases [OECD, 1995b]. However, if exemptions remain, the use of an existing tax system may limit policy effectiveness (certain states in the U.S. allow exemptions against fossil fuel taxes).

The fourth administrative duty, policy enforcement, entails actions that force violators to adhere to legislation. Enforcement also encourages others in the regulated population to maintain desired behavior. During the 1980s, the U.S. Environmental Protection Agency (EPA) attempted to improve enforcement by defining a uniform civil penalty policy and stressing the criminal prosecution of polluters. To remove the advantage of noncompliance, the EPA calculates the potential savings to polluters from noncompliance and allocates enforcement resources accordingly. The EPA also assigns monetary values to violations. This activity, incidentally, poses serious estimation problems, at least when compared to assessing the value of pollution damage [Russell, 1990].

Fifth, the level of legislative control is complicated when pollution crosses national, state, or local boundaries. A water effluent charge, such as in Belgium, France, and Germany, may conveniently fall under local control [OECD, 1995a]. However, by exporting environmental damage, a transcendent form of pollution such as an airborne chemical emission clouds policy jurisdiction. To accurately price environmental damage and to not discourage business activity, countries or states may adopt complementary legislation. As policy jurisdiction increases in scope, administrative cost rises.

Williams [1995] finds that the calculation of optimal carbon tax rates is not complete with an estimation of MED. On the contrary, in a model that relates the time path of global carbon dioxide emissions to the time path of global average temperature from the greenhouse effect, Williams finds that new carbon taxes will raise significant administrative cost. He concludes that these costs would offset the benefits of pricing the global warming externality. As a result, the optimal carbon tax is zero in the U.S., according to Williams, for at least several decades.


Another administrative goal is to minimize the cost of compliance. In the U.S., each environmental tax, which represents a small fraction of the government budget, collects revenue for a cleanup program. Each tax requires its own administrative structure, its own set of forms, and for individual polluters, its own calculation of the tax base. The cost of these calculations remains the same whether the tax base is large or small. As a result, the compliance cost as a fraction of revenue tends to be high when the tax rates are low. In the U.S., each environmental tax rate is less than 1 percent. Thus, these levies entail a relatively large compliance cost per dollar of revenue [Fullerton, 1996]. The compliance cost results from monitoring, acquiescing, evasion, and avoidance.

In the U.S., the importance of monitoring led to Section 313 of Title III of the Superfund Amendments and Reauthorization Act. Established in 1986, this policy requires large manufacturing firms in standard industrial classification codes 20 to 39 to monitor and report more than 600 chemical pollutants annually. These firms, with 10 or more employees, manufacture or process at least 25,000 pounds or use at least 10,000 pounds of one of more than 600 chemicals. In the Toxics Release Inventory National Report, the EPA [1991] tabulates chemical releases into the air, water, and earth. By making data available, this report informs citizens of chemical hazards. Monitoring expense depends on emission characteristics--including flow, concentration, and stability--and abatement technology. The expense decreases with the implementation of advanced abatement techniques [Brashares and Gerardi, 1993].

For the individual polluter, compliance also entails acquiescing, the time and resources allocated beyond the cost of pollution abatement, to the minimization of the tax payment. Acquiescing includes record keeping, seeking professional advice, preparing returns, and sending the tax payment [Graetz et al., 1989]. As an example, the cost of compliance with the Superfund Amendments and Reauthorization Act corporate environmental tax can be 100 percent of total corporate environmental tax revenue [Fullerton, 1996]. This compliance cost accompanies an environmental tax that was not designed to target the burden on polluters or provide an incentive for pollution abatement.

Because the tax payment partially depends on the size of the base, environmental tax systems, by their design and focus, provide incentive for the misrepresentation of pollution activity. An optimal tax analysis of environmental taxation must therefore, in its calculation of compliance cost, include estimations of evasion, the illegal game of tax reduction or nonpayment and avoidance, the legal methods of tax reduction. Evasion of environmental taxes, although not measured quantitatively in the U.S. or elsewhere, reduces efficiency and increases collection costs when agents allocate resources away from production to cheat on tax payments. A misrepresentation of emissions also severs the tie between marginal tax rates and marginal damage, when rates focus on the latter effect. Avoidance activity entails reducing tax liability through the avoided cost of pollution damage. This expenditure, the marginal cost of abatement, may be calculated as a cleanup cost or avoided pollution control. Estimates of abatement cost vary widely [Anderson, 1993]. [6] Some economists, most notably Porter [1991], argue that the abatement expenditure which results from the implementation of environmental policy increases the competitiveness of regulated firms.

Tax policy intends to discourage evasion. For this reason, no tax system can stand alone without an effective enforcement mechanism. The extent that individual polluters comply with environmental tax policy depends on the marginal tax rate, the probably of detection, and the penalty of evasion. Presumably, for compliance increases with lower marginal tax rates, the chance of detection is greater and the penalty of noncompliance is higher. By underreporting pollution, a firm decreases its tax liability if evasion goes undetected, but the polluter pays a penalty if detected. Therefore, the decision to evade tax payment depends on the firm's attitude toward risk, the chances of being caught, and the penalty if detected [Slemrod, 1990]. Additional research must focus on the economic values of evasion and avoidance from environmental taxes, and the extent that the existing penalty structure provides incentive for compliance.


Important policy trade-offs involve revenue. An optimal tax analysis must include five interwoven issues: yield, short-term stability, elasticity, long-term base behavior, and revenue functions. The first issue, yield, determines the level of revenue from each tax base. Worldwide, taxes on output and inputs with environmental implications raise significantly more revenue than emission taxes [OECD, 1995a]. Since the goal of incentive-based policy is to reduce the tax base, no environmental tax implemented for the purpose of pollution abatement will establish a consistent revenue yield. Environmental taxes in the U.S. are set to raise a predetermined level of revenue for cleanup programs. Because these taxes do not provide the incentive for pollution abatement, revenue yields are consistent over time.

The second revenue issue, short-term stability, relates the collection of environmental tax revenue to the business cycle. With sales and income taxes, policy makers intend to minimize the risk, that is, revenue instability. When environmental taxes are intended to induce behavioral response, revenue instability prohibits stable funding for cleanup programs. Revenue instability is a necessary byproduct of effective policy. During economic downturns, a reduction in output, emissions, and sales lowers the tax payment on resource inputs, commodities, and emissions when revenue directly correlates with the level of the tax base.

The third issue, elasticity, refers to the impact of a change in the tax rate on revenue. A trade-off exists between revenue stability and elasticity. In general, the most stable tax base is likely to be inconsistent with an elastic base [Murray, 1997]. The calculations summarized by Lee and Misiolek [1986] show that tax base elasticities for a number of air and water pollutants in the U.S. display great variation: some exceed unity while others do not. Bovenberg and van der Ploeg [1994] conclude that environmental tax rates in a second-best economy are likely to lie below the Pigouvian rate. From the perspective of the revenue authority, such rates may or may not increase revenue.

The fourth revenue issue focuses on the long-term potential, through a lower tax base, of greater environmental quality. Economic theory prescribes the implementation of externality taxes on the polluting activity that generates external cost [Baumol and Oates, 1988]. Policy effectiveness depends on the extent that the tax is linked to environmental damage [OECD, 1993]. A marginal tax rate high enough to impact behavior will encourage agents to substitute away from dirty consumption or production. In particular, by discouraging waste output directly, an emission tax distinguishes between a change in the production of economic output and waste output, but emissions are often expensive to monitor. An output or input tax, such as the gas-guzzler charge in the U.S., is often used instead [OECD, 1995a].

The fifth issue concerns potential revenue uses, including a contribution to the general budget, revenue substitution, or earmarking. While revenue in OECD countries from taxes on petrol, sulfur, and waste disposal contributes to general budgets, total receipts from each levy is low but increasing [OECD, 1994]. Pearce [1991] argues for revenue substitution: by reducing rates on preexisting taxes that distort economic decision making, environmental taxes could gain political acceptance and reduce distributional impacts. However, this policy creates the TI impact as previously discussed.

Public Choice

Environmental taxes in the U.S. (see Footnote 1) raise revenue for specific cleanup programs. These taxes come from different industries responsible for contaminated sites and from finance trust funds to clean up the sites. However, the taxes do not lead to behaviors that discourage contamination [Fullerton, 1996]. As a result of the dual nature of environmental taxes, both environmental regulators and fiscal authorities are involved in policy design. Therefore, a fundamental trade-off exists between revenue adequacy and improvements in environmental quality.

From a public choice perspective, a revenue authority (such as a tax committee in the legislature or Treasury official) may attempt to set marginal tax rates to maximize the revenue stream or raise a predetermined level of revenue for earmarking but not decrease the tax base. In contrast, to provide the incentive for pollution abatement, the environmental regulator may increase tax rates at the margin but sacrifice an adequate revenue stream. The implementation of environmental taxes must entail a compromise that reflects both revenue needs and environmental objectives [Oates, 1995].

Public authorities allocate environmental tax revenue for cleanup programs, recycling centers, and water treatment plants. Oates [1995] cautions against such earmarking: projects could result from the availability of earmarked public funds but not because of economic and budgetary tests. By altering environmental taxes in the U.S. to provide the incentive for pollution abatement, instead of serving as revenue instruments, policy makers could focus on the tax burden and increase efficiency.


An analysis of environmental taxation in an optimal tax framework reveals the important trade-offs of policy design. If revenue neutrality must accompany environmental taxation, an efficiency-reducing TI effect will partially offset the benefits of pollution abatement and RR. All environmental tax policies that intend to impact decision making or fund cleanup programs entail important administrative and compliance costs. Taxes intended to create the incentive for pollution abatement must be accompanied by an appropriate mechanism for enforcement. Taxes designed to raise revenue for cleanup may lead to a steady yield but may not provide the incentive for abatement. Conversely, higher marginal rates may reduce the tax base but may not provide adequate funding for cleanup programs. Enhancing policy effectiveness lies in improving the tradeoffs of policy design.

The optimal tax methodology in Table 1 clarifies and conceptualizes the policy making process, allows a comprehensive assessment of policy objectives, and provides a framework to contrast regulations. The optimal tax methodology also helps to clarify that policy alternatives possess specific strengths and weaknesses.

The identification of appropriate regulation crucially depends on the choice of policy goals, their corresponding attributes, and the trade-offs of policy design. Whether net policy benefits result from environmental taxation depends on the policy scope and the portfolio choices of the regulator. Policy assessment first requires the determination of the relevant costs and benefits of policy objectives. Assessment then necessitates the application of different weights of importance to the attributes of policy goals. The final step of assessment entails the calculation of net policy impacts, which crucially depends on the allocation of public resources.

If the governing authority wants to target the source of environmental damage while maintaining distributional incidence, an incentive policy may serve as an appropriate alternative, and resources may need to be allocated for the subsidization of compliance cost beyond some predetermined level. Raising important administrative, compliance, and efficiency costs, the extent that environmental taxation increases welfare crucially depends on the level and value of pollution abatement.

(*.) Pace University--U.S.A.


(1.) The IRS Statistics of Income program identifies environmental taxes which raise revenue for trust funds that help to clean up pollution problems. The 1980 excise tax on hazardous waste finances the Superfund. The Oil Spill Liability Trust Fund pays for unrecovered claims that result when an oil owner assumes first liability [Barthold, 1994]. Other charges include taxes on chemical inputs to production, which are collected from industries thought responsible for pollution damage, taxes on chemical feedstocks for the Superfund, and taxes on motor fuels for leaky underground storage tanks [Fullerton, 1996].

(2.) For thorough discussions of optimal taxation, see Slemrod [1990], Alm [1996], and Murray [1997].

(3.) The argument for environmental taxation usually begins with estimates of excess burden. This burden is the loss, in addition to tax payment, of private revenue due to reductions in work effort, saving, and investment. Research by Ballard et al. [1985] found marginal excess burden to be $.17 to $.56 per dollar for every 1 percent increase in all distortionary taxes, including taxes on corporate profits, property, social security, motor vehicles, retail sales, and personal income.

(4.) A well-known principle of public finance is that when a wide tax base achieves a target level of revenue, efficiency costs will be small. In addition, the efficiency cost will decrease the more compensated demand or supply is inelastic. The Ramsey principle of commodity taxation endorses tax rates in inverse relation to demand elasticity [Stiglitz, 2000].

(5.) When implemented to enhance environmental quality, environmental taxes reduce the tax base. In a revenue-neutral setting, environmental taxes will not be able to continuously finance lower rates on preexisting taxes. The real wages of workers in regulated industries will decline. An important efficiency cost arises when workers substitute leisure for labor [Qates, 1995; Goulder, 1995, 2000; Bovenberg, 1998].

(6.) Abatement cost information exists for nitrous oxide, carbon dioxide, sulfur dioxide, lead, and many other substances. Anderson [1993] explains that carbon dioxide abatement costs are $1.17 per pound by rail, $2.56 per pound by truck, and $1.54 per pound by barge. Abatement costs vary widely, depending on the type of production and pollutant, means of equipment replacement, emission characteristics, and other factors.


Aim, James. "What is an 'Optimal' Tax System?," National Tax Journal, 49, 1, March 1996, pp. 117-33.

Anderson, Dennis. "Energy-Efficiency and the Economics of Pollution Abatement," Annual Review of Energy and the Environment, 18, 1993, pp. 291-318.

Ballard, Charles L.; Shoven, John B.; Whalley, John. "General Equilibrium Computations of the Marginal Welfare Costs of Taxes in the United States," American Economic Review, 75, 1, March 1985, pp. 128-38.

Barthold, Thomas A. "Issues in the Design of Environmental Excise Taxes," Journal of Economic Perspectives, 8, 1, 1994, pp. 133-51.

Baumol, William J.; Oates, Wallace E. The Theory of Environmental Policy, 2nd ed., New York, NY: Cambridge University Press, 1988.

Bovenberg, A. Lans. "Environmental Taxes and the Double Dividen," Empirica, 25, 1, 1998, pp. 15-35.

Bovenberg, A. Lans; de Mooij, Ruud A. "Environmental Levies and Distortionary Taxation," American Economic Review, 84, 4, September 1994, pp. 1085-9.

Bovenberg, A. Lans; Goulder, Lawrence H. "Optimal Environmental Taxation in the Presence of Other Taxes: General-Equilibrium Analyses," American Economic Review, 86, 4, September 1996, pp. 985-1000.

Bovenberg, A. Lans; van der Ploeg, F. "Environmental Policy, Public Finance, and the Labor Market in a Second-Best World," Journal of Public Economics, 55, 1994, pp. 349-90.

Brashares, Edith; Gerardi, Geraldine. "Issues in the Design of a Tax on Toxic Pollutants," National Tax Association: Proceedings of the Eighty-Sixth Annual Conference, 1993, pp. 139-45.

Carlin, Alan. "The United States Experience with Economic Incentives to Control Environmental Pollution," report, EPA-230-R-92-001 U.S. Environmental Protection Agency, July 1992.

Fullerton, Don. "Why Have Separate Environmental Taxes?," in James M. Poterba, ed., Tax Policy and the Economy, 10, Cambridge, MA.: MIT Press, 1996, Pp. 33-70.

Goulder, Lawrence H. "Energy Taxes: Traditional Efficiency Effects and Environmental Implications," in James M. Poterba, ed., Tax Policy and the Economy, 8, Cambridge, MA: MIT Press, 1994, pp. 105-59.

_____. "Environmental Taxation and the Double Dividend: A Reader's Guide," International Tax and Public Finance, 2, 1995, pp. 157-83.

_____. "Environmental Policy Making in a Second-Best Setting," in Robert N. Stavins, ed. ,Economics of the Environment, 4th ed., New York: W. W. Norton, 2000, pp. 396-427.

Goulder, Lawrence H.; Parry, Ian W. H.; Burtraw, Dallas. "Revenue-Raising vs. Other Approaches to Environmental Protection: The Critical Significance of Preexisting Tax Distortions," RAND Journal of Economics, 28, 4, Winter 1997, pp. 708-31.

Goulder, Lawrence H.; Parry, Ian W.H.; Williams, Roberton C., III; Burtraw, Dallas. "The Cost-Effectiveness of Alternative Instruments for Environmental Protection in a Second-Best Setting," Journal of Public Economics, 72, 3, 1999, pp. 329-60.

Graetz, Michael J.; Dublin, Jeffrey A.; Wilde, Louis L. "United States," in Cahiers de Droit Fiscal International, ed., Administrative and Compliance Costs of Taxation; Deventer, Netherlands: Kluwer Law International, 1989, pp. 311-44.

Hamond, M. Jeff; DeCanio, Stephen J.; Duxbury, Peggy; Sanstad, Alan H.; Stinson, Christopher H. "Tax Waste, Not Work," Challenge, 40, 6, November-December 1997, pp. 53-62.

Kotlikoff, Laurence J.; Summers, Lawrence H. "Tax Incidence," in Alan Auerbach; Martin Feldstein, eds., Handbook of Public Economics, 2, Amsterdam, Netherlands: Elsevier Science, 1987.

Lee, Dwight R.; Misiolek, Walter S. "Substituting Pollution Taxation for General Taxation: Some Implications for Efficiency in Pollutions Taxation," Journal of Environmental Economics and Management, 13, 1986, pp. 338-47.

McKenzie, Fiona. "Policy Formulation for the Management of Hazardous Pollutants," Journal of Environmental Planning and Management, 37, 1, 1994, pp. 87-105.

Muller, Frank; Hoerner, J. Andrew. "Greening State Energy Taxes: Carbon Taxes for Revenue and the Environment," Pace Environmental Law Review, 12, 1, Fall 1994, pp. 5-60.

Murray, Matthew N. "Moving the Retail Sales Tax to a Retail Tax: Optimal Tax Considerations," in Matthew N. Murray; William F. Fox, eds., The Sales Tax in the Twenty-First Century, Westport, CT: Praeger, 1997.

Oak Ridge National Laboratory; Resources for the Future. Estimating Externalities of Coal Fuel Cycles, Washington, DC: U.S. Department of Energy, 1994.

Oates, Wallace E. "Green Taxes: Can We Protect the Environment and Improve the Tax System at the Same Time?," Southern Economic Journal, 61, 4, 1995, pp. 915-22.

Organization for Economic Cooperation and Development. Taxation and the Environment: Complementary Policies, Paris, France: OECD, 1993.

_____. Managing the Environment: The Role of Economic Instruments, Paris, France: OECD, 1994.

_____. Environmental Taxes in OECD Countries, Paris, France: OECD, 1995a.

_____. Implementation Strategies for Environmental Taxes, Paris, France: OECD, 1995b.

_____. Environmental Taxes and Green Tax Reform, Paris, France: OECD, 1997.

Parry, Ian W. H. "Pollution Taxes and Revenue Recycling," Journal of Environmental Economics and Management, 29, 3, November 1995, pp. 64-77.

Pearce, David. "The Role of Carbon Taxes in Adjusting to Global Warming," Economic Journal, 101, July 1991, PP. 938-48.

Porter, Michael A. "America's Green Strategy," Scientific American, 264, 4, April 1991, p. 168.

Repetto, Robert; Dower, Roger C.; Jenkins, Robin; Geoghegan, Jacqueline.Green Fees: How a Tax Shift Can Work for the Environment and the Economy, Washington, DC: World Resources Institute, 1992.

Robison, H. David. "Who Pays for Industrial Pollution Abatement?, "The Review of Economics and Statistics, November 1985, Pp. 702-6.

Russell, Clifford S. "Monitoring and Enforcement," in Paul R. Portney, ed., Public Policies for Environmental Protection, Washington, DC: Resources for the Future, 1990.

Slemrod, Joel. "Optimal Taxation and Optimal Tax Systems," Journal of Economic Perspectives, 4, 1, Winter 1990, Pp. 157-78.

Stiglitz, Joseph E. Economics of the Public Sector, 3rd ed., New York, NY: W. W. Norton, 2000. Terkla, David. "The Efficiency Value of Effluent Tax Revenues," Journal of Environmental Economics and Management, 11, 1984, pp. 107-23.

U.S. Environmental Protection Agency. "Toxics in the Community: National and Local Perspectives. The 1989 Toxics Release Inventory National Report," report, 560/4-91-014, September 1991.

Von Weizsacker, Ernst U.; Jesinghaus, Jochen. Ecological Tax Reform, London, United Kingdom: Zed Books, 1992.

Williams, Michael. "Global Warming and Carbon Taxation: Optimal Policy and the Role of Administration Costs," Energy Economics, 17, 4, 1995, pp. 319-27.
            Optimal Tax Criteria for Environmental Taxation
Criteria                         Emission Tax  Output Tax  Input Tax
  TI effect                           x            x           x
  Revenue-recycling (RR) effect       x            x           x
  Environmental impact                x            x           x
  Relevant costs and benefits         x            x           x
  Direct/indirect effects             x            x           x
  Formal/effective incidence          x            x           x
  Baseline comparison                 x            x           x
  Dimensions of distribution          x            x           x
  Pollution identification            x            x           x
  Policy adaptability                 x            x           x
  Collection                          x            x           x
  Enforcement                         x            x           x
  Pollution source/destination        x            x           x
  Monitoring                          x            x           x
  Acquiescing                         x            x           x
  Evasion              x  x  x
  Avoidance            x  x  x
  Yield                x  x  x
  Short-run stability  x  x  x
  Elasticity           x  x  x
  Base behavior        x  x  x
  Functions            x  x  x
COPYRIGHT 2001 Atlantic Economic Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Atlantic Economic Journal
Geographic Code:1USA
Date:Jun 1, 2001
Previous Article:Economies of Scale and Optimal Capital in Nuclear and Fossil Fuel Electricity Production.
Next Article:The Political Economy of Gunnar Myrdal.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters