by Joachim von Amsberg
The World Bank, Latin America and the Caribbean Region, Country Department I, Natural Resources, Environment and Rural Poverty Division, 1818 H Street, N.W., Washington, DC 20433, USA, Fax (202) 614-0128, Internet: JVONAMSBERG@WORLDBANK.ORG
July 19, 1995
The findings, interpretations and conclusions of this paper are those of the author and should not be attributed to the World Bank, its Executive Board of Directors or any of its member countries.
Cost-effectiveness has emerged as a major consideration in the design of environmental policies. Cost-effectiveness means that with prudent policy design, the same level of environmental improvement can be achieved at a lower cost, which implies real cost savings for the affected polluters. Alternatively, a higher level of environmental quality can be achieved at the same cost, which implies a real gain for the population affected by pollution. Cost-effectiveness thus generates win-win opportunities between polluters and the community at large and has emerged as a concept that can bring the battling parties in the political controversy about pollution control policies together to one table.
Economists have long argued that economic instruments for environmental policy can lead to significant savings in social costs over conventional administrative regulation with emission standard. However, only recently has significant experience with the use of economic instruments been gathered in OECD countries. Most developing countries have on their books environmental standards and policies that are copied from regulations in the United States or Europe. However, very few of these regulations have been enforced systematically. In recent years, policy makers in developing countries have realized that effective environmental controls are a critical element of sustainable economic development. The cost implications of implementing the standards that are theoretically in place are therefore becoming more and more relevant. The question is being asked whether developing countries can afford the same costly path of administrative pollution control that industrial countries have taken but are now slowly correcting or whether there are more cost-effective approaches available. As a result, significant interest in the use of economic instruments has emerged in several developing countries over the last few years.
The experience with the use of economic instruments in the environmental policy of developing countries is extremely limited. There are very few successful examples of significant scale. This paper attempts to gather some of the experiences available in developing countries, enriched by the more ample experiences in the newly industrialized countries of east Asia and the transition economies in Eastern Europe and draw lessons from the experiences presented. The paper does not attempt a comprehensive survey of all non-OECD countries and all instances of the use of economic instruments. Rather, examples were selected for which information was readily available and from which interesting lessons could be drawn. The paper is restricted to the analysis of economic instruments for pollution control policies. Interesting examples of economic instruments can also be found in the management of natural resources, but these examples are excluded from the analysis in this paper.
After a brief conceptual discussion of different economic instruments for pollution control, various examples of recent experiences are briefly described followed by an analysis of the main lessons of the limited available experience. The discussion will show that the dichotomy between administrative and economic instruments of pollution control is often exaggerated. While economic instruments, can in principle, yield significant cost savings, many situations call for a pragmatic combination of administrative and economic instruments. Moreover, detailed design aspects of either instruments are very important in order to avoid adverse impacts. In this sense, economic instruments should be viewed as an important addition to the toolbox of environmental regulators and not as a panacea that can overcome the ineffectiveness of current regulations.
Pollution causes costs that are external to the polluter. They occur in the form of health damages, material and vegetation damages, lost productivity, visibility and property values and other costs to society. Without government interventions, polluters rarely have an incentive to control the pollution caused by their activities. Efficient government policies internalize the external costs of pollution; they make polluters behave as if they themselves bore the costs of pollution.
Traditionally, government regulation has focused on so called "command-and-control" (CAC) instruments, which determine emission standards for every polluting source, either uniformly for all sources or differentiated by source. In either case, the polluter has no choice but to comply with the mandated emission standard. In contrast, economic instruments (EI) or market-based instruments (MBI) change the incentives of polluters without determining a specific level of required pollution control for each polluting source.
In theory, economists have shown very significant cost savings that can be achieved by using EI instead of CAC regulation. The key saving from EI results from the fact that EI can more easily achieve equal marginal abatement costs across pollution sources. This means that sources, which can reduce pollution at a lower cost, will reduce pollution more than those sources, which would occur higher costs for reducing pollution by the same amount. Clearly, if pollution is reduced at the sources where the costs are lower, total social costs of pollution control will be less. EIs leave the polluters with choices about the level of pollution abatement at the individual source and about the technology used to achieve pollution control. Together with the economic incentives of polluters to minimize their costs, this flexibility means that pollution control can be achieved at the lowest cost.
The real cost savings from the use of EI will depend on "how bad" actual CAC regulation is. A bad CAC regulator imposes uniform standards across all sources. A good CAC regulator would take the abatement costs of different sources into account when setting the standards for specific sources. A good CAC regulator would thus try to imitate the results of regulation with EI. However, there are good reasons to believe that even a very sophisticated CAC regulator would not achieve the same cost savings that can be achieved through EI regulation. The first reason is that regulators will always have less information than the polluters about all possibilities for pollution abatement and their costs. The second reason is that EI create a dynamic incentives for polluters to find cheaper ways of pollution control beyond the levels determined by the applicable standards.
The term economic instruments refers to a set of different regulations that attempt to create markets for environmental protection, or at least integrate environmental costs in market prices while leaving polluters the choice about their individual pollution abatement levels and their technology. Table 1 shows a taxonomy of policy instruments for pollution control distinguishing between EI, CAC and direct government provision, on the one hand, and direct instruments that are tied directly to emission levels versus indirect instruments that address pollution through the inputs or outputs to the processes that generates pollution, on the other hand.
Table 1: A Taxonomy of Policy Instruments; Source: Eskeland and Jimenez, 1991a
EI/MBI - Direct Instruments: Effluent charges or subsidies (price based), tradable permits (quantity based), deposit-refund system
EI/MBI - Indirect Instruments: Input/output taxes and subsidies, subsidies to substitutes and to abatement inputs
CAC - Direct Instruments: Emission standards (source-specific)
CAC - Indirect Instruments: Regulation of equipment, processes, input and output
Government Production - Direct Instruments: Cleanup, waste disposal
Government Production - Indirect Instruments: Technological development
Clearly, the textbook dichotomy between CAC regulation and economic instruments exaggerates the real life differences between these two approaches to pollution control. In real life, pollution control policies focus heavily on negotiated agreements between regulators and polluters. In these cases, regulators can apply economic thinking to different degrees.
The case of environmental taxes and charges shows that the distinction between CAC regulation and EI is not always clear. Under CAC regulation, emissions above the standards would be prohibited and punished. Unless the punishment is prohibitive and completely enforced (i.e. prison term or very high fines) a polluter may well decide to accept the fine and continue to pollute above the standard. In this case the fine has a similar effect as a tax or charge which has the objective to influence pollution levels but does not determine a fixed maximum level of pollution.
Under an environmental charge in the proper sense, the intention is to leave the polluter free to set the level of pollution but charge for every unit of pollutants emitted. Thus, if the unit charge is higher than the unit abatement cost, the polluter will voluntarily reduce pollution levels. In order to reduce total costs (abatement costs plus remaining charges), the polluter will seek cheaper ways and better technologies for pollution control.
Even though the terms are not always used consistently, the difference between taxes and charges is generally that taxes are raised with the objective of raising revenues rather than influencing behavior. Taxes, however, do also often influence behavior (see section on gasoline taxes below). Charges, in turn, are meant to influence behavior but also often contribute significantly to government revenues. In many countries, environmental charges are earmarked for the environmental protection or enforcement agency. In these cases, incentives for enforcement agencies to do their job are significantly improved. Earmarked charges also generate the funds urgently needed for building up environmental management capacity.
There are different types of environmental charges: (a) effluent or emission charges are assessed on pollution released into the environment; (b) user charges require polluters to pay discharge and treatment services in proportion to the amount of pollution they generate; (c) product charges are added onto the price of products to reflect the cost of collection, treatment or recycling connected with the production or use of the product; and (d) administrative charges finance systems of licensing and license monitoring which may be necessary to track harmful substances.
It is important to note, that charges fix the price of pollution but do not fix the quantity of pollution. If charges were set equal to marginal damage costs of pollution ("Pigouvian taxes"), market incentives would result in the efficient level of pollution. Often, however, regulators don't know the marginal damage costs of pollution and want to set a maximum level of pollution. Then, escalating charges are used that are prohibitive for emissions above the maximum emission levels that regulators are willing to accept.
A direct charge on the quantity of pollution emitted is obviously the most desirable solution. This requires the government to continuously monitor each emitters' emissions. This is often not feasible, particularly in the case of many small emitters or informal sector emitters. In these cases, the use of indirect instruments such as charges on the inputs to or the outputs from polluting processes may be an attractive alternative. The most common example are fuel taxes, which are particularly popular since it is difficult to monitor continuously the emissions from individual vehicle. Also, there exists a relatively well established correspondence between fuel or energy use and emissions for a given automobile type or a given industrial process.
Recently, the emission elasticity of various fuel prices in the Chilean manufacturing industry have been estimated (Eskeland and Devarajan, 1994). These elasticities determine the response of emissions to changes in fuel prices due to energy conservation and shifts in demand between different fuels. The results show significant responsiveness of emissions and highlight the potentially important role that fuel or energy taxes could play as an instruments of environmental policy.
Governments have widely accepted the "polluter-pays-principle" (PPP) as a distributional rule for the selection of policy instruments. Even though subsidies for pollution control violate PPP, subsidies in various farms (cash-subsidy, tax credits, tariff exemptions, directed credit programs) have been widely applied in many countries.
In theory, a suitably designed subsidy for pollution control that is based on the quantity of emission reduction can be equivalent to charges in the ability to induce polluters to reduce pollution to a desired or efficient level. In practice, however, subsidies are often not well targeted, are not tied to the quantity of pollution reduction, and contribute to the expansion of the polluting activities. The reasons for pollution control subsidies are, thus, more based on political than economic arguments. In particular, subsidies may help overcome resistance against the implementation of tighter pollution control policies if they would otherwise lead to losses or even business failures in the affected industries. Under certain circumstances, a subsidy can reduce the cost of monitoring and enforcement of pollution control policies (Estache 1992). Another problem with subsidies is that they are often fiscally unsustainable.
Sometimes, subsidies can play a useful role to accelerate pollution abatement during the transition from a period of little industrial pollution control to a period of broader compliance (Brandon and Ramankutty, 1993). In a situation of capital market distortions, directed credit lines for pollution control investments have sometimes been justified since small and medium size enterprises, who would otherwise not have access to credit, would be forced out of business by stricter pollution control. However, the arguments for using subsidies are weak and the experience is mixed. In any case, subsidies should be offered for a limited time only, they need to be accompanied by strict enforcement policies, and need to be carefully targeted to industries which could otherwise not meet pollution control targets.
While charges fix the cost of pollution control but leave the total level of emissions to be determined by the market, a system of tradable permits fixes the total amount of emissions from all sources but leaves the price of pollution and the allocation of the total emissions to individual sources to the market. The regulator issues emission permits in the amount of total acceptable pollution, which are then made tradable between emitting sources. Permits can either be reissued periodically for a specific period of time or be issued in perpetuity often with a built in reduction schedule. In a free standing permit system, every polluter needs a permit for any amount of emissions. Alternatively, a permit system can be added to an existing standard. In this case, new polluters or polluters which want to exceed the existing standard would need to buy compensating emission reductions below the standard from other polluters. In a free-standing permit system, the initial allocation of permits can be either auctioned to the highest bidders or issued free of charge in proportion to historic emissions or some other allocation formula.
There are various design variants to permit trading system to ensure that the emissions from trading sources are indeed comparable. For example, trade for atmospheric emissions can be restricted to limit trading that would increase emissions in already heavily polluted areas. Alternatively, a pollution permit can be designed such that it would allow different quantities of emissions in different geographic zones to reflect different ambient conditions, and thus different damage costs.
Deposit-refund schemes are similar to environmental charges but have a built-in mechanism that reduces enforcement costs. These schemes involve the payment of an initial deposit on potentially polluting products. If the products are returned or safe disposal can be demonstrated, thus avoiding pollution, a refund is paid. The most common example is the extension of the historic deposit-refund scheme for glass bottles (where the motivation was for the producer to recover the material value of the glass bottle) to plastic bottles (where the motivation is to promote recycling and waste reduction).
Some countries have recently applied the example of a deposit on bottles to automobiles. Potential future applications would include solid and hazardous wastes where the refund would be paid upon proof of safe disposal. Here, the deposit-refund scheme would help overcome the incentive for illegal dumping that otherwise would result from increasingly costly requirements for safe disposal. Performance bonds and presumptive charges are other instruments that apply a similar mechanism and shift the burden of proof of environmentally sound behavior to the polluter.
The potential of ex-post liability in the future for health and environmental damages caused in the present provides an incentive for firms to reduce pollution even in the absence of ex-ante regulation limiting emissions. Liability schemes are generally implemented by the court system in contrast to the administration of pollution taxes and charges by government agencies. When polluters know with certainty that they will be required to pay compensation for damages resulting from pollution, a liability system can be effective in dealing with environmental problems
A benefit of liability systems is that they are largely independent of government regulatory activities. A drawback is that they can initiate long and expensive litigation, which may not be worthwhile in smaller cases or where there is long-term environmental injury and causation is difficult to establish (Brandon and Ramankutty, 1993). Another problem is that a liability regime only provides sufficient incentives if the liable capital of polluters is sufficient to cover all possible liability claims. Some of the mentioned problems can be overcome by making environmental liability insurance mandatory. In this case, insurance companies would likely take on some functions of environmental monitoring which would allow them to set insurance rates according to the environmental risks of a specific operation.
In several OECD countries, consumer preference for environmental protection has become a powerful market force for industry to use environmentally sound technology. Consumer preferences, expressed through willingness to pay higher prices for environmentally benign products, create the market incentives for producers for more pollution control. Obviously, information dissemination is a critical prerequisite for this mechanism to work.
A relatively low-cost and potentially important regulatory step would be the introduction of public disclosure requirements, for example for emission inventories. The release of information to the public increases pressure by the community on polluters to reduce their emissions. In certain circumstances, they can lead to the internalization of some of the cost of polluters, if the affected community decides to reduce demand for products from polluting facilities or reduce their supply of labor to such facilities.
There are various ways in which traditional instruments of environmental management can be improved by integrating some aspects of economic thinking. Pollution abatement investments in major industrial operations usually follow the negotiation of a compliance plan between the regulator and the regulated. In these negotiations, creative "economic" solutions that follow the concepts of economic instruments without having the characteristics of a formal system can improve the outcome of these negotiations. For example, the investor of a power plant that would be a major emission source, may be allowed to invest in emission reduction measures in the residential sector that would achieve the same emission reduction at a lower cost than tighter emission limits for the plant itself.
Similarly, economic thinking can be integrated with the Environmental Impact Assessment (EIA) process that is now established practice in many countries. A promising approach for making the EIA process more efficient and transparent is the application of shadow prices for environmental damages in the EIA evaluation. One of the main criteria for evaluating an EIA would be whether the investment project is profitable after subtracting its environmental shadow cost. The environmental agency would develop a shadow price list of environmental damages (a monetary amount per ton of each pollutant) from economic analysis, including studies of damage costs or abatement costs done at other locations as a first-order approximation.
A significant amount of experience has been gathered in OECD countries on the use of economic instruments, and their use continues to increase (Tietenberg 1995). In developing countries, on the other hand, the experience is extremely limited. There are very few glorious examples of large scale application of economic instruments. However, on closer analysis, many little examples of the use of economic instruments can be found in a significant number of countries. Some of these experiences are summarized in this section.
Arguably the most important case of how economic instruments have been use effectively in bringing about environmental improvement is the removal of subsidies on goods that contribute to environmental deterioration. Even though the removal of subsidies has often been motivated by considerations of economic efficiency and financial sustainability, their effect on the environment has been important. The environmental benefits that can be achieved by reducing the widespread subsidies on the use of energy, water, and agrochemicals have been amply demonstrated. The removal of these subsidies is a typical win-win policy that results in economic efficiency gains and also improvements in the environment. In 1992, it was estimated that developing countries subsidized energy consumption with US$ 150 billion annually, of which US$ 100 billion were subsidies for electricity consumption. Developing countries use about 20 percent more electricity than they would if consumers paid the true marginal cost of supply. The pollution from the generation of electricity for this excess demand is the direct result of energy price subsidies. Underpricing of electricity also discourages investments in new cleaner technologies and more energy efficient processes (World Bank 1993).
Like many other countries, Indonesia used to heavily subsidize the use of fertilizers and pesticides. The subsidies reached 80% of the full farmgate price of pesticides and constituted 42% of the agriculture and irrigation development budget of the Indonesian government. These subsidies resulted in overuse of agricultural chemicals. Average annual fertilizer consumption in Indonesia was around 100 kg per hectare which was two to three times higher than in comparable Asian countries: 26 kg in Thailand and 50 kg in the Philippines. Apart from the economic efficiency loss of excessive agrochemical usage, environmental damages occurred. The environmental damages from chemical instead of organic fertilizers include excessive soil erosion. Excessive pesticide use contributed to reduced plant resistance, eradication of natural pest predators and pest outbreaks. Furthermore, excessive agrochemicals contributed to significant water pollution problems in Indonesia. The annual economic loss from the subsidies have been estimated at US$ 860 million. Two incidences of environmental damages from excessive agrochemical use have been valued at US$ 180 million (pest population explosion) and US$ 120 million (lost grain variety resistance) respectively (Ruitenbeek 1994).
Effluent charges are the most common economic instrument used in the environmental policies of developing countries. However, in most countries effluent charges are set at very low levels (in relation to abatement costs), and their effectiveness in influencing the polluters' choice of pollution levels is very limited. Moreover, these charges are often not systematically enforced.
In Mexico, the water authority enacted discharge fees for all polluters based on chemical oxygen demand, suspended solids, and volume. No reference is made to heavy metals or other toxic substances. No charge is levied for discharges which comply with the concentration standards. The system is not effectively oriented toward emission control, and the charges are not earmarked for the enforcement agency giving it little motivation to collect the fine so that little has been achieved so far (OECD 1993e).
In 1980, Argentina made an attempt to introduce a discharge fee for industrial effluent (Decree 2125/1980, Cuotas de Resarcimiento por Contaminación). The structure of this fee was that of a two part tariff with a fee for discharges between the permitted level and a maximum allowable discharge, and much higher penalties for discharges above the maximum allowable threshold. There were provisions for increasing the level of fees gradually over 10 years up to the level of treatment costs and for the granting of transitory waivers for up to two years where enterprises were in the course of implementing abatement measures. In practice, the fees were never applied on a wide basis and the system was modified in 1989 to lower the level of fees and to revise the penalties imposed on enterprises exceeding the maximum allowable discharge. Environmental groups sued the Government on the grounds that the Cuota de Resarcimiento amounted to a license to pollute beyond legal limits. The court declared the decree unconstitutional based on formal legal grounds and the issue remains confused in legal terms. It appears that the court regarded the fee as exceeding the powers of the national government to levy taxes and that it could not be justified as payment for a service.
Four states in Brazil have recently applied, or are in the process of applying effluent charges as industrial sewage tariffs based on pollution content. In Rio de Janeiro the charge will be levied on all polluters and will be based on volume and concentration of the effluent, including BOD and heavy metals. Tariff rates, defined as price per pollution unit, are defined to cover the budgetary needs of the State Environment Agency, which is responsible for revenue collection. In Sao Paulo State, an effluent charge (sewerage tariff) was approved in 1983 but has only partially been implemented in some regions, Rates are defined as a function of average sectoral emissions of BOD and suspended particulates. An improved version of the system will allow for rebates for previous payments or payments of less than the average if firms can prove they are cleaner than the average. The tax is collected by the water and sanitation company as a payment for sewerage treatment services. In Parana State the system is similar to that of Sao Paulo and is only now beginning to be implemented. Revenues will cover the cost of collection and treatment of industrial pollutants and the tariff will be based on the volume of effluent plus a pollution-intensity component. Finally, in Minas Gerais State information on effluent is now being collected to implement an effluent charge system similar to the other states (OECD 1993e, Seroa da Motta 1994)
One of the major historical difficulties in implementing such systems has been the poor coordination between the water and sanitation companies and the environmental agencies. Since the structure of the tariff system requires parameters and information, which are more in the domain of the environmental agency (typically pollution intensity and volume) while the implementation of the system is more the responsibility of the water and sanitation companies, cooperation has now become critical for the efficient functioning of the system. The charging systems are essentially managed by the water and sanitation companies, but part of the revenues go to the environmental agencies to cover the costs incurred in monitoring emissions. (OECD 1993e)
In Sao Paulo State, significant reductions in unit pollution have been reported as the result of the sewerage tariff based on pollution load. For the pharmaceutical, food, and milk derivative industries, BOD emissions were reduced by 30%, 42%, and 57% respectively from 1980-82. For suspended particles, the reported reductions were 46%, 43%, and 55% respectively. (Miglino and Harrington 1984)
A major legal issue is the designation of effluent fees as user charges and not as new taxes which would contradict the political efforts to reduce the number of taxes and the constitutional prohibition on new taxes imposed by the states.
Economic instruments have been applied not only toward individual polluters but also between different government jurisdictions. The State of Parana in Brazil has implemented a very interesting example of the use of economic incentives for environmental objectives in their system of fiscal transfers. At least 75% of the value added tax (ICMS) collected by the state is redistributed to municipalities based on the origin of the value added. The remaining revenues are distributed based on other factors. These other factors include, in the State of Parana, a formula for water quality of the streams leaving from headwater municipality. These transfers are meant to compensate headwater municipalities for the foregone economic activities and the services these municipalities provide to downstream locations, primarily the city of Curitiba, by means of clean water. Since there is an explicit formula linking water quality to revenue transfers, municipalities have an explicit incentive to implement tighter water quality controls within their jurisdictions. For the future, it is envisaged that the city of Curitiba will have to make payments for the clean water it receives and pay a charge for effluents released within city boundaries (Tlaiye and Biller, 1994).
Columbia has emission charges on the books since more than a decade (Decree 02 of 1982 for air and Decree 1594 of 1984 for water). The decree for air emission determines the formula for charges as the multiple of the emissions above the standard times the minimum wage rate times a factor determined for each pollutant by the ministry of health. The latter factor has never been determined rendering the policy instrument for air pollutants ineffective. The water effluent charges (tasas retributivas) have been selectively applied, for example by the Corporación Autónoma Regional del Valle de Cauca (CVC). CVC's water quality control program has been based primarily on negotiations with polluters to achieve compliance with standards within 4-5 years supplemented by the pollution charge which contributes to CVC's revenues (Sánchez and Uribe 1994).
A new environmental law (Law 99 of 1993) goes much further in the use of economic instruments for pollution control. This law explicitly requires the use of pollution charges that would be determined by the economic valuation of the social damages caused by the respective pollution. Damage categories to be evaluated include explicitly among others health, visibility, and direct economic damages. This law has so far not been reglemented and the provisions not been implemented. While this is the most explicit example of a Pigouvian tax written in law, questions about its applicability arise considering that, despite many advances, the art of economic valuation of environmental damages is until now rather rudimentary.
BOD discharges from the palm oil industry are the main water pollution problem in Malaysia. Effluent control in the industry is effected through a system of licensing, discharge standards and fees. The regulation allowed one year for the mills to install treatment facilities and then to comply with a schedule of standards. The first stage of effluent standards took effect in 1978. Dischargers were required to pay RM100 per ton of BOD exceeding the discharge standard and RM10 per ton of BOD equal or less the discharge standard. Mills which succeeded in developing technologies for treating BOD were rewarded by being charged lower effluent fees. The legislation includes the power to waive or reduce effluent fees if research for the reduction of effluent is carried out by the respective establishment. The charging of high effluent fees and the waiver of fees for research on effluent disposal or treatment has expedited research with remarkable breakthroughs in the treatment technology. So far, compliance with pollution control in the sector, however, is found to be unsatisfactory. In 1991, out of 112 mills monitored, 75% were found to comply with BOD discharge standards. Remaining problems are attributed to improper management of treatment systems and the use of under-sized systems while expanding milling capacity. (OECD 1993f)
The use of economic instruments for environmental policy in South Korea has only begun very recently. An emission charge (Environmental Quality Improvement Charge) was introduced in 1991 to provide pollution abatement incentives and to help finance investment projects to improve the environment. The charges are imposed on potential polluters of air and water and are calculated in terms of the amount of fuel and water used by the facilities and the amount of emissions be vehicles. In 1993, approximately US$ 45 million was raised for the Environmental Pollution Control Fund (OECD 1993g).
In recent years, Beijing has introduced a range of environmental management systems for reducing industrial pollution, including a pollution discharge permit and fee system. So far market based instruments have not been used widely. (Brandon and Ramankutty 1993) The mainstay of the Chinese environmental policies is regulation in the form of effluent standards combined with a pollution levy system. The system involves charging enterprises a fee for violating the emission standards (non-compliance fee). It is calculated on the basis of the pollutant that exceeds the standard by the greatest amount. The difference between the measured concentration and the standard is multiplied by a factor to arrive at the levy. The levy is doubled if the polluter is found to be misreporting emissions. Most of the fees collected are retained at the local level, and placed into a fund for redistribution to enterprises in the form of loans for pollution abatement investments. Twenty percent of the funds are retained by the environmental agency to support monitoring and enforcement activities.
Since the level of fees is low (about 0.1 percent of enterprises total production costs) and generally far below the marginal costs of treatment, most firms rather pay the fee than reduce emissions. Moreover, the system does not provide an incentive to reduce emissions below the standard. Since the fees are assessed on the pollutant that exceeds the standard by the largest amount, there are no incentives to reduce the other pollutants (World Bank 1992b). Various reform initiatives to make the pollution levy a more effective system of pollution control in China are currently ongoing.
Environmental charges were introduced in Poland in the 1970s. Although fee levels were increased several times during the central planning era, their effects were counterbalanced by the lack of financial motivation of economic actors. Environmental authorities used a political opportunity during the collapse of the old regime to strengthen the role of environmental charges by increasing their real levels, pegging them to the official inflation index and changing collection frequency. As a result some charges provide significant abatement incentives. In most cases, however, charges are below the estimated damages and site- specific damages are not reflected in charges in a satisfactory manner. Charges are levied on a high number of water and atmospheric pollutants, the reliable monitoring of which exceeds available monitoring capacity. As a result levies are frequently challenged by polluters (Zylicz 1994, Lovei 1994b).
The problems at the national scale were reflected in provincial programs. Throughout the 1980s the provincial government of Katowice attempted to improve the city's air quality by levying fees for emissions that exceeded permissible levels. Although the rates were double those set by the national government for the rest of Poland, they were revised infrequently during the 1980s and fell sharply in real terms as prices rose. Furthermore, because industrial plants claimed that they lacked the resources to invest in better environmental controls, emission permits were typically set much too high to achieve reasonable ambient air quality, and enterprises were often exempted from paying fees and fines.
The situation in Katowice has changed radically since 1990. Air quality has improved significantly, and enterprises are considering or actually investing in environmental control. There are three reasons for this change: (a) some of the worst polluters have closed down permanently; (b) the level of fees and fines has been raised more than ten times in real terms, and payment is enforced under a real threat of closure; and (c) the prospect of privatization has meant that enterprises faced real budget constraints and provincial authorities no longer strived to maintain industrial production at the expense of other objectives. (World Bank 1992a)
Ambient standards were established during the former Soviet regime based on the theoretical requirement of zero human health damages from pollution, demonstrating the idealistic objectives of Soviet-type legislation. In practice, the permanent emission limits that were determined to achieve the ambient standards had to be supplemented by more lenient temporary emission limits corresponding to the technological and economic capabilities of the sector.
Environmental damage calculations undertaken in the early 1970s created the basis for the emission charge system that was gradually introduced starting in 1988. During the experimental phase, large differences existed between regions in the methodologies for calculating charges and in the actual charge levels. Emission charged for carbonmonoxide, for example, showed variations in the order of several hundred between different cities.
The national system of charges was introduced in 1991 and included some 300 base rates for air pollutants and almost 150 base rates for water pollutants. The rates reflected the desire to centrally evaluate and mitigate environmental health and other pollution risks. Regional characteristics such as background pollution and the assimilative capacity of the ecosystem were expressed by coefficients determined by national law. Despite the intention to create corrective taxes that would induce optimal pollution levels, charge rates were in practice calculated to yield enough revenues for mitigating damages.
The present system of charges is a combination of non-compliance penalties and emission charges. The base emission charge system applies when discharges are within permitted levels derived from ambient standards. Charges increase considerable (25 times) for discharges between the maximum permitted and temporarily permitted levels and yet even higher (125 times) for emissions beyond the temporarily permitted levels. The level of charges, however, is still very low, even at the highest level, and does not meet the objective of inducing reduction of emissions to optimal levels. Since 1992, agreements between polluters and the environmental protection authorities create the legal basis for the collection of charges. Such agreements specify (i) the permitted level of discharges; (ii) base rates and penalty rates for each pollutant discharged; and (iii) the schedule for charge payments. (Lovei 1994, NAPA 1994, Averchenkov 1994)
Singapore has introduced the Area Licensing Scheme (ALS) in 1975 with the sole motivation of reducing congestion. Even though it was not implemented with the direct objective of pollution control, this experience is recounted here since it provides one of the most successful examples of the management of externalities with economic instruments and provides valuable lessens for environmental pollution control as well. Since that time, an awareness of environmental problems has grown and the positive effects of the ALS on air pollution are now recognized. Revenue generation was not a key motivation even though the scheme has had a considerable surplus of income over expenditures.
The ALS was introduced as part of a package which also included park-and-ride scheme with 13 peripheral car parks, new shuttle services between the car parks and the central business district, bus improvement, a 67 kilometer rapid transit system, a new by-pass and traffic management measures. The ALS covers 720 hectares, and vehicles entering the covered area need to buy a daily (S$ 3) or monthly (S$ 60) windscreen ticket displayed. The scheme has been refined various times.
The ALS has been accompanied by economic measures to restrain the ownership of automobiles by taxing of car ownership and usage. Additional registration taxes reach 150% of car value. Taxes are discounted for the scrapping of older more polluting cars. Since 1990, a "Certificate of Entitlement" is required for the purchase of a car. The number of new certificates is determined by the Government which auctions off these certificates to prospective purchasers.
The ALS scheme and related policies had a dramatic impact almost immediately and continued to be very effective. The model split has changed from 46% in public transport and 43% in private car in 1974 to 63% in public transport and 22% in private car in 1988. (OECD 1993d)
Indirect environmental charges have been mostly applied to mobile sources through the taxing of car ownership and usage. The obvious rationale is that there are many mobile sources, and continuous individual monitoring of emissions is difficult and expensive. While vehicle and gasoline taxes could theoretically be differentiated according to emission levels (and this is increasingly becoming a real possibility), in practice taxation is applied not to emissions but to the cars and gasoline, differentiated by type of equipment or type of gasoline. Even though the taxing of car ownership and usage has often be introduced with non- pollution related objectives in mind (congestion or revenue), their important effect on the pollution intensity of a city's transport system is being recognized increasingly.
In Mexico, the pricing differential between leaded and unleaded fuel has been reduced in favor of unleaded gas, and a gasoline tax has increased fuels prices to 50% above world market prices. Even though the tax was not aimed at pollution in particular, it perfectly served pollution control objectives. There are a large number of other countries which have introduced a tax differential between leaded and unleaded gasoline (including Taiwan, Hungary and others).
In 1985, a program of tax exemption for compressed natural gas (CNG) was introduced in Argentina to promote the replacement of petroleum fuels by CNG. The program was quickly adopted by mid-sized trucks and taxis. By the end of 1994, 210,000 vehicles in the Metropolitan Area of Buenos Aires had been converted to CNG usage. Of the nearly 40,000 official taxis, about 65% use CNG. To date, the CNG program has led to the substitution of about 12% of the diesel use in Buenos Aires, which should correspond to a 6% reduction in particulate emissions. Since the conversion of vehicles to CNG usage in Buenos Aires is based on a relatively simple technology, NOx emissions are reportedly high, and significant NOx emission increases may have resulted from the shift from diesel to CNG. (World Bank 1995c)
Several countries apply tax exemption or other subsidies as an incentive to invest in pollution control. Taiwan, for example, exempts all pollution control equipment from import duties, applies a direct income tax deduction for pollution control investments and allows accelerated depreciation of these investments (OECD 1993b). Low interest loan programs or financing from an earmarked environmental fund for pollution control programs have been introduced in many countries (Taiwan, Brazil, Korea, Thailand, Indonesia, Poland).
The Mexican government set up several trust funds in the 1970s to provide long-term financing sources for priority sectors. FONEI, established in 1971 and administered by the Central Bank, provided financing for industrial projects through the entire network of the Mexican investment and multi-purpose Banks. In the 1980s, FONEI started to promote lending for pollution control and became the principal trust fund responsible for pollution control. FONEI's pollution control loans were eligible for 2-5 percentage points below market interest rates. The majority of FONEI's pollution control loans supported air pollution abatement investments. After FONEI was abolished, the National Finance Institution took over lending for pollution abatement investments (Lovei 1994b).
Several World Bank projects have been designed to channel resources through financial intermediaries for industrial pollution abatement investments in Brazil. Basic environmental regulations, standards, and permit systems had been put in place before lending started. However, monitoring and enforcement mechanisms operated inefficiently, and, as a result, these credit lines through banks played only a marginal role in environmental financing. The success of industrial pollution abatement credit programs strongly depended on the environmental management capabilities and the enforcement of environmental regulations. Inadequate enforcement was one of the main reasons why the credit lines for pollution control investments remained underutilized during the first Sao Paulo project. Improved monitoring and enforcement also reduced the need for subsidies extended through lower than market interest rates (Lovei 1994b).
Economic incentives have been used as "carrots" to induce concerned agencies to respond more positively to water quality concerns. Low interest loans from an Environment Fund are made available to local administrations and private businesses which are required to set up treatment facilities. Other promotional measures include the reduction of import duties to no greater than 10% for equipment used for any treatment. However, between 1984 and 1989, only US$ 5.1 million worth of waste-water treatment equipment was imported under these incentives. (OECD 1993h)
Fiscal incentives for pollution abatement financing have been widely used in India: (a) Soft loans are provided for technology upgrade by the Industrial Development Bank of India and the Industrial Credit and Investment Corporation of India. (b) Depreciation allowances are higher (50 rather than 30 percent) for certain pollution control equipment. (c) Excise taxes are reduced (from 15 to 5 percent( for certain pollution control equipment. (d) The maximum customs duty is reduced from 85 to 40 percent for all pollution control equipment. (e) The income tax exemption is extended to private contributions to environmental organizations. These incentives, however, have only a limited effect on stimulating pollution control measures. Entrepreneurs try to maximize government subsidies and minimize their individual contributions to environmental investments even if that requires the delaying of planned investments (World Bank, 1991).
Several countries have implemented deposit refund schemes for plastic bottles. Taiwan has successfully implemented a deposit refund scheme for PET bottles achieving a return rate of almost 80%. The use of presumptive charges for typical pollution problems is not yet widespread.
To control hazardous wastes from industrial sources, the Thailand Development Research Institute has proposed the creation of an autonomous Industrial Environment Fund. In line with the polluter-pays-principle, the fund would be financed from waste charges that would first be estimated for each industry and later verified by environmental auditing. The charge would be set at a level that covers the cost of transport, treatment, and disposal of hazardous wastes and provides a margin for running the program. A charge of 1,000 baht per ton on the 600,000 tons of industrial hazardous wastes projected for 1991 would raise 600 million baht. This is 0.3% of the GDP originating in the 17,000 industrial plants in Thailand that generate hazardous wastes, or 1.5% of net profits.
The proceeds would be used to establish and operate central treatment and disposal facilities for hazardous wastes collected from factories. Factories would deposit with the fund their waste charges for the entire year. Plants that attained lower waste per unit of output as verified by accredited private environmental auditing firms, would then be eligible for rebates. The operation of the treatment and disposal facilities would be contracted out to private waste management firms through competitive bidding. (World Bank 1992a)
With the exception of the particulate compensation regime for Santiago, Chile, there has been no documented experience of the use of trading systems as pollution control instrument at a significant scale in developing countries until now. In the Santiago Region, industrial emission sources of PM-10 have been addressed by a Decree (No. 04/1992) which provides the policy framework for a transferable pollution rights approach. This decree combines an emission standard with a compensation system to be phased in until 1997. The initial distribution of emission rights is based on gas flows as reported in March 1992. Starting in 1994, any new emission sources or sources which want to exceed the existing standard require compensation from an existing source. However, no trades were reported during that year.
The PM-10 compensation system in Santiago is an innovation at an early stage of implementation. This system of tradable pollution permits has the potential to achieve a desirable level of pollution reduction at the least possible cost. However, attention needs to be paid to the implementation details of this system. In particular, a situation might arise in which implicit emission permits are issued in excess of current emission levels, or trade in permits leads to a migration of emission sources into areas with greater air pollution problems, such as the downtown area. For a permit trading system to work, a stable institutional framework is required. As long as enforcement is weak and there is still uncertainty about the exact definition of the initial pollution inventory on which trades are based, or the measurements of emissions from different processes, the system is unlikely to work effectively.
Additional uncertainty for the PM-10 compensation system is created by the expectation that natural gas from Argentina will be available in Santiago shortly after the last step in the implementation of the emission compensation system in 1997. The administration had not yet taken a clear position on how it will handle pollution sources that would require compensation only for a short period of time between 1997 and the introduction of natural gas. So far, the system does not allow trades between different groups of emission sources (fixed and mobile sources). (World Bank 1994)
The important role of informal regulation and information disclosure in pollution control is increasing being recognized. Resent research on the behavior of polluting industries in developing countries indicates that there is a significant degree of "voluntary" pollution abatement by industries that is not driven by formal regulation but by informal agreements and community pressure. More research on the implicit economic incentives is ongoing and will hopefully lead to better understanding.
Use of environmental certification to improve environmental awareness and improve consumer information is one approach that several governments have selected ("Green Mark" in Taiwan, "Ozone Seal" in Chile).
In Thailand, a point source standard for water pollution was established allowing environmental monitoring agencies to bring proceedings against polluters. Possible penalties for polluters include imprisonment for up to one year. Individuals and NGOs are also allowed to take legal actions against polluters enhancing significantly the effectiveness of the standard.
Research in Bangladesh shows that information dissemination can play an important role in reducing industrial pollution as the dissemination of pollution records of large companies facilitates bargaining between polluters and the affected parties. (Huq and Wheeler 1992).
Obviously, conditions vary widely from one country to the next, and these conditions require careful consideration when applying the experience of OECD and non-OECD countries to draw lessons for the selection of environmental policy instruments in developing countries. Nevertheless some general lessons can be drawn and are presented in this section.
There are no cookbook approaches to the design of environmental policy instruments that can overcome the implementation difficulties encountered in many developing countries. In particular, economic instruments in general are no panacea for overcoming problems such as the lack of enforcement of environmental regulations. Even though economic instruments hold great promises for achieving more cost effective environmental regulation, their design and applicability needs to be evaluated with care in every detailed case. Driven by the criteria of (a) economic efficiency, (b) administrative costs, and (c) political acceptability, this section illustrates the considerations that should guide the selection of the appropriate policy instrument in a specific country for a specific pollution problem.
Interestingly, many of the concerns raised against the use of economic instruments in the environmental policy of developing countries are not really specifically directed at economic instruments but can be characterized, more generally, as the obstacles that any more effective environmental policy would have to overcome and that explain the failure of environmental policies in general in many developing countries (OECD 1993e):
Resistance against stricter environmental controls based on the belief that this would hurt economic growth and industrial competitiveness;
the inertia of the political bureaucracy that has not yet internalized the need to react to the increasing concern for pollution control and the lobbying of interest groups which would be economically hurt by the implementation of pollution control policies;
the lack of information on the damages of pollution in contrast to more readily available information on the costs of pollution control linked with lack of public participation in decision making;
the short-term planning horizon of political systems that have no interest in benefits that accrue only after the tenure of the current administration;
the underbudgeting of environmental institutions;
the poor legal framework for environmental management, which involves fragmentation and overlap in the regulation of pollution and the assignment of regulatory responsibility linked with lack of coordination between sectoral agencies with environmental agencies.
Other concerns are directed specifically against the use of economic instruments. These concerns include the following and will be addressed in detail in the following discussions:
The requirement for strong environmental or sectoral agency;
the loss of control by the environmental agency in the case of instruments that require coordination with or implementation by other agency such as the Economics or Finance Ministry in the case of environmental taxes;
resistance by polluters who expect more bargaining power under a system of incomplete enforcement of non-enforced standards compared to a system of economic incentives.
The choosing of the right policy instrument rarely is a black and white choice between economic instruments and CAC regulation. In practice, a combination of both approach will often be selected. Moreover, in the comparison and the design of different policy instruments, details matter a great deal. While there is no doubt about the theoretical superiority of the perfectly designed and implemented economic instrument over the current, highly imperfect, CAC practice, a poorly designed economic instrument may not fare as well in comparison.
Arguably, the first step in any program of economic incentive-based environmental policy should be the screening for subsidies for environmentally harmful activities. Because of the attractiveness of these measures and the prevalence of such subsidies in many developing countries, this analysis should precede policies to internalize external costs. There should be little controversy about the implementation of typical win-win policies through changes in economic incentives. The main policy of this kind is the removal of any subsidies on goods associated with pollution such as energy, agrochemicals and water usage. The removal of subsidies improves economic efficiency, increases revenues, and improves the environment. Social objectives pursued with subsidies can be more effectively achieved by targeted cash subsidies and lump-sum transfers.
In developing countries, resources available to the government and the private sector for pollution control are generally limited. This tighter resource constraint should mean that cost-effectiveness considerations are even more important in developing countries which can ill afford the very costly detour of exclusive CAC regulation made by many OECD countries over the last decades.
The key consideration that motivates the use of economic instruments is cost-effectiveness. Most recent empirical work on pollution control costs has consistently shown, that the instruments chosen and the design of these instruments for achieving a desired level of ambient quality have a much greater impact on total costs of pollution control than the chosen level of ambient quality. Tietenberg (1988) reviews nine studies where applied CACs and EIs are calibrated to reach the same ambient level. For seven of these studies, the ratio of EI costs to CAC costs is 1/4 or lower. For two studies, the ratio is even 1/14 or lower. Even if these potential cost savings exaggerate real cost savings because CAC instruments and their cost-effectiveness can be improved, and EI's cost effectiveness is limited by real-life complications and design mistakes, the case for EIs based on cost-effectiveness remains very strong.
The available evidence also suggests that, at the relatively low levels of pollution control prevailing in most developing countries, the transparency and predictability of environmental policy as well as the policy instruments chosen are much more important determinants of the total costs of pollution control than the chosen level of ambient environmental quality. Companies in developing countries more often complain about unpredictable environmental policies, erratic and arbitrary enforcement, and institutional confusion than about the burden imposed on them by a consistent policy regime with very tough objectives.
The effectiveness of different economic instruments for environmental policy clearly depends on the economic, political, legal, and institutional framework of a given country. Relatively little can be said about general country conditions that favor either CAC or EI regulation. In contrast, the effectiveness of each individual policy instruments needs to be evaluated in the context of a specific country. The following comments suggest a few important considerations in this respect.
Many EIs will work better in an economic environment in which polluters are used to making economic choices. A permit trading system will be implemented more easily in countries that have experience with formal trading systems in other sectors (active stock market, auctioning systems for import or export quota).
The efficiency and cost-effectiveness of economic instruments is based on the assumption of profit maximizing polluters, which reduce pollution if there is money to be made. Therefore, economic incentives often fail in an environment in which polluters follow objectives other than profit maximization. For example, companies in the former socialist regimes of Eastern Europe had production goals and where often quite insensitive to fines or charges on pollution since incentives were driven by output rather than profit. Similar experience has been made with public enterprises in market economies where these enterprises are not run according to commercial management practices.
When polluters operate in non competitive markets (for example as monopoly provider), polluters may be less sensitive to economic instruments since they can pass on cost increases to consumers. This concern is particularly relevant to developing countries where markets are often not competitive. These situations need to be evaluated in each individual case since the optimal environmental instrument depends on other regulations imposed on the polluter (for example, price regulation of monopoly providers).
An environmental liability regime depends heavily on an effective legal system. Where access to the court system is limited, courts lack experience in environmental matters, costs of litigation are high, and the quality of the legal process is in doubt, a major reliance on environmental liability is unlikely to be the right emphasis. An analysis of liability cases in non-environmental areas may give a good indication of whether environmental liability can be expected to be an effective instrument in a given country. In most developing countries this is unlikely to be the case. The legal system may also be a factor for the effectiveness of environmental charges and trading systems if the enforcement of the system requires involvement of the courts. Where possible, instruments should be designed such that the government agencies are given sufficient enforcement powers to make the systems work.
The functioning of economic instruments depends on the economic environment in which decision makers are placed. In particular, economic stability, or lack thereof, may influence the responsiveness of polluters to economic instruments. Economic instability often leads to a short-term focus and a higher implicit discount rate of decision makers. With such higher discount rate, abatement investments may not be made if the potential saving in pollution charges would occur only at some time in the future.
Economic, legal and institutional stability is also required to generate the trust of polluters that the implemented policies are sustainable and that is required to make polluters invest money according to the desired economic incentives. Polluters are unlikely to invest in pollution control or to buy pollution permits if they are not certain that in the future the charges will be levied and the permits honored. If uncertainty about the future control regime is large, polluters are more likely to pay the charges than to invest in reduced pollution levels.
Another common example of how macroeconomic conditions influence environmental policy has been inflation. Inflation has often rendered environmental charges ineffective after some time unless there was an automatic mechanisms implemented for maintaining the real value of these charges.
Depending on circumstances in a specific country, economic instruments such as charges may be an administratively simple solution if they can be added to already exiting charges or taxes such as pollution based charges as an addition to existing sewage charges, differentiation of existing gasoline charges by fuel type, and differentiation of vehicle registration fees by emission levels.
Subsidies that are sufficiently high to induce changes in the behavior of polluters are fiscally unsustainable in countries with significant budget deficits. Moreover, the misuse of subsidies is common in countries that do not have in place very tight monitoring and enforcement systems.
If environmental agencies had perfect information available, they could develop sophisticated CAC regulation that would try to mimic the outcome of the application of EIs and be, at least statically, efficient. In developing countries, the information base on which the design of pollution control policies is based is often very weak. Emission inventories are often incomplete or non-existent. The lack of information in the hands of the government highlights the potential benefits from the use of EI since CAC of a government with weak information is likely to be more inefficient than CAC instrument design based on good information. The advantage of EIs is that detailed knowledge of the abatement costs of each polluter by the regulator are not required for efficiency. Since information is likely to be less available in developing countries, the advantage of EIs over CACs should weigh greater in developing countries.
Under certainty, the effects of pollution charges and a tradable permit system would be equivalent. The choice between both, therefore, depends on the nature of the uncertainty about pollution damage and abatement costs. Pollution charges fix a ceiling for the maximum cost imposed on polluters by environmental regulation since the total cost will be less or equal to current emissions times the charge rate. On the other hand, if the regulator does not know the marginal abatement cost curves, the results in terms of pollution reduction are uncertain. For example, if a current standard is replaced by a low charge, pollution levels could increase. Thus, when concern about the total cost to polluters overrides concerns about the actual pollution levels, a charge (that can later be adjusted if the result on emissions is undesirable) may be more appropriate.
An emission trading permit system fixes the total amount of emissions but leaves uncertainty about the total cost of the system to the polluters. Thus, when the optimal or desirable level of pollution control is known, when irreversible environmental damages or threshold effects are suspected above a certain level of pollution, or when there are other reasons to limit emissions to a specific level, a trading system would be preferable over charges.
Even though the establishment of a well running system of EIs is an institutionally challenging task, the universal and effecive implementation of comprehensive CAC regulations would be even more daunting and is therefore almost never accomplished in a satisfactory manner. Since EIs lead to the revalation of important information, they are generally less complex than an equally efficient system of CAC regulations.
In cases where monitoring of individual emissions is difficult or costly (many small sources, or informal sector sources that cannot be regulated), indirect instruments that address pollution problems at the level of the input or output to pollution generating processes may be attractive. The stronger the link between the taxed input/output and the amount of pollution, the more effective is the indirect instrument. Especially fuel taxes, have proved to be an effective instrument.
Clearly, indirect instruments have the potential to distort incentives. For example, the experience with subsidies for pollution control equipment (tax or tariff exemptions, directed credit) has been mixed at best. These instruments favor capital intensive end-of-pipe solutions to pollution control that are often not cost- effective. Also, they do not by themselves ensure the continuous operation of the subsidized equipment. This is a serious problem in cases were operation costs of abatement equipment are high. The targeting of subsidies has also proved to be a difficult problem that casts doubt on the wisdom of wide application of fiscal support for pollution control investments.
The difficulty of monitoring emissions often leads regulatory agencies to focus on "visible solutions" such as end of pipe equipment installation. These solutions are not always cost effective, and more effort is needed to change incentives to the reduction of emissions through operational modifications and process changes rather than investment in visible pollution control equipment alone.
Both, pollution charges and trading systems, can be used to generate the revenues to cover the costs of administering the system. The revenue-generating function of a system of charges is obvious. The trading system can generate revenues through the auctioning of permits or a combination of the permits with a small charge to cover the system costs. Such a charge does not reduce the efficiency of a trading system. If the political will for effective enforcement exists, key progress can be achieved by earmarking a share of charges or penalties for the monitoring and enforcement agencies. Thus, these agencies can be funded to fulfill their mandate and have an increased incentive to do so diligently.
Ultimately, enforcement requires not only technical expertise but also political backing. If environmental agencies don't have the political strength to enforce regulations that are on the books, technical improvements alone will be insufficient. However, clever choice of instruments can make the job of enforcing environmental regulations much easier. In many developing countries, government environment institutions are weak. This means that enforcement of any government regulation (CAC or EI) is often lacking. Therefore, the use of instruments that are simple and have elements of self-enforcement is more important in those countries with weak enforcement agencies.
Neither CAC nor economic instruments can work without enforcement. If enforcement is selectively applied to formal enterprises only, both types of instruments imply the danger that formal enterprises are pushed into informality and may actually increase emissions. Regulation of any nature (CAC or EI) should only be imposed if the political will and commitment exists to enforce the regulation. Otherwise, selectively enforced regulations will reduce transparency and credibility of the system and encourage rent-seeking behavior. Arguably, a system with selective and relatively lenient regulations that are systematically enforced is a better basis for future strengthening than a system with complete coverage and tight regulations but without systematic enforcement.
In some cases, decentralization or privatization of monitoring and enforcement can improve the situation by giving the responsibilities to entities with better information than central government environment agencies. A typical example is a sewage company which is accountable to the regulator for the effluents from its sewage network and has been given the authority by regulators to enforce rules for discharges into the sewers and possibly fine violators.
Clearly, the design of instruments can contribute to making enforcement easier. Due to the information asymmetry between regulator and polluter (polluters know more about their emissions than the regulators), the social cost of enforcement can be reduced by reversing the burden of proof and putting it on the polluter rather than the regulator. Presumptive charges can be applied based on high estimates of pollution from a given industry. It is then up to the polluter to demonstrate actual discharges below the presumed levels in order to claim a refund. The related deposit-refund systems could be applied much more widely to overcome enforcement and monitoring problems. The experience with self-reporting of emissions has generally been good if the system includes spot-checks and stiff penalties for false emission reports.
The details of the design of EIs matter a great deal. If the spatial dimension of air pollution is neglected in an emissions permit trading program and emissions in already heavily polluted areas rise, the trading system can make a bad situation worse. A well intentioned charge on solid or hazardous waste can lead to the uncontrolled, illegal dumping of substances, which may induce a deterioration rather than improvement of the environmental situation. In order to not discredit EIs, great care should be taken to avoid these and other design errors.
Instrument design should focus on simple implementation procedures and clear institutional assignment of responsibility. Details of the regulatory regime often determine the assignment of responsibilities between agencies. Care should be taken to equip environmental agencies with sufficient authority to implement or at least effectively coordinate policies that normally require actions by other government agencies, such as the Ministry of Finance for taxation.
The administrative costs of any system for pollution control are of obvious concern to governments with limited budget resources. In principle, the administrative costs of an EI system should be less than that of an equally effective CAC system with much higher information requirements. On the other hand, a permit trading system poses the important question of transactions costs. These transactions costs depend mostly on the detailed design of the system and facilities for information exchange between potential buyers and sellers of pollution permits. Significant experiences are available from OECD countries and summarized elsewhere (i.e. Tietenberg 1995). Similarly, care needs to be taken that trading systems are designed for a sufficient number of participants to ensure competitive price formation for emission permits.
The political acceptability of any pollution control regime depends, among other factors, on its distributional effects. Due to poor design, lack of political acceptability has often been an obstacle for effective environmental policy, whether through effective enforcement of CAC regulation or sufficiently high charges. At least theoretically, most distributional effects of environmental policy can be avoided through the clever design of economic instruments.
Emission charges have to be sufficiently high to induce behavior changes. Especially in cases where the supply of pollution abatement is very inelastic, charges have to be high. This will mean that the charge may create a real burden for the polluter, and the government will collect revenues much higher than the costs of administering the system. If so desired, some or all of the revenues collected from these charges can be returned to the polluters through rebates on other taxes leaving polluters in aggregate unharmed. In order to not offset the incentive effect, tax rebates must not be related to pollution abatement of the individual polluter. Instead, tax rebates could be granted based on the average revenues from pollution charges redistributed by a formula based on output, value added or some other non-pollution related formula.
Emission permit trading schemes have implicit the separation between the efficiency effect of the tradability and the equity effect of the initial allocation of permits. Emission permit trading systems can reduce distributional changes by allocating the initial rights free of charge in proportion to past pollution quantities. In the case of inelastic supply of pollution abatement, a trading system may be more attractive since it avoids the associated high pollution charges.
For political acceptability, it will be important that a system of economic instruments be simple, transparent, and easily understandable. Especially, in designing first steps for the introduction of EIs, it would be advisable to select an application that is simple enough to allow a positive demonstration effect. For example, for the introduction of a first emission permit trading program, emissions of ozone depletion substances may be more suitable since the emission sources are more homogenous and enforcement is simpler than particulate or gas emissions which originate in heterogeneous sources and are much more difficult to monitor.
New environmental agencies are often having a hard time acquiring the trust and the confidence of the private sector and other government agencies. A key strategy to build trust for a weak institution is to focus the limited resources on priority policy areas but deal with these areas well. In the long term, it is more effective to start with limited regulations for priority areas but implement and enforce these regulations rigorously than to pretend action on a larger number of issues but fail in the implementation and enforcement. Environmental agencies can also improve their political standing by explicitly including cost- effectiveness considerations in the design of their programs. If environmental agencies take the cost of compliance with their regulations seriously into account, they are more likely to earn the trust of the business community, Economics Ministries and other agencies, which will ultimately provide them with the political clout to implement and enforce their programs.
Political resistance against the more widespread use of economic instruments may also come from the environmental regulators themselves. Agencies which are mainly staffed by environmental engineers often prefer to get involved themselves in technical choices at the plant level and may not appreciate the merits of an incentive based system of EIs. Ultimately, it is important to recognize that EIs can raise the revenues required for better overall environmental management and improvements in the environmental agencies themselves.
The following table summarizes some of the main requirements for specific instruments of pollution control policy:
Pollution Charges fix aggregate cost of emission control.
Profit maximizing polluters
Source monitoring and enforcement
Agreement on maximum aggregate cost of pollution control
Tradable Pollution Permits fix quantity of aggregate emissions.
Profit maximizing polluters
Source monitoring and enforcement
Experience with formal trading/auctioning systems
Agreement on maximum aggregate emissions
Liability Regime establishes ex-post liability for environmental damages.
Effective and accessible legal system
Good information on pollution effects and damages
Indirect Charges modify price of input/output of polluting process.
Knowledge of a relatively fixed relationship between input/output and pollution
Efficient CAC Standards fix emissions from each individual source.
Source monitoring and enforcement
Agreement on maximum aggregate emissions
Complete information on abatement cost curves by each polluter
Averchenkov. A. et.al. 1994. The System of Environmental Funds in the Russian Federation. Paper presented at the OECD conference on Environmental Funds, St. Petersburg, Oct 27-28.
Brandon, Carter and Ramesh Ramankutty. 1993. Toward an Environmental Strategy for Asia. World Bank Discussion Paper No. 224. Washington DC.
Eskeland, Gunnar and Emmanuel Jimenez. 1991a. Curbing Pollution in Developing Countries. In: Finance & Development: March, pp. 21-24.
Eskeland, Gunnar and Emmanuel Jimenez. 1991b. Choosing Policy Instruments for Pollution Control: A Review. Policy Research Working Paper WPS No. 624. Washington DC.
Eskeland, Gunnar and Shantayanan Devarajan. 1994. Taxing Bads by Taxing Goods: Pragmatic Approaches to Environmental Protection. World Bank. mimeo.
Estache, Antonio and Taylor. 1992. Water Pollution Control: Which Level of Government Should Do What in Brazil? World Bank Policy Research Working Paper. Washington DC.
Huq, Mainul and David Wheeler. 1992. Pollution Reduction without Formal Regulation: Evidence from Bangladesh. The World Bank, mimeo.
Liebenthal, Andres. 1995. Economic Instruments for Pollution Management: An Overview of the International Experience with Implications for Brazil. World Bank. mimeo
Lovei, Magda. 1994a. Environmental Financing: The Experience of OECD Countries and Implications for Transition Economies. World Bank. mimeo.
Lovei, Magda. 1994b. Pollution Abatement Financing: Theory and Practice. World Bank. mimeo.
Miglino, L.C.P. and J.J. Harrington. 1984. The Impacts of Tariffs on the Generation of Industrial Effluents (in Portuguese). Revista DAE, Vol. 44 No. 138, Sao Paulo.
NAPA. 1994. The Environment Goes to Market: The Implementation of Economic Incentives for Pollution Control. National Academy of Public Administration. Washington DC.
OECD. 1993a. Integrating Environment and Economics: The Role of Economic Instruments. Group on Economic and Environment Policy Integration. mimeo.
OECD. 1993b. The Use of Economic Instruments in Environmental Protection - The Taiwan Experience. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Tin-Bai Pan. mimeo.
OECD. 1993c. Use of Economic Instruments in Environmental Policy in Hong Kong. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document. mimeo.
OECD. 1993d. The Singapore Area Licensing Scheme. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Keith Buchan. mimeo.
OECD. 1993e. The Experiences of Brazil, Mexico, Chile, and Argentina in the Use of Economic Instruments in Environmental Policy. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Sergio Margulis. mimeo.
OECD. 1993f. The Standard cum Charge Approach in Environmental Policy: The Malaysian Experience. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Khalid Abdul Rahmin. mimeo.
OECD. 1993g. The Use of Economic Instruments for Environmental Protection in Korea: Performances and Difficulties. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Hosaeng Rhee. mimeo.
OECD. 1993h. Economic Instruments for Water Resource Management in Thailand. Informal Workshop on the Use of Economic Instruments in Environmental Policies. Room Document prepared by Mingsarn Kaosa-ard. mimeo.
Ruitenbeek, Jack, Surna Djajadiningrat and Cynthia Cartier. 1994. Macroeconomic Policies and Economic Instruments for Environmental Management in Indonesia. Environmental Management Development in Indonesia Project (EMDI).
Sánchez Triana, Ernesto and Eduardo Uribe Botero. 1994. Instrumentos Económicos para el Control de la Contaminación Industrial. in: Planeación y Desarrollo, XXV, No. 1, Bogota.
Seroa da Motta, R. 1994. Uso de Instrumentos Economicos para Gestao Ambiental: Teoria e Práctica no Brasil. mimeo.
Tietenberg, Tom. 1988. Environmental and Natural Resource Economics. Harper Collins.
Tietenberg, Tom. 1995. Applying Emissions Trading to the Control of Air Pollution in Santiago: Lessons from the US Experience. Colby College. mimeo.
Tlaiye, Laura and Dan Biller 1994. Successful Environmental Institutions: Lessons from Colombia and Curitiba, Brazil. LATEN Dissemination Note #12, The World Bank.
World Bank. 1991. India: Industrial Pollution Control Project. Report No. 9347-IN.
World Bank. 1992a. World Development Report 1992: Development and the Environment. Washington DC.
World Bank. 1992b. China: Environmental Strategy Paper. Report No. 9669-CHA.
World Bank. 1993. Energy Efficiency and Conservation in the Developing World. A World Bank Policy Paper. Washington DC.
World Bank. 1994. Chile: Managing Environmental Problems - Economic Analysis of Selected Issues. Report No. 13061-CH.
World Bank. 1995a. National Environmental Strategies and Action Plans: Key Elements and Best Practice. Washington DC. mimeo.
World Bank. 1995b. World Development Indicators 1995. Washington DC.
World Bank. 1995c. Argentina: Managing Environmental Pollution - Issues and Options. Draft Report No. 14070.
Zylicz, Tomasz. 1994. Taxation and Environment in Poland. In: Taxation and the Environment in European Economies in Transition. OECD, Paris.