Curbing Pollution in Developing Countries

International Monetary Fund. External Relations Dept.
Published Date:
January 1991
  • ShareShare
Show Summary Details

Pollution of the air, land, and sea poses serious risks to present and future generations, but policies aimed at controlling pollution—especially in developing countries—are far from adequate and could be improved at a low cost. Those that are in place tend to be either poorly designed or not enforced. General economic policies often exacerbate the problems, failing to take into account possible effects on the environment.

As developing countries begin to redesign policies to control pollution more effectively, they will need to find ways to minimize possible adverse effects on other critical objectives—-such as growth, revenue raising, and equity. They will also have to keep in mind important administrative, technological, and institutional constraints, drawing heavily on a wide array of fiscal tools to complement more traditional pollution control instruments.

The need for reform

Government intervention is generally needed to prevent or alleviate pollution, because of a fundamental failure of the market to take into account the interests of those that are being hurt. Since pollution usually affects the well-being of many individuals who have not had any role in the activities causing it, government action should be aimed at making polluters behave as if they have the interests of the victims at heart. This can be accomplished either through regulatory or price intervention. Possibilities include a tax on emissions, a limit on the amount of pollution, subsidies to cleaner alternatives, and, at times, the assignment and enforcement of property rights (see box).

In some developing countries today, policies to control pollution levels do not even exist. For example, imports of heavily polluting used cars, although high in number, are often not subject to any emission constraints. For most, however, these policies are in place—in fact, the standards for individual polluters and environmental quality are frequently similar to those in Europe and the United States—but they have not been effective. One reason, as noted in the World Bank’s Annual Report on the Environment 1990, is that monitoring, enforcement, and regulatory capacities have tended to be weak in these countries. In Mexico, for instance, the influence of regulations has been limited by the resources of the enforcement agency and the low level of fines. In Colombia, laws have included formulas for calculating a tax on discharges to water, but no apparatus has been in place to monitor and bill polluters. In India, inefficient legal processes have reduced the disincentive effects of lawsuits against polluters. Even in developed countries, subsidized and mandated pollution control equipment has often been ineffective. For example, a recent study of monitoring and enforcement of pollution regulation in the United States found that during the 1970s and 1980s, much more public attention was focused on getting pollution equipment in place than ensuring its proper maintenance and use.

Another reason is that pollution control policies often are not well designed, meaning that they would have been both weak and un-necessarily costly even if they had been implemented vigorously. Emission regulations, the most popular form of intervention, frequently provide no incentives for polluters to choose least-cost abatement options. In Brazil, for example, studies show that some industrial polluters could reduce emissions at much lower costs than others. Incentives—as opposed to rigid regulations—would have been able to exploit these differences by allowing polluters flexibility in finding where and how pollution can be reduced at the lowest cost. Even in industrial countries, surprisingly little use has been made of incentives, despite many studies demonstrating that savings would often be in the range of 50 to 90 percent of the costs of existing controls.

Often exacerbating the deficiencies of environmental policies is the fact that general economic policies may have environmental effects, and these are not considered when the economic policies are formed. Subsidized charges for energy (particularly fossil fuels), water, pesticides, and fertilizers, as well as selective investment incentives to industries, may be costly ways of reaching other objectives if they contribute simultaneously to pollution problems. Economic interventions must take into account whether the stimulated activities are good or bad for the environment.

Possible approaches to reform

What options are open to governments? Typically, there is a much wider range of policy instruments at their disposal than those now being systematically exploited (see table). Traditionally, analysts have proposed “direct” instruments, which attempt to correct directly for the imposition of pollution on others by, for example, levying charges, issuing permits, or setting standards based on emissions. When such constraints are source specific and nontradable, they are termed command and control, since they mandate where and how to abate. They contrast with market-based instruments, which provide abatement incentives to polluters, who can then choose among options. A long-standing theoretical result—which has also been confirmed by empirical analysis—is that taxes or tradable quotas on emissions will achieve a given emission reduction at the lowest possible cost. Abatement will be cheap if polluters face an emission tax, because only polluters with low abatement costs will choose to reduce emissions, while those with higher costs will prefer to pay the charge.

In addition, there are “indirect” instruments—those that are not directly related to emissions, but that nonetheless have important environmental effects, sometimes unintentionally. Public finance policies, through their impact on relative prices, will often have a strong impact on pollution. Other policies, such as those related to macro or trade balances are also potentially important.

Minimizing trade-offs with other objectives. In assessing the relative merits and design of these policy instruments, developing countries should strive to avoid, or at least minimize, potential trade-offs with other objectives, such as economic growth, revenue mobilization, and equity. This will also facilitate adoption and implementation.

Some pollution control policies may actually complement efforts to meet growth and narrowly defined efficiency objectives. For example, reducing energy subsidies to industries may lower pollution and enhance efficiency by subjecting firms to fair competition and forcing them to economize in the use of energy. Subsidy reductions, in turn, may also contribute to net public revenue, and employment may be generated if energy and labor are substitutes. Trade-offs are more likely to be avoidable if resource allocation has been extensively distorted due to past policies—the situation in Eastern Europe and the Union of Soviet Socialist Republics. Heavily subsidized prices for certain key polluting commodities could have encouraged overutilization and waste. A first step may then be to remove subsidies and move to marginal cost pricing for such items as water supply, electricity, fossil fuels, pesticides, and fertilizers.

A taxonomy of pollcy instrumen
Direct instrumentsIndirect instruments
Market-based incentives:Effluent charges, tradable permits, deposit refund systemsInput/output taxes and subsidies, subsidies to substitutes and to abatement inputs
Command and control:Emission regulations (source specific, nontransferable quotas)Regulation of equipment, processes, input and output
Government production or expenditures:Purification, cleanup, waste disposal, enforcement and agency expendituresTechnological development
Note: A direct instrument addresses the level of damages or emissions directly, whereas indirect instruments work via other variables.
Note: A direct instrument addresses the level of damages or emissions directly, whereas indirect instruments work via other variables.

When public budgets are tight, the case is strengthened for environmental taxes (or auctioned tradable permits) rather than regulation. Developing countries often use tax instruments to a point where distortions are very costly—export taxes discourage cash crops, sales taxes distort consumption, and so on. For them, using pollution control instruments that generate revenue would be particularly useful, since this would enable governments to lower other tax rates. While pollution taxes are only a fairly recent innovation and not yet significant in terms of general revenue, a recent study by the Organization for Economic Cooperation and Development shows that environmental taxes have been a significant source of funding selected environmental expenditures in developed countries. The magnitude of the revenue potential of pollution-related taxes in developing countries still has to be seen in practice. But since fuel constitutes 10 to 20 percent of imports for many, there is no doubt that taxation of fuels—which could then be motivated by environmental, as well as fiscal objectives—holds substantial revenue potential. Some countries where intensive pesticide use is subsidized at a rate of 44 percent of retail cost could save hundreds of millions of dollars per year and reduce health risks as well.

The possible trade-offs with equity and efficiency also have to be assessed, especially when the major expenditure items and revenue sources of poor households are concerned. Often, these trade-offs can be avoided altogether. For example, on the production side, many studies of individual industries have estimated that energy and labor are sub-stitutable, indicating that price-induced energy efficiency can increase employment and labor income. On the demand side, many studies indicate that the use of petroleum fuels (except possibly kerosene) and electricity vary greatly with income levels in developing countries, making these potential tax bases for distributive reasons, too. If a policy has adverse consequences for the poor, it can still be attractive if there is scope for compensating measures, such as lowering the price for other essential items.

Adjusting to administrative and technological constraints. These constraints are likely to be different, at least in degree, from those confronted by developed countries. In particular, one must take into account the government’s ability to monitor pollution damage and enforce regulations. For direct instruments to be effective, the regulator has to be both technologically capable and sufficiently accountable to withstand pressure from interested parties, since policy instruments based on environmental monitoring have monetary implications for the polluters.

The theoretical case for market-based instruments (such as effluent charges or tradable emission permits) is very strong, as these exploit information held by the polluters to seek out the cheapest ways to reduce pollution. Under such a regime, polluters facing high costs of pollution abatement would prefer to pay the charges, while those facing lower costs would prefer to avoid the charges by reducing emissions. Command and control measures cost more because the regulator must decide where and how emission reductions should take place. The theoretical case for direct rather than indirect instruments is also strong. Actual monitoring of emissions allows perfect incentives for all actions that can reduce emissions, whereas indirect instruments can only imperfectly mimic such an incentive scheme. For example, a fuel tax can act as an effluent charge in the absence of continuous emission monitoring, but it will only imperfectly mimic a tax on actual individual emissions. In particular, it will fail to induce any actions that can reduce emissions per unit of fuel consumed (such as installation of filters and catalytic converters).

Environmental property rights

The environment could be effectively protected through a reassignment of property rights—giving the rights of clean air and water to consumers and letting them sell as few or as many pollution permits as they like. But one problem with this approach is that without monitoring, enforcement of rights to open access resources such as clean air and water is virtually impossible; polluters will have all rights de facto. Another problem is that since everyone can enjoy the general air quality, each individual will have excessive incentives to sell the rights to polluters. Thus, pollution rights must be sold collectively, not individually.

Under special conditions, polluters and victims can negotiate socially optimal pollution levels without government intervention. According to an argument attributed to economist Ronald Coase, as long as negotiations are not too costly and bargaining works, these levels can be achieved, regardless of the initial allocation of rights. For example, in the Philippines, soil sediments caused by a single logger threatened tourism development in a bay. In such a case, direct negotiation between polluter and victim would have been feasible. Similarly, transborder pollution issues between two sovereign nations, such as acid rain, are often resolved without the involvement of a supranational governing entity. But this solution tends not to be efficient when a large number of people are involved and information about their stakes is imperfect. In Mexico City, air pollution from 30,000 industrial firms and 2.6 million motor vehicles affects 20 million inhabitants.

But in the real world, monitoring each polluting source (and enforcing taxes or regulations related to such monitoring) is often costly or impossible, and thus it might be more effective to use selective taxes or regulations to stimulate pollution control indirectly. Such policies could include subsidies to pollution control equipment and public transport, taxes on polluting inputs, and noneconomic regulatory approaches (e.g., fuel efficiency standards). These indirect instruments can often be implemented without reliance on technically and institutionally vulnerable monitoring and enforcement functions.

The problem, of course, is not only a question of technological feasibility but also whether institutions are strong enough to cope with pressure from special interest groups and corruption. Indirect instruments may thus be more efficient even though, for instance, fuel consumption is only an imperfect indicator of emissions. Moreover, these measures—whether they be taxes on polluting inputs, or take the form of tax concessions and subsidies to low-polluting products and processes—can be administratively easy to carry out if they can be grafted onto existing systems of indirect taxation.

It may turn out that for some activities, emissions and damages are affected mostly by inherently unobservable actions, such as whether combustion machinery is adjusted and used carefully. Under such circumstances, there will often be no good substitute for monitoring emissions. In general, however, policymakers will be able to find appropriate indirect instruments, and the key in their use will be the identification of polluting sectors and instruments affecting their mode of operation. For instance, if household use of energy is found to be a major contributor to pollution (as in Ankara and Beijing), changing the availability or relative prices of alternative energy sources will probably be the major instrument. In addition, some improvement may come from reducing the consumer prices of selective appliances (e.g., improved ovens and stoves).

One weakness of indirect instruments such as fuel taxes or subsidies to less-polluting equipment may be undesired incentive effects and distortions in behavior. A fuel tax, for instance, treats all users of the fuel as if they pollute equally per unit of fuel. The cement industry, which retains practically all the sulfur content of its fuels, would then be taxed at too high a rate, and its use of high-sulfur fuels would be unnecessarily discouraged. The problem could be overcome, however, by instruments that differentiate according to user characteristic, if they are not too costly to administer. In the above example, sulfur taxes paid on fuels could be refunded when regulators are convinced that the shipment was used in cement production.

Taking the institutional dimension into account. A major issue here is jurisdictional—whether it arises because the desired tool for pollution control is national (possibly a gasoline tax) while the problem is strictly local, or because a municipality or state is polluted by activities outside its authority. A good example of the latter would be Paraiba, the river that runs through Rio de Janeiro from São Paulo. A solution would be to give responsibility for water quality to a higher federal authority that would monitor discharges and enforce regulations affecting them. The constitution, however, gives the states this authority, and it remains to be seen whether negotiations between the two states can induce abatement within São Paulo in order to benefit Rio.

Another concern is the ability and willingness of governments to take firm actions when the interests of strong and influential private groups are put in jeopardy. Many analysts argue that industrialized countries have chosen costly environmental regulation rather than more efficient instruments because established industries are favored, while costs are born by unorganized consumers and potential competitors. In developed and developing countries alike, excessive depletion of resources, such as forests, is often explained by the fact that well-placed groups can take advantage of such exploitation, while the benefits from sustainable management would go to others.


In a country facing severe revenue constraints and a weak institutional framework, much of the pollution control work in the early phases should be done by taxing fixed inputs (e.g., combustion machinery) and variable inputs (e.g., fuels), according to the level of expected emissions. These instruments may be more effective than traditional regulatory approaches, which rely heavily on monitoring and enforcement. Moreover, they should contribute to—rather than burden—public budgets. Indeed, studies performed in developed countries clearly show that an extensive regulatory framework brings little relief if the combined effect of monitoring, enforcement, and the size of penalties does not support compliance. An indication that developing countries must deal with the same constraints comes from the mixed experience with efforts to enforce regulations on safety standards and overloading of motor vehicles. The stakes for the involved personnel are very high and the task is immense.

Although we have focused primarily on local or national pollution problems, the same principles apply to transnational issues. When several nations are involved, agreement about emission levels will have to come about in the context of negotiations, with international transfers playing a facilitating role—as is the case with the recently established global fund to help phase out the use of chlorofluorocarbons (CFCs) in developing countries (see article on page 24 in this issue). Each nation, in turn, will then have to find the most cost-effective way of dealing with the problem within its own borders.

Recommendations about how to intervene depend heavily on assumptions regarding behavioral responses, and a common one is the assumption of competitive behavior. This, of course, warrants some caution, especially since in many developing countries, the role of prices and competition is often not clear. For instance, when prices are controlled, or the firms involved are parastatals, this should be taken into account. Another cautionary note is that the benefits and costs of pollution abatement are often uncertain, and policy recommendations may hinge on which type of uncertainty is more important. Carefully thinking through how individuals and firms will respond to policy instruments holds the potential to greatly reduce the costs and enhance the efficiency of intervention, in turn determining the affordability of environmental protection.

Other Resources Citing This Publication