Journal Issue
Share
Article

Mineral Taxation, Market Failure, and the Environment

Author(s):
International Monetary Fund. Research Dept.
Published Date:
January 1993
Share
  • ShareShare
Show Summary Details

The literature on mineral taxation (Hotelling (1931), Burness (1976), Levhari and Liviatan (1977), Fisher (1981), and Heaps (1985)) has concentrated on the effects of these taxes on resource allocation, revenue yield, and stability, conservation, and depletion dates. The general conclusion is that taxes such as a Brown tax, a resource rent, or a pure profits tax are nondistortionary, whereas franchise taxes, a specific tax on output, ad valorem taxes, or property taxes are distortionary.

These studies, with the notable exception of Schulze (1974), however, did not take into account the external effects (or environmental externalities) of mineral extraction. As a result, they did not consider the appropriate corrective taxes required to ensure that mining firms account for the social costs that they may impose on others. Nor have they examined the effects of mineral taxes on the environment.

This paper examines the effects of mineral taxes on the environment in a framework that incorporates current as well as cumulative environmental externalities, and proposes an appropriate corrective tax for extractive firms that generate such externalities. An important finding is that neutral taxes, such as the resource rent tax, need to be combined with an appropriate corrective tax, such as the one proposed in this paper, in order to ensure efficient resource allocation. In the absence of such corrective taxes, neutral taxes may perform worse than some nonneutral taxes because the latter taxes act as proxies that partially offset the impact of environmental externalities on resource allocation. Finally, the paper examines the impact of mineral taxes on issues of environmental concern.

I. Extraction of Mineral Resources and Environmental Degradation

The extraction of mineral resources often involves the use and degradation of environmental assets and, in some cases, can affect sustainable development. When the market fails, in the presence of externalities or monopoly, the environmental degradation associated with mineral extraction leads to inefficient resource allocation. Furthermore, public policies that provide explicit or implicit subsidies to certain inputs or outputs when there are environmental externalities often intensify the externalities. Finally, policymakers have had to consider the sustainability of extraction rates when formulating economic policy. For example, for countries that are dependent on extractive industries for their foreign earnings, devaluation to correct the balance of payments may lead to unsustainable rates of extraction.1

Environmental Externalities

Public concern with environmental degradation from mineral extraction stems from two related sources.2 First, it can arise from the use of privately or publicly owned environmental resources without any external effects. For example, land that could have been used for recreational, agricultural, or other uses often has to be forgone or degraded, and rendered less suitable for other uses. However, to the extent that no external effects are involved, public policy intervention is unwarranted.

Second, public concern over environmental degradation may stem from spillover effects from mineral extraction into the environmental or other assets of other economic agents. For example, in coal and other mining operations, dust particles emanating from the mines can blow into surrounding downwind areas, which are not owned by mining firms. In other cases, acid runoff in processes that use mercury, such as gold mining, can downgrade the quality of water streams. In these cases, not only are privately owned environmental assets degraded by their owners, but the spillovers increase the extent of the degradation.

These environmental externalities depend, among other things, on the type of mineral being extracted, on whether mining is onshore or offshore, and on the methods and technologies of extraction used. Environmental externalities in onshore mineral extraction take the form of surface water pollution, pollution of aquifers, land degradation, and air pollution. During wet seasons, excess water combined with underdeveloped tailings and earth road cuts can create waste streams that are heavy with suspended solids. Turbidity and sediment can turn the water courses of the mined areas into serious environmental problems. In addition, landslides, landslips, or landbursts pose potential hazards for surrounding land and habitats. In dry seasons (and for coal mining in any season) the surrounding downwind area is polluted by sulfur dioxide, arsenic, and dust particles.

The extent of these environmental externalities often depends both on rates of extraction and the cumulative amount already extracted. The extent of road cuts and waste emissions into streams or the air thus depends on the rates of extraction; and the amount of tailings and the risks of landslips or slides depend on cumulative amounts of mineral ore already extracted.3

Monopoly

Market structure can affect the extent and amount of environmental degradation caused by an extractive firm, primarily because it can influence the rate at which minerals are extracted and the cutoff points of extracting resources. For example, if the extent of environmental externalities from mineral extraction depends on rates of extraction or cumulative amounts of resources extracted, then it also tends to depend on the elasticity of demand for minerals as well as the type of extraction costs. In particular, if the elasticity of demand is constant, and extraction costs are zero, then the rate of environmental externalities of a monopolist would be the same as that of competitive firms.4 But in the more realistic cases where the elasticity of demand increases over time, or where extraction costs are positive, the optimal extraction rate for a monopolist during earlier periods (and, hence, the rate of environmental degradation) will tend to be lower than that of firms under competition. In other words, a monopolist pursues a more conservationist policy, which means a lower rate of environmental externalities in earlier periods than under competition. At some point, however, the reverse may be true because a monopolist, like firms under competition, eventually exhausts the resource.5 Therefore, total cumulative environmental externalities are the same under both market conditions.

The situation is more complicated in the more general case where extraction costs depend on cumulative amounts of extraction. In this case the resource is not necessarily exhausted and, therefore, cumulative amounts of environmental damage tend to be lower for a monopolist than for competitive firms.6

Policy Failures

When mining and exploration activities are associated with environmental externalities, some tax or nontax policies may aggravate environmental degradation by encouraging excessive rates of extraction by granting explicit or implicit subsidies. The overall resulting effect on the environment reflects the amount of environmental externalities that would have occurred in the absence of this policy-induced behavior, and the additional amount that is generated by the effects of these policy measures over and above the amount that would have occurred in their absence.

Some countries have provided generous tax incentives for investment in mineral resources, including income tax holidays, unlimited loss carryovers beyond tax holiday periods, exemptions (especially of capital goods) from domestic sales taxes and import duties, and generous depletion and depreciation allowances. Granting tax holidays for investments in mineral sectors serves to lower pretax required rates of return on investment in these sectors, prompting higher rates of investment, and greater rates of mineral extraction and environmental externalities than would be the case without the incentives. Similarly, granting exemptions from sales taxes and import duties on capital goods provides implicit subsidies on equipment, including heavy machinery, which is especially associated with environmental degradation in mining.

Tax laws in a number of countries also provide depletion allowances far in excess of cost recovery and grant more generous investment allowances than provided to other sectors.7 These tax provisions lead to resource misallocation and rates of extraction and, hence, rates of environmental degradation that may be higher than they would have been without the provisions. These tax laws also tend to encourage capital intensity, which, in labor surplus countries, frustrates that use of labor-intensive production that would help to alleviate poverty and reduce the stress on environmental resources.

Sustainability

In recent years interest in, and concern with, sustainability in exploitation of mineral resources have increased. It is therefore important to examine how, and to what extent, tax policies affect sustainability of rates of extraction of these resources. Analysis of sustainability issues is, however, complicated by the large number of definitions of the concept and by the absence of a consensus on an appropriate definition. For example, sustainable growth and development are sometimes defined in terms of nondeclining output, consumption, or utility or in terms of achieving minimum levels of consumption. (See Pezzey (1989) for an exhaustive survey of sustainability definitions.) Sustainability of rates of mineral extraction, which is related to, but not necessarily derived from, sustainability of overall economic growth and development, is often defined in terms of nondeclining resource stocks, or maintaining the stock of capital, defined to include the stocks of man-made and natural resource stocks (see Solow (1974)).

In the case of extraction of mineral resources by an individual country, the issue of sustainability may be somewhat less complicated for two reasons. First, mineral resources contribute to welfare only indirectly as inputs in production of goods and not directly as final consumption goods.8 Therefore, imposing a sustainability constraint on the rate of extraction for an individual country is unnecessary because when the resource is exhausted, the country can always use substitutes or import from other countries. Second, the critical issue for an individual country has less to do with preserving its resources than with ensuring that the competitive rents from the extraction of the resources are invested in a way that maintains the capital stock intact. Viewed in this way, sustainability of mineral resource extraction for an individual country should be addressed as a macroeconomic and investment issue rather than as a tax policy, microeconomic, or sectoral issue.

II. Mineral Resource Extraction, Environmental Externalities, and Corrective Taxes

A number of studies have examined the behavior of a mining firm and the effects of mineral taxation on rates of extraction, recoverable reserves, grade selection, or depletion dates. However, these studies assumed that the firm’s mining activities have no external effects on other economic agents.9 An important exception is Schulze (1974), who incorporated cumulative environmental externalities in his model and concluded that when the amount of resource extracted has negative externalities, and corrective measures are not taken, then competitive markets lead to higher-than-optimal rates of exploitation of resources.

Schulze analyzed the impact of cumulative environmental externalities but ignored current environmental externalities associated with rates of extraction, which are equally or, in some cases, even more significant than cumulative externalities. Indeed, so important are current externalities that much of the static analyses on externalities has focused on them to the exclusion of cumulative externalities (see Baumol and Oates (1988)). In addition, while recognizing the importance of cumulative environmental externalities, Schulze’s analysis ignored the impact of cumulative extraction on the firm’s extraction costs. His analysis is, therefore, appropriate for special cases where extraction costs do not depend on cumulative amounts of extraction, even though environmental externalities do.10 In many other cases, especially in mining, both extraction costs and environmental damage are affected by cumulative amounts extracted.

In this section the effects of both types of environmental externalities discussed above are analyzed under the assumption that a firm’s costs of extraction depend on cumulative extraction. An appropriate Pigovian type of tax is derived that addresses these environmental externalities. This extension is important for two reasons. First, it suggests that the nature of corrective taxes for extractive industries may be different from those of other industries and from those that have been suggested in the literature.11 Second, it provides a framework for evaluating mineral taxes that are implemented primarily for revenue and other fiscal reasons but that may in fact act as proxies for corrective taxes for environmental externalities.

Assume that an extractive firm producing current output q(t) at time t inflicts environmental externalities on other economic agents, the extent of which depends on the rate of output as well as cumulative output from time 0 to time t, so that dx(t)/dt = q(t). Total cumulative output, x(T), at the termination of production cannot exceed x¯, the total stock of mineral reserves in place at time 0. Total production cost, C(q, x), and environmental externalities, D(q,x), are increasing functions of the rate of output and of cumulative output. If the firm sells its output at price P(t), its current profits, after paying an appropriate Pigovian tax to compensate society for environmental externalities that it causes, are Π = PqC(q,x) – D(q,x).

The objective of the firm is to maximize V, the present value of current and future profits net of corrective taxes for environmental externalities, which the government requires the firm to pay, or

subject to

where α(t) = e–rt, r is the rate of interest, q is the control variable, and x is a state variable. Carrying out the maximization and after some manipulation, it can be shown that (see Muzondo (1992))

This equation will be used to examine a number of special cases. First, ignoring all types of environmental externalities and the effects of cumulative output on costs of extraction, equation (4) reduces to

which implies the Hotelling result that marginal profits should increase at the rate of interest. Alternatively, this result can be interpreted as requiring that marginal profits at time t are equal to discounted marginal profit in any future period. Under these assumptions the mine is exhausted, x(T)=x¯, and output at the terminal point is zero, q(T) = 0. These are, however, very restrictive assumptions for real world mining firms.

Second, when extraction costs of the firm increase with cumulative output but extraction does not involve any environmental externalities, equation (4) becomes

As noted by Gordon (1967) and Levhari and Liviatan (1977), equation (6) indicates that when the effects of cumulative output on extraction costs are taken into account, the Hotelling rule that marginal profits must increase at the rate of interest no longer holds. Specifically, when extraction costs increase with cumulative output, Cx > 0, then marginal profits will increase at less than the rate of interest. This result follows from the fact that under these assumptions an increase in current output at time t gives rise to additional extraction costs (represented by the last term on the right-hand side of equation (6)) from time t through to the terminal period, T. In other words, an extractive firm facing a cost function that increases with cumulative output must take into account not only the current costs of extraction, and the fact that its production plans are subject to a resource constraint, but also the effects of current output on extraction costs from the current to the terminal period. Since marginal profits at the terminal time are zero, {(P(T) – Cq(T)} = 0, and positive for all other times, {P(t) – Cq(t) > 0} for all t < T, it follows that marginal profits for an extractive firm decline as the rate of extraction proceeds.

Finally, if there are environmental externalities for which the extractive firm is made to pay, this will affect marginal profits in two ways. First, current environmental externalities will affect the level of marginal profits but not their rate of growth. Second, cumulative environmental externalities will reduce the rate of growth of marginal profits over time. To see this, assume that extractive costs are not affected by cumulative output, but that there are negative current (but no cumulative) environmental externalities. Under these assumptions equation (4) becomes

Since Dq(t) > 0 for all q(t)> 0, comparing equation (7) with equation (5) shows that the level of marginal profits is reduced by current environmental externalities. However, their rate of growth remains r, the rate of interest. If, however, negative cumulative (but no current) environmental externalities are assumed, equation (4) can be written as

Since Dx(t) > 0 for all q(t) > 0, it follows that the rate of growth of marginal profits in equation (8) is less than in equation (6). This result says that a mining firm (that is made to pay for cumulative environmental externalities) will reduce its rate of extraction, in part because of the burden of future tax liabilities that will be imposed on account of the entire profile of cumulative environmental externalities in the future caused by an increase in current extraction.

Imposing a corrective tax that takes into account both current and cumulative externalities also increases the cutoff point for mineral extraction, and therefore reduces the amount of recoverable reserves12 and slows down the rate of extraction of minerals. Since these effects are offsetting, their net impact on depletion dates is ambiguous.

One widely recognized difficulty encountered in implementing environmental taxes is the complexity of computing them, mainly because of the large amount of information required for their design. This difficulty has often been discussed in the context of environmental taxes analyzed within a static framework when only current environmental externalities are considered.13 The analysis here shows that the problem of computing environmental taxes is considerably more complex when cumulative externalities are also taken into account.

The appropriate corrective tax when current and cumulative externalities are present is given by

The complexity involved in computing an optimal environmental tax, or a dynamic Pigovian tax, arises from the difficulty in obtaining information required for computing marginal current environmental externalities, Dq(t), at each point in time, and cumulative externalities, Dx(t); and, even more so, from the fact that to compute the tax at time t involves computing the present value of all future marginal cumulative externalities arising from a unit increase in cumulative output. This present value, represented by the second term on the right-hand side of equation (9), cannot be assumed to be constant over time since

that is, the second term on the right-hand side of equation (9) approaches zero as the terminal period is approached. In other words, a dynamic Pigovian tax reduces to its static equivalent as terminal production is approached.

It is interesting to note that a dynamic Pigovian tax has similarities with taxes designed primarily for revenue and other (other than correcting for environmental externalities) purposes. The first term on the right-hand side of equation (9) is, for example, similar to a specific tax on output. The second term, which declines over time, is similar to a sliding specific tax that is imposed on remaining reserves of an extractive firm and vanishes when mining stops. That these taxes are similar to a dynamic Pigovian tax is important for policy purposes because it provides some justification for output taxes that are otherwise condemned as being distortionary.

When mineral extraction involves current as well as cumulative environmental externalities, an appropriate corrective tax has two elements—one that depends on current rates of extraction, and one that depends on cumulative output. The first element of the tax is similar to a specific tax on mineral output; the second element declines over time and vanishes when mining stops and is, therefore, similar to a sliding specific tax imposed on the remaining amount of reserves. These results suggest that in mining, where environmental taxes are rarely imposed, but specific taxes are popular with governments, such taxes have a redeeming feature: they can be considered proxies for current environmental externalities.

III. Mineral Resources Taxation, Environmental Degradation, and Sustainable Growth

In taxing mineral resources governments often seek to maximize tax revenues, ensure that mineral taxes are neutral, minimize variability and uncertainty of tax revenues, and avoid delays in receipt of tax revenues from mineral resources.14 Beyond these objectives, much of the literature on mineral taxation has focused on issues of particular concern to conservationists, such as the effects on rates of extraction, recoverable reserves, and depletion dates. Interest in these issues dates back to Gray (1914) and Hotelling (1931) and is motivated, in part, by conservationists’ concern that extraction of mineral resources in free markets may be too rapid and could be inconsistent with sustainable economic growth and development. However, the existing literature on mineral tax policy has generally ignored another concern of environmentalists—the direct and indirect effects of mineral taxes on environmental externalities.

This section provides a review of the effects of various mineral taxes on rates of extraction, recoverable reserves, and depletion dates. Consistent with the analysis in the previous section and studies by Levhari and Liviatan (1977), Conrad and Hool (1981), Heaps (1985), and others, it will be assumed that firms in the mining sector operate under resource constraints, and that costs of extraction increase with cumulative extraction. The effects of these various taxes on recoverable reserves, depletion dates, and current and cumulative environmental externalities are summarized in Table 1.

Table 1.Summary of Economic and Environmental Effects of Various Forms of Mineral Taxation
TaxRecoverable

Reserves
Rate of ExtractionReserve

Depletion Dates
Rate of Environmental

Externalities
Cumulative

Environmental

Externalities
Profits tax
Flat rate with no depletion allowancesZeroZeroZeroZeroZero
Flat rate with cost depletion allowancesIncreasedIncreasedAmbiguousIncreasedIncreased
Flat rate with percentage depletion allowancesIncreasedZero, increased, or decreased depending on prices rising at rate equal to, less than, or greater than interest rateAmbiguousZero, increased, or decreased depending on prices rising at rate equal to, less than, or greater than interest rateIncreased
Flat rate with “excessive” depreciation, investment, and exploration allowancesIncreasedIncreasedAmbiguousIncreasedIncreased
Progressive rate with no depletion allowancesIncreasedReducedExtendedReducedIncreased
Rate increasing over timeReducedIncreasedAdvancedIncreasedReduced
Rate decreasing over timeReducedReducedAmbiguousReducedReduced
Resource rent taxaZeroZeroZeroZeroZero
Resource rent tax with interest rate less than investor’sReducedReducedAmbiguousReducedReduced
Resource rent tax with interest rate greater than investor’sIncreasedIncreasedAmbiguousIncreasedIncreased
Brown taxZeroZeroZeroZeroZero
Franchise taxReducedIncreasedAdvancedIncreasedReduced
Taxes on output
Per unit output taxReducedReducedAmbiguousReducedReduced
Flat rateb
Rate increasing at rate less than rate of interestReducedReducedAmbiguousReducedReduced
Rate increasing at rate greater than rate of interestReducedIncreasedAdvancedIncreasedReduced
Rate equal to rate of interestReducedUnchangedAdvancedUnchangedReduced
Ad valorem tax
Flat rateReducedZero, reduced, or increased, depending on prices rising at rate equal to, greater than, or less than interest rateAdvanced except when prices rising less than interest rate, when it is ambiguousZero, reduced, or increased, depending on prices rising at rate equal to, greater than, or less than interest rateReduced
Property taxesIncreasedIncreasedAmbiguousIncreasedIncreased

Assumes that the interest rate that is used in discounting the net cash flow of the extractive firm is equal to that of the firm. It is also assumed that the cash flow patterns of mineral projects undertaken by the firm exclude those in which a cumulative positive net cash flow in earlier years is followed by subsequent negative net cash outflow. If any of these assumptions is not satisfied, the tax is no longer neutral.

A flat rate per unit of output tax is equivalent to a once-and-for-all downward shift in the demand curve: the cutoff point at which marginal profits equal zero implies a lower rate of output; total cumulative output is lower as the result of an imposition of the tax.

Assumes that the interest rate that is used in discounting the net cash flow of the extractive firm is equal to that of the firm. It is also assumed that the cash flow patterns of mineral projects undertaken by the firm exclude those in which a cumulative positive net cash flow in earlier years is followed by subsequent negative net cash outflow. If any of these assumptions is not satisfied, the tax is no longer neutral.

A flat rate per unit of output tax is equivalent to a once-and-for-all downward shift in the demand curve: the cutoff point at which marginal profits equal zero implies a lower rate of output; total cumulative output is lower as the result of an imposition of the tax.

Profits Tax

In static analyses it is well established that imposing a pure profits tax does not change the allocative decisions of a firm. This conclusion is also true for an extractive firm facing a resource constraint, whether or not costs increase with cumulative output (see Burness (1976)). Such a tax, therefore, has no effect on rates of extraction and on current and cumulative environmental externalities, recoverable reserves, or resource depletion dates. When a firm causes environmental externalities, a combination of a pure profits tax and a dynamic Pigovian tax would result in efficient resource allocation.

In practice, a pure profits tax is difficult to implement, particularly in mining. In addition to the difficulties of computing economic depreciation, a pure profits tax creates problems involving the treatment of depletion allowances, exploration, and development costs. Granting depletion allowances under a profits tax, for example, often induces firms to shift production toward early periods. Imposing a profits tax with a provision for cost depletion allowances (that is, a fixed nominal value for each unit of output) effectively raises output price by a constant (equal to τθ/1 – τ, where τ is the rate of the tax, and θ is the fixed dollar allowance per unit of output when profits are positive).15 The effects of the tax on output are the opposite of a per unit output tax and could be conceived as a per unit of output subsidy to the firm. Therefore, the tax induces mining firms to increase rates of extraction and increase current and cumulative environmental externalities. Recoverable reserves are also increased by these provisions of the tax. The effects of such a depletion allowance on depletion dates are ambiguous, since the rates of extraction and recoverable reserves both increase. If, however, extraction costs are independent of cumulative output, the tax has no effect on recoverable reserves and amounts of cumulative environmental externalities, whereas reserve depletion dates would be advanced to the present.

Imposing a profits tax with provisions for percentage depletion (that is, a fixed proportion of the current value of output regardless of actual cost) effectively raises the price (by τθ/1 – θ, where θ is a fixed proportion of output) and has the same qualitative effects as a negative ad valorem tax or subsidy.16 As can be seen from Table 1, the effects of the tax are the opposite of those of an ad valorem tax discussed below.

The tax structures of many countries provide for accelerated depreciation, immediate expensing of exploration costs, and investment tax incentives. Imposing a profits tax with these provisions has similar qualitative effects as the effects of a tax with cost depletion provisions. To see this, assume that accelerated depreciation is proportional to output. The effect of imposing a profits tax under this assumption is to effectively raise output price in the same way that cost depletion does.

Some countries impose progressive profits taxes with a view to capturing windfall returns or higher proportions of rents accruing in mineral resource exploitation. Often the tax provisions have a base rate and a growth component that depends on the level of profits. Such a tax lowers rates of extraction and, therefore, rates of environmental externalities; it increases amounts of recoverable reserves and, therefore, cumulative amounts of environmental externalities. Since recoverable reserves increase at the same time as rates of extraction are lowered, reserve depletion dates are extended by the imposition of such a tax.

These results assume a profits tax rate that is fixed over time. If the tax rates increase over time, then rates of extraction and environmental externalities are faster, recoverable reserves and cumulative environmental externalities are reduced, and reserve depletion dates are advanced (see Table 1).

A dynamic Pigovian tax should still be imposed even when a profits tax contains provisions that make the tax distortionary. However, the choice between such a tax package and other distortionary packages, such as a specific tax on output combined with a dynamic Pigovian tax, is no longer clear cut.

Resource Rent Tax

A resource rent tax seeks to tax the net present value of a mining firm as it is realized. Under the tax, when a firm’s net cash flow is negative, no tax is applicable, and a negative cash flow is accumulated at an interest rate chosen by the government until the cumulative net cash flow turns positive when the tax is applied at a flat rate.17

A resource rent tax will not affect a firm’s allocative decisions, assuming (1) the government’s chosen interest rate is the same as the firm’s discount rate; and (2) the cash flow pattern of a firm excludes cases in which the accumulated net cash flow in earlier years is followed by subsequent negative net cash flow with no further positive net cash flow. If the government’s chosen interest rate is lower than the firm’s discount rate, the firm’s tax liability is higher than would be the case under assumption (1). In this case, recoverable reserves and rates of mineral extraction as well as current and cumulative environmental externalities are lower than would be the case under assumption (1).18 The impact on resource depletion dates is, however, ambiguous.

If assumption (2) does not hold, the firm’s tax liability is higher than that indicated under this assumption. The effects of the tax on resource allocation under these assumptions is to reduce recoverable reserves and rates of mineral extraction compared to what obtains under assumption (2); the effect on resource depletion dates is again ambiguous, and current and cumulative amounts of environmental externalities would be reduced to below what obtains under assumption (2).

The fact that a resource rent tax is based on cash flow rather than income means that no distortions arise from application of the tax because of depreciation rules that differ from economic depreciation. Even though the choice of an appropriate interest rate is crucial, and not simple, this tax has many desirable properties and has often been recommended as the appropriate tax for mineral resources.19

A combination of a resource rent tax that satisfies assumptions (1) and (2) and a dynamic Pigovian tax is preferable to a package that includes a specific tax on output. However, when assumptions (1) and (2) are not satisfied, the resource rent tax also becomes distortionary and its superiority over a specific tax on output cannot be assumed a priori.

Brown Tax

Under a Brown tax the cash flow is subject to a flat rate tax. Unlike a resource rent tax under which a firm accumulates net negative cash flows, under a Brown tax the firm receives negative taxes (subsidies) when the net cash flow is negative. The tax has one major advantage over a resource rent tax: there is no need to choose an interest rate, since there are no accumulated negative net cash flows. Its major disadvantage is that it entails great revenue risk to governments where, as is often the case, cash flows from firms’ mining operations are uncertain.20 Apart from its effect in reducing risk, the Brown tax is neutral and therefore has zero impact on all the variables of interest in this paper. In the presence of environmental externalities as described in this paper, this tax combined with the dynamic Pigovian tax would be the economist’s ideal mineral resource tax.21

Franchise Tax

A franchise tax is similar to a lump-sum tax in static analysis, except that in the dynamic context the tax depends on time. Thus, although in static analysis it is well known that a lump-sum tax does not change a firm’s allocative decisions, the effects of a franchise tax for an extractive firm are to increase rates of extraction, reduce recoverable reserves, and advance depletion dates (see Burness (1976)). The effects of a franchise tax are to increase rates while reducing cumulative environmental externalities over the life of a mine.22

If extraction costs are independent of amounts extracted, then the mine is exhausted at the terminal period, and the tax has no impact on recoverable reserves. Similarly, cumulative amounts of environmental externalities are the same with or without the tax, even though their time pattern would be different. The effects of the tax on rates of extraction and, therefore, on rates of environmental externalities and depletion dates are the same as under the assumption of costs that increase with cumulative extraction.

From an economic point of view, a franchise tax loses the neutrality appeal of a lump-sum tax in static analysis. Making a judgment on the desirability of the tax is more complex when environmental and conservationists’ concerns are taken into account. For example, under the assumption of extraction costs that increase with cumulative amounts of extraction, rates of environmental externalities are higher with than without the tax. At the same time, cumulative environmental externalities are lower with than without the tax. Similarly, although conservationists may condemn the tax because it increases rates of extraction, they would welcome the fact that reserves left in place are greater with than without the tax. If extraction costs are independent of cumulative extraction, the effects of the tax pose no such dilemmas, since cumulative environmental externalities and recoverable reserves are the same with or without the tax. Rates of environmental externalities are higher with the tax than without it, and depletion dates are advanced to the present.

Taxes on Output

Taxes on mineral output include fixed payment per unit of gross output, fixed payment per unit of certain quality of output, and fixed proportion of the price of a mine’s output. Mineral taxes on output can also be applied on sliding scales, which reduce tax payments to zero as extraction costs reach a certain threshold—after a number of years of extraction, or when extraction ceases. Specific or ad valorem taxes are quite popular forms of taxing minerals or petroleum, primarily because of their administrative simplicity. In static analysis the general conclusion is that the taxes induce firms to restrict output and encourage high grading—that is, extraction of high-grade ores bypassing lower-grade ores. However, as shown below, in dynamic analysis, when stocks of resource ores are given, the effects of these taxes vary depending on specific assumptions about the nature of the tax. For simplicity, only two forms of the taxes are examined here—specific and ad valorem.23

Specific Taxes

The effects of a specific tax on output at a rate s on an extractive firm operating under a resource constraint depend, in part, on how τ, and its rate of change, is related to r, the rate of interest. Imposing a flat rate per unit output tax on an extractive firm reduces rates of extraction and recoverable reserves, but its effects on depletion dates are ambiguous. The tax reduces the rate of extraction because the marginal tax liability is constant for each period, but the present value of future marginal tax liabilities is smaller than for the present period. Therefore, an extractive firm can increase its net present value by postponing production into the future. Recoverable reserves are reduced because the tax increases the cutoff point. It has ambiguous effects on reserve depletion dates because, on the one hand, it reduces rates of extraction and, on the other, it increases reserves left in place. The tax is a favorite for environmentalists and conservationists because it reduces both rates and cumulative environmental externalities while increasing reserves left in place.

If extraction costs are independent of amounts extracted, so that reserves are always exhausted, then the effect of the tax is to extend depletion dates. All its other effects are the same as those discussed under the assumption of costs that increase with cumulative extraction.

The effects of the tax when rates change over time depend on whether the rate of change is greater, equal to, or less than the rate of interest. If the rate of increase of the tax is greater than the rate of interest (τ˙/τ>r), then rates of extraction are increased, recoverable reserves reduced (assuming increasing costs), and reserve depletion dates advanced. Rates of environmental externalities are increased while cumulative externalities fall, because with nominal marginal tax liabilities increasing at a rate greater than the rate of interest, the firm can increase its net present value by reducing its tax liability through shifting extraction to the present, which is less heavily taxed in real (that is, present value) terms. If extraction costs are independent of amounts extracted, then reserves will be exhausted with or without the tax, and the effect of the tax on cumulative environmental externalities is zero.

If the tax rate increases at rates equal to the interest rate, then the tax has no effect on rates of extraction, but recoverable reserves will be reduced so that reserve depletion dates are advanced. Rates of environmental externalities are unchanged by the tax while cumulative environmental externalities are lower. The effects of the tax (assuming extraction costs are independent of cumulative output) are the same as when the tax rate increases at a rate greater than the rate of interest. The effects of the tax when the tax rate increases at rates lower than the rate of interest are similar (and for similar reasons) to the effects of a flat rate tax.

As already noted, a combination of a specific tax on output with a dynamic Pigovian tax is inferior to a package that combines a dynamic Pigovian tax with the neutral taxes discussed above. However, in practice, the conditions for neutrality, in particular, of a profits tax or a resource rent tax are not satisfied, which means the choice between these taxes and specific taxes has to be based on other grounds. In addition, dynamic Pigovian taxes, or any other corrective taxes, are rarely imposed, and this further complicates judgment of specific taxes and even a Brown tax. This implication arises because without an appropriate corrective tax, a Brown tax would lead to excessive rates of mineral extraction. At the same time, a specific tax on output if combined with a sliding element that depends on amounts of remaining reserves is similar to a dynamic Pigovian tax and would have similar qualitative effects in internalizing environmental externalities. This paper takes the view that specific taxes should be explicitly designed to internalize environmental externalities, and a resource rent tax, profits tax, or Brown tax should be designed to capture resource rents. In other words, these taxes should be viewed as complementary rather than substitutes.

Ad Valorem Taxes

The effects of an ad valorem tax at rate τ differ from that of a specific tax, primarily because tax liability in this case varies with output price rather than the rate of output. Generally, the effects of the tax on rates of extraction are the opposite of those of changes in prices. For example, the effect of a once-and-for-all increase in price (assuming extraction costs are independent of the amount extracted) is to increase rates of extraction and rates of environmental externalities, while advancing depletion dates (see Levhari and Liviatan (1977)). The effects of an ad valorem tax, however, are to reduce rates of extraction and environmental externalities and to advance depletion dates. The effects of the tax on rates of extraction and environmental externalities remain the same when extraction costs increase with amounts extracted. At the same time, under the same assumption the tax reduces recoverable reserves and cumulative environmental externalities, while its effects on resource depletion dates are ambiguous.

The effects of continuously increasing output prices depend on whether the rate of increase in prices is greater, equal to, or less than the rate of interest. For example, resource extraction would cease if prices were rising faster than the rate of interest, since the resources would be more valuable in place than on the market. Extraction would be shifted into the future, however, if prices were rising at a slower pace than the rate of interest (see Levhari and Liviatan (1977)). The effects of imposing an ad valorem tax when prices are rising faster than the rate of interest would be to induce the firm to carry out some extraction in the current period. If, however, the tax is imposed when prices are increasing at a slower rate than the rate of interest, the effect is to shift extraction into the future. In either case the effect of the tax is to reduce recoverable reserves and cumulative amounts of environmental externalities while resource depletion dates are advanced in the first case and are ambiguous in the second.

Property Taxes

In principle, this tax is imposed on the remaining mineral reserves at the end of each period. A particular case of the tax is one in which the tax is imposed on the capitalized value of the firm or on the present value of all the assets of the firm. In practice, the base of the tax can vary substantially from this, because it is difficult to estimate the capitalized value of a mining firm, and because mining firms tend to report exploration costs as proxies for the base of the tax.24 Because the base of the tax declines over time, the nominal value of a constant rate property tax also declines over time. The effect of the tax is, therefore, to induce faster rates of extraction and higher rates of environmental externalities than would be the case in its absence, since, other things remaining the same, faster extraction rates reduce the firm’s tax liabilities. In this sense, the tax has the opposite effect of an ad valorem tax.

A property tax increases recoverable reserves and cumulative amounts of environmental externalities. To appreciate this result it is useful to remember that since total available reserves are fixed, the firm can reduce its tax liabilities at the terminal point by extracting beyond the nontax cutoff point because it receives the equivalent of a negative ad valorem tax on each unit extracted. The effect of the tax on depletion dates is, however, ambiguous because, on one hand, it induces higher rates of extraction, while, on the other, it increases recoverable reserves. The tax is not liked by environmentalists (because it increases both rates and cumulative amounts of environmental externalities) or conservationists (because it shifts extraction from the future to the present and reduces the reserves left in place).

The discussion here suggests that there are complementarities as well as trade-offs between mineral tax policies and environmental concerns. There are also considerable uncertainties about the effects of certain mineral taxes on the environment and related issues. From the point of view of tax policy, the ideal mineral resource tax is a Brown tax, a neutral resource rent tax, or an income tax with true economic depreciation and true depletion allowance provisions. Such taxes have the advantage of raising revenues in a nondistortionary manner; that is, in the absence of environmental and other externalities, they create the least amount of distortions in rates of extraction, recoverable reserves, and depletion dates. However, environmentalists could contend that the resulting rates of extraction may not be sustainable; they would prefer taxes that reduced the rates of extraction, recoverable reserves, and extended depletion dates. From this perspective, specific and ad valorem taxes may be considered more environmentally friendly than the more neutral taxes.25

If mining operations are associated with environmental externalities as analyzed in this paper, then economists’ and environmentalists’ concerns will coincide at least up to the point of urging the imposition of a dynamic Pigovian tax. Fiscal economists, in addition, would point to the complementarity of these taxes between economic efficiency, environmental protection, and improvement of fiscal management for governments strapped for budgetary resources. Furthermore, in the absence of environmental taxes, distortionary taxes such as specific taxes that are common in the mining sectors of some countries can be seen in a more favorable light in cases where mining operations inflict environmental externalities. Even though the rates of such taxes may not equal the rates of the dynamic Pigovian taxes analyzed in this paper, their imposition provides partial offsets for otherwise uncorrected environmental externalities.

The discussion also points to some uncertainties about the effects of various taxes on depletion dates, which complicates attempts to rank them according to environmental friendliness. Consider, for example, a comparison between a flat rate per unit of output tax and a progressive profits tax with no depletion allowances (see Table 1). On the one hand, imposing either tax reduces rates of environmental externalities. On the other hand, imposing a specific tax reduces cumulative amounts of environmental externalities, while a progressive profits tax increases them. However, this result does not necessarily imply that a specific tax is more environmentally friendly, because its effect on depletion dates is ambiguous, whereas a progressive tax extends them. Similar difficulties arise in ranking a franchise tax and a property tax (see Table 1).

These taxes have been analyzed as if they were alternatives. In practice, the taxes are often presented as a package. An assessment of the net effects of these packages would shed more light on their effects in actual cases. However, such an assessment is beyond the scope of this paper. At the same time, however, one package that could simultaneously address the quest of economists for nondistortionary taxes and the concerns of both environmentalists and economists with environmental externalities could be either a Brown tax, a neutral resource rent tax, or a pure profits tax, on the one hand; and a specific tax on output that incorporates a sliding specific tax on output that reduces to zero at the terminal period, on the other. For this reason, (distortionary) specific taxes on mineral output may in fact improve resource allocation at the same time as they protect the environment.

IV. Concluding Remarks

This paper has examined the effects of mineral taxes on the environment from two perspectives. First, an appropriate framework for environmental taxes—the dynamic equivalent of the Pigovian taxes in static analyses—was developed. The framework takes it into account that mining firms face resource constraints, their extraction costs increase with cumulative extraction, and mining activities are associated with current and cumulative externalities. Second, the paper examined the effects of various taxes that are often imposed to achieve several fiscal objectives but not specifically directed toward correcting for environmental externalities. On the basis of this examination the following conclusions may be drawn.

Changes in a Brown tax, or a neutral resource rent tax (in which the rate of interest used in discounting the net cash flow is equal to a mining firm’s discount rate) have neutral effects on the environment. These taxes are the preferred forms of taxing mineral resources. However, in the presence of environmental externalities, they need to be supplemented with dynamic Pigovian taxes if economic efficiency, as well as environmental protection, is to be achieved.

Many countries have environmental regulations governing the operations of mining firms, and few impose appropriate environmental taxes. No doubt this lack partly reflects governments’ preference for a command and control approach, rather than the use of economic instruments to control environmental degradation. But it may also reflect the considerable complexity and extensive information requirements associated with the design and implementation of appropriate environmental taxes for mining firms. The absence of applied research on practical designs of environmental taxes may be a further impediment to efforts to implement these taxes.

Even though usually imposed largely for revenue purposes, and often condemned as distortionary, per unit taxes on mineral output have similar qualitative effects in restraining environmental externalities as static Pigovian taxes. This is not to suggest that these taxes are a perfect substitute for static Pigovian taxes, let alone the dynamic version discussed in this paper. They are not, in part because their rates, which are often based on revenue and other fiscal considerations, are likely to differ from those based on the need to “internalize” environmental externalities.

However, if environmental externalities are significant and they are related to rates of mineral extraction, then, in the absence of any environmental taxes, per unit of output taxes could be viewed as proxies for static Pigovian taxes. If these taxes are supplemented by taxes on remaining reserves, then they resemble, at least in form, dynamic Pigovian taxes. Viewed in this light, these taxes may be less distortionary than often suggested in theoretical analyses. Profits taxes with the rate decreasing over time, resource rent taxes with the rate of interest less than the investor’s discount rate, and a per unit of output tax with a rate increasing at a rate less than the rate of interest are all environmentally friendly. If, in a reform package, these taxes were replaced by, for example, a Brown tax, or a neutral resource rent tax, then it might be necessary to implement appropriate environmental taxes as part of the reform package; otherwise, environmental degradation would worsen after the tax reform.

There is complementarity between economic and environmental objectives when taxes, such as a flat rate profits tax with provision for cost depletion, a flat rate profits tax with provisions for excessive depreciation allowances, a resource rent tax with an interest rate that is greater than the investor’s discount rate, or a property tax, are replaced by a Brown tax or a neutral resource rent tax. In such cases, the tax substitution reduces environmental damage at the same time as it improves resource allocation. Because these taxes are distortionary and environmentally damaging, they should be the first target for elimination or reduction in a reform package in which economic and environmental considerations are important.

The trade-off between economic and environmental concerns is less clear for the other taxes—a flat rate profits tax with percentage depletion, a franchise tax, a per unit of output tax with a rate greater than the rate of interest, and a flat rate ad valorem tax. Specifically, replacing such taxes with a Brown tax, or a neutral resource rent tax will improve resource allocation, but the effects on the environment may be mixed. For example, elimination of a franchise tax would increase the rate of environmental degradation at the same time as it reduced the cumulative environmental externalities; the environmental impact of eliminating a flat rate, an ad valorem tax, or a flat rate profits tax with percentage depletion is even less certain.

REFERENCES

Although there may be some obvious cases where these policies may lead to unsustainable growth and development strategies, in general, their net effects on sustainability are complex and often ambiguous (see Muzondo and others (1990)).

The terms “environmental degradation,” “pollution,” and “environmental externalities” are often used interchangeably. In this paper “environmental degradation” and “pollution” refer to the use or consumption of any environmental assets whether or not there are external effects. “Environmental externalities” refers to external effects associated with the use or consumption of environmental assets. All forms of environmental degradation may be of concern to environmentalists. However, from an economic point of view it is the degradation associated with externalities that requires public policy intervention.

None of the many descriptions of environmental damage from mineral extraction have made the analytical distinction between the rate and the amount of cumulative environmental degradation. As shown in Section II below, this distinction is important and calls for different corrective measures from those of standard static analysis.

Under the stipulated assumptions the equilibrium rate of output for a monopolist requires that MRt= MRt + 1/1 + r, where MRt is the marginal revenue at time t, and r is the interest rate. With a constant demand curve, MRt = βPt, where 1/1 — β is the elasticity of demand. It follows that the equilibrium rate of output for a monopolist implies a price such that Pt= Pt+1/1 + r, which is also the equilibrium condition under competition. For further discussion of an extractive monopolist firm, see Stiglitz (1976). See also Stiglitz and Dasgupta (1982) for a discussion of market structure and resource depletion.

This assumes that costs are independent of the cumulative level of extraction.

In addition, a monopolist may, because of its size, have readier access to capital markets than individual competitive firms. If this means that the cost of capital is lower for the monopolist than for firms under competition, this would further strengthen the bias in favor of conservation by the monopolist. It should be stressed, however, that although a monopolist may pursue a more conservationist policy than firms under competition, from a social welfare point of view, the monopolist’s rate of extraction is not necessarily preferred to that of firms under competition.

For a detailed discussion of taxation of nonrenewable resources, see Church (1981), and for a survey of mineral resource taxation for a selected group of countries, see Muzondo (1992, Appendix II).

The exception is ornamental minerals such as gold and diamonds.

Burness (1976) analyzed the effects of resource taxation under the assumption that the costs of resource extraction are not affected by the cumulative amount of extraction—an assumption that leads to the conclusion that resources are completely exhausted at the terminal period. Gordon (1967), Levhari and Liviatan (1977), and Heaps (1985), among others, incorporated the effects of cumulative extraction in their models and modified earlier conclusions. For example, as against the standard Hotelling conclusion that a flat-rate severance tax always prolongs the depletion date of a mine, Levhari and Liviatan show that the impact of the tax is ambiguous. These studies have not, however, incorporated the effects of environmental externalities in their models. Other studies, including those by Anderson (1972) and Vousden (1973), have examined the importance of environmental externalities, resource conservation, and other environmental issues in the context of macroeconomic growth models.

Clearing of forests and park land (which have scenic and recreational values) for agricultural, human settlements, or other economic uses may represent such cases: production costs (that is, costs of clearing) need not increase with the amount of forest already cleared; however, the scenic and recreational value diminishes with the amount of forest already cleared.

The analysis may, however, be relevant for cases beyond extractive industries. For example, it would seem to apply to cases discussed by Pearce (1976) where, beyond a certain level, economically optimal waste emissions into an environmental medium diminish the medium’s assimilative capacity, and this in turn increases marginal environmental damages.

The cutoff point for mineral extraction refers to a point where extraction ceases, or when marginal profits are zero.

See Baumol and Oates (1988) for a suggested way of getting around the problem of designing environmental taxes.

For a discussion of the objectives of mineral resource taxation, see Garnaut and Clunies-Ross (1983).

For a derivation of this result, see Conrad and Hool (1981).

For similar conclusions and derivations of this result, see Conrad and Hool (1981).

For a detailed discussion on tax, see Garnaut and Clunies-Ross (1983).

The case in which the chosen interest rate is higher than the firm’s discount rate is the reverse of that when it is lower.

It has been applied in Papua New Guinea and Tanzania; see Garnaut and Clunies-Ross (1983).

For a detailed comparison between a Brown and a resource rent tax, see Garnaut and Clunies-Ross (1983).

Perhaps because of its significant revenue risk to governments, many governments do not favor the tax.

The difference in the effects of the lump-sum tax in static analysis and the franchise tax in dynamic analysis is, of course, due to discounting. Since the nominal value of the franchise tax paid is the same in each period irrespective of the level of output, an extractive firm can minimize the present value of its tax liability by reducing the number of periods in which extraction takes place through accelerating extraction from the future toward the present. In this sense the tax is avoidable and, strictly speaking, not a lump-sum tax.

That is, the distinction between a specific tax based on gross value and that based on certain quality of output is ignored. For an analysis that makes the distinction between a specific tax based on gross value and one based on certain quality of output, see Conrad and Hool (1981).

For a more detailed discussion of the property tax, see Burness (1976).

However, as noted in Section I, sustainability or conservation in the extraction of mineral resources should, ideally, be considered a macroeconomic and investment policy issue rather than a micro or sectoral policy issue.

Other Resources Citing This Publication