Property Rights
Efficient property rights systems have four characteristics:
Resources generate Ricardian scarcity rents which can also be viewed as opportunity costs: allocation of a unit of resource to any one use precludes other uses of it; use of a unit of resource at any one time period precludes using it in any other time period. These opportunity costs are sometimes referred to as "marginal user costs."
Mainstream microeconomic theory assumes perfect markets (many buyers and sellers, full information precluding price discrimination, etc.) that achieve Pareto-optimal outcomes. However economists have identified various categories of market failure, where violations of one or more key assumptions of the perfect market model prevent markets from achieving a Pareto-optimal outcome. Market failures arise in cases of...
An externality occurs when one party generates costs or benefits for some other party but doesn't account for those costs or benefits. A positive externality generates an inadvertent benefit for someone else: for example, in the process of making honey for himself, a bee-keeper's bees pollinate neighboring crops, improving crop yields on neighboring farms, although the farmers don't share the benefits of the higher yields with the bee-keeper. A negative externality generates an inadvertant cost for other people: for example, a paper mill dumps pollutants upstream from a trout hatchery and reduces the hatchery's productivity, but doesn't account for the damages to the hatchery in its own accounting.
An pollution externality increases the social costs of production beyond the private costs borne by the firm. In the graph below, a firm's output generates a pollution by-product. Without controls, this firm produces to the point where its marginal private cost equals its marginal revenue (Demand). The firm enjoys free use of the environment for waste disposal; the environmental costs are borne by society. The social optimum is where the marginal social cost (marginal private cost plus marginal pollution damage cost) equals marginal revenue (Demand). The social optimum level of the firm's output Q* is less than the private optimum level Q'. The social optimum price per unit P* is higher than the private optimum price P'.
The lack of clear, enforceable property rights in the use of the river, or more specifically, the lack of "exclusivity" in property rights, yields a situation where the market fails to achieve a socially optimal outcome. The polluting firm generates too much output, sells it too cheaply, and produces too much pollution. In producing too much, it bids other resources away from more efficient uses. It can afford to sell its output more cheaply because society is basically subsidizing the firm with free waste disposal services. Free waste disposal gives the firm no incentive to reduce its waste flow at all--no investment in waste management, recycling or re-use. The true cost of the firm's product includes the environmental damage associated with its production. In fact, pollution externalities typically generate market distortions that ripple throughout the economy.
Economists are careful to distinguish true externalities from "pecuniary externalities." Unlike a true externality, a pecuniary externality only affects prices; it doesn't affect consumers' utilities or firms' productivities directly. For example, a large firm moves into an area and bids up local wages, increasing labor costs for the other firms in the area. The other firms may complain, but this is how efficient labor markets are supposed to work. Pecuniary externalities do not imply market failure, and are not true externalities.
A Corrective Tax or Subsidy
The Cambridge economist A.C. Pigou proposed correcting this market failure by imposing a tax on the polluter's output corresponding to the marginal external damage cost. This presumes that the victims of the externality are entitled to a clean environment, so the polluter should pay compensation for the damage he causes. The pollution tax thus "internalizes" the externality, making the firm's marginal cost schedule coincide with the marginal social cost of production. Facing marginal tax $T per unit of output, the profit-maximizing firm reduces its output to the socially optimal level Q*, charges P* and earns an after-tax marginal revenue (P*-T) which just covers marginal private cost. For simplicity, the tax of $T per unit could be applied to the firm's total output, or the pollution tax schedule could follow the marginal damage cost schedule; the firm's corrected incentives are the same in either case.
The equivalent outcome could be achieved by subsidizing the firm's pollution abatement in line with the marginal damage schedule. This presumes the polluter is entitled to pollute, and should be compensated to forego that right. In this case, the subsidy for pollution abatement creates an opportunity cost of pollution for the polluter: each unit of pollution he emits is a subsidy payment forgone, so he will again reduce output to the socially optimal level Q*, charge P* and earn total revenue (P*+S)Q*. The subsidy per unit could be paid on all of the firms output, or paid in accordance with the marginal damage cost schedule; again, the firm's corrected incentives are the same in either case.
The tax and subsidy solutions may generate different incentive problems. The pollution tax paid by the firm might be distributed to the victims, but this weakens the victims' incentive to move away from the pollution or find other ways to reduce their exposures to the pollution. Furthermore, the provision of compensation might conceivably attract additional pollution-tolerant "victims" to the area, who could then extort additional compensation from the polluting firm. In fact, Ronald Coase proposed that the victims of the externality should be taxed to prevent exactly this kind of behavior. In most cases this should not be necessary: the harm suffered by the victim should provide sufficient incentive to avoid it; the victim should not be compensated or taxed.
There is one special case where taxing the victim is appropriate: suppose the victim has the ability to shift the harm from the externality to a third party. In such cases, the victim becomes a secondary generator of an externality, and should pay a Pigovian tax reflecting the externality he imposes on the third party.
A pollution abatement subsidy paid to the polluting firm may create adverse incentives as well; if the subsidy is based on the environmental threat from the firm, the firm has an incentive to maximize that threat, perhaps targeting its pollution to the most vulnerable victims.
The Coase Theorem and Transactions Costs
Suppose a paper mill discharges waste into a river rather than paying $100 to dispose of it in some environmentally benign way. Suppose that a fish hatchery downstream loses $60 in fish because of the paper mill's emissions. What will happen?
A Pigovian tax on the paper mill makes the paper mill bear the full social cost of its activities, but the mill keeps on polluting, since it's cheaper to pay the $60 tax rather than pay the $100 treatment cost. The government collects the tax revenue, but doesn't ompensate the hatchery. The hatchery still has an incentive to relocate to a less polluted river if this is more profitable.
Ronald Coase's famous 1960 article "The Problem of Social Cost" (Journal of Law and Economics 3:1-44) takes a different view of this problem and points out some important limitations of the Pigovian solution. If the hatchery sues the paper mill for damages, the court will merely clarify the parties' property rights to the river: either the paper mill has the right to dump waste into the river, or the hatchery has the right to a clean river. But the court's decision will not determine whether or not the paper mill will continue dumping its waste into the river! This is for the hatchery and paper mill to negotiate between themselves.
If the court finds for the paper mill, the mill continues dumping and the hatchery continues suffering damages of $60. The hatchery would be willing to offer the mill up to $60 not to pollute, but since the alternative treatment costs $100, the paper mill would decline. But if the court finds for the hatchery, the paper mill would be willing to offer the hatchery compensation up to $100 for putting up with its waste, and since this exceeds the hatchery's damage costs, the hatchery would accept any compensation over $60 and the paper mill would continue polluting as before. So the amount of pollution the mill dumps in the river is determined by the relative costs of the parties, not by the court's decision. The court's decision merely determines who pays who.
Coase's "theorem" is summarized as "In the absence of transactions costs, the generator of an externality and the victim will negotiate to the same externality level regardless of how the peoperty rights are assigned."
The stipulation regarding transactions costs is important. This implies a small number of victims who can collectively negotiate with the firm reasonably efficiently. In contrast, a large number of victims is unlikely to negotiate efficiently, since no individual victim is likely to have sufficient incentive to shoulder the burden of mobilizing the entire group.
In effect, there is a $40 economic surplus available for distribution between the polluter and the victims, so negotiations between the parties will proceed as long as the transactions costs are less than $40. More specifically, negotiations will proceed as long as each party's share of the transactions costs is less that his or her expected share of the $40 total gain from the negotiation. Coase argues that a Pigovian tax should be unnecessary! Here's why:
Suppose
the polluter and victim have marginal benefit and damage functions as shown.
In the absence of negotiation, the polluter's output is A, where his own
marginal benefit goes to zero. If the parties engage in Coasian bargaining,
they will negotiate a socially optimal level of output B, where the marginal
damage to the victim equals the marginal benefit to the polluter.
An environmental regulator, perceiving that the victim is still suffering
some pollution damage, might now impose a Pigovian tax of $T on the polluter,
equivalent to the remaining marginal damage suffered by the victim.
This effectively shifts the polluter's marginal benefit schedule downward
to MB-T, so the final level of output is C, less than the social
optimum. The sequence of bargaining and tax imposition could be reversed,
but the outcome is the same. In this case Pareto efficiency is achieved
by either the Pigovian tax or the Coasian bargain, not by both.
The conclusion here is that the Pigovian tax is only appropriate where
Coasian bargaining fails to reduce the externality to the socially optimal
level on its own.
As noted above, Coasian bargaining typically fails in cases where there are large numbers of victims. Conversely, Pigovian tax/subsidy schemes may fail in small-numbers cases where parties may have incentives to engage in strategic behaviors, such as polluters threatening more harm to increase the abatement subsidy payment, or victims threatening to expose themselves to additional harm to increase the polluter's tax burden or their own compensation.
We can immediately see some general correlaries from Coase's theory.
Since transactions costs prevent Pareto-improving bargaining, any technology,
institution or market mechanism that reduces transactions costs must be
Pareto-improving. The Coase theorem ought to apply to politics as
well: rational political bargaining betwen interest groups should result
in socially efficient policies. If transactions costs remain high,
it may be appropriate to inquire why: is there some third party that benefits
from keeping the two parties apart?