Externalities and The "Coase Theorem"
Externalities and The "Coase Theorem"
Externalities and The "Coase Theorem"
The “Coase Theorem” has been one of the most influential contributions to come from
economics in the last fifty years. Its influence on the law has been especially profound.
An Externality Example:
Suppose one party engages in an activity that produces a benefit for itself, but which
damages other parties. Let’s suppose that the first party is a firm, whose production
activity generates profit but also pollutes the air or water, adversely affecting one or more
nearby residents, or negatively affecting the profitability of one or more nearby firms. In
the simplest case, let’s suppose that it’s just a single consumer who suffers damages from
the firm’s production. Let x denote the level of the externality-generating activity, in this
case the firm’s production; let B(x) denote the resulting benefit to the firm (its profit);
and let D(x) denote the damage to the consumer (for example, the amount he would be
willing to pay to eliminate the effects the externality imposes on him). We’re assuming here
that both the benefit and the damage are measured in dollars, and we’ll assume that the
net social benefit of operating the activity at level x is v(x) = B(x) − D(x). The socially
(Pareto) optimal level of the activity is therefore the level x at which B 0 (x) = D0 (x), i.e.,
where the marginal benefit equals the marginal damage. Coase emphasized that the optimal
level of the damaging activity is not zero, and he described how the two parties should be
expected to bargain to an outcome in which they agree that the activity will be operated at
the optimal level, with one party paying compensation to the other — just as we described
in our public-good example of the water level in the lake.
In fact, there is no formal difference between this situation and our earlier two-person public
good problem. Figure 1, below, depicts the “Kolm diagram” for this situation. Note that
in the core allocations the externality-generating activity is always operated at the Pareto
level, but that both the direction and the size of the compensation depend upon which party
has the right to set the level of x. In this example, then, the Coase Theorem could easily be
formalised as an actual theorem.
But the simplicity of the example also allows us to see a number of caveats, any of which
would change the Coase result, in some cases by only a little, but in others it would be
significantly changed:
(1) If the objective of either party has income effects, then the level of the activity will not
be determinate, and in particular it will depend upon which party has the right to set the
level of x and would therefore receive compensation. This effect would not typically be large.
(2) If there is a large number of individuals on either or both sides of the “market,” then
the bargaining or transaction costs can become prohibitively large, making it unlikely that a
core (or Pareto) outcome will be achieved by bargaining.
(3) If there is a large number of individuals on either side of the market, then the core, unlike in
the no-externalities case, is not small. The various core allocations will differ considerably in
the distribution of the compensations paid and/or received, further increasing the difficulties
in even achieving a core allocation.
Why has the “Coase Theorem” been so influential? Most likely it’s because Coase’s article
(“The Problem of Social Cost”) was published in the Journal of Law & Economics, and was
written in the style of law journals — no mathematical symbols or equations; every idea
presented as an actual externality that had arisen in legislative or court cases (albeit always
a two-party externality); and descriptions and citations of actual legal cases and case law on
every page. It’s had an enormous impact on the legal profession. In fact, by some measures
it’s the most-cited law review article of all time, from 40% to 80% more citations than the
second-most-cited.
The effect of Coase’s paper has been that when microeconomics predicts that an outcome
will be inefficient (or not in the core), we’re likely to ask ourselves whether the participants
would themselves devise some means to overcome the inefficiency. In both the legal and
economics professions, the focus now tends to be on the barriers to efficient outcomes, and
why the barriers can’t be overcome. This focus on barriers to efficiency, and on whether the
barriers are likely to be overcome without intervention, is not restricted to externalities and
public goods. Other examples are incomplete markets for dealing with uncertainty; adverse
selection; and moral hazard.
Incidentally, Coase himself did not describe this idea as a theorem, and was apparently not
happy that is was described that way. His paper pointed out that the felicitous outcome he
was describing would often not be achieved, due to the attendant transaction costs.
2
Figure 1