Ronald Coase, The Problem of Social Cost
Two-page summary
Ronald Coase’s 1960 paper The Problem of Social Cost is fundamentally about how society should deal with situations in which one person’s or firm’s activity adversely affects another. The conventional economic treatment, associated especially with A. C. Pigou, viewed these effects as externalities. A factory emits smoke that damages its neighbors; therefore the factory imposes a social cost not reflected in its private costs. The natural policy response is to make the factory bear that cost, through taxation, liability, regulation, or prohibition.
Coase argues that this way of framing the problem is incomplete. The central insight is what he calls the reciprocal nature of the problem. If A’s activity harms B, preventing A from undertaking the activity also imposes a cost on A. The economic question therefore is not simply how to prevent A from harming B. It is which arrangement avoids the more valuable harm.
His examples make the point concrete. Cattle wandering onto a neighboring farmer’s fields destroy crops. Preventing the cattle from wandering preserves crops but may make cattle raising more costly. A factory polluting a stream may destroy fish, but preventing the pollution may reduce valuable industrial production. A confectioner’s machinery interferes with a neighboring doctor, but silencing the machinery reduces confectionery production. The economic problem is consequently one of choosing among competing uses of scarce resources.
Coase first considers an artificial benchmark: a world in which market transactions cost nothing. Suppose legal rights are clearly specified. If the cattle raiser has the right to let his cattle stray, the farmer can pay him to reduce the herd or erect a fence when preventing crop damage is worth more than the cattle production forgone. Conversely, if the cattle raiser is legally liable for crop damage, he will compare the cost of additional cattle with the damage they cause. Under the assumptions Coase specifies, bargaining leads resources toward the use with the highest value.
This argument later became known as the Coase theorem: with clearly defined rights and zero transaction costs, bargaining can produce an efficient allocation regardless of the initial assignment of rights. But this formulation can obscure Coase’s larger purpose. The zero-transaction-cost world is essentially a theoretical device. His real interest is what happens when we leave that imaginary world and confront actual institutions.
In reality, transactions are costly. Parties have to discover who is affected, communicate, negotiate agreements, draft contracts, monitor behavior, enforce agreements, resolve disputes, and sometimes organize large numbers of people. Once these costs are introduced, many bargains that would theoretically increase total value will never occur.
This changes the role of law and institutions fundamentally. If bargaining is costly, the initial legal assignment of rights can affect not merely the distribution of income but what economic activities actually take place. Courts, legislatures, regulations, firms, and contractual arrangements become alternative mechanisms for coordinating activities.
Coase therefore rejects the automatic inference that observing an externality establishes a case for government intervention. Government action itself has costs. Regulation requires information and administration and can make mistakes. Courts have costs. Markets have transaction costs. Firms have organizational costs. There is no institution that solves the problem costlessly.
The relevant comparison is consequently between alternative institutional arrangements as they actually operate, not between an imperfect market and an idealized government solution. Sometimes liability will work best; sometimes regulation; sometimes bargaining; sometimes organizing activities within one firm; and sometimes tolerating the externality may cost less than eliminating it.
This is why Coase’s argument is broader than the proposition usually labeled the Coase theorem. His lasting contribution is an institutional way of thinking:
Define the conflicting uses → establish the relevant rights → examine possible bargains → identify transaction costs → compare alternative institutional arrangements → choose the arrangement whose total benefits exceed its total costs by the greatest amount.
That reasoning connects The Problem of Social Cost directly with Coase’s earlier The Nature of the Firm. In both papers, transaction costs explain why the institutional structure of economic activity matters. Markets, firms, contracts, courts, and governments are alternative ways of organizing transactions, and the costs of using them determine which arrangements make sense.
Coase ultimately asks economists to stop treating factors of production merely as physical objects. What matters economically is the right to perform particular actions. Legal rules determine those rights, and therefore law inevitably enters the economic allocation of resources. The task is not simply to eliminate “externalities,” but to compare the total social product obtainable under different arrangements of rights and institutions.
That is the enduring message of the paper.
The noisy confectioner and the doctor — one-page treatment
The case Coase discusses, Sturges v. Bridgman (1879), is particularly useful because it overturns the intuitive way most people initially think about nuisance.
A confectioner had operated machinery on his premises for many years. A doctor subsequently occupied neighboring premises. For some time there was no conflict. Eight years after moving in, however, the doctor constructed a consulting room at the end of his garden, immediately adjacent to the confectioner’s kitchen. The confectioner’s machinery created noise and vibration that interfered with the doctor’s work, including his ability to examine patients. The doctor sued and obtained an injunction preventing the offending use of the machinery.
The conventional interpretation is straightforward:
Confectioner → creates noise → harms doctor → confectioner should stop.
Coase says that this misses the economic problem.
Suppose the confectioner earns an additional $10,000 from operating the noisy machinery while the noise costs the doctor $20,000 in lost income. Stopping the machinery destroys $10,000 but prevents $20,000 of damage. The higher-value outcome is to stop the machinery.
Now reverse the numbers. Suppose the machinery produces $20,000 of additional confectionery income while the doctor suffers only $10,000 of loss. Prohibiting the machinery now destroys more value than it preserves.
Notice what has happened. The physical facts have not changed. The confectioner still creates the noise. But knowing who physically causes the nuisance does not tell us what the economically desirable solution is.
Coase then introduces property rights and bargaining.
If the doctor has the legal right to quiet, the confectioner must stop. But if continuing the machinery is worth $20,000 to the confectioner while quiet is worth only $10,000 to the doctor, there is room for a bargain. The confectioner could pay the doctor, say, $15,000 for permission to continue. Both are better off.
Now suppose the confectioner has the legal right to make the noise. If quiet is worth $20,000 to the doctor but operating the machinery is worth only $10,000 to the confectioner, the doctor could pay the confectioner $15,000 to stop. Again both are better off.
Thus, with costless bargaining, the activity with the greater economic value survives regardless of which party initially receives the legal right. The legal decision changes who pays whom—and therefore the distribution of wealth—but not the efficient use of the resources.
But now introduce transaction costs, and the result becomes much more interesting.
Imagine instead that the noisy activity affects 500 neighboring households. Identifying everyone’s loss, determining legitimate claims, negotiating with every household, preventing strategic holdouts, drafting agreements, monitoring compliance, and enforcing the resulting contracts may cost more than the economic gain obtainable from bargaining.
The theoretical bargain then doesn’t happen.
At that point, the legal rule matters greatly. Giving the doctor a right to quiet and giving the confectioner a right to make noise can produce different real-world outcomes because private bargaining may be too expensive to rearrange the rights afterward.
This is the part of Coase that is sometimes lost when the paper is reduced to “people can bargain away externalities.” Coase’s practical lesson is almost the opposite:
Once transaction costs exist, institutions matter.
The policymaker or court therefore needs to think beyond “Who caused the nuisance?” The better questions are: What are the alternative arrangements? What value is lost under each? Can the parties realistically bargain? What would bargaining cost? And which legal or institutional arrangement minimizes the total social cost?
The noise example is therefore one of the clearest demonstrations in the paper of Coase’s broader approach to transaction-cost economics.
Reference
Coase, Ronald H. (1960). “The Problem of Social Cost.” Journal of Law and Economics, 3, 1–44.