It is obvious to both casual and formal students of the subject that something is rotten with the state of economics. In the physical world, the presence of rot is often evident long before its source is clear. If the economics profession were a barrel of apples, an approaching bystander could sense spoilage from a considerable distance. In these cases, it is just as important not to discard a healthy apple as it is to cull a spoiled one. To help us tell the difference, we need to examine a specimen of each, side by side.
The Concept of Social Cost and the “Coase Theorem”
No better illustration of what is right with economics exists than that of Nobel Laureate Ronald Coase’s famous article, “The Problem of Social Cost,” published in 1960. Coase was writing in opposition to two long- and strongly held beliefs. First was the tradition in the common law of nuisance that required government to assign and enforce property rights to enforce equity. Second, economists have long supported the use of taxes as a means of discouraging producers from imposing costs on third parties.
Coase developed a hypothetical example involving two neighboring landowners, a farmer and a cattle rancher. The rancher’s cattle were allowed to roam at will and trampled the crops of the farmer. The traditional legal approach in such cases is for the law to specify the legal rights, remedies and obligations of all parties. For over half of the 20th century, economists followed the lead of A. C. Pigou, who considered such activities to be an external cost; that is, a cost imposed on the farmer by the rancher even though the farmer was not directly involved in the production or consumption of cattle at the wholesale level. In Coase’s example, an additional head of cattle would cause $300 worth of damage to the farmer’s crops.
According to Pigou, the fact that the rancher’s production activities creates costs not just for himself and his business but for a third party makes this case special. The costs felt by the rancher are the private costs of production, while those felt by third parties are the external costs. The sum of both these types of costs is the social cost of production.
According to Pigou, the solution to the problem was to “internalize the externality;” e.g., make the external cost part of the decisionmaking process by levying a tax equal to its magnitude on the production of cattle. He claimed that this would cause ranchers to reduce their production of cattle to an appropriate extent. To continue Coase’s example, suppose that the additional head of cattle returns $200 in revenue to the rancher. The $300 per-unit tax will make it too expensive for the rancher to raise it; thus, cattle ranching will stop short of this level of production.
But Coase demonstrated that this same outcome would occur even without the Pigovian tax. Suppose, for example, the law allows the rancher’s cattle to roam and does not require damages to be paid to the farmer. In that case, the farmer has an incentive to pay the rancher any amount less than $300 in order to prevent the raising of that marginal head of cattle. And the rancher has an incentive to accept any amount greater than $200 in order to forego that production, since he will net more money by doing so. By paying some amount of money between $200 and $300 to the rancher, the farmer can make them both better off than by watching his crops be destroyed by the marginal head of cattle. It is clear that the marginal head of cattle will not be raised.
Alternatively, suppose that the law stipulated that the rancher was liable for damages caused by his cattle. Now the rancher would be forced to pay $300 to the rancher if that marginal head of cattle were raised. But the rancher would earn only $200 from the additional production. So why would he do it? The answer is that he wouldn’t; the result – in terms of cattle raised and crops left undestroyed – would be the same. The assignment of property rights (or a Pigovian tax) affects the distribution of income or wealth between rancher and farmer, but not the allocation of resources between ranching and farming.
Coase’s result has become known as the “Coase Theorem.” (Interestingly enough, it got its name from George Stigler, who advertised the generality of the result and named it in Coase’s honor.)The Theorem says that – under the right circumstances – Pigou’s distinction between private cost and social cost does not exist because there will be no external costs. In practical terms, it argues against using Pigovian taxes to internalize externalities. For one thing, the taxes would distort other markets since it would be necessary to raise the tax revenue by taxing the production or consumption of some other good. And in real life, individuals like farmers and ranchers have a pretty good idea what their actual costs are and have very strong incentives to negotiate to reach the best outcome. But people in government have almost no idea what the particular costs of hundreds of thousands of individual businesses are and have little or no incentive to levy the correct taxes necessary to duplicate the Coase result.
There is one catch to the Coase Theorem, though. The people involved – “rancher” and “farmer” – have to be able to get together and negotiate; those are the “right circumstances” stipulated above. Economists formalize this by saying that “transactions costs are zero.” This is an overstatement, but the costs of successful negotiation must be low enough to make it feasible.
The Coase Theorem and the Bad Apples of Economics
The Coase Theorem (like the work of Ronald Coase in general) is one of the most-often cited of all economic propositions. Professional economists like it because it yields a categorical solution to a wide variety of problems, and the solution is economic in character. They also like it because they find its baggage – the assumption of “zero transactions costs”- very appealing.
This assumption is likewise categorical and precise. It makes the problem at hand easier to deal with. And – this is the heart of the matter – it avoids the complications of not assuming transactions costs to be zero. The existence of positive transactions costs makes the Coase conclusion problematic. In order to resolve the issue, the economist must inquire into the nature and size of transactions costs. This stirs up a hornet’s nest of pesky potential problems. Now the economist has stepped out of his neat, clean, determinate world of mathematical precision and into a messy, dirty, indeterminate world of subjective, qualitative evaluation and argument. Ugh!
The fact that, in reality, transactions costs are never zero hasn’t killed economists’ love affair with the Coase Theorem. It is a lot easier to explain away an assumption made “for purposes of exposition,” “for clarity,” because “economists deal in clean theoretical concepts that allow them to exclude extraneous ideas” or for some nobler academic purpose than it is to grapple with the grubby detail of real-world markets. Data are hard to come by and, even when available, often inaccurate. The public is impatient with qualification and ambivalence. Academic promotion and tenure are dependent on publication, and ambiguity does not dress a submission for success in the world of academic journals.
“Zero transactions costs” and categorical results puts the economist in control. It nourishes his fantasy of being a philosopher king who manipulates reality in behalf of humanity. The uncertainty of positive transactions costs kicks the economist off his pedestal and down into the dirt with ordinary mortals. It is humbling as well as professionally risky.
Why Coase is not a Fan of the Coase Theorem
Ironically, Ronald Coase himself does not approve of the promiscuous resort to the Coase Theorem. That does not mean that he repudiates it; after all, he originated it and defended it in debate against the best minds at the University of Chicago in 1960. But much as Alfred Nobel regretted the uses to which the world put his invention of dynamite, Coase bemoans the fact that the economics profession has misused his Theorem.
Coase analyzed production in a world of zero transactions costs in order to stimulate research into the importance of transactions costs in actual markets – much as a science teacher might contrast an environment of zero gravity with one encumbered by gravitational pull. His emphasis on the subject began a quarter-century earlier with the publication in 1937 of “The Nature of the Firm.” In this, his first major article, he attributed the existence of business firms to the existence of transactions costs.
Why don’t households either produce all goods and services internally or purchase them directly from each other? Because it is usually (though not always) cheaper to organize a good’s production under a single rubric which provides a common source of output. It is the number, size and scope of transactions costs – the “running around” and “arranging” that households would have to do in the absence of business firms – that makes it so.
The last thing in the world Coase wanted was to transport economic theory to an alien planet where there were no transactions costs. Here on Earth, transactions costs are ubiquitous. For that matter, so are “external costs.” When a firm builds a factory, there is every possibility that the ensuing production process will impinge on the lives of others through the creation of effects such as smoke, discharges and noise.
One possibility is that markets may reflect these effects, just as predicted by the Coase Theorem. Another possibility is that the costs of transactions between the business owning the factory and the surrounding inhabitants (and other, more distant, parties) preclude the negotiations necessary to a Coase-Theoretic outcome. But this second set of circumstances does not make a conclusive, or even a prima facie, case for action by government. The high cost of transactions does not bode well for an attempt by uninformed government regulators to change the outcome without imposing costs greater than the benefits available.
As a practical matter, Coase sees “a prima facie case against intervention,” bolstered by “the studies on regulation which have been made in recent years in the United States, ranging from agriculture to zoning, which indicate that regulation has commonly made matters worse… .” But viewing the matter formally, Coase recognizes that assuming zero transactions costs and assuming that positive transactions costs make government intervention mandatory are two different ways of assuming the problem away. If economists study actual markets, they must study transactions costs as an empirical driving force in those markets.
The Costs of Government Action
Foremost among transaction costs is the cost of government action. Throughout the 20th century, economists have taken a weirdly asymmetrical view of the private sector and government. In the example above, the “external costs” cited are commonly viewed as somehow foreign or extrinsic to the production process, as if they could be omitted at will or at whim. Thus, their very presence makes the motives and intentions of the producer suspect. Government is regarded as a benign organic unity, miraculously untainted by self-interest, whose objective function is some inadequately defined conception of aggregative welfare.
Coase casts a gimlet gaze upon the role in which economists have placed themselves. They are the philosopher kings, making government policy on their academic blackboard. “All the information needed is assumed to be available and the teacher plays all the parts. He fixes prices, imposes taxes and distributes subsidies (on the blackboard) to promote the general welfare.”
Alas, “there is no counterpart to the teacher within the real economic system. There is no one who is entrusted with the task that is performed on the blackboard. In the back of the teacher’s mind (and sometimes in the front of it) there is, no doubt, the thought that in the real world the government would fill the role he plays. But there is no single entity within the government which regulates economic activity in detail, carefully adjusting what is done in one place with what is done elsewhere. In real life, we have many different firms and government agencies, each with its own interests, policies and powers.”
Coase derides “blackboard economics,” noting that while government intervention may be costless on the blackboard, it is both highly costly and questionably motivated in reality. Meanwhile, the “external costs” that are singled out so invidiously by economists are part and parcel of production. The real choice we face is how to minimize the costs of production when both external costs and transaction costs are included. To leave them out would be tantamount to rejecting production outright.
Ronald Coase, Centenarian
Ronald Coase was born on December 28, 2010. Although his first seminal article was published in 1937, he didn’t obtain his doctorate until 1951. His crowning work on social cost did not appear until 1960, when he was almost fifty years old. Thus, he was a late bloomer.
His famous articles number barely a dozen. He employs no abstruse mathematical or statistical tools. Yet his standing is that of a giant in the profession.
Coase practices what he preaches. In recent years, he has organized a society to study the markets and institutions of China. His book on the economies of China and Vietnam, whose publication was expected in the summer of 2010 but is still delayed, has been long-awaited by his legion of colleagues and admirers.
Ronald Coase is now well into his 102nd year.