Market Failure and Public Goods

“Market failure” is the standard economic argument that a free market, left to itself, produces inefficient outcomes in specific, well-defined cases — public goods, externalities, natural monopoly, and asymmetric information — and that government action can improve on them. It is the most respectable case for the state in purely economic terms, made not by socialists but by mainstream neoclassical economists. This entry states each market-failure argument fairly and gives the Austrian and libertarian replies: that the framework judges the real market against an imaginary benchmark, that many “failures” are really failures to define property rights, that monopoly is usually the state’s own creation, and that the government proposed as the remedy has failures of its own.

The framework

In standard welfare economics the benchmark of success is perfect competition, under which, given certain conditions, markets reach a Pareto-efficient allocation — no one can be made better off without making someone worse off. A “market failure” is a systematic departure from that ideal of the specific kinds catalogued below, and it is taken to create at least a prima facie case for corrective intervention: a tax, a subsidy, regulation, or direct government provision. The canonical statements are A. C. Pigou on externalities, Paul Samuelson on public goods, George Akerlof on asymmetric information, and — on the other side of the argument — Ronald Coase on transaction costs and bargaining over external costs. Four types of failure do most of the work.

Public goods

A public good is non-excludable (you cannot easily stop non-payers from enjoying it) and non-rival (one person’s use does not diminish another’s). National defense, clean air, and basic research are the stock examples. The non-rivalry half is Samuelson’s 1954 definition: collective consumption goods are those “which all enjoy in common in the sense that each individual’s consumption of such a good leads to no subtraction from any other individual’s consumption of that good”.

His negative claim is the one that does the work. Where goods are private, each person is “motivated to do the signalling of his tastes needed to define and reach the attainable-bliss point” — but for collective consumption:

“However no decentralized pricing system can serve to determine optimally these levels of collective consumption. Other kinds of “voting” or “signalling” would have to be tried.”

— Paul A. Samuelson, “The Pure Theory of Public Expenditure”

It is worth being exact about what that establishes, because the usual gloss overshoots it. Samuelson argues that prices cannot reveal the optimal level, and leaves what should replace them open — other kinds of voting or signalling “would have to be tried”. He does not show that a taxing state is the unique or the best remedy; the step from “the price system will not do this” to “therefore the state must” is one his readers supply.

The reply. Rothbard makes the case in Man, Economy, and State; Power and Market only gestures at it, dismissing the “collective goods” defence of state provision as a “fallacious theory” that “is refuted elsewhere” and footnoting the longer treatment. He argues that the “collective goods” category is far narrower than claimed and that much of it is in fact excludable and privately provided; that the argument would justify state provision of almost anything with spillover benefits; and that there is no way to know the “efficient level” of a good once market prices for it are abolished. The hardest case — large-scale defense — is treated in the criticisms of anarcho-capitalism, where it is the single strongest objection to a stateless order.

Externalities

An externality is a cost or benefit falling on third parties who are not part of a transaction — pollution is the classic negative case, a well-kept garden the positive one. Where costs are external, the argument runs, the market over-produces the harmful activity, and a corrective (“Pigouvian”) tax is needed to align private with social cost.

The name is Pigou’s because the argument is. He begins from a qualified observation: in general, “industrialists are interested, not in the social, but only in the private, net product of their operations”. Self-interest equalises the private returns without equalising the social ones, and the conclusion follows:

“When there is a divergence between these two sorts of marginal net products, self-interest will not, therefore, tend to make the national dividend a maximum; and, consequently, certain specific acts of interference with normal economic processes may be expected, not to diminish, but to increase the dividend.”

— A. C. Pigou, The Economics of Welfare

Note what the claim is and is not. Pigou does not argue that markets fail generally; he isolates the cases where a wedge opens between private and social net product, and it is only there that interference is expected to pay. In this chapter the corrective he names is bounties and taxes in general terms, not a rate calibrated to the external cost. The libertarian reply below contests the diagnosis of the wedge, not the logic that follows from it.

The reply. The Austrian and property-rights answer is that most externalities are really missing or unenforced property rights: pollution is a trespass or nuisance — an invasion of someone’s person or property — best handled by defining and defending those rights and letting the parties bargain or litigate, not by having a regulator guess the “right” tax. It links the problem back to property rights.

Coase’s 1960 paper is the usual authority for that reply, and it supports half of it. His first move is to deny that harm runs in one direction, which is what makes the tax framing look natural:

“The question is commonly thought of as one in which A inflicts harm on B and what has to be decided is: how should we restrain A? But this is wrong. We are dealing with a problem of a reciprocal nature. To avoid the harm to B would inflict harm on A. The real question that has to be decided is: should A be allowed to harm B or should B be allowed to harm A? The problem is to avoid the more serious harm.”

— R. H. Coase, “The Problem of Social Cost”

That is a real blow to Pigou’s framing, though not the one it is often taken for. Coase does not deny that A harms B; he denies that restraining A settles the economics, because preventing the harm to B imposes a cost on A. What the tax framing hides is the comparison — two incompatible uses, and a question of which is worth more.

The half the libertarian retelling tends to drop is the other one. Costless bargaining is his simplifying case, not his conclusion — he introduces the section that abandons it by calling the zero-cost assumption “a very unrealistic assumption”. Once transactions cost something:

“In these conditions the initial delimitation of legal rights does have an effect on the efficiency with which the economic system operates. One arrangement of rights may bring about a greater value of production than any other. But unless this is the arrangement of rights established by the legal system, the costs of reaching the same result by altering and combining rights through the market may be so great that this optimal arrangement of rights, and the greater value of production which it would bring, may never be achieved.”

— R. H. Coase, “The Problem of Social Cost”

So Coase does not establish that defining rights and letting the parties bargain reaches the efficient outcome. He establishes that where transaction costs obstruct rearrangement, the rights the legal system starts with can determine what gets produced — a reason to care a great deal about how it assigns them, not an argument that any assignment will do. Critics press there: millions of diffuse victims of, say, global emissions cannot realistically bargain, and that is where the externality objection keeps its force. What Coase adds to that dispute is a rule for conducting it — compare the institutional arrangements actually available rather than a market against an ideal — and the rule cuts against a regulator assumed to be costless just as it cuts against a market assumed to be frictionless.

Natural monopoly

A natural monopoly is said to arise where economies of scale are so large that one firm can serve the whole market more cheaply than several — utilities and networks are the usual examples — so that competition is wasteful or self-eliminating, justifying a regulated or state-owned provider.

The reply. The libertarian response is that durable monopoly is typically a creature of government rather than of the market. Rothbard’s move is definitional before it is empirical — “The only viable definition of monopoly is a grant of privilege from the government” — which settles the usage rather than the frequency. How often durable monopoly arises without a grant is a separate, empirical question, and a definition cannot answer it. Free-market economists add that a market can remain contestable — that the threat of entry disciplines even a single incumbent. The argument carries its own conditions: entry and exit have to be genuinely free and reversible, which is exactly what is disputed for utilities and networks, where the sunk investment is the whole point. The firmer point is the statutory one, and the mail is its clearest case: the postal monopoly Rothbard describes was created by law, not by a cost curve. This ties natural-monopoly claims to the analysis of crony capitalism: a franchise sold as a remedy for market failure is often the privilege that created the monopoly.

Asymmetric information

If one side of a transaction knows more than the other — the used-car seller, the borrower, the insured — markets are said to fail. Akerlof’s 1970 paper states the mechanism: where buyers judge quality by a market statistic, “there is incentive for sellers to market poor quality merchandise, since the returns for good quality accrue mainly to the entire group whose statistic is affected rather than to the individual seller. As a result there tends to be a reduction in the average quality of goods and also in the size of the market.” His used-car illustration gives the result its name: “bad cars drive out the good because they sell at the same price as good cars”. Regulation and disclosure mandates are the usual policy responses. Licensing is on Akerlof’s own list of counteracting institutions, alongside guarantees, brand names and chains — “Licensing practices also reduce quality uncertainty. For instance, there is the licensing of doctors, lawyers, and barbers” — which is worth noting before either side claims his authority for it.

The reply. Akerlof himself puts both routes on the table, which is usually dropped by whichever side is quoting him. Because social and private returns diverge here, he writes, “in some cases, governmental intervention may increase the welfare of all parties. Or private institutions may arise to take advantage of the potential increases in welfare which can accrue to all parties.” He qualifies the second route in the next sentence, and the qualification is not one libertarians usually quote: “By nature, however, these institutions are nonatomistic, and therefore concentrations of power — with ill consequences of their own — can develop.” The libertarian case takes the second route anyway: markets generate their own remedies for information problems — repeat purchase, brands and reputation, word of mouth, and consumer-research organizations — precisely because sellers profit by credibly signaling quality. They add that mandatory licensing tends to be captured by incumbents to raise rivals’ costs rather than to protect consumers (again the crony-capitalism pattern). Fraud, where it occurs, is a violation of rights the law should punish, not a market outcome to be managed.

The deeper objection to the framework

Beyond the case-by-case replies, the Austrian criticism is that the whole framework rests on a flawed benchmark. Perfect competition is not a state real markets could occupy; the free market is a discovery process, not an equilibrium, and judging it against a frictionless ideal guarantees it will always register as a failure. The same knowledge problem and calculation problem that afflict a central planner afflict the regulator who is supposed to compute the optimal tax or output. And public choice adds that the government offered as the cure is itself run by self-interested actors subject to capture — so a demonstrated market imperfection does not establish that intervention will improve matters. The honest comparison is between imperfect markets and imperfect states, not between real markets and an ideal planner.

Where it is contested

The libertarian replies are on firmest ground where a statutory origin can be pointed at — the postal monopoly is the clean case — and on information, where Akerlof himself allows that private institutions may capture the available gains. They are hardest pressed on large-scale defense and on global, diffuse externalities such as greenhouse emissions, where the number of parties makes the bargaining Coase describes impractical and the free-rider structure Samuelson describes is at its most acute. Both are judgements about where the arguments bite, not claims about where the profession stands. The Austrian response to both is that intervention’s own failures may still be worse, and that the calculation and knowledge problems cut against confident correction. This is the live frontier of the debate, and neither side of it is settled here.

Sources

The Pure Theory of Public Expenditure (Full Text) — Samuelson 1954: the non-rivalry definition of a collective consumption good, and the claim that no decentralized pricing system can determine its optimal level. OCR of an image scan; quoted spans are continuous prose away from the equations, and each was audited against the unstripped OCR — the definition, the signalling sentence and the impossibility passage are contiguous, with no page number or running head inside any of them
The Market for ‘Lemons’ (Full Text) — Akerlof 1970: the adverse-selection mechanism, the bad-cars-drive-out-good result, and the even-handed conclusion in his section I that either governmental intervention or private institutions may capture the available gains. OCR of an image scan; the mechanism passage, the policy sentence and the bad-cars clause were audited the same way and are each contiguous
The Problem of Social Cost (Full Text) — Coase 1960: the reciprocal-harm argument against Pigou’s framing, and the passage on positive transaction costs in which the initial delimitation of legal rights does determine efficiency. OCR scan. Both quoted spans were audited against the unstripped extraction, which retains the running heads and page numbers: the reciprocal-harm passage sits wholly on p. 2 and the transaction-cost passage wholly on p. 16, with no page break, page number or running head inside either
The Economics of Welfare — Part II, Chapter IX: Divergences Between Marginal Social Net Product and Marginal Private Net Product - A. C. Pigou’s own statement of the externality argument: the private/social net product distinction, and the conclusion that where the two diverge, specific interference may raise the national dividend rather than lower it
Power and Market: Government and the Economy (Full Text Aggregate) - Rothbard on collective goods, monopoly as a government grant, and the critique of intervention as a remedy for market imperfection
Man, Economy, and State: A Treatise on Economics (Full Text Aggregate) - Rothbard’s fuller treatment of collective goods, neighborhood effects, and external benefits

Galtland Index

Explorer

Market Failure and Public Goods

Market Failure and Public Goods

The framework

Public goods

Externalities

Natural monopoly

Asymmetric information

The deeper objection to the framework

Where it is contested

See Also

Sources

Graph View

Table of Contents

Backlinks