Public Goods

What is a public good?

A public good has two defining characteristics, and both must hold:

▶

• Non-rivalrous: one person's consumption does not reduce the amount available to anyone else. My watching a lighthouse beam leaves the beam undiminished for every other ship.
• Non-excludable: it is impossible, or prohibitively costly, to prevent people who have not paid from consuming it.

Standard examples: national defence, street lighting, flood defences, lighthouses, the legal system.

"Public good" is a technical term about the good's characteristics, not about who provides it. A good funded and produced by government is not automatically a public good. State schools and public hospitals are both rival and excludable, so they are not public goods in the economic sense, even though they are publicly provided. Getting this right is a frequent discriminator.

The classification grid

Two contrasts matter most:

• Public goods versus common pool resources. Both are non-excludable, but public goods are non-rival while common pool resources are rival. That difference is why they fail in opposite directions: public goods are under-provided because nobody will pay, while common pool resources are over-used because everyone consumes without cost. Confusing the two is one of the most damaging errors in Unit 2.
• Public goods versus merit goods. A merit good (education, healthcare) is under-consumed because consumers undervalue its private benefit or cannot afford it, and it generates positive externalities. It is usually rival and excludable, so it is not a public good. Merit goods are under-consumed; public goods are not provided at all.

Quasi-public goods are the reason most real examples are argued about. A road is non-rival and non-excludable when empty, but becomes rival once congested and excludable once tolling technology exists. So the classification is not a fixed property of the object, it depends on the circumstances and the available technology. Saying so about the specific good in the question is a strong analytical move, and it is usually what a question naming a road, a beach or a park is inviting.

Demerit goods complete the set: goods over-consumed because consumers underestimate the private harm, and which generate negative externalities, tobacco, alcohol, gambling. They belong with externalities in 2.8, not here, but the pairing is worth holding: merit goods are under-consumed, demerit goods over-consumed, public goods not produced.

The free-rider problem

Because a public good is non-excludable, anyone can consume it whether or not they contributed. The rational individual therefore waits for others to pay and consumes for free.

1. Since the good is non-excludable, each consumer can benefit without paying
2. each has an incentive to understate their willingness to pay and free-ride
3. if everyone reasons this way, no revenue is collected
4. no profit-seeking firm will supply it
5. the market provides nothing at all, despite the good being genuinely valued by society.

This is the crucial distinction from other market failures. With an externality, the market produces the wrong quantity. With a pure public good, the market typically produces zero, which is a complete market failure, and a total welfare loss equal to the entire social benefit that could have been created.

Note that everyone behaves rationally. The failure is structural, not a matter of selfishness. Each individual's reasoning is correct given what everyone else is doing, and the collective result is worse for all of them, the same logic as the prisoner's dilemma, and the same reason it cannot be solved by asking people to be more public-spirited.

Government responses

• Direct provision, funded from general taxation. Taxation solves the free-rider problem by making contribution compulsory rather than voluntary. This is the standard response for defence, policing and flood defences.
• Contracting out: the government funds the good but pays private firms to produce it, aiming to combine guaranteed provision with private-sector efficiency.
• Making the good excludable where technology allows, converting it into a club good. Electronic tolling turns an uncongested road into something a private firm could charge for.

The difficulties

• How much to provide? There is no market price to signal how much society values the good, so the government must estimate demand through cost–benefit analysis, surveys or political processes, all imperfect.
• Opportunity cost. Resources devoted to a public good are unavailable elsewhere.
• Government failure. Provision may be inefficient, politically driven, or captured by particular interests.
• Valuation is genuinely hard. How much is a marginal unit of national defence worth? The absence of an answer is not a failure of effort; it follows from the good's characteristics.

Cost–benefit analysis is the standard tool, and its limits are examinable. It attempts to value all social costs and benefits, including those with no market price, and to discount future ones to present value. The judgements inside it, what a life is worth, what discount rate to apply to benefits fifty years away, drive the answer, which is why two honest analyses of the same project can disagree.

Worked example

A coastal town is considering building a flood defence wall.

Why the market will not provide it. Once the wall exists, it protects every property in the town, nobody can be excluded, and one household's protection does not reduce another's. So each resident reasons that if their neighbours fund it; they will be protected anyway; and if the neighbours do not, one household's contribution is insufficient. Everyone waits. No wall is built, even though the total value of protection to the town exceeds the cost.

The response. The government funds it from taxation, which removes the option of free-riding, and uses cost–benefit analysis to decide whether the wall's total social benefit exceeds its cost, and how high to build it.

Evaluation. Provision is justified where the estimated social benefit exceeds the cost, but the estimate is uncertain: valuing avoided flood damage decades ahead, and the wellbeing of future residents, involves significant judgement. There is an opportunity cost, the same funds could build a hospital. And the benefits are geographically concentrated while the taxation is broad, raising an equity question about who pays.

Check whether it is genuinely pure. If the wall protects only the seafront properties, the benefit is not evenly non-rival across the town, and a targeted local levy becomes defensible. Most real "public goods" are impure in some such way, and identifying how is usually where the top marks sit.

Real-world examples

• Vaccination programmes are often cited here but belong with merit goods: the vaccine dose itself is rival and excludable. What is non-rival and non-excludable is the herd immunity it produces, a genuine public good arising from a private one, which is precisely why the market under-provides it.
• Basic scientific research is close to a pure public good: once published, knowledge is non-rival and hard to exclude anyone from. That is the standard justification for public research funding, and for patents, which create artificial excludability to restore the incentive.
• Street lighting and flood defence remain the cleanest textbook cases because neither exclusion nor rivalry is realistically achievable.

Common exam mistakes

• Defining a public good as "a good provided by the government".
• Giving only one of the two characteristics. Both non-rivalry and non-excludability are required.
• Confusing public goods with merit goods (rival and excludable, under-consumed), the single most common error here.
• Confusing public goods with common pool resources (rival, non-excludable, over-used).
• Saying the market under-provides a public good when for a pure public good it provides none.
• Treating the classification as fixed, when congestion or new technology can change it.
• Treating government provision as free of cost or of difficulty.
• Presenting cost–benefit analysis as objective.

Exam technique

Define both characteristics explicitly and apply each to the specific good in the question. Show why this good is non-rival and why this good is non-excludable. That application is worth more than a memorised definition.

Then run the free-rider chain to its conclusion: non-excludable → free-riding → no revenue → no private provision → complete market failure.

Where the good is impure, a road, a park, a beach, say so and say under what conditions it changes category. That is analysis rather than recall.

For evaluation, the strongest routes are: how much to provide in the absence of a price signal, opportunity cost, government failure, whether the good is genuinely pure, the assumptions inside any cost–benefit analysis, and who pays versus who benefits.

Quick revision

• Public good = non-rivalrous AND non-excludable.
• Not a synonym for "government-provided".
• Free-rider problem → no revenue → the market provides zero → complete market failure.
• The free-rider problem is structural, not selfishness, each person's reasoning is individually correct.
• Merit good: rival, excludable, under-consumed. Demerit good: over-consumed. Common pool resource: rival, non-excludable, over-used.
• Quasi-public goods change category with congestion or technology.
• Response: direct provision funded by taxation, which makes contribution compulsory.
• Difficulty: no price signal, so the optimal quantity must be estimated, and CBA is full of judgements.

Not the topic you were looking for? Describe what you are stuck on in your own words and we will take you to the notes that answer it.