8.6 Efficient provision of public goods

Efficient Provision of Public Goods

Public goods are non-excludable and non-rivalrous, which means markets on their own will almost always underprovide them. Efficient provision asks: how much of a public good should society produce, and how do we pay for it? The answer hinges on Samuelson's condition, aggregate demand through vertical summation, and the Lindahl equilibrium.

Concept and Principles

Efficient provision of a public good means producing the quantity that maximizes social welfare. Because everyone consumes the same unit of a public good simultaneously (non-rivalry), you can't find the market equilibrium the way you would for a private good. Instead, you add up every individual's marginal benefit from an additional unit and compare that sum to the marginal cost of producing it.

This leads to Samuelson's condition: the sum of all individuals' marginal rates of substitution (between the public good and a private good) must equal the marginal rate of transformation.

$\sum_{i=1}^n MRS_i = MRT$

Why does this differ from private goods? For a private good, efficiency requires each individual's $MRS$ to equal the $MRT$ separately ($MRS_i = MRT$ for all $i$). For a public good, because everyone shares the same unit, you sum the $MRS$ values across all consumers. This is the key distinction to internalize.

The goal is Pareto efficiency: no reallocation can make someone better off without making someone else worse off. Private markets fail to reach this point for public goods because individuals can free-ride, consuming the good without paying.

Mathematical Framework

The Samuelson condition can be restated in terms of marginal benefits and costs. The optimal quantity $Q^*$ satisfies:

$\sum_{i=1}^n MB_i(Q^*) = MC(Q^*)$

where $MB_i(Q^*)$ is individual $i$'s marginal benefit at quantity $Q^*$, and $MC(Q^*)$ is the marginal cost of production.

Vertical summation of demand curves. This is where public goods diverge sharply from private goods:

• For private goods, you sum individual demand curves horizontally (add up quantities at each price).
• For public goods, you sum individual demand curves vertically (add up willingness to pay at each quantity).

You sum vertically because every consumer enjoys the same quantity, so the relevant question is: what is the total value society places on one more unit?

Worked example. Suppose there are two individuals with marginal benefit functions $MB_1(Q) = 10 - Q$ and $MB_2(Q) = 8 - Q$, and the marginal cost is $MC(Q) = 6$.

1. Sum the marginal benefits vertically: $MB_{total}(Q) = (10 - Q) + (8 - Q) = 18 - 2Q$

2. Set the sum equal to marginal cost: $18 - 2Q = 6$

3. Solve: $2Q = 12$, so $Q^* = 6$

4. At $Q^* = 6$: $MB_1 = 4$ and $MB_2 = 2$. These are the Lindahl prices for each individual, and they sum to $6 = MC$.

Lindahl equilibrium. A Lindahl equilibrium assigns each individual a personalized price (called a Lindahl price) equal to their own marginal benefit at the efficient quantity. Each person pays exactly what that unit is worth to them, and the sum of all Lindahl prices covers the marginal cost. As the example above shows, person 1 pays 4 per unit and person 2 pays 2 per unit. In theory, this achieves efficiency and voluntary participation. In practice, it requires knowing everyone's true preferences, which brings us to the next set of problems.

Conditions for Efficient Public Good Provision

Information and Preference Revelation

The biggest obstacle to efficient provision is figuring out how much people actually value the public good. Because public goods are non-excludable, individuals have a strategic incentive to free-ride: understate their true willingness to pay, hoping others will fund the good anyway.

Several mechanisms try to solve this:

• Lindahl pricing works in theory but requires truthful preference reporting, which people have no incentive to provide voluntarily. If you know your stated valuation determines your tax share, you'll underreport.
• Vickrey-Clarke-Groves (VCG) mechanism uses a tax design that makes truthful reporting each person's dominant strategy. Each individual's tax depends on the cost their participation imposes on others (the "pivotal" cost), so lying can only make them worse off. The mechanism is strategy-proof but can run a budget deficit, limiting its practical use.
• Contingent valuation surveys ask people hypothetical questions about willingness to pay (e.g., "How much would you pay to preserve this wetland?"). These are widely used for environmental goods but suffer from hypothetical bias, since people may overstate values they'll never actually have to pay.

Accurate preference aggregation also requires political and institutional structures capable of translating revealed preferences into policy.

Cost and Funding Considerations

Once you know the efficient quantity, you still need to fund it. The marginal cost of serving one additional user of a pure public good is zero (that's what non-rivalry means), but the total cost of producing the good is not.

Common funding approaches and their trade-offs:

• Taxation is the most common method. The challenge is raising revenue without creating large deadweight losses. Lump-sum taxes are efficient but regressive; income or consumption taxes are more equitable but distort behavior.
• Benefit principle suggests those who benefit more should pay more, which aligns with Lindahl pricing in spirit. Toll roads are a rough real-world approximation, though tolls also introduce excludability (converting the good into a club good).
• Congestion pricing (e.g., London's congestion charge) applies user fees to goods that become rivalrous at high usage levels, helping manage demand for quasi-public goods.
• Public-private partnerships can fund infrastructure projects by combining government oversight with private-sector efficiency, though they introduce their own incentive problems around risk-sharing and contract design.

The core tension: any tax that isn't lump-sum creates some distortion, so funding public goods always involves an equity-efficiency trade-off.

Public vs. Private Good Provision

Objectives and Characteristics

The table below highlights the structural differences:

A private firm producing national defense would undersupply it because it can't charge each beneficiary. Private goods don't face this problem: if you don't pay for your groceries, you don't eat. The positive externalities inherent in public goods are exactly what makes market provision break down.

Market Dynamics and Intervention

The free-rider problem is the central market failure for public goods. Because no one can be excluded, each person has an incentive to let others pay. This leads to systematic underprovision (or zero provision) by private markets.

Government intervention is the standard response: taxation funds national defense, public parks, and basic research. But not all public goods require top-down provision:

• The Coase theorem suggests that if property rights are well-defined and transaction costs are low, private parties can negotiate efficient outcomes without government involvement. A gated community hiring private security is a small-scale example. For large-population public goods (clean air, national defense), transaction costs make Coasean bargaining impractical because coordinating millions of people is effectively impossible.
• The Tiebout model proposes that local public goods can be efficiently provided through competition among jurisdictions. People "vote with their feet" by choosing to live in communities whose tax-and-services bundle matches their preferences. This works best when there are many jurisdictions, people are mobile, and there are no spillovers across borders. It doesn't apply to national or global public goods.

Challenges in Efficient Public Good Provision

Preference Aggregation and Information Problems

Even with well-designed mechanisms, several deep problems remain:

• Free-riding persists in most real-world settings. Experimental evidence shows people do contribute to public goods voluntarily, but typically well below the efficient level, and contributions tend to decay over repeated interactions.
• Arrow's Impossibility Theorem proves that no voting rule can simultaneously satisfy a small set of reasonable fairness criteria (unrestricted domain, Pareto, independence of irrelevant alternatives, non-dictatorship) when aggregating preferences over three or more alternatives. This means there's no perfect democratic procedure for deciding public good levels.
• Preference intensity is lost in simple majority voting. A passionate minority that deeply values a public good can be outvoted by an indifferent majority, leading to underprovision relative to the efficient level.
• Behavioral biases complicate valuation. Present bias causes people to undervalue long-term public goods (climate mitigation). Loss aversion makes people resist paying for goods whose benefits are diffuse and hard to see.

Political and Long-term Considerations

Political realities introduce further distortions:

• Political business cycles can cause over-provision of visible public goods (new construction) near elections and under-provision of less visible ones (infrastructure maintenance).
• Rent-seeking diverts resources from productive public good provision toward lobbying. Interest groups may push for public goods that benefit them disproportionately rather than those with the highest social return.
• Time inconsistency plagues long-term projects. Politicians may commit to a 30-year infrastructure plan but face pressure to redirect funds to short-term priorities once in office.
• Intergenerational equity is especially difficult. Public goods like climate stability or debt-free fiscal policy benefit future generations who have no voice in current decisions. Standard cost-benefit analysis, which discounts future benefits, tends to undervalue these goods.
• Global public goods like climate change mitigation and ocean conservation face a tragedy-of-the-commons dynamic at the international level. No global government can enforce contributions, so each country has an incentive to free-ride on others' efforts.