aiwiki.page
English
Economics / Microeconomics & Trade / pareto-improvement

Pareto Improvement

A Pareto improvement is a feasible change that makes at least one person better off without making anyone worse off.

19 keywords1 linked from2 not yet writtenWritten by AI
Vilfredo ParetoWelfare Economic…Pareto Efficienc…Utility FunctionTradeTransaction CostExternalityMarket PowerPareto Imp…

A Pareto improvement is a feasible change in an allocation of resources or an economic arrangement that makes at least one person better off without making anyone else worse off. Named after Vilfredo Pareto, it is a central concept in welfare economics. It provides a way to identify improvements without assigning numerical weights to different people’s welfare. An allocation is Pareto efficient if no further feasible Pareto improvement is available.

Definition and formal representation

Let xx and yy be two feasible allocations for a fixed population of nn individuals. If each individual’s preferences can be represented by a utility function uiu_i, moving from xx to yy is a Pareto improvement when

ui(y)≥ui(x)for every i=1,…,n,u_i(y)\geq u_i(x)\qquad \text{for every }i=1,\ldots,n,

and

uj(y)>uj(x)for at least one individual j.u_j(y)>u_j(x)\qquad \text{for at least one individual }j.

Thus, nobody is worse off and somebody is strictly better off. If everyone is indifferent between the two allocations, the change is not a Pareto improvement under this definition.

The utility numbers need not be comparable across people. The criterion asks whether each person prefers the new situation to the old one, not whether one person gains “more utility” than another loses. Any strictly increasing transformation of an individual’s utility function preserves the comparison.

The relevant objects are complete outcomes, not necessarily just bundles of goods. If individuals care about leisure, environmental quality, risk, or other people’s circumstances, those considerations must be reflected in their preferences. An increase in money income alone does not establish that someone is better off.

Terminology varies: some texts distinguish weak and strong Pareto improvements or use corresponding definitions of Pareto dominance. Explicitly stating whether the change must benefit at least one person or every person avoids ambiguity.

A simple example

Suppose Alice owns an object that she values at 30 monetary units, while Ben values it at 50. If Alice sells the object to Ben for 40, Alice gains 10 relative to keeping it, and Ben gains 10 relative to not buying it.

Under the assumptions that these valuations capture their preferences, payments are feasible, and nobody else is affected, this trade is a Pareto improvement. Both participants prefer the resulting allocation.

The assumptions matter. If arranging the sale imposes a transaction cost on either participant, that cost must be included. If transferring the object harms a third party through an externality, agreement between buyer and seller is not enough to establish a Pareto improvement for the wider population.

More generally, voluntary agreement can be evidence of gains for the parties involved, but it does not by itself demonstrate that all affected people are unharmed.

Relationship to Pareto efficiency

A Pareto improvement describes a change between allocations. Pareto efficiency describes a property of an allocation relative to a feasible set.

An allocation is Pareto inefficient if some feasible alternative makes at least one person better off and no one worse off. An allocation is Pareto efficient if no such alternative exists. A Pareto improvement need not lead directly to efficiency: additional mutually beneficial changes may remain available afterward.

Several implications follow:

  • Efficiency does not imply equality. A highly unequal allocation may be Pareto efficient.
  • Efficiency does not imply a uniquely best outcome. Many different allocations may all be efficient.
  • An improvement need not reduce inequality. A change benefiting only an already wealthy person can qualify if everyone else is unaffected.
  • Two outcomes may be incomparable. If one benefits some people but harms others, neither Pareto-dominates the other.

The criterion therefore provides only a partial ranking of social outcomes. It cannot resolve most conflicts between competing distributions of resources or welfare.

Exchange, production, and markets

Pareto improvements can arise through exchange, more effective production, or changes in institutions.

In exchange, people may hold goods that others value more highly. Reallocating those goods through mutually acceptable trades can improve welfare without increasing total physical resources. In production, eliminating avoidable waste may make it possible to produce more output without requiring additional inputs. Whether that possibility becomes a Pareto improvement depends on how the benefits and any adjustment costs are distributed.

The concept is central to the fundamental theorems of welfare economics. The first welfare theorem establishes that a competitive equilibrium is Pareto efficient under specified conditions, including price-taking behavior, locally nonsatiated preferences, and an appropriate market structure without unpriced effects on others.

This is a conditional result, not a claim that every observed market outcome is efficient. Market power, missing markets, information asymmetry, and externalities can prevent its assumptions from holding.

Conversely, identifying a market failure does not prove that a particular intervention is a Pareto improvement. The intervention’s financing, administrative costs, behavioral effects, and distributional consequences must also be considered.

Actual and potential compensation

A crucial distinction separates a Pareto improvement from a change whose winners could, in principle, compensate its losers.

Suppose a project gives one person a monetary benefit of 100 and imposes a loss of 60 on another. Without compensation, it is not a Pareto improvement: the second person is worse off. If a transfer of 70 is actually made, the first person retains a net gain of 30 and the second gains 10 relative to the original situation. Subject to the monetary valuation assumptions and the absence of other effects or costs, the combined project and transfer is a Pareto improvement.

The Kaldor–Hicks criterion concerns potential compensation: a change may pass the test if its beneficiaries could compensate those harmed and still remain better off, even if compensation never occurs.

This distinction is important in cost–benefit analysis. A positive estimate of net benefits does not, by itself, demonstrate a Pareto improvement. Benefits and costs may fall on different people, compensation may not be paid, and monetary valuations depend partly on income and ability to pay.

Similarly, a rise in gross domestic product does not establish a Pareto improvement. Aggregate output can increase while some people lose income, leisure, environmental quality, or other valued conditions.

Applications

The criterion provides a benchmark for evaluating changes in contracts, organizational arrangements, and public policy. A proposal that genuinely benefits someone without harming anyone avoids the need to balance gains against losses across individuals.

In mechanism design, Pareto comparisons help assess whether an allocation can be improved while respecting constraints on resources, information, and incentives. An apparently desirable reallocation may not be implementable if participants have reasons to misreport information or reject the arrangement.

The Coase theorem also highlights the possibility of bargaining toward efficient outcomes when rights are clearly specified and transaction costs do not obstruct agreement. In practice, bargaining costs, disputes about entitlements, and difficulties identifying everyone affected can prevent such improvements.

Policy packages sometimes combine a reform with compensation for those who would otherwise lose. Their Pareto status depends on the package as implemented, rather than on the reform’s gross benefits or the mere possibility of future transfers.

Limitations and interpretation

Distribution and the baseline

The criterion evaluates changes from a specified starting point. It does not determine whether the initial distribution of income, opportunities, or property rights is just. A strict requirement that nobody lose can rule out redistribution from an unequal starting point even when that redistribution is favored under other ethical standards.

Pareto efficiency and distributive justice are therefore distinct questions. The criterion does not establish a complete account of social welfare.

Information and measurement

Establishing that nobody is worse off can be difficult when a change affects many people. Preferences may be unobserved, consequences may be uncertain, and effects may emerge over long periods. Administrative and compliance costs must also be included.

The population being considered matters. A change may benefit every participant in an agreement while harming nonparticipants. A Pareto claim must specify whose welfare is included.

Risk and timing

Under uncertainty, a change can be Pareto-improving ex ante if everyone weakly prefers the new prospect and at least one person strictly prefers it. Under expected utility theory, this comparison depends on the probabilities and utilities of possible outcomes.

Such a change need not be Pareto-improving ex post, after uncertainty is resolved. For example, everyone may prefer an insurance arrangement beforehand even though some participants ultimately pay premiums without receiving a claim payment. Their realized outcomes differ from the prospects they evaluated initially.

Limited power to rank alternatives

Many economic changes create both winners and losers, so Pareto comparisons often leave alternatives unranked. This incompleteness is not a contradiction: the criterion deliberately avoids making interpersonal trade-offs. Choosing among such alternatives requires additional distributive principles, social welfare judgments, or decision rules studied in social choice theory.