Pareto Optimality and Welfare Economics

Conceptions of Social Justice and General Efficiency

Social justice is conceptualized based on a variety of core societal values. These foundational values include rights (such as human rights), needs, equity/equality, and merit.

Efficiency, in general terms, describes the extent to which society or an agent is able to maximize output and/or minimize effort through the use of a machine, method, strategy, or process.

In physical sciences, efficiency is defined as the ratio of useful work performed by a machine to the total energy consumed. In agriculture, efficiency is measured by quantitative yield, such as the amount or number of rice produced per unit of land or unit of labor using a given farming method.

Evaluating efficiency becomes significantly more complicated when applied to public-sector decisions. Public policy evaluation primarily relies on two popular conceptions of efficiency: Pareto Optimality and the Kaldor-Hicks Criterion.

The Pareto Criterion and Pareto Optimality

The Pareto criterion functions as a precise technique for comparing and ranking alternative states of an economy. Under the strict Pareto criterion, if it is possible to make at least one person better off in moving from state $A$ to state $B$ without making anyone else worse off, state $B$ is ranked higher by society than state $A$.

A Pareto optimum (also termed Pareto efficiency) defines a state of the economy from which it is impossible to make any single individual better off without making at least one other person worse off. If society finds itself in a position from which no further Pareto improvement can be made, the economy has attained a Pareto optimum. Conversely, if an economy is not at a Pareto optimum, there exists inherent inefficiency within the economy.

Graphically, Pareto optimality is evaluated using a utility space diagram that plots individual utilities, such as $\text{Utility Person A}$ on the horizontal axis and $\text{Utility Person B}$ on the vertical axis. A Utility Possibility Curve corresponds to all possible combinations of utility for individuals $A$ and $B$ given a fixed production possibility frontier. Any starting point located inside the curve represents a Pareto-inefficient position, from which a bounded set of Pareto improvements can lead to optimal allocations along the curve.

Pure Exchange Economy and Consumer Preferences in the Edgeworth Box

To analyze exchange efficiency, consider a pure exchange economy consisting of $2$ individuals ($\text{Adam}$ and $\text{Eve}$) and $2$ commodities ($\text{Apples}$ and $\text{Figs}$) in fixed supply, such as on a desert island. An Edgeworth Box depicts the distribution of these goods between the two individuals, where the origin for $\text{Adam}$ is located at the bottom-left corner ($0$) and the origin for $\text{Eve}$ is located at the top-right corner ($0'$). An arbitrary point inside the box, such as point $g$, provides an initial allocation of goods ($\text{A}_g$ apples and $\text{E}_g$ fig leaves) and establishes an initial level of utility for both individuals.

Consumer preferences across goods are represented by indifference curves. The magnitude of the slope of an indifference curve measures the consumer's marginal rate of substitution ($MRS$) between two goods. The marginal rate of substitution ($MRS$) is defined as the maximum amount of a good that a consumer is willing to give up in order to obtain one additional unit of another good.

As an illustrative example of consumer preferences between clothing ($C$) and food ($F$), the magnitude of the $MRS$ falls progressively from $6$ (between points $A$ and $B$) to $4$ (between points $B$ and $D$), to $2$ (between points $D$ and $E$), down to $1$ (between points $E$ and $G$). This decline in the $MRS$ along an indifference curve reflects a diminishing marginal rate of substitution, which causes the indifference curve to be convex.

Reallocation, Contract Curve, and Efficiency Conditions

Reallocating goods between individuals can generate welfare gains. Starting from an initial allocation $g$ in an Edgeworth Box, moving along $\text{Eve}$'s indifference curve to allocation $h$ keeps $\text{Eve}$'s utility completely unchanged while increasing $\text{Adam}$'s utility. Continuing to reallocate goods along $\text{Eve}$'s indifference curve eventually leads to allocation $p$. Once allocation $p$ is reached, it is impossible to raise $\text{Adam}$'s utility any further while keeping $\text{Eve}$'s utility unchanged.

An allocation is formally defined as Pareto efficient if the only way to make one person better off is to make another person worse off. Pareto efficient allocations serve as the standard for evaluating the economic desirability of resource distribution, whereas Pareto inefficient allocations are wasteful. A Pareto improvement is defined specifically as a reallocation of resources that makes one person better off without making anyone else worse off.

There are multiple Pareto efficient allocations within an Edgeworth Box, such as allocations $p$, $p_1$, $p_2$, $p_3$, and $p_4$. Among these Pareto efficient allocations, some provide $\text{Adam}$ with higher utility than others, while the opposite ones provide $\text{Eve}$ with higher utility. The locus of the entire set of Pareto efficient points in an Edgeworth Box is called the contract curve.

Each Pareto efficient point along the contract curve occurs where an indifference curve of $\text{Adam}$ is tangent to an indifference curve of $\text{Eve}$. Mathematically, the absolute values of the slopes of $\text{Adam}$'s and $\text{Eve}$'s indifference curves are equal at these points. Because the absolute slope of the indifference curve represents the marginal rate of substitution ($MRS$), the condition for Pareto efficiency in exchange requires that the marginal rate of substitution between the two goods must be equal for both consumers:

$MRS_{\text{Adam}} = MRS_{\text{Eve}}$

Limitations of the Pareto Criterion

The pure Pareto criterion suffers from severe theoretical and practical limitations. The criterion breaks down entirely if even a single individual is made worse off by a policy change. As a consequence, many economic alternatives are simply not comparable under this framework.

Distinct Pareto optima that correspond to different underlying income distributions are not Pareto comparable with one another. Therefore, the Pareto criterion does not provide a mechanism for choosing between alternative income distributions, which inherently causes the criterion to favor the status quo.

Additionally, not all first-best Pareto-optimal choices are socially superior to second-best (Pareto-inefficient) choices. A second-best choice may possess an income distribution that society deems vastly superior to that of a first-best choice.

Potential Pareto Criterion and Kaldor-Hicks Compensation

The Potential Pareto Criterion, also known as the Kaldor-Hicks Compensation Test, was developed as a modification of the pure Pareto criterion to overcome its limitations. It relies on the compensation principle and forms the theoretical foundation for ranking alternative states of the economy in Cost-Benefit Analysis ($\text{CBA}$).

Under the Potential Pareto Criterion, State $A$ is preferable to State $B$ if it is potentially possible for the gainers in State $A$ to compensate the losers and still remain better off. Actual payment of compensation is explicitly not required.

Consider a numerical example where gainers in State $A$ receive $100\text{ units}$ of benefit while losers are penalized by $90\text{ units}$ of cost. If the gainers potentially compensate the losers by paying $90\text{ units}$, the gainers still retain a net positive benefit of $10\text{ units}$. Thus, State $A$ is ranked superior to State $B$ under the potential criterion.

In actual market economies, compensation is rarely direct; instead, it is administered through various forms of transfer payments. The overarching purpose of potential compensation in policy evaluation is to separate the efficiency components of a proposed policy change from its equity components.

Graphically, consider a Utility Possibility Frontier ($\text{UPF}$) plotting $\text{Utility A}$ against $\text{Utility B}$. A move from an initial inefficient state $I$ to any point on the frontier between $S_0$ and $S_1$ represents a pure Pareto improvement, making all points on the $S_0S_1$ segment Pareto-superior to $I$. A move from $I$ to an efficient point $E$ outside the $S_0S_1$ segment is not a Pareto improvement in itself because one individual suffers a utility loss relative to point $I$. However, the Potential Pareto Criterion ranks the Pareto-optimal allocation $E$ superior to allocation $I$ because potential compensation would allow society to redistribute utility from $E$ to reach the Pareto-improving range $S_0S_1$

Cost-Benefit Analysis versus Welfare Analysis

Standard Cost-Benefit Analysis ($\text{CBA}$) assigns equal weight to net benefits for every individual, regardless of their underlying income or wealth. Conversely, welfare analysis represents an alternative concept of efficiency that assigns different weights to the net benefits accrued by different groups of people.

When evaluating public investment projects, international institutions such as the World Bank sometimes assign extra weight to the net benefits received by very poor individuals. Weighting net benefits inversely relative to the income or wealth of the recipients is the defining feature of welfare analysis.

The underlying theoretical rationale for welfare analysis is that an extra dollar spent by a rich individual on opera tickets increases overall societal welfare by a smaller amount than an extra dollar spent by a poor individual on bread.

Citizens and political theorists in democratic societies actively debate whether the state should redistribute income from the rich to the poor. Libertarians typically oppose government income redistribution and favor policies approved by Pareto efficiency or standard cost-benefit analysis. In contrast, socialists typically favor state redistribution and approve policies evaluated through welfare analysis.