Fact-checked by Grok 2 weeks ago

Indirect utility function

The indirect utility function, often denoted as v(\mathbf{p}, w), represents the maximum level of utility a consumer can achieve given a vector of prices \mathbf{p} and or w, formally defined as v(\mathbf{p}, w) = \max_{\mathbf{x}} u(\mathbf{x}) subject to the \mathbf{p} \cdot \mathbf{x} \leq w, where u(\mathbf{x}) is the direct function and \mathbf{x} is the bundle. This function arises in theory by substituting the Marshallian demand functions—the optimal choices—back into the direct function, transforming utility from a function of quantities into one of observable market variables like prices and . Key properties of the indirect utility function include homogeneity of degree zero in prices and income, meaning v(\lambda \mathbf{p}, \lambda w) = v(\mathbf{p}, w) for any \lambda > 0, which reflects that scaling all prices and income proportionally does not change the maximum attainable utility. It is also non-decreasing in income (strictly increasing under local non-satiation of preferences) and non-increasing in prices, ensuring that higher income allows for at least as much utility while price increases reduce it. Additionally, the function is continuous in \mathbf{p} and w (assuming continuity of the direct utility) and quasi-convex, implying that the set of price-income pairs yielding utility no better than a given level is convex. Roy's identity further connects it to demand theory by relating the partial derivative with respect to a price to the corresponding Marshallian demand: x_i(\mathbf{p}, w) = -\frac{\partial v / \partial p_i}{\partial v / \partial w}. In economic analysis, the indirect utility function plays a central role in , enabling the measurement of consumer welfare changes from price or income shifts, such as through compensating or equivalent variations. It establishes duality with the , where the minimum expenditure to achieve a level \bar{u} at prices \mathbf{p} is the in a sense, facilitating derivations of Hicksian demands and cost-of-living indices. This framework is foundational for empirical applications in demand estimation and policy evaluation, linking theoretical preferences to market data without directly observing .

Definition and Basics

Formal Definition

The indirect utility function, denoted as v(\mathbf{p}, m), represents the maximum level of utility a consumer can achieve given a vector of prices \mathbf{p} for and a m. Formally, it is defined as the solution to the 's utility maximization problem: v(\mathbf{p}, m) = \max_{\mathbf{x} \geq \mathbf{0}} \, u(\mathbf{x}) \quad \text{subject to} \quad \mathbf{p} \cdot \mathbf{x} \leq m, where \mathbf{x} is the consumption bundle, u(\cdot) is the direct utility function representing preferences over \mathbf{x}, and the dot denotes the inner product. This definition assumes the direct utility function u(\cdot) is continuous, strictly increasing, and quasi-concave, ensuring the existence of a unique interior solution to the maximization problem under standard conditions in theory. The concept originated in the late 19th century with Antonelli's work on integrability and functions, where he first introduced the indirect utility function as a means to express in terms of prices and income. It was further developed in the early by in his foundational analysis of consumer budget constraints and substitution effects, laying groundwork for modern theory. The framework was formalized in contemporary terms through Hendrik S. Houthakker's extension of , which connected observable data to the existence and properties of representations, including indirect forms. The term "indirect" arises because v(\mathbf{p}, m) evaluates based on economic variables—prices and —rather than directly on the unobservable quantities consumed, providing a bridge from to revelation via the optimal .

Interpretation in Consumer Theory

The , denoted v(p, m), represents the maximum level of or that a can achieve given a of prices p and m, reflecting the outcome of optimal choices under the . This measure serves as a key index in economic analysis, enabling comparisons of across different price environments or levels, and it underpins evaluations such as cost-of-living adjustments in indexing. For instance, changes in v(p, m) can quantify the effects of price shifts, informing adjustments to benefits or policies to maintain equivalent living standards. In , the encodes choices by linking observed demand behavior directly to underlying preferences, without requiring direct observation of the derived from specific bundles. It provides an axiomatic foundation for inferring preference orderings from and , where axioms of revealed favorability—analogous to those in standard —ensure consistency with rational maximization. This approach allows economists to test whether observed demands are consistent with a well-behaved indirect utility function, facilitating empirical validation of solely through market . The interpretation of the indirect utility function relies on several core assumptions in consumer theory, including local non-satiation, which ensures that more consumption is always preferred, preventing indifference at any finite bundle. Convexity of preferences guarantees that the consumer's is well-defined and that indifference curves bow toward the origin, supporting the existence of an interior solution to the maximization problem. Additionally, the framework assumes Walrasian , where consumers exhaust their and respond continuously to and changes, ensuring the indirect utility captures genuine optimization behavior. Regarding utility measurement, the operates within an ordinal , preserving the ranking of orderings across situations but not assigning values that imply interpersonal or absolute comparisons of satisfaction levels. Any monotonic of the underlying direct utility function yields an equivalent indirect utility that maintains these ordinal properties, emphasizing relative rather than absolute assessments in consumer theory.

Mathematical Derivation

From Direct Utility Maximization

The indirect utility function arises from the 's primal problem of utility maximization subject to a . Consider a with a continuous, strictly increasing, and strictly quasi-concave utility function u(\mathbf{x}), where \mathbf{x} = (x_1, \dots, x_n) is the of consumed, facing prices \mathbf{p} = (p_1, \dots, p_n) > \mathbf{0} and income m \geq 0. The problem is to maximize u(\mathbf{x}) subject to \mathbf{p} \cdot \mathbf{x} \leq m and \mathbf{x} \geq \mathbf{0}. Assuming an interior solution exists, form the \mathcal{L}(\mathbf{x}, \lambda) = u(\mathbf{x}) + \lambda (m - \mathbf{p} \cdot \mathbf{x}). The conditions are \frac{\partial u}{\partial x_i} = \lambda p_i for all i = 1, \dots, n, along with the budget exhaustion condition \mathbf{p} \cdot \mathbf{x} = m. These conditions, together with the second-order conditions ensured by strict quasi-concavity of u, yield the unique Marshallian functions \mathbf{x}(\mathbf{p}, m) that solve the . Substituting the optimal demands back into the utility function gives the indirect utility function v(\mathbf{p}, m) = u(\mathbf{x}(\mathbf{p}, m)), which represents the maximum attainable given prices and . The strict quasi-concavity of u and positivity of prices guarantee the existence and uniqueness of this solution under standard continuity assumptions. By the applied to the , the partial derivative of the indirect utility with respect to is \frac{\partial v}{\partial m} = \lambda, where \lambda > 0 is the interpreting the of . This follows directly from differentiating v(\mathbf{p}, m) while holding \mathbf{p} fixed, as the indirect effect through demands cancels out, leaving only the effect on the .

Properties of the Indirect Utility Function

The indirect utility function v(\mathbf{p}, m), arising from the solution to the consumer's utility maximization problem, exhibits several essential mathematical properties that align with the axioms of consumer preferences, such as continuity, monotonicity, and convexity. A key property is homogeneity of degree zero in prices and income: v(t\mathbf{p}, tm) = v(\mathbf{p}, m) for any scalar t > 0. This means that proportional increases in all prices and income leave the maximum attainable utility unchanged, reflecting the absence of money illusion in rational consumer behavior. The property follows directly from the homogeneity of degree zero in the Marshallian demand functions, as the optimal consumption bundle \mathbf{x}(\mathbf{p}, m) satisfies \mathbf{x}(t\mathbf{p}, tm) = \mathbf{x}(\mathbf{p}, m), yielding the same utility evaluation. By Euler's theorem applied to this homogeneous function, \sum_i p_i \frac{\partial v}{\partial p_i} + m \frac{\partial v}{\partial m} = 0. The indirect utility function is non-increasing in each price p_i: \frac{\partial v}{\partial p_i} \leq 0. Under the standard assumption of local non-satiation, this effect is strict (\frac{\partial v}{\partial p_i} < 0), as higher prices reduce purchasing power and force the consumer to a lower indifference curve. This can be shown using Roy's identity, \frac{\partial v / \partial p_i}{\partial v / \partial m} = -x_i(\mathbf{p}, m), where x_i > 0 due to local non-satiation and \frac{\partial v}{\partial m} > 0, implying the negative price derivative. Shephard's lemma provides an indirect link through duality, confirming the consumption response to prices. It is also strictly increasing in income: \frac{\partial v}{\partial m} > 0. Greater income expands the budget set, enabling access to higher utility levels consistent with monotonic preferences. This marginal utility of income is positive because any additional resources allow for a preferred consumption bundle under local non-satiation. Finally, the indirect utility function is quasiconvex in prices and income, meaning the sublevel sets \{(\mathbf{p}, m) \mid v(\mathbf{p}, m) \leq \bar{v}\} are convex for any \bar{v}. This property ensures smooth and consistent welfare adjustments to changes in economic conditions, preserving the convexity of preferences in the dual space.

Connections to Other Economic Functions

Relationship with Expenditure Function

The expenditure function, denoted e(\mathbf{p}, u), is defined as the minimum expenditure required to achieve a given utility level u at prices \mathbf{p}, formally expressed as
e(\mathbf{p}, u) = \min_{\mathbf{x}} \mathbf{p} \cdot \mathbf{x} \quad \text{subject to} \quad u(\mathbf{x}) \geq u.
This function serves as the dual problem to the underlying the indirect utility function v(\mathbf{p}, m).
A fundamental duality theorem establishes the inverse relationship between these functions: for any prices \mathbf{p} and m, the income equals the expenditure needed to achieve the indirect level, m = e(\mathbf{p}, v(\mathbf{p}, m)); conversely, for any u, the indirect from the expenditure to achieve u recovers the original , u = v(\mathbf{p}, e(\mathbf{p}, u)).
The proof relies on the equivalence of optimization solutions: the bundle \mathbf{x}^* that maximizes subject to the also minimizes expenditure subject to the constraint, assuming of preferences and .
This duality implies that the expenditure function can be obtained by inverting the indirect utility function: solving v(\mathbf{p}, m) = u for m yields e(\mathbf{p}, u), allowing recovery of the minimum cost to attain a specific from observable s and income-derived utility levels. In , the facilitates the construction of true cost-of-living indices, such as the Konüs index, which measure the ratio of expenditures needed to maintain a reference level across price changes, often derived from estimates of the indirect utility function to assess adjustments.

Roy's Identity and Duality

Roy's identity establishes a fundamental link between the indirect utility function and the Marshallian functions in consumer theory. It states that the demand for the i-th good, x_i(\mathbf{p}, m), can be recovered from the indirect utility function v(\mathbf{p}, m) as x_i(\mathbf{p}, m) = -\frac{\partial v(\mathbf{p}, m) / \partial p_i}{\partial v(\mathbf{p}, m) / \partial m}, where \mathbf{p} is the price vector and m is income. This identity, named after the French economist René Roy who first outlined it in his seminal work on across , provides a bridge between theoretical maximization and practical . The identity arises from the applied to the constrained underlying the indirect utility function. Consider the optimization \max_{\mathbf{x}} u(\mathbf{x}) subject to \mathbf{p} \cdot \mathbf{x} \leq m, with \mathcal{L} = u(\mathbf{x}) + \lambda (m - \mathbf{p} \cdot \mathbf{x}). The first-order conditions imply \partial u / \partial x_j = \lambda p_j for each good j. Differentiating v(\mathbf{p}, m) = u(\mathbf{x}(\mathbf{p}, m)) with respect to p_i via the chain rule gives \partial v / \partial p_i = \sum_j (\partial u / \partial x_j) (\partial x_j / \partial p_i). Substituting the first-order conditions simplifies the terms, yielding \partial v / \partial p_i = -\lambda x_i, while \partial v / \partial m = \lambda. Dividing these partial derivatives thus recovers the x_i. This derivation highlights the identity's reliance on the optimality conditions of the expenditure-constrained problem. In empirical applications, facilitates the estimation of demand systems by allowing researchers to specify and estimate an indirect utility function directly from observable data on prices and expenditures, then derive demands without needing quantity data for each good. This approach has proven valuable in econometric models where quantities may be unobserved or costly to measure, enabling welfare analysis and policy simulations from utility estimates alone. A dual counterpart, , similarly links the to Hicksian demands.

Examples and Applications

Cobb-Douglas Utility Example

The Cobb-Douglas utility function provides a concrete illustration of the indirect utility function, representing preferences where the between goods is constant at unity. Consider a two-good where the direct utility function is given by u(x_1, x_2) = x_1^\alpha x_2^{1-\alpha}, with $0 < \alpha < 1, indicating the relative preference weight on good 1. To derive the indirect utility function v(\mathbf{p}, m), first solve for the Marshallian demand functions by maximizing u(x_1, x_2) subject to the budget constraint p_1 x_1 + p_2 x_2 = m. The first-order conditions yield the demands x_1(p_1, p_2, m) = \frac{\alpha m}{p_1}, \quad x_2(p_1, p_2, m) = \frac{(1-\alpha) m}{p_2}. These demands allocate the budget share \alpha to good 1 and $1-\alpha to good 2. Substituting these optimal demands back into the direct utility function gives the indirect utility: v(p_1, p_2, m) = \left( \frac{\alpha m}{p_1} \right)^\alpha \left( \frac{(1-\alpha) m}{p_2} \right)^{1-\alpha}. Simplifying step by step, v(p_1, p_2, m) = (\alpha m)^\alpha p_1^{-\alpha} \cdot ((1-\alpha) m)^{1-\alpha} p_2^{-(1-\alpha)} = \alpha^\alpha (1-\alpha)^{1-\alpha} m^{\alpha + (1-\alpha)} p_1^{-\alpha} p_2^{-(1-\alpha)} = \alpha^\alpha (1-\alpha)^{1-\alpha} \frac{m}{p_1^\alpha p_2^{1-\alpha}}. This expression shows that indirect utility increases linearly with income m and decreases with prices, with elasticities equal to -\alpha for p_1 and -(1-\alpha) for p_2. The indirect utility satisfies key properties of the general form. For homogeneity of degree zero, scaling prices and income by t > 0 yields v(t p_1, t p_2, t m) = \alpha^\alpha (1-\alpha)^{1-\alpha} \frac{t m}{(t p_1)^\alpha (t p_2)^{1-\alpha}} = \alpha^\alpha (1-\alpha)^{1-\alpha} \frac{t m}{t^\alpha p_1^\alpha t^{1-\alpha} p_2^{1-\alpha}} = \alpha^\alpha (1-\alpha)^{1-\alpha} \frac{t m}{t m} \cdot \frac{1}{p_1^\alpha p_2^{1-\alpha}} = v(p_1, p_2, m). An increase in income m raises v proportionally, reflecting constant marginal utility of income in this case, while a rise in p_1 reduces v by a factor proportional to p_1^\alpha, illustrating the .

Empirical and Policy Applications

In empirical economics, the indirect utility function is frequently estimated using techniques on household-level to test and impose theoretical restrictions on structures. For instance, the translog form of the indirect utility function, introduced by Christensen, Jorgenson, and Lau, allows for flexible estimation of patterns and effects across , often applied to cross-sectional surveys like the Consumer Expenditure Survey. Similarly, the Almost Ideal Demand System (AIDS), derived from a specific indirect utility specification by Deaton and Muellbauer, facilitates of the indirect utility function through and expenditure shares, enabling tests of homogeneity and in from sources such as the Family Expenditure Survey. These approaches, including Gorman polar forms for linear Engel curves, have been widely used to analyze aggregate consumer behavior and validate in macro models. In , the indirect utility function underpins welfare measurements such as (), defined as the adjustment needed to restore after a change, calculated via duality as CV = e(p', u) - m, where e is the recovering the original u at new prices p' and initial m. This metric has been applied to evaluate tax reforms, such as the 1986 U.S. Tax Reform Act, where estimated indirect utilities from household data quantified distributional impacts and deadweight losses across groups. For subsidies, particularly in developing economies, changes in v(p, m) help assess equivalent variation—the equivalent of shifts from subsidized goods like food or fuel—informing programs such as India's Distribution System to target poverty alleviation. These applications leverage to link estimated demands back to , ensuring path-independent welfare rankings in partial equilibrium settings. Despite these advances, empirical implementation faces challenges from unobserved heterogeneity, where individual-specific factors like tastes or constraints vary across households, biasing estimates of the indirect utility function in standard demand systems. Addressing this requires random coefficient models or nonparametric methods to incorporate latent preference variation, as in Barten scale adjustments that rescale prices for demographic differences. Extensions integrate the indirect utility into general equilibrium frameworks, such as models for trade policy, to capture economy-wide price feedbacks, though this amplifies data demands. In , deviations from standard assumptions—like or reference dependence—have prompted hybrid models that modify v(p, m) to include psychological parameters, tested on experimental or field data from nudge interventions. Recent developments emphasize computational applications of the indirect utility function in dynamic settings, such as life-cycle models where v(p, m) is solved iteratively via value function methods to simulate responses over time horizons. For example, in heterogeneous models, on indirect utilities accommodates prices and incomes, aiding forecasts of savings behavior under shocks like reforms. In measurement, ordinal rankings derived from estimated v(p, m) enable dominance tests across distributions, adjusting for indices to compare living standards in spatial or temporal contexts, as seen in analyses of global poverty trends using surveys. These tools, often implemented in software like or , prioritize scalable algorithms for large datasets while preserving theoretical consistency.

References

  1. [1]
    [PDF] Lectures 3—4: Consumer Theory
    Value of (CP) = welfare of consumer facing prices p with income w. The value function of (CP) is called the indirect utility function. Definition. The indirect ...
  2. [2]
    [PDF] 4. Duality in Consumer Theory
    Can we use the indirect utility function? If prices and incomes change from p0, w0 to p1, w1 , then utility increases by ∆U = v(p1, w1) − v(p0, w0). But ...
  3. [3]
    [PDF] Economics 326: Marshallian Demand and Comparative Statics
    Sep 17, 2012 · 6 Indirect Utility Function. Definition: Plug in the demand functions back into the utility function. Then the utility function is a function ...
  4. [4]
    [PDF] Duality - Santa Fe Institute
    6 Historical Notes. Duality ideas appeared very early in the marginal revolution. Antonelli, for instance, introduced the indirect utility function in 1886.
  5. [5]
    [PDF] Slutsky's 1915 Article: How It Came to be - Department of Economics
    Slutsky challenges the "traditional" definition of substitutes and complements according to the sign of the second cross partial deriva- tives of the utility ...
  6. [6]
    Revealed Preference and the Utility Function - Semantic Scholar
    Professor Samuelson's "revealed preference" approach' has proved to be a useful basis for deriving a considerable part of the static theory of consumer's ...
  7. [7]
    [PDF] Chapter 3: The Theory of the Cost of Living Index
    Apr 28, 2020 · The paper discusses the economic approach to the construction of a consumer cost of living price index (due to Konüs) as well as the ...
  8. [8]
    Revealed favorability, indirect utility, and direct utility - ScienceDirect
    This paper studies the axiomatic foundations of the indirect utility function, based upon a revealed preference approach.
  9. [9]
    [PDF] ECON 825 NOTES II: CONSUMER THEORY Fall 2007 I. The Utility ...
    Assumptions on Utility Functions. 1. Continuous and increasing. 2. Indifference curves are convex to the origin. 3. Local nonsatiation. 1 ...
  10. [10]
    [PDF] Consumer Choice 1
    The Indirect Utility Function. If can be represented by a utility function we can define the indirect utility function v(p,w) = u(x(p,w)) v now represents ...
  11. [11]
    [PDF] Preferences and Utility - UCLA Economics
    The indirect utility function has three attractive properties. 1. The indirect utility function is homogenous of degree zero in prices and income. That is ...
  12. [12]
  13. [13]
    [PDF] INDIRECT UTILITY FUNCTION U*(Px,Py,M) = max {U(x, y)
    Solve the indirect utility function for income: u = U*(Px,Py,M) ⇐⇒ M ... SLUTSKY EQUATION. Link between Marshallian and Hicksian demands. Equal if u ...Missing: 1915 | Show results with:1915
  14. [14]
    [PDF] Chapter 4: Topics in Consumer Theory - Nolan H. Miller
    That is, an indirect utility function of the Gorman form can be separated into a term that depends on prices and the consumer's identity but not on her wealth, ...
  15. [15]
    [PDF] Lecture 14: Topics in Demand Theory - P.J. Healy
    Sep 30, 2020 · Use the function µ to define an indirect utility function w by w(p, m) = µ(p. ∗. ;p, m). 3. Invert the demand function to give (p, m) as a ...
  16. [16]
    [PDF] An Almost Ideal Demand System - American Economic Association
    trary direct or indirect utility function or, more rarely, a cost function. For examples, see Christensen, Jorgenson, and Lau; W. Erwin Diewert (1971); or ...
  17. [17]
    [PDF] Specification and Testing in Applied Demand Analysis
    For rather more than two decades, applied econometricians have been estimating systems of demand equations which have been explicitly or implicitly derived.
  18. [18]
    [PDF] Exact Consumer's Surplus and Deadweight Loss - Jerry A. Hausman
    Sep 28, 2002 · Exact Consumer's Surplus and Deadweight Loss. By JERRY A. HAUSMAN*. Consumer's surplus is a widely used tool in applied welfare economics ...
  19. [19]
    [PDF] NBER WORKING PAPER SERIES THE EMPIRICAL ANALYSIS OF ...
    Tax reform can have a major impact on the net incomes and welfare of families, on a par with significant changes in the annual increment to GNP. It is ...
  20. [20]
    [PDF] Prices and Welfare - World Bank Open Knowledge Repository
    Feb 16, 2016 · that an indirect utility function ... differentiable utility functions, this theorem allows approximating changes in utility due to price.
  21. [21]
    Unobserved Preference Heterogeneity in Demand Using ... - jstor
    Jun 30, 2017 · We then model unobserved preference heterogeneity in consumer demand as utility functions with random Barten scales. These Barten scales ap-.
  22. [22]
    [PDF] Unobserved Preference Heterogeneity in Demand Using ...
    By standard revealed preference theory, the direct and indirect utility functions are therefore also nonparametrically identified up to an unknown monotonic ...
  23. [23]
    [PDF] Generalized Compensation Principle;
    May 11, 2022 · We derive a formula for the compensating tax reform and its impact on the government budget when only distortionary tax instruments are ...Missing: subsidies | Show results with:subsidies
  24. [24]
    [PDF] Taxation and Economic Efficiency - University of California, Berkeley
    This paper analyzes the distortions created by taxation and the features of tax systems that minimize such distortions (subject to achieving other ...
  25. [25]
    [PDF] Deep learning for solving dynamic economic models.
    Moreover, we can define indirect accuracy measures to reflect the economic. 36 significance of accuracy. For example, we can express approximation errors in ...
  26. [26]
    [PDF] Measuring Welfare and Inequality with Incomplete Price Information
    Measuring changes in household welfare is valuable in many contexts, both to evaluate the impacts of policies and to assess changes in well-being across time ...
  27. [27]
    [PDF] demandsys — Estimation of flexible demand systems - Stata
    Jul 17, 2024 · A limitation of the generalized translog model is that we cannot solve the indirect utility function for a closed-form expression for the cost ...