General Equilibrium Theory

General equilibrium theory is the mathematical framework in business and economics theory that analyzes how prices and quantities are determined simultaneously across all interconnected commodity, factor, and financial markets in an economy. In contrast to Alfred Marshall's partial-equilibrium supply and demand analysis—which isolates a single market under the ceteris paribus assumption that prices in all other markets remain fixed—general equilibrium theory explicitly models cross-market feedback loops across consumers, firms, and resource endowments. Originated by Léon Walras in Éléments d'économie politique pure (1874) and placed on rigorous topological foundations in 1954 by Kenneth Arrow, Gérard Debreu, and Lionel McKenzie using Kakutani's fixed-point theorem, the Arrow–Debreu–McKenzie framework provides the microeconomic foundation for modern welfare economics, international trade models, and dynamic stochastic general equilibrium (DSGE) macroeconomics.

Primitives of a Private Ownership Economy

Consider a competitive economy with L distinct commodities (indexed l = 1, …, L, differentiated by physical characteristics, geographic location, date of delivery, and state of nature), I consumers (indexed i = 1, …, I), and J producers (indexed j = 1, …, J):

1. Consumers: Each consumer i is characterized by a consumption set X_i ⊂ ℝ₊^L, a continuous and locally non-satiated utility function U_i: X_i → ℝ (with diminishing marginal utility), an initial endowment vector of commodities ω_i ∈ ℝ₊^L, and profit ownership shares θ_ij ≥ 0 in each firm j such that ∑_i=1^I θ_ij = 1.
2. Producers: Each firm j possesses a production possibilities set Y_j ⊂ ℝ^L (where negative entries denote factor inputs and positive entries denote outputs, often parameterized via a Cobb–Douglas production function). Given a price vector p = (p₁, …, p_L) ∈ ℝ₊^L, firm j selects a production plan y_j^*(p) ∈ Y_j to maximize profit Π_j(p) = p · y_j.
3. Budget Constraint: Consumer i selects a consumption bundle x_i^*(p) ∈ X_i to maximize U_i(x_i) subject to the wealth constraint derived from endowments and distributed firm profits:

Walrasian Equilibrium and Excess Demand

Define the aggregate excess demand function z(p) = (z₁(p), …, z_L(p)) as total demand minus total initial endowments and total net production:

A Walrasian (Competitive) Equilibrium is a price vector p^* ∈ ℝ₊^L and an allocation ((x₁^*, …, x_I^*), (y₁^*, …, y_J^*)) such that z(p^*) ≤ 0 for all l = 1, …, L (with z_l(p^*) = 0 for every commodity with a positive price p_l^* > 0).

Homogeneity of Degree Zero and Walras's Law

Two structural properties govern the aggregate excess demand function z(p):

1. Homogeneity of Degree Zero: Multiplying all prices by a positive scalar λ > 0 leaves every consumer's budget set and every firm's profit-maximizing plan unchanged (z(λp) = z(p)). Consequently, only relative prices matter in a pure exchange or real production economy, allowing prices to be normalized onto the (L − 1)-dimensional unit simplex Δ^L−1 = {p ∈ ℝ₊^L : ∑_l=1^L p_l = 1} or expressed relative to a single numéraire commodity (p₁ ≡ 1).
2. Walras's Law: Because local non-satiation implies that every consumer exhausts their budget constraint with equality, summing all I budget constraints yields the identity that the aggregate market value of excess demand is identically zero at all price vectors (whether or not p is an equilibrium):

An immediate algebraic corollary of Walras's Law is that if L − 1 markets clear (z_l(p) = 0 for l = 1, …, L − 1), then the L-th market with p_L > 0 must also clear automatically (z_L(p) = 0).

Arrow–Debreu Existence Theorem via Kakutani's Fixed-Point Theorem

Walras originally argued that an equilibrium exists because the system contains L − 1 independent equations in L − 1 relative price unknowns, but counting equations and unknowns does not guarantee a real, non-negative solution. Arrow and Debreu (1954) proved that if consumption sets and production sets are closed and convex, preferences are continuous, convex, and locally non-satiated, and each consumer possesses a strictly positive interior endowment (ω_i ≫ 0), then a continuous price-adjustment mapping g: Δ^L−1 → Δ^L−1 defined by:

satisfies the hypotheses of Brouwer's (for single-valued z(p)) or Kakutani's Fixed-Point Theorem (for upper hemicontinuous convex-valued excess demand correspondences), guaranteeing the existence of at least one fixed point p^* = g(p^*) where z(p^*) ≤ 0.

The Edgeworth Box and Contract Curve

In a two-consumer (A, B), two-good (X₁, X₂) pure exchange economy, the feasible allocations are represented geometrically in the Edgeworth–Bowley box of dimensions ω₁ = ω_1A + ω_1B by ω₂ = ω_2A + ω_2B. Tangency points between Consumer A's and Consumer B's indifference curves satisfy equal Marginal Rates of Substitution (MRS₁₂^A = MRS₁₂^B), tracing out the contract curve (the Pareto set). At a Walrasian equilibrium, both consumers optimize against the same price ratio p₁^* / p₂^*, guaranteeing MRS₁₂^A = p₁^* / p₂^* = MRS₁₂^B; in a production economy, this also equates to the Marginal Rate of Transformation (MRT₁₂, the marginal opportunity cost ratio).

This equivalence yields the First and Second Fundamental Theorems of Welfare Economics: (1) under local non-satiation and complete markets free of externalities or information asymmetry, every Walrasian equilibrium is Pareto efficient; and (2) under convex preferences and production sets, any Pareto-efficient allocation can be supported as a Walrasian equilibrium following suitable lump-sum endowment redistributions.

Uncertainty, Time, and CGE / DSGE Models

Arrow (1953) and Debreu (1959) extended static general equilibrium to intertemporal choice and uncertainty by indexing commodities by their date and state of nature (Arrow–Debreu state-contingent securities, each paying 1 unit of numéraire if state s occurs and 0 otherwise), laying the foundation for modern asset pricing and the Black–Scholes model. In empirical policy analysis, Herbert Scarf's (1967) simplicial fixed-point algorithms enabled Computable General Equilibrium (CGE) models used to evaluate tariff changes under comparative advantage and carbon Pigouvian tax policies, while macroeconomic extensions yielded Real Business Cycle (RBC) and New Keynesian Dynamic Stochastic General Equilibrium (DSGE) frameworks governing monetary policy and fiscal policy.

Why is partial equilibrium analysis sometimes insufficient for policy evaluation?

Partial equilibrium analysis examines a single market in isolation, assuming all other market prices and household incomes remain constant. When a policy shock affects a major sector—such as an economy-wide carbon tax, a comprehensive trade tariff, or a national minimum wage—it alters factor prices (wages and capital returns), shifts demand for substitute and complementary goods across dozens of industries, and feeds back into the original market. Only general equilibrium captures these multi-market feedback loops.

What does the Sonnenschein–Mantel–Debreu (SMD) theorem prove about aggregate excess demand?

The Sonnenschein–Mantel–Debreu theorem (proved between 1972 and 1974) demonstrates that even when every individual consumer maximizes a strictly quasi-concave utility function, the resulting aggregate excess demand function z(p) inherits only three general properties on compact price domains: continuity, homogeneity of degree zero, and Walras's Law (p · z(p) = 0). Because individual income effects can offset substitution effects in complex ways, uniqueness and global tâtonnement stability of general equilibrium require additional structural conditions such as the gross substitutes property.

What is the role of the numéraire good in a Walrasian equilibrium?

Because multiplying all prices by any positive constant leaves every real budget constraint unchanged (homogeneity of degree zero), the general equilibrium system determines only L − 1 relative price ratios rather than absolute nominal price levels. Setting the price of one chosen commodity equal to unity (p₁ = 1)—designated as the numéraire—anchors the relative price scale so that all other prices p_l / p₁ are expressed in units of Commodity 1.

1. Walras, L. (1874). Éléments d'économie politique pure, ou théorie de la richesse sociale. L. Corbaz & Cie.
2. Arrow, K. J., & Debreu, G. (1954). "Existence of an Equilibrium for a Competitive Economy." Econometrica, 22(3), 265–290. https://doi.org/10.2307/1907353
3. McKenzie, L. W. (1954). "On Equilibrium in Graham's Model of World Trade and Other Competitive Systems." Econometrica, 22(2), 147–161. https://doi.org/10.2307/1907539
4. Debreu, G. (1959). Theory of Value: An Axiomatic Analysis of Economic Equilibrium. Yale University Press. https://cowles.yale.edu/publications/monographs/m17-theory-value
5. Scarf, H. (1967). "The Approximation of Fixed Points of a Continuous Mapping." SIAM Journal on Applied Mathematics, 15(5), 1328–1343. https://doi.org/10.1137/0115116