Continuity of equilibria in spaces of Bochner and Gel'fand economies

This paper establishes the continuity of equilibrium correspondences on a dense subset of infinite-dimensional Bochner and Gel'fand economies modeled on Banach lattices, providing a unified framework that extends classical results to diverse economic models without requiring differentiability assumptions.

The Gist

Imagine the economy as a giant, complex machine. In this machine, there are millions of "agents" (people or companies), each with their own unique desires (preferences) and starting resources (endowments). The goal of economists is to find the "equilibrium"—a state where everyone is happy with what they have, no one wants to trade anymore, and supply matches demand.

For a long time, economists have known that if you tweak the machine slightly (change a person's taste or a resource amount), the equilibrium should change only slightly. This is called continuity. If the machine is not continuous, a tiny measurement error could cause the entire system to crash into a completely different, chaotic state.

This paper by Matías Fuentes is like a master mechanic upgrading the tools used to check if this machine is stable. Here is the breakdown in simple terms:

1. The Problem: The Machine Got Too Big and Complex

Previous studies could only check the stability of this economic machine if the "commodities" (goods) were simple and finite, like apples, oranges, and bananas (think of a standard grocery list).

However, real-world economics often deals with infinite or continuous goods:

• Time: Resources available every second for the next 1,000 years.
• Uncertainty: Outcomes for every possible future scenario.
• Differentiation: Every unique version of a product (like every specific shade of blue paint).

When you move from a finite list to an infinite one, the math gets messy. The "space" where these goods live becomes a Banach Lattice (a fancy mathematical structure that handles order and size in infinite dimensions). Previous tools couldn't handle these infinite spaces without making very strict, unrealistic assumptions (like assuming everything is "smooth" enough to be differentiated, like a smooth curve).

2. The Solution: A New, Universal Toolkit

Fuentes introduces a new way to model these economies using two powerful mathematical concepts: Bochner and Gel'fand integrals.

• The Analogy: Imagine you are trying to measure the total weight of a cloud.
  • Old way: You tried to weigh every single water droplet individually (impossible if there are infinite droplets).
  • Bochner/Gel'fand way: You treat the cloud as a continuous flow. You don't count the drops; you integrate the "density" of the cloud over the sky. This allows you to handle the "infinite" nature of the resources without breaking the math.

The paper proves that even in these massive, infinite-dimensional spaces, the economic machine is stable. Specifically, it shows that for a "dense" (very large and representative) group of economies, if you nudge the parameters slightly, the equilibrium shifts smoothly.

3. The "No Smoothness" Rule

A major breakthrough in this paper is that it doesn't require the economy to be "smooth."

• The Old Rule: To prove stability, mathematicians usually had to assume that people's preferences were like a perfectly smooth, differentiable curve (like a polished marble slide). If the curve had a sharp corner (a "kink"), the old math said, "We can't guarantee stability here."
• The New Rule: Fuentes shows you don't need that smoothness. Even if the preferences are jagged, bumpy, or have sharp corners (like a crumpled piece of paper), the equilibrium is still continuous for most economies. This is a huge deal because real human preferences are rarely perfectly smooth.

4. Two Ways to Look at the Machine

The paper uses two different lenses to look at the economy, proving the result holds for both:

1. The Distributional View (The Big Picture): Instead of tracking Person A, Person B, and Person C, the economist looks at the "cloud" of all agents. It's like looking at a crowd from a helicopter; you see the density and flow, not individual faces.
2. The Individualized View (The Close-Up): This tracks specific agents on a map. The paper proves that the "Big Picture" view and the "Close-Up" view are consistent. If the crowd is stable, the individuals are stable, and vice versa.

5. What This Actually Covers

The paper claims this new toolkit works for a wide variety of specific economic scenarios that were previously hard to analyze, including:

• Infinite Time Horizons: Planning for resources that stretch on forever.
• Monopolistic Competition: Markets with many firms selling slightly different products.
• Financial Equilibria: Markets dealing with complex, continuous streams of future payments.
• Asymmetric Information: Situations where some agents know more than others.

The Bottom Line

Think of this paper as upgrading the "stability test" for the global economy. Before, the test only worked for simple, finite, and perfectly smooth economies. Now, Fuentes has built a test that works for infinite, complex, and jagged economies.

He proves that for almost all realistic economic scenarios (a "dense subset"), the system is robust. If you change the inputs slightly, the outcome changes slightly. You don't need to assume the world is perfectly smooth to know that the economic machine won't fall apart when you tweak the dials.

Technical Summary

Technical Summary: Continuity of Equilibria in Spaces of Bochner and Gel'fand Economies

Problem Statement
The paper addresses the continuity of equilibrium correspondences in general equilibrium theory within infinite-dimensional settings. While the existence of competitive equilibria is well-established, the stability of these equilibria under perturbations of exogenous parameters (characteristics of agents) remains a critical theoretical concern. In finite-dimensional spaces, continuity is often studied under "regularity" assumptions requiring differentiability of demand functions. However, in infinite-dimensional commodity spaces—essential for modeling financial markets, dynamic growth, and commodity differentiation—such differentiability assumptions are often untenable. Furthermore, previous extensions to infinite dimensions have been restricted to specific topological spaces (e.g., separable Banach spaces with non-empty interiors in the positive cone) or specific integral definitions (Bochner integrals over norm-compact sets). The paper seeks to establish continuity results for a broader class of locally convex spaces without relying on differentiability, utilizing both Bochner and Gel'fand integrals to define aggregate endowments.

Methodology
The study employs a distributional approach to modeling economies, where an economy is defined as a Borel probability measure on a space of characteristics (preferences and endowments). The commodity space $L$ is modeled as a Banach lattice, and the price space is its dual or predual $L'$.

1. Topological Framework: The analysis utilizes a unified framework accommodating various topologies: the norm topology ($\|\cdot\|$), the Mackey topology ($\tau$), and the weak topology ($w$). The commodity space $L$ and price space $L'$ are paired via duality $\langle L, L' \rangle$. The paper considers cases where $L'$ is the topological dual $L^*$ or a predual of $L$.
2. Integral Definitions: The paper distinguishes between two types of economies based on how aggregate endowments are defined:
   • Bochner Economies: Total endowments are defined via Bochner integrals of strongly measurable functions, applicable when $L' = L^*$.
   • Gel'fand Economies: Total endowments are defined via Gel'fand integrals of weakly* measurable functions, applicable when $L$ is a dual space ($L = (L')^*$).
3. Space of Economies: The set of economies is endowed with the weak topology (Prohorov metric), making it a Polish space. The equilibrium correspondence maps an economy to the set of equilibrium allocations and prices.
4. Continuity Analysis: The authors utilize tools from fixed-point theory and measure theory, specifically:
   • Baire Category Theorem: To identify dense residual subsets where continuity holds.
   • Upper and Lower Hemi-continuity: The equilibrium correspondence is shown to be upper hemi-continuous generally, and lower hemi-continuous on a dense subset.
   • Individualized Representations: The paper links distributional economies (measures on characteristics) to individualized economies (measurable mappings from an atomless agent space) using Skorokhod's theorem and saturation properties.

Key Assumptions
The main continuity results rely on the following assumptions:

• (PR) Preferences: Preferences are continuous, and for every agent, either the bundle is locally non-satiated or it is a satiation point.
• (PC) Compactness: The space of preferences is compact.
• (IPVE) Individually Positively Valued Environments: For any price vector, almost all agents have strictly positive value for their endowments.
• (EP) Endowments-Prices: A condition ensuring the lower semi-continuity of the value of endowments under weak convergence of prices and endowments sequences.
• (SB) Satiated Bundles: If a bundle is satiated, it must be greater than or equal to the endowment.

Key Results

1. Continuity on a Dense Residual Set: The primary result (Theorem 5.5) establishes that the equilibrium correspondence is continuous on a dense residual subset (a "large" set in the sense of Baire category) of the domain of economies that admit equilibria. This holds without requiring differentiability of demand functions.
2. Unified Treatment of Integrals: The results apply to both Bochner and Gel'fand economies. This extends previous work (e.g., [7]) which was limited to separable Banach spaces with non-empty interiors. The current framework includes non-separable spaces like $L^\infty$, $\ell^\infty$, and spaces of measures $M(K)$.
3. Individualized Economies: The paper demonstrates that the continuity results for distributional economies translate to individualized economies (measurable mappings from agent spaces). It shows that atomless economies can be approximated by sequences of economies with finitely many agent types, and conversely, limits of such sequences can be atomless.
4. Applications to Specific Models: The framework is applied to:
   • Neoclassical economies with continuous commodities ($C(K)$).
   • Economies with non-renewable resources over infinite horizons ($c_0$, $\ell^1$).
   • Economies with renewable resources and continuous time horizons ($\ell^\infty$, $L^\infty$).
   • Models with commodity differentiation ($M(K)$).

Significance and Claims
The paper claims to provide a unified analytical treatment applicable to a substantially broader class of locally convex spaces than previously available. By avoiding differentiability assumptions, the results extend the scope of equilibrium continuity from "regular" economies to "essential" economies in infinite-dimensional settings.

The authors emphasize that their approach:

• Removes the necessity of separability and non-empty interior assumptions in the positive cone, which were restrictive in prior literature.
• Accommodates diverse topological structures (norm, Mackey, weak) within the same framework, allowing for the analysis of economies with infinite planning horizons, monopolistic competition, and asymmetric information.
• Demonstrates that the continuity of the equilibrium correspondence does not depend on the specific choice of integral (Bochner vs. Gel'fand) as long as the appropriate topological conditions are met.

The study concludes that while the existence of equilibria is a prerequisite, the continuity of the equilibrium correspondence is a robust property holding on a dense subset of the space of admissible economies, thereby reinforcing the explanatory power of general equilibrium theory in complex, infinite-dimensional environments.