NATURALLY OCCURRING PREFERENCES AND GENERAL EQUILIBRIUM: A LABORATORY STUDY
通过实验测量个体风险偏好，检验Sonnenschein-Mantel-Debreu定理指出的聚合病理（如均衡不存在或多重均衡）在真实经济中是否出现。发现同质经济中病理频繁，但随多样性增加而消失；实际交易中一般均衡预测多数准确，仅对先验分类为脆弱的失败。
Abstract We examine whether economies constructed using experimentally measured preferences (for risk) tend to suffer from aggregation pathologies (like nonexistence and multiplicity of equilibrium) cautioned by the Sonnenschein–Mantel–Debreu theorem. We show that aggregation pathologies should be expected to arise frequently in homogeneous exchange economies, but dwindle and eventually disappear as economies grow diverse. When subjects actually trade, general equilibrium predictions are accurate in most cases, and the failures occur only in economies a priori classified as fragile to preference instability. Our study uses individual‐level experimental data to address longstanding questions in general equilibrium theory that are unanswerable using prior methods.