Intrinsic Geometry and the Stability of Minimum Differentiation

We develop a framework for horizontal differentiation in which firms compete on a product manifold representing the feasible combinations of characteristics. This approach generalizes both the Hotelling line and Salop circle to any Riemannian space, allowing for a unified analysis of product space. We show that equilibrium existence and stability are governed by intrinsic geometric properties, specifically curvature, symmetry and dimension. We prove that negative curvature and high intrinsic dimension act as stabilizers of minimum differentiation equilibria, moving the analysis beyond the symmetry-induced instabilities found in simpler, fixed domains. By characterizing curvature as a measure of consumer heterogeneity, we demonstrate that intrinsic geometry is a fundamental determinant of competitive outcomes.

💡 Research Summary

The paper develops a unified geometric framework for horizontal product differentiation by modeling the set of feasible products as a smooth, compact, path‑connected Riemannian manifold M instead of the traditional line or circle. Consumers are assumed to be uniformly distributed with respect to the manifold’s volume measure, and their choice follows a multinomial logit specification in which utility depends on price and the intrinsic geodesic distance to a firm’s location. The parameter β controls the degree of rationality: as β →∞ the model converges to the deterministic Voronoi competition of classic Hotelling, while finite β introduces “soft” competition that guarantees the existence of pure‑strategy Nash equilibria under mild conditions.

Firms simultaneously choose price pᵢ and location yᵢ to maximize expected profit Πᵢ = (pᵢ − c)·Λᵢ, where Λᵢ is the aggregate demand derived from the logit probabilities. Social welfare is defined as the sum of consumer and producer surplus, which is equivalent to minimizing total social cost (production plus transportation). In a symmetric equilibrium with equal prices, welfare maximization reduces to minimizing the average geodesic distance between firms and consumers.

The core theoretical contribution is a geometric stability condition for the “minimum differentiation” (concentrated) equilibrium, where all firms locate at the same point (the geometric median). Two opposing forces are identified: a centrifugal force proportional to the variance of marginal profit gradients that pushes firms apart, and a centripetal force given by the average Hessian of the transport cost that pulls firms together. The authors prove that concentration is sustainable if and only if the intrinsic convexity of the market space (captured by curvature) dominates the dispersive pressure. Crucially, the centripetal force scales positively with the dimension d of the manifold, implying that higher‑dimensional product spaces tend to stabilize clustering.

Negative curvature (hyperbolic geometry) expands distances exponentially, making small product deviations appear large to consumers; this dramatically raises differentiation costs and therefore stabilizes the concentrated equilibrium. Positive curvature (spherical geometry) contracts distances, lowering differentiation costs and encouraging dispersion. The paper formalizes these insights in Table 1 and Figure 1, linking curvature, volume‑measure distortion, and economic incentives.

The authors also prove a general existence theorem (Theorem 1): on any compact, strongly geodesically convex manifold, there exists a threshold β₀ such that for all β < β₀ the simultaneous price‑location game admits at least one pure‑strategy Nash equilibrium. This extends the classic Hotelling result (where pure equilibria may fail) to a broad class of manifolds, including all Euclidean, hyperbolic, and Hadamard spaces.

A further contribution is the “Partial Concentration Theorem.” In multi‑attribute markets, some attributes may be “stable” (the market is symmetric and firms concentrate) while others are “unstable” (periodic or divergent differentiation). The theorem shows that equilibrium stability can decouple across attributes, allowing firms to cluster on core specifications while diversifying on peripheral features without destabilizing the overall market structure.

Methodologically, the paper introduces a retracted gradient‑ascent dynamics for numerical exploration of equilibria on complex manifolds, providing a practical tool for analyzing stability when analytical solutions are intractable.

Overall, the study demonstrates that intrinsic geometry—specifically curvature, dimensionality, and symmetry—constitutes a fundamental determinant of competitive outcomes in product markets. By embedding consumer heterogeneity directly into the manifold’s metric, the framework unifies and extends the Hotelling line, Salop circle, and numerous higher‑dimensional or curved‑space extensions under a single, mathematically rigorous theory of spatial competition.