Bertrand Competition Paradox
The Bertrand paradox demonstrates that two competing firms producing homogeneous goods with constant marginal costs and simultaneous price setting will drive the equilibrium price down to marginal cost, earning zero economic profit.
Bertrand competition paradox is a counterintuitive proposition in industrial organization economics, first published by French mathematician Joseph Bertrand in an 1883 book review of Antoine Augustin Cournot's work. Bertrand argued that in real-world retail and commercial commerce, competing enterprises choose prices rather than production quantities. In a duopoly where two identical firms produce an undifferentiated homogeneous good, possess unbounded production capacity, and post prices simultaneously, consumers purchase from the lower-priced vendor.
The resulting game theoretic equilibrium constitutes the paradox: despite having only two competitors in the entire industry, the unique Nash equilibrium price collapses to marginal cost, eliminating all economic profits. The market outcome mimics the allocative efficiency of perfect competition with infinite firms. Economists label this result a paradox because observed duopolies in modern business rarely engage in zero-profit price wars, prompting extensive research into the structural assumptions—capacity constraints, product differentiation, search frictions, and repeated interaction—that resolve the paradox in real economies.
Mathematical Formulation and Discontinuous Demand
Consider two identical firms, Firm 1 and Firm 2, producing an identical substitutable good. Both firms face a constant marginal cost c and zero fixed costs. Firms announce their selling prices p1 and p2 simultaneously and non-cooperatively.
The profit function for Firm 1 is discontinuous with respect to price:
Suppose both firms post a price p strictly above marginal cost (p > c). Each firm earns half of the duopoly profit. However, either firm can capture the entire market by undercutting its competitor's price by an infinitesimal amount (setting price to p - epsilon). Because doubling sales volume more than compensates for the negligible price concession, each firm faces an irresistible unilateral incentive to undercut its rival.
This undercutting cascade halts only when price equals marginal cost (p1 = p2 = c). At this point, neither firm can lower its price without incurring operating losses, and neither can raise its price without losing all sales to its rival. Thus, the unique pure strategy Nash equilibrium is p1* = p2* = c, yielding zero economic profit for both competitors.
Resolving the Paradox: Theoretical Extensions
Because empirical oligopolies routinely sustain positive operating margins, economists identified four foundational mechanisms that break the zero-profit Bertrand outcome:
Capacity Constraints and Edgeworth Cycles
Francis Ysidro Edgeworth demonstrated that if firms face physical production bottlenecks and cannot satisfy the entire market demand alone, the zero-profit equilibrium collapses.
If Firm 1 lacks sufficient capacity to supply all consumers, Firm 2 can charge a price above marginal cost to capture the residual unsatisfied consumers. In a landmark 1983 paper, David Kreps and José Scheinkman proved that in a two-stage game where firms first invest in production capacity and subsequently compete in Bertrand prices, the equilibrium yields capacities and prices that replicate Cournot duopoly outcomes. Capacity pre-commitments soften price competition, demonstrating how capital investments shield firms from Bertrand commoditization.
Product Differentiation and Monopolistic Competition
In modern markets, consumer goods are rarely identical commodities. When goods exhibit horizontal or vertical differentiation (such as brand reputation, quality, or geographic proximity), consumer demand functions become continuous rather than step-discontinuous.
If Firm 1 lowers its price slightly, it attracts price-sensitive consumers but does not strip Firm 2 of brand-loyal patrons. Because each firm retains market power over its core customer segment, both firms set prices strictly above marginal cost in equilibrium, earning positive profits proportional to the degree of consumer product differentiation.
Dynamic Price Wars and Edgeworth Cycles
When firms compete dynamically across sequential periods, Eric Maskin and Jean Tirole demonstrated that Bertrand price competition can produce cyclical pricing dynamics known as Edgeworth cycles.
In an Edgeworth cycle, competing retailers progressively undercut each other by small decrements to steal market volume. As prices approach marginal cost and profits approach zero, a firm abruptly resets its price to a high monopoly baseline, accepting temporary volume loss. The rival firm then follows by undercutting the new high price, initiating a fresh descent. These asymmetric price fluctuations are widely observed in retail gasoline markets, where fuel stations cycle between gradual daily price cuts and abrupt price restorations.
References
- Bertrand, J. (1883). Review of "Théorie mathématique de la richesse sociale" and "Recherches sur les principes mathématiques de la théorie des richesses". Journal des Savants, 67, 499–508.
- Kreps, D. M., & Scheinkman, J. A. (1983). Quantity precommitment and Bertrand competition yield Cournot outcomes. The Bell Journal of Economics, 14(2), 326–337. https://doi.org/10.2307/3003636
- Maskin, E., & Tirole, J. (1988). A theory of dynamic oligopoly, II: Price competition, kinked demand curves, and Edgeworth cycles. Econometrica, 56(3), 571–599. https://doi.org/10.2307/1911701
- Dastidar, K. G. (1995). On the existence of pure strategy Bertrand equilibrium. Economic Theory, 5(1), 19–32. https://doi.org/10.1007/bf01213642
- Baye, M. R., & Morgan, J. (1999). A folk theorem for one-shot Bertrand games. Economics Letters, 65(1), 59–65. https://doi.org/10.1016/s0165-1765(99)00118-400118-4)