The measurement of conjectural variations in an oligopoly industry

Using independently derived estimates for the market demand elasticity and firm marginal cost, this paper measures the conjectural variations (cv's) of the eight largest U.S. steel firms for the years 1920 to 1972. Comparisons are then made between the measured cv's and those predicted by certain industry conduct hypotheses. Specifically the hypotheses are those for competitive behavior, Cournot behavior, imperfect collusion, and industry profit maximization (perfect collusion). One of the two extreme theories of firm behavior, industry profit maximization, is rejected, but the acceptance or rejection of the other theories depends on the assumptions made about the cost structure of the sample firms.     

Bertrand Competition Under Uncertainty

We look at a Bertrand model in which each firm may be inactive with a known probability, so the number of active firms is uncertain. The model has a mixed-strategy equilibrium, in which industry profits are positive and decline with the number of firms, the same features which make the Cournot model attractive. Unlike those in a Cournot model with similar uncertainty, Bertrand profits always increase in the probability that firms are inactive. Profits decline more sharply than in the Cournot model, the pattern found empirically in Bresnahan and Reiss [1991].     

Testing oligopolistic behavior

This paper develops an empirical model of short-run pricing behavior for firms in a homogeneous product oligopoly. It is based on the variable profit function and related output supply and factor demand equations and is used to analyze the behavior of firms in the coffee roasting industry. A parametric test of the dominant firm model is developed and applied. The hypothesis of dominant firm behavior and Cournot behavior are rejected. The hypothesis of price-taking behavior is rejected for the largest two firms in the industry but is not rejected for all other firms. Using the estimated pattern of firm behavior, an index of overall industry performance is constructed.     

Price and quantity experimentation: A synthesis

This paper compares experimentation about product differentiation in a linear setting under four market structures: quantity-setting and price-setting monopoly, Cournot and Bertrand duopoly. Quantity-setting firms always experiment by raising their quantities and the monopolist experiments relatively more than the duopolists. A price-setting monopolist does not experiment. The value of information to Bertrand duopolists may be positive or negative depending on the degree of product differentiation. When information is valuable, price-setting duopolists experiment by lowering prices. A numerical example indicates that the intensity of experimentation is higher in a Cournot duopoly than in a Bertrand duopoly.     

Target pricing: Demand-side versus supply-side approaches

The practice of target pricing has been a key factor in the success of Japanese manufacturers. In the more commonly known demand-side approach, the target price for the supplier equals the manufacturer's market price less a percent margin for the manufacturer but no cost-improvement expenses are shared. In the supply-side approach, cost-improvement expenses are shared and the target price equals the supplier's cost plus a percent margin for the supplier. Using a general oligopoly and Cournot duopoly models, we characterize the equilibrium and optimal policy for each approach under various conditions. We find that sharing cost-reduction expenses allows the manufacturer using the supply-side approach to attain competitive advantage in the form of increased market share and higher profit, particularly in industrial conditions where margins are thin and price sensitivities are high.     

Is there a U-shaped relation between competition and investment?

We consider a two-stage game with cost-reducing investments followed by a linear differentiated Cournot duopoly. With competition inversely parameterized by the extent of product differentiation, investment in the subgame perfect equilibrium is typically minimal for intermediate levels of competition. Laboratory experiments partly confirm the U-shape in a reduced one-stage version of the game. In the two-stage version, there is no evidence for positive effects of moving from intermediate to intense competition.     

Horizontal mergers with synergies: Cash vs. profit-share auctions

We consider takeover bidding in a Cournot oligopoly when firms have private information concerning the synergy effect of merging with a takeover target and bidders can influence rivals' beliefs through their bids. We compare cash and profit-share auctions, first- and second-price, supplemented by entry fees. Since non-merged firms benefit from a merger if synergies are low, bidders are subject to a positive externality with positive probability; nevertheless, pooling does not occur. Unlike cash auctions, profit-share auctions are not revenue equivalent, and the second-price profit-share auction is more profitable than the other auctions.     

The role of obsolescence and inventory costs in providing commitment

This paper modifies the Cournot and Stackelberg models to allow the firms to sell less than they have produced and to store the unsold portion of their output as inventory. Even if one of the firms can choose its output before the other, if it is costless to carry unsold goods forward then the Cournot outcome emerges. We show, however, that costs of carrying the inventory forward endow the firms with a limited degree of commitment to sell what they have produced. This partially restores the first-mover advantage. If the firms choose their outputs simultaneously, the Cournot outcome emerges and has locally consistent conjectures.     

Common ownership,market power,and innovation

I examine the effects of overlapping ownership on market power when there are external effects across firms. This is done in an oligopoly model with cost-reducing innovation with technological spillovers where firms have an overlapping ownership structure based largely on López and Vives (2019). The model allows for Cournot competition with homogeneous product and for Bertrand with differentiated products as well as for strategic effects of R&D investment. It derives positive testable implications and normative results to inform policy.     

The intertemporal behavior of economic profits

The intertemporal behavior of economic profits is examined. Tests show that profit rates are not independent of their initial level. A set of tests for the nature of the profit adjustment mechanism is proposed and a stock market-based profit measure is employed. Despite intertemporal dependence, we find evidence of profit rate adjustment patterns roughly consistent with neoclassical theory for negative profits. In some, but not all cases, the probability of maintaining positive profits appears less than the probability of moving toward zero profits.     

Horizontal mergers, entry, and efficiency defences

In standard models of Cournot competition, it is well-known that if large-scale entry is impossible, then any merger failing to create technological synergies must harm consumers through a higher price level. This paper shows that this is true irrespective of entry conditions: any profitable Cournot merger failing to generate synergies must raise price, even if large-scale entry is possible or if the merger allows the avoidance of fixed cost duplication.     

The incentives for takeover in oligopoly

We present a model of takeover where the target optimally sets its reserve price. Under relatively standard symmetry restrictions, we obtain a unique equilibrium. The probability of takeover is only a function of the number of firms and of the insiders' share of total industry gains due to the increase in concentration. Our main application is to the linear Cournot and Bertrand models. A takeover is more likely under Bertrand competition if goods are substitutes, and more likely under Cournot competition if goods are complements.     

Vertical Market Participation

Firms that operate at both levels of vertically related Cournot oligopolies will purchase some input supplies from independent rivals, even though they can produce the good at a lower cost, driving up input price for nonintegrated firms at the final good level. Foreclosure, which avoids this strategic behavior, yields better market performance than Cournot beliefs.     

A Review of: “MODELING THE SUPPLY CHAIN” BY JEREMY R SHAPIRO Duxbury Press,Pacific Grove,California, 2001, xv + 586 pp. ISBN 0-534-37363-1. List: $71.95.

In this paper, we measure economic profit in an inflationary environment. Inflation adjusted economic profit requires two principal alterations to Economic Value Added (EVA[rgrave]). First, because inflation increases the replacement cost of non-current assets above book value, one must make a downward adjustment to EVA[rgrave] to correctly measure economic profit. Second, we add a term to EVA[rgrave] to represent a manager's ability to increase product price as costs increase. This adjustment is positive because this ability adds value to a firm. When a firm's assets are largely current, the first adjustment is not necessary, and therefore, the second dominates. In this case, EVA[rgrave] underestimates economic profit. Economic profit remains positive when a firm's rate of return on invested capital, after tax and after depreciation, is less than the weighted average cost of capital by as much as the rate of inflation. Even when inflation is modest, unadjusted EVA[rgrave] has the potential to seriously misrepresent the operating performance of a firm.     

Endogenous capacities and price competition: The role of demand uncertainty

We analyze the role of demand uncertainty in markets of fixed size, in which firms take long-run capacity decisions prior to competing in prices. We characterize the set of subgame perfect Nash equilibria under various assumptions regarding the nature and timing of demand uncertainty. In order to prove equilibrium existence, we identify a sufficient condition for the capacity choice game to be submodular. This condition resembles the standard downward-sloping marginal revenue condition used in Cournot games. A robust conclusion of the analysis is that equilibrium capacity choices are asymmetric, even when firms are ex-ante identical. Concerning the equivalence between the capacity-price game and the Cournot game, we find that with inelastic demands, the equilibria of the former belong to the equilibrium set of the latter. However, as compared to the Cournot game, the capacity-price game leads to lower prices and generates price dispersion.     

The incentives for cost reduction in a differentiated industry

This paper investigates how the incentives for cost reduction in a differentiated industry depend upon the degree of product substitutability. When goods are imperfect substitutes, both Cournot and Bertrand competition result in underinvestment in the sense that a social planner would be willing to pay more for a given cost reduction than a profit-maximizing firm. Overinvestment may occur when the goods are sufficiently close substitutes. Similarly, Cournot competition provides a stronger incentive to innovate than Bertrand competition if the degree of substitutability is low, and a weaker incentive if this degree is high.     

Competition and investment — A unified approach

Using a simple but general two-stage framework, this paper identifies the circumstances under which increasing competition leads to more cost-reducing investments. The framework can, for instance, capture increasing substitutability for different types of oligopoly models or changes from Cournot to Bertrand competition. The paper identifies four transmission mechanisms by which competition affects investment. For a firm with lower initial marginal costs (higher efficiency), a positive effect of competition on investment is more likely. Positive spillovers support a negative effect of competition on investment. The relation between competition and investment is not affected in an unambiguous way by the level of pre-existing competition.     

MARKET MAKING OLIGOPOLY*

This paper analyzes price competition between market makers who set costly capacity constraints before they intermediate between producers and consumers. The unique equilibrium outcome with pure strategies at the capacity stage is the Cournot outcome. The paper thus provides a rationale for Cournot‐type competition between market makers. This contrasts with previous findings in the literature, where due to the absence of capacity constraints that are set ex ante the Bertrand result typically obtains.     

When is it rational for firms to acquire silent interests in rivals?

The cartelizing effects of firms' shareholding in rivals have been well established theoretically, yet empirical instances of such are rate to nonexistent. It is argued here that if the stock market is efficient in the sense that share prices reflect post-share trading product market equilibria, then acquiring shares in rivals is not subgame perfect for Cournot oligopolies but is subgame perfect for some Bertrand oligopolies. Acquiring a silent interest in a rival is an example of the “fat cat ploy”. That the stock market is not a cartel maker is consistent with Cournot having wide empirical application, but not with Bertrand.     

Strategic incentives when supplying to rivals with an application to vertical firm structure

We consider a vertically integrated input monopolist supplying to a differentiated downstream rival. With linear input pricing, at the margin the firm unambiguously wants the rival to expand—unlike standard oligopoly with no supply relationship—for either Cournot or Bertrand competition. With a two-part tariff for the input, the same result holds if downstream choices are strategic complements, but is reversed for Cournot with strategic substitutes. We analyze vertical delegation as one mechanism for inducing expansion or contraction by the rival/customer.