Sequential horizontal mergers
Introduction
Past mergers in an industry act as a driving force for future mergers, according to business advisors.1 In contrast to this observation, however, the theory of horizontal mergers is mostly focussed on a single merger seen in isolation. Our aim is to help filling this gap between fact and theory. We propose a shift of focus to cases of multiple mergers, in particular to situations where merger decisions are interrelated over time. We suggest a way of separating out a strategic motive to merge in cases where one's own merger affects future merger decisions by rivals. We find that introducing such a strategic motive may have significant effects on how mergers should be evaluated, with respect both to profits and to competition policy.
The starting point of our analysis is the effect a merger has on the profits earned by non-merging firms in the same industry; this effect is positive if there is no cost saving to gain for the merging firms but may be negative for sufficiently large savings on variable costs following a merger. Suppose that two different mergers are about to form and that these formations take place in sequence. The firms involved in the first merger may want to encourage or discourage the second merger, depending on how the latter affects the profits of non-merging firms. The effect of a merger on the profitability of subsequent mergers is, on the other hand, ambiguous. We argue that a merger that is unprofitable in isolation may be carried through if it encourages a subsequent merger that has a positive effect on the first group, or if it discourages one that has a negative such effect. In the same vein, a merger may not be carried through, even if it is profitable in isolation, if the merger would encourage a subsequent merger with a negative side effect or if it would discourage one with a positive side effect.
We suggest that the general considerations above can be phrased in terms of Fudenberg and Tirole's (1984) taxonomy of business strategies, and we set out to do so in Section 3below, after discussing some related literature in Section 2. In Section 4, we present an analysis of one particular and simple model, in which a sequence of merger decisions is followed by Cournot competition with linear demand and costs, and find that all four different merging strategies that we introduce can be established as equilibrium outcomes.
Our analysis also has a bearing on merger policy. According to Farrell and Shapiro (1990), a proposed merger should be permitted if its external effect, i.e., the combined effect on consumers and non-merging firms, is positive. They specify a general condition for a single merger in isolation to have a positive external effect. In Section 5, we show that, when merger decisions are interrelated over time, their general condition for a single merger is no longer a sufficient criterion for permitting only welfare-increasing mergers. A motive for merger may be to affect rivals' future merger decisions, either to trigger price-increasing mergers or to stop price-reducing mergers. By ruling out such motives, the Farrell–Shapiro criterion applied to an isolated merger may therefore overestimate its actual external effect.
We conclude in Section 6, while an appendix contains details of calculations reported in 4 A specific model, 5 The Farrell–Shapiro criterion reconsidered.
Section snippets
Related literature
The literature on horizontal mergers has two main themes: the implications of a merger on profits and welfare, respectively. With regard to merger profitability, Stigler (1950) points out that the new firm created by a merger typically produces less than the combined, pre-merger production of its constituent firms. Unless there are large cost savings associated with the merger, this reduction in quantity increases the industry price. The non-merging firms will then expand their production and
A general discussion
We start out in this section with introducing the necessary notation in Section 3.1before proceeding to the analysis in Section 3.2.
A specific model
In this section, we show how all four possible business strategies of Proposition 2 are attainable in a simple oligopoly model when one allows for cost savings following merger. We apply the model of Salant et al. (1983). There are initially n identical firms in a homogenous-goods industry, and firm j has the cost functionwhere F>0 denotes fixed costs, c∈(0, 1) (constant) marginal costs, and qj quantity produced by firm j. To ensure that πj≥0, we assume that F is sufficiently small.
The Farrell–Shapiro criterion reconsidered
Our general discussion in Section 3can be related to the question of whether competition authorities should permit a proposed merger. Since any proposed merger is presumably privately profitable, a sufficient condition for a merger to increase welfare is that the non-participants, i.e., the non-merging firms and the consumers, jointly are better off as a result of the merger. This external-effect approach was first proposed by Farrell and Shapiro (1990). They analyzed, however, only a single
Concluding remarks
In this paper, we have tried to spell out some consequences of merger decisions being interrelated over time, paying particular attention to the profitability, strategic nature, and welfare implications of mergers in such a perspective. The present theoretical work should be seen in the light of the empirical fact that mergers occur in waves (Town, 1992). However, as should be clear, ours is not a theory of merger waves. Rather, we have introduced a set-up for analysing the sequential nature of
Acknowledgements
We would like to thank two anonymous referees, as well as Eirik Kristiansen, Halvor Mehlum, and participants at the 1995 EARIE Congress in Nice and the 1995 EEA Congress in Prague for valuable comments. Our research has been partly financed by the Research Council of Norway through the Foundation for Research in Economics and Business Administration.
References (35)
- et al.
Competitively cost advantageous mergers and monopolization
Games and Economic Behavior
(1991) - et al.
Shared ATM networks and banking competition
European Economic Review
(1994) - et al.
Competitive-share group formation in rent-seeking contests
Public Choice
(1995) Mergers and malls
Journal of Industrial Economics
(1994)Did antitrust policy cause the Great Merger Wave?
Journal of Law and Economics
(1985)Endogenous structures of association in oligopolies
RAND Journal of Economics
(1995)- et al.
The foreclosure effects of vertical mergers
Journal of Institutional and Theoretical Economics
(1991) Corporate mergers in international economic integration
Vertical vs. horizontal integration: Pre-emptive merging
Journal of Industrial Economics
(1995)- et al.
Incentives to form coalitions with Bertrand competition
RAND Journal of Economics
(1985)