National versus international mergers in unionized oligopoly.

We analyze how the presence of trade unions affects the pattern of mergers in an international oligopoly and the welfare implications thereof. We find that wages for the merger participants are always lower when they merge internationally, rather than nationally. Using a model of endogenous merger formation, we find that the firms will merge internationally in equilibrium. There are more international mergers than socially preferred, unless products are close substitutes. A "national champion" policy of promoting domestic mergers rather than international ones is nevertheless never optimal.

1. Introduction

* International mergers increasingly shape the industrial structure of developed and developing economies alike. (1) This is probably a natural development. At some stage, domestic economies of scale are exhausted. In addition, economic integration means that not only trade but also the market for corporate control is liberalized. The question remains, though, whether firms can also have strategic reasons for choosing an international rather than a national merger. The purpose of this article is to apply an international oligopoly model to analyze how the interplay between the labor market and the product market may affect firms' merger decisions. Could it be that firms merge internationally rather than nationally to curb the market power of trade unions? If so, will we observe a higher number of international mergers than would be optimal seen from a welfare point of view, or perhaps that international mergers supplant domestic ones to an excessive degree?

To analyze such questions, a natural starting point would be the existing models on mergers and merger policy in open economies. (2) However, most of the existing literature is about domestic mergers with spillovers on foreign agents, often focusing on the interplay between merger policy and trade policy. In contrast, we focus on firms' choice between a domestic and a cross-border merger. (3) Horn and Persson (2001 a) suggest that cooperative game theory could be used to pinpoint which industry structure will materialize when many different mergers are possible. We apply this method to solve for the equilibrium market structure when we allow for any two-firm merger in a situation with four firms initially. (4) As a robustness check, we also describe two versions of a noncooperative acquisition game that yield the same prediction about market structure as the cooperative framework. (5)

The novel feature of the present article is the focus on the interaction between market power in the product market and in the labor market. Already, Brander and Spencer (1988), Davidson (1988), Dowrick (1989), and De Fraja (1993) have suggested that oligopoly power in the product market might be an important reason why trade unions have the potential to influence wage setting. (6) Empirical studies suggest that mergers in the product market--which leads to higher concentration--may in fact influence wages. But the picture is mixed. Some studies find that a merger leads to higher wages, while others find the opposite result or no effect at all. (7) Unfortunately, there are few theoretical studies that can guide us on how mergers should be expected to affect wages. Our article helps to fill this gap by showing how different types of mergers can have distinctly different effects on wages and in turn on profits and welfare.

A core idea in our article is that an international merger can tilt the power balance between employers and workers. We study an international Cournot oligopoly with two domestic and two foreign firms, where wages are set by monopoly trade unions. The analysis rests further on the assumption that it is easier for workers to organize within, as opposed to across, national borders. (8) This notion is most conveniently implemented by letting trade unions be national by assumption: any firm operating in a given economy meets the wage claims of the relevant national union. As long as there are national unions--or at least that unions within a nation cooperate more easily than unions in different countries--then an international merger, as opposed to a national one, will imply that the merged firm meets two uncoordinated unions.

Since we model market power both in the input (labor) and output market, a merger will change both output prices and wages. A national merger makes market shares less sensitive to wage changes, which gives the unions an incentive to raise wages, and more so for the firms not taking part in the merger. An international merger, on the other hand, has a distinctly different effect on the unions' wage setting. An international merger would imply that the merged firm is served by two different unions, each producing input to one of the merged firm's two products. Then the merged firm can partly replace sales of one of its products by increasing the sales of the other product. Since an international merger leads to such a flexibility, it triggers increased competition between the unions. As a result, the unions compete more fiercely on wages. (9) We also allow for the possibility of exogenous merger synergies in the form of nonlabor cost savings for the merger participants. The presence of such cost savings improves the competitive position of the merged entity, which tends to increase wages for the merger participants and lower wages for the outside firms. However, due to the effects of different types of merger on union rivalry, wages are always lower for the merger participants if they merge cross-border rather than domestically.