Mergers and acquisitions (M&As) have been extensively used by firms as vehicles for growth and competitiveness within the context of the global economy (U.N., 2000; U.N., 2007). Re- stricting attention to horizontal M&A, Gugler et al. (2003) identify that over the period 1981-1998, nearly 16,500 were realized around the world (with each deal’s worth being over $1 million), where profits increased by attaining efficiency gains. The question that easily arises is, where do these efficiency gains arise from?
This paper explores how do pre-integration long-run decisions create endogenous efficiency gains that result in profitable horizontal M&As; and what are the welfare effects of these M&As. In particular, it attempts to address the following three questions.
First, how do the pre-integration investments in cost-reducing R&D create endogenous efficiency gains? This question is motivated, on the one hand, by the variety of the sources of efficiency gains addressed in the literature and the scant empirical evidence regarding these sources (for a survey see Röller et al., 2001), on the other hand. Straume (2006) argues that the available evidence suggests that “the most commonly indicated among these sources is the presence of pre-integration cost asymmetries”. This implies that the integration entails the reallocation of production, from the plants with the relatively high marginal costs to those with the relatively low, without increasing the joint technological capabilities. However, in the bulk of the relevant literature, the pre-integration cost asymmetries are assumed to be exogenous. The purpose of this paper is precisely to study how firms invest strategically in cost-reducing R&D, prior to the decision for integration, so as to establish endogenous cost asymmetries and exploit efficiency gains.
Second, how does the method of payment through which the integration is realized affect the firms’ R&D investments and integration incentives? Empirical evidence suggests a great variety on the methods of payment in M&As (see Faccio and Masulis, 2005 and the references therein). On the other hand, recent advances in modelling the integration process are divided in two strands. The first one restricts attention on “the limits of monopolization” through non-cooperative acquisition transactions where each firm posts bids for the other firms and commits to a purchase price for its own (Kamien and Zang, 1990; Ziss, 2001; Gonzàlez-Maestre and López-Cuñat, 2001; Inderst and Wey, 2005). The second one, the endogenous mergers strand, focuses on “whether a particular market structure can be the outcome of a merger process” (Banal-Estañol et al., 2008). In particular, it compares all the possible mergers in order to find the equilibrium one (Barros, 1998; Horn and Persson, 2001; Lommerud et al., 2006; Straume, 2006). Besides Barros (1998), this latter strand focuses on the equilibrium post- merger firm’s profits while their sharing across participants, although crucial, is not explored. Our contribution is to point out that it is precisely the rule of sharing of the integrated firm’s profits across participants that determines the firms’ R&D investments and integration incentives.
Third, do the efficiency gains exploited by integration paricipants create any wealth gains for consumers? Moreover, should these efficiency gains be incorporated in the merger control? These question are crucial and have been motivated by the ongoing debate regarding the policy goals of merger control (Neven and Röller, 2005). On the one hand, Williamson (1968) along with Farrell and Shapiro (1990) argue that competition authorities should assess M&As on the basis of total welfare, implying that they should be allowed, even if their effect on consumers alone is negative. On the contrary, the US Department of Justice Merger Guidelines and the European Merger Guidelines assess M&As on the basis of their effects on consumers surplus.
To address the above questions, we consider a three-firm Cournot oligopoly industry with homogenous products and initially symmetric marginal costs across competitors. The timing of moves is as follows. In stage one, firms invest in perfectly substitutable cost-reducing R&D activities. In the second stage, they decide whether to integrate or not, under the constraint that the antitrust authority does not allow a monopoly to be formed. At the final stage, firms compete in quantities.
We formalize two different types through which the integration can be formed, depending on the rule of sharing of the integrated firm’s profits across participants: in the acquisition- type, the bidder firm, by paying a purchase price, acquires the target firm. This type of integration captures the stylized facts reported by Cramton (1998) according to which, target firms employ several tactics so as to maximize the purchase price to receive. In the present paper, the target firm invests in R&D so as to maximize its purchase price, but this induces wasteful duplication of R&D activities. In the merger-type instead, participants share the integrated firm’s profits based on a pre-merger exogenously determined rule, while one of them abstains from investing in R&D and avoids the wasteful duplication of R&D activities. The type of integration is assumed to be exogenously given. The above formalization implies that although firms are initially symmetric, they do not invest equal amounts in R&D, ending up with asymmetric marginal costs.
Then, if the integration takes place, regardless of the type chosen, the integrated entity produces all its output with the minimum of the marginal costs of its constituent parts while the less efficient one shuts down (Barros, 1998; Straume, 2006). It is precisely this reallocation of production that induces the efficiency gains for the integrated entity. On the other hand, the post-integration higher concentration in the output market implies a relatively higher market share and price for the firm that does not participate in the integration. This is the “business stealing” effect outlined by Stigler (1950) and Salant et al., (1983) which negatively affects the profits of the integrated firm. In this environment, we propose a candidate equilibrium configuration, assigning a special role to each firm. Then, we check whether, or not, it survives all possible deviations, in roles and configurations. This means that, given the R&D investments, all firms are allowed to choose whether to integrate and with whom.
In the acquisition-type of integration, the bidder’s R&D investments increase in the target’s investments so as the former to maximize its post-acquisition profits, net from the purchase price. The target firm’s R&D investments induces wasteful duplication of R&D activities and increases the cost of the integration. Yet, the post-acquisition firm’s profits are further decreased by the business stealing effect. On the contrary, the aforementioned profits are positively affected by the efficiency gains through the rationalization of production. We find that for the rationalization of production effect to dominate both the wasteful duplication of R&D and the business stealing effects, the R&D investments must be sufficiently effective in the reduction of the marginal cost. Moreover, in the unique stable acquisition, the most efficient firm in the industry acquires the least efficient one. This happens for two reasons. Firstly, because of the business stealing effect, the firm that has been assigned to be the outsider has no incentives to deviate towards becoming either the bidder or the target. Secondly, the proposed bidder’s R&D investments are so high that its purchase price is not affordable by any firm.
In the merger-type of integration, there is no wasteful duplication of R&D activities, since one of the merger participants abstains from investing in R&D. This implies that the stable merger requires a relatively less effective R&D than the stable acquisition. We find that two firms have incentives for merger, if the one that invests in R&D receives a sufficiently high share of the post-merger profits, while the firm that abstains from investing is effectively compensated to shut down. Note that the only stable merger is formed between the two least efficient firms in the industry, contrary to the acquisition. Intuitively, the firm that has been assigned to be the outsider, in the candidate equilibrium configuration, has no incentives to deviate towards participating in any merger configuration, due to the business stealing effect.
Interestingly, the welfare effects of horizontal integrations depend crucially on the type of integration chosen and the effectiveness of the R&D investments that determines the magnitude of the efficiency gains. We find a significant contradiction between private and social incentives for horizontal integrations in the sense that the effectiveness of the R&D investments that guarantees a welfare-enhancing integration, under both types, is higher than that required for private profit-enhancing. Moreover, the respective effectiveness that guarantees a welfare- enhancing acquisition is higher than that for a welfare-enhancing merger. The latter underlines the dead-weight loss caused by the target firm’s strategic behavior in the acquisition, that increases the integration’s cost. We also find that whenever a horizontal integration increases total welfare, it is because of the increased industry profits. Therefore, from a competition policy perspective, we argue that policy measures for the assessment of horizontal integrations should carefully be designed, taking into account the relative efficiency of the participating firms and the transaction through which the integration will take place.
Our paper contributes to the literature on horizontal M&As with efficiency gains. Similarly to us, a recent branch of this literature has focused on endogenous mergers with efficiency gains arising from cost asymmetries across participants. In particular, Barros (1998) and Straume (2006) demonstrate that the number of equilibrium mergers and the identity of each equilibrium’s participants depend on the level of cost asymmetry among firms. For moderate cost asymmetries, they find that the equilibrium merger occurs among the least and the most efficient firm. If instead the cost asymmetries are large, Barros (1998) finds that the merger takes place between the two most efficient firms, while in Straume (2006), the merger occurs among either the less and the most efficient firm or the two least efficient firms. Besides the problem of multiple equilibria present in these papers, they do not explore how did the cost asymmetries arise. Undertaking this task in the present paper, we demonstrate that it is precisely the pre-integration decisions and the sharing of the integrated firm’s profits across participants that leads to endogenous cost asymmetries and efficiency gains where a unique stable integration is formed.
Our paper also contributes to the literature for the effects of M&As decisions on R&D investments. A number of papers within this literature deals with the R&D investments problem of firms in the prospect of horizontal M&As (Stenbacka, 1991; Wong and Tse, 1997; Canoy et al., 2000; Socorro, 2004). However, they restrict attention to R&D incentives solely. In the present paper instead, we broaden the analysis and explore the endogenous efficiency gains, the relative efficiency of the integration’s participants and the integration’s welfare effects.