The effect of mergers on the incentive to invest in cost-reducing innovations

Abstract

Both mergers and innovation are central elements of a firm's competitive strategy. However, model-theoretical analysis of the merger-innovation link is sparse. The aim of this paper is to analyze the impact of mergers on innovative activities and product market competition in the context of incremental process innovations. Inefficiencies due to organizational problems of mergers are accounted for. We show that optimal investment strategies depend on the resulting market structure and differ significantly from insider to outsider. In our linear model mergers turn out to increase social surplus.

Keywords:

• horizontal mergers
• innovation
• research joint venture
• market structure

JEL Classification :

• L13
• L22
• G34
• O31

Acknowledgements

I am grateful to Norbert Schulz for continuous guidance, encouragement, and assistance. I would also like to thank the Editor Christiano Antonelli, Peter Welzel, Liliane Karlinger, two anonymous referees, seminar participants at the 2nd BGPE research workshop, and participants of the 1st conference of the research network on innovation and competition policy for helpful comments.

Notes

While this is true for the US, innovation defence is a smaller issue in Europe. We refer to Cefis et al. (2007) for a comparison of the role of innovation in merger policies between the US and the EU.

This knowledge level is also called absorptive capacity (see Cohen and Levinthal 1989).

Huck, Konrad, and Müller (2004) provide similar arguments for modeling horizontal mergers that way and derive comparable results. However, their argumentation is a bit different as they use an endogenous timing approach. In our setup, there is no preference for either of these two motivations for the internal structure change due to the merger.

The parameter k has to be that high to assure that innovations are non-drastic in all cases, i.e., to guarantee interior solutions. We derive this barrier in the appendix. In the following, we refer to k as innovation costs while K(c) is called total innovation costs.

The lower bound for these initial marginal costs is dependent on the innovation cost parameter k and is derived in the appendix.

This assumption is not necessary in our model. However, some results concerning changes in total profits may change quantitatively, depending on how a merger or an R&D joint venture affects fixed costs.

However, the innovation cost parameter k can be reinterpreted to allow for time preferences.

Second-order conditions are satisfied here and in all other cases (see appendix).

In order to improve readability, we just provide the intuition and interpretation of the lemmas. All proofs are deferred to the appendix.

This can be easily verified looking at the profits of insider and outsider for identical marginal costs and no investment in innovation.

This result is in line with Daughety (1990) who states that mergers and increases in concentration can be socially beneficial if an asymmetric situation occurs.

An overview of empirical evidence for this type of competition in multidivisional firms is given in Creane and Davidson (2005) and in Huck, Konrad, and Müller (2004).

This scenario can also be interpreted as a situation where firms merge but output decisions are delegated to managers like in the CD case. However, the managers now move simultaneously rather than sequentially.

The superior performance of R&D joint ventures may be caused by less integration problems of the R&D department. This is addressed in Section 3.

Of course, there are other reasons to merge, like diversification or disciplining of management, see e.g., Matsusaka (1993), that are not included in our model.

There are exceptions for the outsider's investment in cases D and CD for a certain parameter range (proofs are available from the author upon request). However, the insider's effect is dominant.

We restrict α by 2 to assure that total innovation costs after integration are not higher than the sum of total innovation costs of the two former independent R&D departments (for a given cost reduction).

Calculations are basically the same as before and are given in the appendix. The innovation decision in the joint venture case is still identical to case D.

Without integration problems, identical marginal costs for the insider and outsider are equivalent to identical investment. With integration problems, to assure identical marginal costs, insiders total investment have to be higher by the factor α. To assure comparability, we only compare resulting marginal costs of the insider, meaning effective investment.

For we have the highest lower barrier c ⁰≥0.25.