In the latest ThoughtLeaders4 Competition quarterly magazine, European experts Gianmarco Calanchi and Josep Peya discuss the renewed debate – sparked by the Draghi Report and the European Commission’s merger-guidelines consultation – over whether mergers among innovators promote or hinder innovation. The authors outline the framework emerging from a decade of economic research, which has evolved from early presumptions of reduced innovation absent efficiencies to more recent, richer and balanced findings.
The views and opinions expressed in this paper are those of the author and do not necessarily reflect the views of Econic Partners or its clients.
Introduction
Nearly a decade after the landmark Dow/DuPont and Bayer/Monsanto decisions,[1] the debate over whether mergers among (established)[2] innovators foster or hinder innovation has regained centre stage. Two key developments reignited interest: the Draghi Report’s[3] proposal of an “innovation defence” for mergers that may reduce competition short-term but boost long-term innovation; and the European Commission’s recent consultation on merger guidelines, which acknowledged that innovation outcomes vary by context and sought to build an appropriate assessment framework.[4] Meanwhile, scholars have been hotly debating the “presumption” – based on two seminal articles[5] which underpinned the Dow/Dupont and Bayer/Monsanto decisions – that mergers between innovators reduce innovation absent efficiencies. Some contributions have identified additional factors that, in certain context, may overturn this presumption.
It is therefore time to take stock and outline the framework for assessing the impact of mergers on innovation that is emerging from this literature. Clearly, while innovation is a key consideration, the ultimate question is whether mergers can harm consumer welfare, and they can do so even if they enhance innovation. Therefore, agencies must weigh innovation effects against price impacts, but this is beyond the scope of this article. We also focus on “product” innovation, i.e., the introduction of new or better products, which seems the real focus of the recent discourse. We cannot capture all the subtleties of the emerging literature: our purpose is rather to paint the direction of travel in broad strokes.
Ability
Mergers can increase the parties’ ability to innovate by removing roadblocks that hinder innovation – such as limited scale, financial constraints,[6] or lack of key assets like technology or IP rights. These issues have emerged, for example, in a number of telecom mergers.[7] Mergers may also hinder innovation by introducing new roadblocks – for example, exhausting the acquirer’s financial resources in the acquisition, at the expense of innovation. Importantly, even if incentives to innovate decline, improved ability alone may still lead to increased innovation efforts.[8]
Incentives – A Presumption of Anticompetitive Effects in Early Contributions
But the focus of the current debate is on innovation incentives. During the Dow/DuPont and Bayer/Monsanto cases, two influential articles[9] argued for a “presumption” against mergers between competing innovators, absent “efficiencies”. Other contributions corroborated these results in somewhat different settings.[10] The authors identified two key channels through which a merger between established innovators affects innovation:[11]
- Innovation externality. A firm’sinnovation efforts are monetised throughthe additional sales won from the other party and third parties (or by better protecting its own sales from rivals). With the merger, each party gains ownership of the other party’s sales even without innovating, so the incentives to innovate become lower. This effect is stronger when the parties are close competitors.
- Price coordination effect. To the extent that the merger reduces competition, the parties’ profits can be higher both in the innovation and no-innovation scenarios. If the merger increases post-innovation profits more than pre-innovation profits, then it raises incentives to innovate, otherwise it reduces them. The effect on innovation is therefore ambiguous.
Incentives – The Richer Framework Emerging from Recent Contributions
In principle, which effect dominates is ambiguous, but Federico et al. (2018) finds that the innovation externality tends to dominate in concentrated markets.[12] So, in concentrated markets, this result would support the presumption that mergers reduce innovation, absent efficiencies.
However, other contributions identified three key additional effects to account for.
The most interesting one by far is the parties gaining access to each other’s innovations.[13] Before the merger, each party would pursue a separate line of research and (absent spillovers discussed below) would not share its results with others. With the merger, each party can access the innovation results of the other party. Sharing results means that whenever one party succeeds in innovating, the other party does too.[14] In other words, post-merger the parties can take two bites at the cherry, thereby increasing their chances to succeed at innovating.[15]
This effect has been described as an “efficiency”, part of a potential “efficiency defence”.[16] Now, taking a step back, the idea behind separating efficiencies from a “no-efficiency benchmark scenario” has always been to distinguish “endogenous” effects directly deriving from joint ownership, which agencies can assess using similar tools as other unilateral affects, from “exogenous” efficiencies, which may or may not arise, are very case specific and harder for agencies to assess since key information lies with the parties.
Gaining access to each other’s innovations is clearly an “endogenous” direct effect of joint ownership, much like accessing each other’s sales. With the merger, each party acquires ownership of two important market outputs which were previously enjoyed individually: (i) R&D outputs and (ii) sales (and associated revenues). Sharing sales gives rise to the innovation externality, which reduces the incentives to innovate; sharing R&D advances gives each party the extra chance to succeed through the other party’s line of research, thereby increasing innovation. The net effect on innovation is ambiguous: innovation declines if the innovation externality trumps the benefits from accessing each other’s innovations, otherwise it increases.
It is not clear it is even possible to construct an internally consistent benchmark scenario excluding this effect. It would mean assuming that the parties would share price information, cost functions and demand elasticity information to jointly set prices; would share information on their R&D processes, their current and expected R&D efforts and R&D cost functions to jointly set R&D efforts; but would not share their R&D advances or assume they would own such results when jointly setting their R&D efforts, despite in fact owning them. This is different from “exogenous” efficiencies like the reorganisation of the R&D process (by, e.g., removing duplications, optimising work across the teams, sharing equipment, etc.) and the parties’ production processes.
Another important factor is the market expansion effect.[17] Innovations may give rise to market expansions, i.e., increase market demand. Advances in areas like generative AI not only shift demand between suppliers but also increase total adoption. So, by innovating, a firm may not only increase its chances of stealing sales from others, but also the size of the pie. Market expansion increases post-innovation profits both with and without the merger. To the extent that sharing knowledge among the parties increases the probability of innovating, then the market expansion is greater with the merger.[18]
Finally, internalisation of spillovers.[19] Firms may be unable to fully appropriate the results of their innovations if there are spillovers to competitors (e.g., if they can easily copy without legal repercussions). A merger internalizes knowledge spillovers between the parties, thereby helping them capture more of the benefits of their R&D, which strengthens their incentives to innovate.
Conclusions
When all five effects are considered, the framework that emerges supports a balanced view of innovation effects of mergers which are neither necessarily detrimental nor beneficial to innovation and require a case-by-case assessment.
This article was originally published in ThoughtLeaders4 Competition quarterly magazine.
[1] European Commission (“EC”), case M.7932 – Dow/DuPont, Decision of 28 July 2017; EC, case M.8084 – Bayer/Monsanto, Decision of 29 May 2018.
[2] As opposed to the acquisition of nascent innovators, which triggered a related debate about killer and reverse killer acquisitions that is beyond the scope of this article.
[3] Mario Draghi. “The future of European competitiveness. Part B: In-depth analysis and recommendations.” 2024.
[4] EC’s Review of the Merger Guidelines: The In-Depth Consultation. Available at: https://competition-policy.ec.europa.eu/mergers/review-merger-guidelines_en.
[5] Giulio Federico, Gregor Langus, and Tommaso Valletti. “A simple model of mergers and innovation.” Economics Letters, vol. 157, issue C, 2017, pp. 136-140 (Federico et al., 2017); and Giulio Federico, Gregor Langus, and Tommaso Valletti. “Horizontal mergers and product innovation.” International Journal of Industrial Organization, vol. 59, 2018, pp. 1-23 (Federico et al., 2018).
[6] Financial constraints are sometimes dismissed assuming firms can rely on external funding if projects are profitable. This overlooks real-world financial market imperfections like information asymmetries that can severely limit access to external capital.
[7] For example, in Vodafone/Three (Vodafone/CK Hutchison JV), the parties argued they were sub-scale; in WIND/Tre (Hutchison 3G Italy/WINDJV, M.7758), the parties argued that WIND was financially constrained and Tre was sub-scale.
[8] For example, if the parties cannot innovate pre-merger but can innovate post-merger.
[9] Federico et al. (2017) and Federico et al. (2018).
[10] Massimo Motta and Emanuele Tarantino. “The effect of horizontal mergers, when firms compete in prices and investments.” International Journal of Industrial Organization, vol. 78, no. 2, 2021.
[11] This is the nomenclature used in Federico et al. (2018). Different scholars have grouped these effects differently and used different labels.
[12] Tommaso Valletti. “The innovation theory of harm in merger control: Some clarifications.” Economics Letters, vol. 255, 2025. (Valletti, 2025).
[13] This effect was first described in relation to process (rather than product) innovation by Arijit Mukherjee. “Merger and process innovation.” Economics Letters, vol. 213, 2022. We believe this is an important factor to consider for product innovation too.
[14] That is if innovation is fully relevant to the other party; partial relevance only partly increases the other party’s probability of success; this effect disappears if the innovation is irrelevant to the other party – but this calls into question whether they were competing in R&D in the first place.
[15] While they may individually invest less after the merger, their overall probability of success tends to increase. Third parties may react by increasing R&D investments, thereby increasing their chances of innovating.
[16] For example, Valletti (2025) explicitly addresses this point, indicating that in his view this effect represents an efficiency.
[17] Marc Bourreau and Bruno Jullien. “Mergers, investments and demand expansion.” Economics Letters, vol. 167, 2018, pp. 136-141.
[18] Ioannis Kokkoris and Tommaso Valletti. “Innovation Considerations in Horizontal Merger Control.” Journal of Antitrust Enforcement, vol. 16, no. 2, 2020, pp. 220-261; Valletti, 2025.
[19] Ángel L. López and Xavier Vives. “Overlapping Ownership, R&D Spillovers, and Antitrust Policy.” Journal of Political Economy, vol. 127, no. 5, 2019, pp. 2394-2437. See also, for example, An and Zhao, 2019, finding that learning through doing through knowledge spillovers were a plausible explanation for decreased costs following the Boeing-McDonnell Douglas merger.
