Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Overlapping Ownership, R&D Spillovers and Antitrust Policy `ngel L. L(cid:243)pez1 Xavier Vives2 1UniversitatAut(cid:242)nomadeBarcelona 2IESEBusinessSchool June 2017 Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Overlapping ownership arrangements (OOAs) are prevalent in the form of cross-shareholding agreements among (cid:133)rms or common ownership by investment funds: Minority shareholdings are widespread in many industries (e.g. automobiles, airlines, (cid:133)nancial, energy, steel sectors). Common ownership has grown tremendously in the last three decades, with proportion of US public (cid:133)rms where investors hold large stakes in the same industry going from under 10% in 1980 to 60% in 2010. Preliminary evidence of OOAs reducing price competition in the airline and banking industries has raised antitrust concerns. There is also a debate about whether and why innovative activity and business dynamism have abated recently (CEA (2016), Obama(cid:146)s executive order to promote competition). Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Questions and objectives The paper analyzes the interaction of OOAs and R&D activity in the presence of technological spillovers and derives testable predictions and antitrust implications from an oligopoly model. OOAs lessen competitive pressure but may have a bene(cid:133)cial e⁄ect on investment provided there are positive spillovers across (cid:133)rms. OOAs help to internalize the spillover externality, especially important for highly innovative industries. Should antitrust authorities limit the (cid:147)partial(cid:148)mergers that result from OOAs in innovative industries? The analysis may help elucidate whether the documented increase in OOAs has outrun its social value. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Approach We consider a general symmetric model of cost-reducing R&D investments with spillovers in Cournot oligopoly (which subsumes received R&D literature) and with overlapping ownership: which includes both Silent Financial Interests or Passive Invesments and Control Interests (as in Salop and O(cid:146)Brien (2000)) as well as distinguishing between stock acquisitions made by investors and those made by (cid:133)rms. In our central scenario (cid:133)rms make simultaneously their R&D and output decisions. Approach aids tractability while helping to capture the imperfect observability of (cid:133)rms(cid:146)R&D investment levels (Cohen (1995), Griliches (1995)). We test the robustness of the results in a two-stage competition model and to Bertrand competition with product di⁄erentiation. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Questions How do R&D and output levels vary with overlapping ownership interests? What are the key determinants of the socially optimal degree of overlapping onwership? How does it depend on structural parameters (such as demand and cost conditions, industry technological opportunity, and level of spillovers)? How does it depend on the objective of the competition authority (to maximize TS or rather CS)? Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Positive results (I) We show that OOAs may increase TS, and even CS, in industries where investment in R&D is important and spillover e⁄ects are su¢ ciently high. For demand not too convex, increases in overlapping ownership increase (decrease) R&D and output for high (low) enough spillovers while it increases R&D but decreases output for intermediate levels of spillovers. The two thresholds that partition the regions for spillovers are increasing in the level of market concentration. We derive testable predictions: a positive relationship between overlapping ownership and R&D should be found in industries with high enough spillovers, low enough concentration and demand not too convex; the positive association should extend to output in industries with high e⁄ectiveness of R&D. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Normative results (II) The range of spillovers is typically partitioned in three regions with, respectively, optimally: no overlapping ownership for low levels of spillovers; positive overlapping ownership for the TS and CS standards for high levels; and positive overlapping ownership for the TS standard only in an intermediate region. Optimal overlapping ownership is increasing in extent of spillovers. We (cid:133)nd conditions for a cartelized Research Joint Venture (RJV) to be optimal. The CS standard is always more stringent than the TS standard. The TS socially optimal degree of overlapping ownership increases with the number of (cid:133)rms, the elasticity of demand and of the innovation function, and the intensity of spillover e⁄ects. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Antitrustconcerns on OOAs There is growing interest in assessing competitive e⁄ects of OOAs: Rapid growth of common ownership with stakes in competing (cid:133)rms. Some notorious cross-ownership cases cases: Ryanair(cid:146)s acquisition ofAerLingus(cid:146)s stock. Northwestpurchase of14% ofthe common stock ofContinental. In the US OOAs examined under Clayton Act (S. 7) and Hart-Scott-Rodino Act: Institutional investors can hold up to 15% without need to notify to the antitrust authority. OOAs can be challenged if they substantially lessen competition. Proposals on how to dealwith OOAs: Elhauge (2016), Baker(2016), Posner et al. (2016)). The European Commission proposes to extend the scope of the Merger Regulation to be able to intervene in OOAs cases. Innovation is a central focus for antitrust policy and merger control: In 2008-14, 36% of DOJ/FTC challenged mergers were identi(cid:133)ed to harm innovation; 76% in high R&D intensity industries. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Literature: e⁄ects ofoverlapping ownership Reynolds and Snapp (1986), Bresnahan and Salop (1986): non-controlling minority shareholdings result in less output/higher prices (even if equity interests are small). Farrell and Shapiro (1990) show that passive (cid:133)nancial stakes may be welfare increasing in asymmetric oligopolies. Evidence on anticompetitive e⁄ects of OOAs: Azar et al. (2015) (cid:133)nd that airline ticket prices in the US are about 10% higher on the average route than would be with no partial ownership interests. Similar results for banking (Azar et al. (2016)). GutiØrrez and Philippon (2016) (cid:133)nd that underinvestment in the US since the 2000s is driven by (cid:133)rms owned by quasi-indexers and in industries with more concentration/common ownership. Evidence of e¢ ciency gains: He and Huang (2017): Through common ownership US public (cid:133)rms increase their market share (1980-2010), improve innovation productivity(patentsper$spendinR&D)andoperatingpro(cid:133)tability. Evidence of transmission mechanism: Anton et al. (2016): industries with more common ownership have less relative performance manager compensation. Introduction FrameworkandEquilibrium WelfareAnalysis TheTwo-StageModel ConcludingRemarks Bertrandcompetitionwithproductdi⁄erentiation Introduction Literature: R&D and spillovers Very extense literature on the e⁄ects of cooperation in R&D with spillovers(BranderandSpencer(1984), Spence(1984), Katz(1986), d(cid:146)Aspremont and Jacquemin (1988), Leahy and Neary (1997)). A main objective of this literature is to examine potential underprovision of R&D and the welfare e⁄ects of moving from a non-cooperative to a cooperative regime in R&D: Result (Leahy and Neary (1997)): when spillovers are positive, R&D cooperation leads to more output, innovation, and welfare. Bloom et al. (2013), with panel of U.S. (cid:133)rms from 1981-2001, (cid:133)nd that: technology spillovers are present in all sectors but with greater importance in high-tech industries such as computers, pharmaceuticals, and telecommunications, and that the socially optimal level of R&D is between two and three times as high as the level of observed R&D. Negative relationship found between spillovers and patent protection levels in a range of industries (Griliches 1990, Galasso and Schankerman (2015)).