Corporate Innovation in a World of Common Ownership

Corporate Innovation in a World of Common Ownership Bin Qiu

September, 2019

Abstract This paper tests how common ownership affects corporate innovation. I provide evidence that common ownership impedes innovation measured by patent grants and citations through reduced R&D expenditures. The results are robust to alternative measures of common ownership. They also hold after addressing endogeneity issues with the difference-in-differences technique. Further analysis shows that common ownership affects post-grant patents by lowering the likelihood that a co-owned firm gets involved in patent litigation and by accelerating the settlement of lawsuits between co-owned firms.

Missouri Western State University, Craig School of Business. E-mail: bqiu@missouriwestern.edu.

1. Introduction

Intra-industry diversification by a small group of large institutional investors--such as BlackRock, Vanguard, State Street Advisors, and Fidelity--creates significant ownership in natural competitors. The resulting structure of overlapping institutional ownership in the same industry is referred to as common ownership. In recent decades, common ownership has increased dramatically possibly due to a wide acceptance of the notion that an investor should hold a diversified portfolio and the popularity of passive investing strategies. The proportion of American publicly-traded companies owned by institutional blockholders that co-own at least 5% of the common equity of other firms in the same industry has risen from less than 10% in 1980 to approximately 60% in 2014.

Common owners want to maximize portfolio values instead of individual firm values. One way to achieve this is to enable portfolio companies to internalize between-firm externalities. Patent and R&D races are one instance where externalities are imposed on one another. Because common ownership possessed by a small set of large diversified institutional investors reduces product market competition and enhances market power (He and Huang, 2017; Azar, Schmalz, and Tecu, 2018), firms could reduce R&D expenditures as a result of attenuated competition. My empirical analysis confirms this conjecture and documents a negative relation between same-industry common ownership and corporate innovative output. Inter-firm legal litigation poses another obvious example of externalities (Hansen and Lott, 1996). Common ownership should discourage commonly-owned firms from suing each other due to patent infringement. My empirical analysis on patent litigations agrees with this prediction. Common ownership correlates with less likelihood of being litigated; for the firms that are indeed involved in patent litigations, common ownership is negatively associated with the time that it takes to resolve the legal case.

To address endogeneity concerns, I employ the difference-in-differences analysis to establish the causal effect of joint ownership on corporate innovation.

1

Specifically, the acquisition of the Citi Group's Asset Management division by Legg Mason in 2005 generated variation across firms in common ownership. I exploit this event to study the "before" and "after" treatment effect.

This paper is closely related to the literature studying institutional investors' impact on corporate innovation. Aghion, Van Reenen, and Zingales (2013) argue that a large share of institutional shareholders is instrumental in facilitating corporate innovation as these shareholders tend to pursue a long-run objective. Bena, Ferreira, and Matos (2014) find that foreign ownership increases firm innovation output. Brav, Jiang, and Tian (2014) show that hedge fund activism leads to more efficient use of innovative resources and human capital. Yang (2016) establishes that institutional dual ownership of a firm's debt and equity lead to fewer but more valuable patents. Geng, Hau, and Lai (2016) provide evidence that institutional ownership overlaps across firms with patent complementarities help mitigate holdup and correlate with more innovative success.

This paper differs from Geng, Hau, and Lai (2016) in that they look at common ownership across firms in the same technology space (i.e., firms with upstream and downstream patents), while I examine common ownership across firms in the same product market space (i.e., same industry). A firm's position in technology space and product market space are typically different. For example, IBM, Apple, Motorola, and Intelare close to each other in technology spaceas revealed by their patenting. However, they are in different product markets. Specifically, IBM and Apple produce PC desktops, while Intel and Motorola mainly produce semi-conductor chips not computer hardware (Bloom, Schankerman, and Reeman, 2013). Therefore, although Geng, Hau, and Lai (2016) find a positive effect of technology-complementary common ownership on patent success, I observe a negative effect of same-industry common ownership on patent grants. Furthermore, I examine the impact of common ownership on post-grant patents, thereby providing a more comprehensive understanding of the effect of the ownership structure on the lifecycle of corporate patents.

2

Several other concurrent papers also study the impact of common ownership on corporate innovation. Kostovetsky and Manconi (2017) find that firms joining the Russell 1000 (2000) index receive more citations from Russell 1000 (2000) firms and fewer from Russell 2000 (1000) firms. In contrast, we find that commonly-owned firms receive fewer citations from all other firms, no matter these firms are in the Russell indexes or not. Borochin, Yang, and Zhang (2019) differentiate the type of common ownership and find that common ownership by focused, long-term dedicated institutional investors promotes innovation while common ownership by diversified, short-term transient investors discourages innovation. I argue that their results are consistent with mine because common ownership is mainly caused by the diversification (i.e., indexing) of institutions, therefore the negative effect should dominate. An incomplete paper by Anton, Ederer, Gine and Schmalz (2018) argues that the effect of common ownership on corporate innovation revolves around the interaction between technological spillovers and product market spillovers. Although they only show correlation evidence but not a causal link, their conclusion that the effect could vary from industry to industry does not contradict with the overall negative effect I observe.

A related line of research explores the effect of an investor owning multiple firms on corporate governance. Edmans, Levit, and Reilly (2016) find that common ownership strengthens governance through both voice and exit. Although my findings about the effect of multiple holdings on corporate innovation do not always require shareholder intervention, it is consistent in spirit with the conclusion of Edmans, Levit, and Reilly (2016). Firms reduce innovation to ease competition with their natural competitors that are also owned by the firms' institutional investors. This anti-competitive effect of common ownership benefits the shareholders at the expense of consumers. However, a comprehensive look into how same-industry common ownership affect innovation input, innovation output, and post-grant patents is missing from previous research. This paper aims to fill this gap.

The remainder of the paper proceeds as follows. Section 2 reviews the literature

3

and develops hypotheses. Section 3 describes the data and presents summary statistics. Section 4 details the methodology used and discusses the main results. Section 5 tests the impact of common ownership on patent litigations. Finally, Section 6 concludes the paper.

2. Literature and Hypotheses

Azar et al. (2018) find that common ownership across firms in the same industry resulted from portfolio diversification impairs competition in the airline industry. Azar, Raina, and Schmalz (2016) show that common ownership concentration impedes competition in the banking industry. Aslan (2018) shows that common ownership increase product price in consumer goods industries. Furthermore, Schumpeter (1912) and Knott and Posen (2003) show that innovation increases with the degree of industry competition. In other words, patent and R&D races are less necessary if product rivalry is attenuated. After all, those races are risky to participating firms because "the winner takes all (i.e., the winner harvests all the profits by patenting)" and the losers' R&D investments are largely wasted. This reasoning leads to my first hypothesis1:

H1) Due to its anti-competitive effect, same-industry common ownership reduces corporate innovation2.

My major measure of common ownership is the difference between the modified Herfindahl-Hirschman Index (MHHI), derived in O'Brien and Salop (2000), and the market share-based Herfindahl-Hirschman Index (HHI, or H(market shares)), following Azar et al. (2018). The traditional HHI does not consider common ownership. To allow for this component, O'Brien and Salop (2000) propose using the

1 Two more hypotheses are made in Section 5. 2 As pointed out by Azar et al. (2018), the anticompetitive effect of common ownership does not require monitoring on the part of institutions. Kang, Luo, and Na (2018) provide evidence that institutional investors with multiple blockholdings in the same industry conduct more effective monitoring over CEOs because of enhanced capabilities and incentives. Therefore, presumably common ownership could result in more innovation if the institutions prefer. However, the institutions may prefer otherwise since innovation increases competition among commonly-held firms. Hence, our hypothesis could still hold when institutions monitor.

4

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download