Dynastic control without ownership: Evidence from post-war Japan

https://doi.org/10.1016/j.jfineco.2021.06.018Get rights and content

Abstract

Dynastic-controlled firms are led by founding-family CEOs while the family owns an insignificant share of equity (defined as less than 5%). They represent 7.4% of listed firms in post-war Japan, include well-known firms such as Casio, Suzuki, and Toyota, and are often grouped with widely held firms in the literature. These firms differ in key performance measures from both traditional family firms and non-family firms, and evolve from the former as equity-financed growth dilutes family ownership over time. In turn, the transition from dynastic control to non-family status is driven by a diminution of family legacy and talent.

Introduction

Can business dynasties continue to exercise control over firms they founded, even when their ownership stakes become insignificant? We define dynastic-controlled2 firms as those in which a member of the founding family serves as the CEO3 while the family owns less than 5% of equity. Despite anecdotal evidence on the presence of dynastic-controlled firms in other advanced economies including the U.S., we provide the first systematic documentation on their prevalence, persistence, and performance, based on the universe of publicly listed firms in post-war Japan.4

The Japanese governance system is ideal for studying dynastic control for two reasons. First, unlike the US, Japan does not permit dual-class voting shares, so the one-share-one-vote rule applies. Second, unlike many other Asian countries, pyramidal ownership structures are, as a rule, absent in Japan.5 Thus, voting control and ownership go hand in hand in Japan, and a loss in ownership is strictly correlated with a loss in voting control.

This study makes four distinctive contributions to the literature on family firms. First, we show the prevalence of dynastic-controlled firms is non-trivial – they represent 7.4% of all listed firms, and 16.3% of all firms incorporated as family firms, on Japanese exchanges between 1955 and 2000. In IPO time, such firms represent 10.1% of all firms incorporated as family firms that survive 10 years after their IPO, and 20.7% of those that survive 20 years after their IPO. To further illustrate the importance of dynastic-controlled firms, we highlight the cases of Casio, Suzuki, and Toyota Motor in Section 2.

Second, we provide an extended literature review of 112 empirical papers on family firms. The literature relies on binary definitions of family firms (a firm either is or is not a family firm), and more than half of the 135 definitions in our survey use ownership as the sole criterion to define a family firm–the most common minimum ownership threshold is 20% of equity for publicly traded firms. The rest use a variety of definitions including management control and board presence. Three out of four definitions group dynastic-control firms with widely held non-family firms, and the rest categorize them as family firms together with firms in which the family has significant ownership with or without the CEO position.

Third, we compare dynastic-controlled firms with two groups of firms: first, traditional family firms, in which the founding family retains significant ownership (>5%) and the CEO's position; and second, ex-family firms, in which the founding family's ownership is less than 5% and it has relinquished the CEO position permanently. We find dynastic-controlled firms have better accounting performance than ex-family firms, but underperform traditional family firms.

Our final contribution is to analyze the factors that drive traditional family firms to evolve into dynastic-controlled firms and, eventually, the factors that drive dynastic-controlled firms to become ex-family firms. We find the former is driven by a growth-induced need for finance, whereas the latter is correlated with diminution of family resources, such as legacy, education, and talent.

Our documentation of the prevalence and persistence of dynastic-controlled firms provides a new perspective on the, Berle and Means (1991) thesis that the modern widely held firm is fated to be taken over by a cadre of professional managers as the founding family's ownership declines over time to minuscule levels. Our paper weakens this conjecture by establishing that many families are able to retain control of their firms well after their ownership has become insignificant.

The rest of the paper is organized as follows. In the next section, we provide case studies of Casio, Suzuki, and Toyota Motor to illustrate three different ways that families retain control when their ownership is materially diluted. In Section 3, we provide a short summary of the literature review of 112 studies covering 135 family-firm definitions and show how extant literature categorizes dynastic-controlled firms. The full survey is included as an Internet Appendix. Section 4 describes our data. Section 5 documents the prevalence of dynastic-controlled firms among publicly listed firms in post-war Japan. In Section 6, we show dynastic-controlled firms are different from both traditional family firms and non-family firms along widely followed performance metrics. Section 7 identifies factors driving the transition from traditional family firms to dynastic-controlled firms, and eventually from dynastic-controlled firms to ex-family firms. We conclude in Section 8.

Section snippets

2.Case studies: Casio, Suzuki and Toyota Motor

Three well-known Japanese companies, Casio, Suzuki, and Toyota Motor, illustrate how founding families maintain management control through talent, advanced governance mechanisms, and board control in situations in which they have very little ownership. Fig. 1 shows the evolution of family ownership and top management from 1960 to 2019 for the three firms. In each case, the ownership stake of the founding family was either never significant, or reduced to an insignificant level (defined as less

Categorization of dynastic-controlled firms in the family-business literature14

In this section, we describe the large variety of family-firm definitions employed in the literature and show how existing studies have failed to recognize dynastic-controlled firms as a separate class. Definitions matter in generating even the most basic insight about family firms. For example, Anderson and Reeb (2003) find superior performance for family firms relative to non-family firms. However, subsequent papers contest this result: Villalonga and Amit (2006) and Miller et al. (2007) show

Data sources

We construct a dataset of all companies that went public in the post-war period in Japan (after the stock exchanges re-opened in 1949). Ownership data are from the Development Bank of Japan database for 1981 through 2000, and our accounting data are from 1955 through 2000. The Toyo Keizai database provides information on stock prices and board composition from 1989 through 2000. We exclude a small number of the firms for which financial or ownership data are missing. The final sample covers

The prevalence and persistence of dynastic-controlled firms

In this section, we analyze the prevalence and persistence of dynastic-controlled firms in post-war Japan. We start with charting the number of dynastic-controlled firms as a fraction of all listed firms over this period. We then analyze the share of family-listed firms that become dynastic-controlled over time.

Panel A in Fig. 2 describes the prevalence of dynastic-controlled firms in calendar time over our sample period. The solid black line depicts the share of dynastic-controlled firms out

Dynastic-controlled firms: comparisons with traditional family firms and non-family firms

In this section, we compare dynastic-controlled firms with the two groups the literature has grouped them with–traditional family firms and non-family firms–and show dynastic-controlled firms differ from both groups along widely used performance measures.

To identify our reference groups, we categorize family firms into three groups: (i) traditional family firms, as described earlier (the founding family owns more than 5% of the shares and a family member is the CEO); (ii) dynastic-controlled

Transition into and out of dynastic control

In this section, we study the determinants of the origins and loss of dynastic control. We begin by documenting how traditional family firms evolve into dynastic-controlled firms as the founding family's ownership erodes over time. We then examine how dynastic-controlled firms eventually lose management control and evolve into widely-held non-family firms.

8. Conclusion

Using the universe of publicly listed firms in post-war Japan, we find that founding families continue to exercise control over their companies even in the absence of material share ownership. We define such firms as dynastic-controlled, and find they represent 7.4% of all listed firms during 1955–2000. We document that they differ in accounting performance and valuation metrics both from traditional family firms and from widely-held firms, the two groups that the existing literature has

References (36)

  • B. Villalonga et al.

    How do family ownership, control, and management affect firm value?

    J. Financ. Econ.

    (2006)
  • R. Anderson et al.

    Founding family ownership and firm performance: evidence from the S&P 500

    J. Financ.

    (2003)
  • J.S. Ang et al.

    Agency costs and ownership structure

    J. Financ.

    (2000)
  • S. Belenzon et al.

    Eponymous entrepreneurs

    Am. Econ. Rev.

    (2017)
  • M. Bennedsen et al.

    The Chocolate Factory

    (2013)
  • M. Bennedsen et al.

    Toyota: A Family Heir Steers Carmaker Out of Crisis

    (2016)
  • M. Bennedsen et al.

    Inside the family firm: the role of families in succession decisions and performance

    Q. J. Econ.

    (2007)
  • A. Berle et al.

    The Modern Corporation and Private Property

    (1991)
  • Cited by (7)

    • Are non-family successors all the same? Inside-promoted vs. outside-sourced

      2021, Journal of Corporate Finance
      Citation Excerpt :

      This could introduce challenges, especially when analyzing the Asian stock markets, because the controlling family may still exert influence over the firm's operation despite lacking significant ownership. Bennedsen et al. (2021) provide clear cases detailing well-known Japanese companies, which show that controlling families can still exert a strong influence on operations even when their shareholdings are highly diluted. This phenomenon also exists in the Taiwan market.

    • Business families: An introduction

      2023, De Gruyter Handbook of Business Families
    View all citing articles on Scopus

    We are grateful for excellent comments and suggestions from an anonymous referee, the editor, Hsi-Mei Chung, Royston Greenwood, Masaharu Hanazaki, Katsuyuki Kubo, Yi-Chun Lu, Danny Miller, Hideaki Miyajima, Randall Morck, Jun-Ichi Nakamura, Daniel Wolfenzon, and seminar participants at the Entrepreneurship Theory and Practice conference at the University of Alberta, Peking University, the Columbia/INSEAD conference on Family Firms, the Conference on Family Firms in Lehigh University, the 5th Global Corporate Governance Colloquia (GCGC) Conference, the European Finance Association meetings, Copenhagen Business School, Development Bank of Japan, Tokyo Institute of Technology, Waseda University, and INSEAD Brown Bag (economics and finance). This project benefited from financial support from the Singapore Ministry of Education Academic Research Fund Tier 1 grants (R-315–000–111–112 & R315–000–086–133), the visiting scholar program of the Research Institute of Capital Formation at the Development Bank of Japan, JSPS KAKENHI grant No. JP15K03563, the Danish Finance Institute (DFI), and the Danish National Research Foundation under the Niels Bohr Professorship. We are also thankful to Jiayi Wei of Tsinghua University for excellent research assistance.

    View full text