Elsevier

Journal of Corporate Finance

Volume 46, October 2017, Pages 461-476
Journal of Corporate Finance

For love and money: Marital leadership in family firms

https://doi.org/10.1016/j.jcorpfin.2017.09.004Get rights and content

Highlights

  • Family firms led by couples perform better than other family firms

  • Marital leadership works best in contexts under greater complexity and agency conflicts

  • Marital leadership makes firms invest during uncertain times, it reduces employee turnover and improves labor productivity

Abstract

We investigate how leadership by couples affects the profitability of family firms. Using comprehensive data from Italy, we show that family firms led by married couples perform significantly better than other family firms. This result is robust to several estimation techniques, including matching, instrumental variables and transition analyses. Marital leadership works best when firm operations are complex and knowledge-based, as well as when the firm is subject to agency conflicts due to weak competition and poor legal infrastructure. Studying the mechanisms behind these results, we show that the presence of couples at the top of the firm enhances the willingness to invest during uncertain times, reduces employee turnover, and improves labor productivity.

Introduction

A rich body of empirical research in corporate finance has demonstrated that family ownership is a common form of corporate control around the world (e.g. Anderson and Reeb, 2003, Faccio and Lang, 2002, La Porta et al., 1999). Family firms have many distinctive traits, the most prominent being that family members often sit in governance and leadership positions. In such firms, family and business dimensions are intimately intertwined: family objectives and demographic characteristics have a profound impact on the company (e.g. Bennedsen et al., 2007, Bennedsen et al., 2008); conversely, involvement in the business tends to affect family dynamics and preferences (e.g. Broussard et al., 2015, Lindquist et al., 2015).

Recent works have begun to investigate the complex interplay between family and family business, studying how family characteristics shape the governance and performance of family firms (Bennedsen et al., 2006, Bennedsen et al., 2007, Bennedsen et al., 2008, Bertrand et al., 2008, Bunkanwanicha et al., 2013, Mehrotra et al., 2013). We contribute to this literature by examining how an under-explored but relatively common leadership model of family firms, namely leadership by a husband and wife couple, affects corporate profitability.1

A sizeable body of research has debated whether family or outside executives are most successful at leading family firms. Many scholars have argued that family members would incur fewer principal-agent agency conflicts as compared to widely-held companies (Anderson and Reeb, 2003). However, it has also been noted that executives drawn from within a single family may be taken from too small a talent pool to be effective (Mehrotra et al., 2011, Mehrotra et al., 2013). Some family firms have addressed this shortcoming by having multiple family leaders jointly running the firm (Miller et al., 2014). Although that combination of manpower does modestly enhance the talent pool, it may occasion conflict as, for example, siblings compete for rents and power for their separate family branches (e.g. Eddleston and Kellermanns, 2007, Brannon et al., 2013). Thus, there remains a good deal of controversy over which kind of leadership works best in family firms (Anderson and Reeb, 2003, Miller et al., 2007, Sraer and Thesmar, 2007, Villalonga and Amit, 2006). A more general debate concerns the breadth of executive teams (Bandiera et al., 2014) and co-CEO leadership arrangements, which are receiving increasing attention in the literature.2 Some studies highlight the human capital advantages of having multiple CEOs (Arena et al., 2011, O'Toole et al., 2002) while others find these arrangements to be conflictual (Krause et al., 2015).3

Responding to these debates, we analyze how couples in co-CEO positions or CEO and executive chairman positions influence the effectiveness of shared leadership. Married couples share a common future, and frequently, common offspring. Thus their interests tend to be well aligned (Luo and Klohnen, 2005, Mare, 1991, Pencavel, 1998), and their incentive to be good stewards of the business is significant (Marshack, 1994, Matzek et al., 2010). Moreover, marriage ties may facilitate monitoring (Ashraf, 2009) and an efficient allocation of resources.4 Thus, couples at the top of firms may be conducive of low agency costs, including those caused by intra-family conflicts, and therefore serve as a basis for establishing its associated financial and administrative implications. Couples also bring diverse experiences and network relationships to a business (Bunkanwanicha et al., 2013, Mehrotra et al., 2011, Mehrotra et al., 2013, Schjoedt et al., 2013). Taken together, these arguments suggest that leadership by couples may lead to superior firm performance. From an opposite perspective, it can be argued that when a couple leads the family business, family conflicts and communication problems between partners will directly feed back into the firm, worsening the quality of corporate decision-making at the expense of profitability (Byron, 2005, Fincham and Beach, 1999).

Due to these opposing factors, the benefits and costs of having married couples at the top of a company have been debated by practitioners and the business community.5 Some academic works (e.g. Belenzon et al., 2016, Dahl et al., 2015, Dyer et al., 2012) have begun to study this topic, but have yielded mixed results.6 Moreover, these empirical assessments have been made using very small businesses (e.g. having a median of 8 employees and $450,000 in total assets; Belenzon et al., 2016) and studying marital involvement in ownership rather than leadership roles. Small entrepreneurial ventures have peculiarities that may compromise the generalizability of existing findings for established firms. For instance, in small ventures married couples are typically the sole owners and managers of the firm, the family relies on the company as the major source of income, and critically, the pool of employees is small and almost entirely drawn from the family itself and that can explain the lower labor costs and superior cash and profits (the major finding of previous studies). These businesses constitute a limited domain for assessing agency implications or managerial effectiveness under more challenging administrative conditions.

Our study addresses this research gap by establishing the performance effect of co‑leadership by couples in large, established family enterprises, and by exploring contextual variations behind this performance result. In this way, we offer contributions going beyond existing works on marital ownership in new ventures. First, within major organizations managerial tasks and strategies are far more complex and challenging, and marital leadership is not simply a product of family economic necessity. Second, the ownership of large organizations is typically more developed than in new ventures (where, as noted, the married leaders are often the sole owners), and this makes leadership by couples potentially distinct from joint ownership.7 Third, in established businesses managerial talent is more prized (Gabaix and Landier, 2008), and agency costs become more relevant as governance arrangements grow in complexity (Ang et al., 2000).

We study marital leadership in Italy, which represents a useful context given the rich array of leadership options adopted by family enterprises there (Miller et al., 2014). These variations provide an ideal opportunity for examining how well couples fare compared to a multiplicity of other kinship and non-family leadership combinations. Estimating a variety of regression models on a comprehensive panel of 1900 companies for the period 2000–2012 (numbering 13,000 observations), we find that family firms led by couples exhibit significantly higher operating profitability, as measured by return on assets, vis-à-vis other family firms. OLS estimates indicate that this outperformance is approximately one percentage point, corresponding to a 19% increase vis-à-vis average profitability. We confirm this result using matching techniques as well as instrumental variables based on historical and geographic variations in gender roles and the value of marriage. Moreover, to strengthen the causal interpretation of our results, we employ time changes showing that firm performance improves when firms appoint a couple to the top position, but that finding reverses when firms abandon such a leadership model.

Providing a fine-grained comparison of leadership models, we show that couples perform better than lone or multiple non-family leaders, as well as companies with lone or multiple family leaders. We also show that companies led by couples outperform those led both by family members from the same generation, and family members from different generations (regardless of the gender composition of such teams).

One interpretation of our findings is that couples at the top benefit from a better alignment between co‑leaders and thus an agency advantage as compared to non-family leaders, who require active monitoring together with strong incentives (Cai et al., 2013), and also as compared to co‑leadership by siblings, who cater to different family branches and tend to be subject to conflicts (Bertrand et al., 2008). In this circumstance, the performance increase should be more pronounced in industries facing modest competitive pressures, which exacerbate agency conflicts (e.g. Guadalupe and Perez-Gonzalez, 2011, Giroud and Mueller, 2011), and in areas characterized by slower resolution of legal disputes. Consistent with this argument, we find that leadership by couples improves performance mostly in concentrated industries and in areas where judiciary efficiency is low. These findings may suggest that couples are only valuable in rent-centric contexts. Yet, in analyzing variations in industry R&D and firm operations we also find significant performance improvements in more complex and knowledge-based contexts. By contrast, we do not find that leadership by couples increases performance more in industries dependent on the public sector, where connections and networking tend to be more valuable (e.g. Amore and Bennedsen, 2013). In analyzing demographic characteristics, we show that couples at the top are most effective when their co-tenures are long and when the age differential between leaders is small. These variations suggest that co‑leaders may take time to learn how to work well together and benefit from skill complementarity, and also that larger age gaps within a couple may signal disparate priorities and reflect power imbalances that impair shared leadership.

Finally, we focus on firm policies to document the various mechanisms driving our performance results. We adopt the news-based index of policy uncertainty developed by Baker et al. (2016) to study how leadership by couples influences the negative effect of nationwide policy uncertainty on investment activities (e.g. Gulen and Ion, 2015). Our results indicate that the investment decisions of family firms with couples at the top are significantly less sensitive to policy uncertainty than other family firms. Moreover, we provide evidence of human resource practices characterized by lower wage costs, but also greater labor productivity and job stability in response to industry shocks, suggesting a superior ability to enforce implicit contracts with employees (e.g. Bach and Serrano-Velarde, 2015, Ellul et al., 2015, Sraer and Thesmar, 2007). These findings are again consistent with the notion that marital leaders bring about benefits to the company by extending the time-horizon of corporate decision-making.

Section snippets

Sample

We use a representative panel data set of listed and private family firms in Italy for the period 2000–2012. Family firms are defined as private companies in which a family owns at least 50% of equity, or 25% if the firm is listed on the stock market (e.g. Amore et al., 2014, Andres, 2008). The sample contains all family firms with sales above €50 million in any year between 2000 and 2012.8

Firm performance

To establish the effect of marital leadership on firm profitability, we estimate the following regression:Yit=α+βMarital leadershipit+γZit+λj+μt+εitwhere the dependent variable is the ROA of firm i at time t.

The key explanatory variable is a dummy equal to one if firm i at time t is classified as having a couple at the top and zero otherwise (Marital leadership). The vector Zit includes the host of controls related to firm characteristics and family involvement in governance and executive

Industry and firm characteristics

In this section, we explore firm and industry variations in order to document more directly how marital leadership can improve operating performance. We start by analyzing industry differences in product market competition. In concordance with prior literature (e.g. Giroud and Mueller, 2011), we proxy for product market competition using the Herfindahl-Hirschman index (HHI) computed using family and non-family firms in each 3-digit industry and year.27

Investment and policy uncertainty

Building on theoretical works dating back to Bernanke (1983), a large body of research has established that policy uncertainty has a detrimental effect on investment decisions (see Bloom, 2014 for a review). Real options provide a useful setting to understand the investment-uncertainty nexus. In this setting, firms making an irreversible investment face a trade-off between the advantages of early commitment against the advantages of waiting to get more information. By increasing the likelihood

Conclusion

Dating back to Becker, 1973, Becker, 1974 scholars have argued that marriage relationships can have a profound influence on the lives of individuals and their economic incentives. Recent works have shown that marriage may even affect financial decision-making of top managers (Hilary et al., 2017, Lu et al., 2016, Love, 2010, Nicolosi, 2013, Roussanov and Savor, 2014). Family business brings about an interesting overlap between family and firm dimensions, and thus provides an ideal context to

References (76)

  • V. Mehrotra et al.

    Adoptive expectations: rising sons in Japanese family firms

    J. Financ. Econ.

    (2013)
  • D. Miller et al.

    Are family firms really superior performers?

    J. Corp. Finan.

    (2007)
  • G. Nicolosi

    Demographics of dividends

    J. Corp. Finan.

    (2013)
  • B. Villalonga et al.

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

    J. Financ. Econ.

    (2006)
  • A. Abadie et al.

    Large sample properties of matching estimators for average treatment effects

    Econometrica

    (2006)
  • A. Alesina et al.

    Goodbye Lenin (or not?): the effect of communism on people

    Am. Econ. Rev.

    (2007)
  • M.D. Amore et al.

    Local political uncertainty, family control and investment behavior

    J. Financ. Quant. Anal.

    (2017)
  • M.D. Amore et al.

    Gender interactions within the family firm

    Manag. Sci.

    (2014)
  • R. Anderson et al.

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

    J. Financ.

    (2003)
  • J. Ang et al.

    Agency costs and ownership structure

    J. Financ.

    (2000)
  • M.P. Arena et al.

    It takes two: the incidence and effectiveness of co-CEOs

    Financ. Rev.

    (2011)
  • N. Ashraf

    Spousal control and intra-household decision making: an experimental study in the Philippines

    Am. Econ. Rev.

    (2009)
  • L. Bach et al.

    CEO identity and labor contracts: evidence from CEO transitions

    J. Corp. Finan.

    (2015)
  • S. Baker et al.

    Measuring economic policy uncertainty

    Q. J. Econ.

    (2016)
  • O. Bandiera et al.

    Span of Control and Span of Attention

    (2014)
  • G. Becker

    A theory of marriage: part I

    J. Polit. Econ.

    (1973)
  • G. Becker

    A theory of marriage: part II

    J. Polit. Econ.

    (1974)
  • S. Belenzon et al.

    Married to the firm? A large-scale investigation of the social context of ownership

    Strateg. Manag. J.

    (2016)
  • M. Bennedsen et al.

    Do CEOs Matter?

    (2006)
  • M. Bennedsen et al.

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

    Q. J. Econ.

    (2007)
  • B. Bernanke

    Irreversibility, uncertainty and cyclical investment

    Q. J. Econ.

    (1983)
  • N. Bloom

    Fluctuations in uncertainty

    J. Econ. Perspect.

    (2014)
  • G. Bobonis

    Is the allocation of resources within the household efficient? New evidence from a randomized experiment

    J. Polit. Econ.

    (2009)
  • D.L. Brannon et al.

    The varying effects of family relationships in entrepreneurial teams

    Enterp. Theory Pract.

    (2013)
  • N.H. Broussard et al.

    (Why) do self-employed parents have more children?

    Rev. Econ. Househ.

    (2015)
  • L.L. Bumpass et al.

    Differentials in marital instability

    Am. Sociol. Rev.

    (1972)
  • P. Bunkanwanicha et al.

    The value of marriage to family firms

    J. Financ. Quant. Anal.

    (2013)
  • H. Cai et al.

    Family ties and organizational design: evidence from Chinese private firms

    Rev. Econ. Stat.

    (2013)
  • Cited by (45)

    • Family agents

      2023, Journal of Family Business Strategy
      Citation Excerpt :

      Different from Krause et al. (2015), our data cover not only listed firms but also privately-held firms and it allows to distinguish between family and non-family firms using information on firms’ ownership structure (see below for details). In addition, our data goes beyond existing studies on shared leadership (e.g., Amore et al. 2017) in that it contains governance and ownership information for both family and non-family firms. Finally, our data allows to measure the fragmentation of leadership positions and thus complement existing studies on the fragmentation of family ownership (Miller et al., 2022; Bertrand et al., 2008).1

    • Your heart is where your treasure is: Family chairman and tax avoidance in family-controlled firms

      2023, Journal of Business Research
      Citation Excerpt :

      The propensity-score matching (PSM) method is also used for a robustness test. Following Amore et al. (2017), we first employ the logit model to estimate the propensity score for family chairman using all control variables in the baseline regression. After that, unrepeated one-to-one nearest-neighbor matching is conducted.

    • Family business, servitization, and performance: Evidence from Portugal

      2022, Technological Forecasting and Social Change
    • Corporate social responsibility and corporate financial performance: The role of executive directors in family firms

      2022, Finance Research Letters
      Citation Excerpt :

      A missing value appears if a company did not have a SER rating. Consistent with previous studies (Amore et al., 2017), we identified family-controlled firms (hereafter, family firms) as those firms in which one or two closely related families control at least 25% of the shares. Executive directors represents a continuous variable that accounts for multiple executive directors on the board, computed as the ratio of executive members to board size.

    View all citing articles on Scopus

    We are grateful to Stefano Palmisano and Fabio Quarato for outstanding research assistance, as well as to an anonymous reviewer, the editor (Jeff Netter), Morten Bennedsen, Marco Cucculelli, Pierre Chaigneau, Kelly Chen, Paolo Colla, Andrew Karolyi, Vikas Mehrotra, Orsola Garofalo, Anna Grosman, Marta Lopes, Daniele Paserman, Carlo Salvato, Xiaowei Xu for comments and suggestions. We also thank seminar participants at IESE Business School, Instituto de Empresa, Loughborough University, University of Groningen, and conference participants at the Frontiers in Finance Conference (University of Alberta), Society of the Economics of the Household Annual Conference (San Diego), Asian Finance Association Conference (Seoul), Academy of Management (Atlanta), and 1st Bocconi Research Day. All errors remain our own.

    View full text