Governance Structure and Firm Performance in Private Family Firms

• Date: 01 November 2015
• Published date: 01 November 2015
• DOI: http://doi.org/10.1111/jbfa.12170
• Author: Limei Che,John Christian Langli

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 42(9) & (10), 1216–1250, November/December 2015, 0306-686X

doi: 10.1111/jbfa.12170

Governance Structure and Firm

Performance in Private Family Firms

LIMEI CHE∗AND JOHN CHRISTIAN LANGLI

Abstract: Although a large proportion of firms are family owned and most family firms are

private, our understanding of private family firms is limited. Using confidential information

on family relationships between board members, CEOs, and shareholders, this is the first

study to provide large-scale evidence on the association between governance structure and

firm performance in family-controlled private firms. Our sample is unique as it covers almost

all private limited liability firms in Norway, spans 11 years, traces firm ownership to ultimate

owners, and identifies family relationship using data on kinship, marriage, and adoption. The

results show a U-shaped relationship between family ownership and firm performance. Higher

ownership of the second largest owner, higher percentage of family members on the board,

stronger family power, and smaller boards are associated with higher firm performance. In

addition, the positive association between the ownership of the second largest owner and firm

performance also occurs when the second largest owner is a member of the controlling family,

but the association is stronger when the second largest owner is a non-family member. We

further test the relative importance of these test variables and find that ownership structure

is more associated with firm performance than board structure.

Keywords: Governance structure, firm performance, private family firms, ownership structure,

board structure

1. INTRODUCTION

A large proportion of businesses are family owned worldwide.1Family firms play

an essential role in society, both in terms of economic contribution and social

∗The authors are at BI Norwegian Business School, Oslo, Norway. The authors are grateful for the

valuable and constructive comments from the anonymous reviewer and the editor Andrew W. Stark. We

acknowledge comments from Jeff Downing, Ole-Kristian Hope, Franz Kellermanns, Mervi Niskanen, Mattias

Nordqvist, Marleen Willekens, Han Wu, and the participants at the 22nd Nordic Academy of Management

Conference, the 37th EAA Annual Congress, the Nordic Accounting Conference 2014, and the workshop

at BI Norwegian Business School. We are grateful for the data provided by the Center for Corporate

Governance Research (CCGR) at BI Norwegian Business School and the Norwegian Tax Administration.

(Paper received December 2014, revised version accepted November 2015).

Address for correspondence: Limei Che, BI Norwegian Business School, 0442 Oslo, Norway

e-mail: limei.che@bi.no

1 ‘Research suggests that 80% of all businesses in the United States are family owned (Daily & Dollinger,

1992) and family businesses contribute between 50% and 60% of US gross domestic product (Francis, 1993;

Upton, 1991). Similar findings have been reported in the UK (Stoy Hayward and The London Business

School 1989, 1990), Western Europe (Lank, 1995), and Australia (Smyrnios and Romano, 1994; Smyrnios

et al., 1997). Providing further evidence of the contribution of family business to the economy, La Porta

et al. (1999) and Shleifer & Vishny (1986) find that the ownership structure of even large public companies

C

2015 John Wiley & Sons Ltd 1216

GOVERNANCE AND PERFORMANCE IN PRIVATE FAMILY FIRMS 1217

responsibility. As the majority of family firms are private, how private family firms

are organized, and how they perform are of great importance. There have been

increasing calls for research that enhances our understanding of private family firms

(Chrisman et al., 2007). Miller et al. (2011, p. 22) note that any study of public family

corporations is biased as it does not reflect the behavior of private companies and

many family businesses are private. Most prior studies have focused on public family

firms probably due to easier data accessibility. The few studies that do focus on private

family firms usually employ small samples (Westhead and Howorth, 2006; and Sciascia

and Mazzola, 2008), which could be subject to sample selection bias.2Although many

issues have frequently been studied for public (family) firms, we cannot take for

granted that the evidence for public firms is valid for private small and medium-

sized firms. The role of boards is different in small or medium-sized private firms with

concentrated ownership and more involvement of the owners in running the business

compared to large public firms (Eisenberg et al., 1998; Huse, 2000; and Chin et al.,

2004). Hence, studies employing high-quality data to explore governance issues in

private family firms are warranted.

This paper focuses on private limited liability firms that are controlled by families

and employs a unique and confidential dataset.3It examines how firm performance is

associated with ownership and board structure in these firms. Although it is likely that

governance variables are related to firm performance, not all are equally important. It

is useful and interesting to understand which governance variable is most associated

with firm performance. Hence, this paper also investigates the relative importance of

factors related to ownership and board structure for firm performance. Our dataset

is unique as it covers almost all private limited liability firms in Norway, spans 11

years from 2001 to 2011, traces firm ownership to ultimate owners, and identifies

family relationship between owners, board members, and CEOs using data on kinship,

marriage, and adoption spanning four generations and extending out to third

cousins.

The board of directors and ownership structure are among the main governance

mechanisms that could affect firm performance (Jensen and Meckling, 1976; and

Blair, 1995). While some studies find a positive relationship between family ownership

and firm performance in public family firms (Anderson and Reeb, 2003; Lee, 2006;

and Maury, 2006), studies on private family firms are sparse in comparison and find

no association between family ownership and firm performance in general (Westhead

and Howorth, 2006; and Sciascia and Mazzola, 2008).4As ownership distribution

is characterized by controlling stockholders who are more often families, usually the founder or their

descendants’ (Carey et al., 2000, p. 37).

2 While little research has been carried out on the relationship between ownership and board structure and

firm performance within private family firms, there are relatively more studies comparing firm performance

between (private) family firms and (private) non-family firms. For example, Arosa et al. (2010) study how

ownership concentration can influence firm performance between private family and private non-family

firms. They show that it matters whether the firms are in the first generation or subsequent generations.

Miller et al. (2007) distinguish family firms where at least two family members are involved in the business

from lone founder firms and find that only lone founder businesses outperform.

3 There are two types of limited liability firms in Norway: private limited liability firms (AS) and public

limited liability firms (ASA). All listed firms must be registered as ASA, but an ASA does not need to be

listed on a stock exchange. As of 2010, there were 280 ASAs and 204,000 ASs.

4 There is mixed evidence on the relationship between family ownership and firm per formance in public

firms. While some studies find positive evidence (Anderson and Reeb, 2003; Lee, 2006; Maury, 2006), others

C

2015 John Wiley & Sons Ltd

1218 CHE AND LANGLI

could determine the power of different stakeholders (Salancik and Pfeffer, 1980) and

ownership structure could potentially affect firm performance (Jensen and Meckling,

1976), it is surprising that family ownership is not associated with firm performance in

private family firms. Firms with different levels of family ownership, say 51%, 80%,

and 100%, are likely to have different firm performance, ceteris paribus, because

the distribution of ownership between the controlling family and outside owners

could affect, for example, to what extent the controlling family expropriate minority

owners, and to what extent outside owners are motivated and powered to curb the

expropriation of the controlling family.

As for the relationship between board characteristics and firm performance, there

is no consensus on how firm performance is associated with board characteristics,

although it is frequently studied (Eisenberg et al., 1998; and Anderson and Reeb,

2004). Furthermore, most of the studies focus on medium or large public firms (Chin

et al., 2004). Hence, the relationship between ownership and board structure and firm

performance in private family firms merits further exploration.

Our hypotheses development is guided by insights from both agency theory and

stewardship theory.5Using these two theoretical perspectives provides a broader

framework to investigate the governance issues in family firms (Klein et al., 2005). Lee

and O’Neill (2003) stress the complementarity of agency and stewardship theory. We

develop hypotheses on two factors related to ownership structure (family ownership

and the ownership of the second largest owner), and three factors related to board

characteristics (the percentage of board members belonging to the controlling family,

board size, and family power). We regress firm performance, measured by return

on assets, on these test variables together with control variables, adjusting for serial

correlation and heteroskedasticity.

The results show that the fraction of shares owned by the controlling family has

a U-shaped relationship with firm performance. Firms with relatively low (e.g., 51%)

and very high (e.g., 100%) family ownership have higher firm performance compared

to firms with family ownership in-between (e.g., 80%). Firm performance is positively

related to the ownership of the second largest owner, consistent with the argument

that a strong second owner has the motivation and power to curb the potential

expropriation of the majority owner and thus increases firm value (Bennedsen and

Nielsen, 2010). It is interesting to note that this applies no matter whether the second

largest owner is a member of the controlling family or not. The positive association

between the ownership of the second largest owner and firm performance is weaker

when the second largest owner is a member of the controlling family than when the

second largest owner is a non-family member.

Firm performance is positively associated with a high percentage of board members

coming from the controlling family and strong family power.6While agency theory

show no or negative relationship between family ownership and firm performance in public family firms

(Stewart and Hitt, 2012).

5 Agency theory assumes that the agent (the manager) acts opportunistically and will maximize his/her

own utility at the expense of the interests of the principal (the owner) unless the principal takes measures

that incentivize or discipline the agent to act in the interests of the principal (Jensen and Meckling, 1976).

Stewardship theory assumes that the goal of the principal and agent is aligned and that the agent acts as a

good steward in the interests of the principal (Donaldson, 1990).

6 Family power is an indicator variable that equals 1 if the Chair of the Board and the CEO are members of

the controlling family and the CEO is a board member, and 0 otherwise (see Section 3(ii)).

C

2015 John Wiley & Sons Ltd