1. Introduction

Studies of family firms as a form of governance structure have defined the family firm concept and analyzed family firm characteristics. They have also explored performance differences between family and non-family firms and examined the causes of these differences (Westhead & Howorth, 2006; Barontini & Caprio, 2006; Odehnalová & Olśevićová, 2009; Ayranci, 2014)

An analysis of the research on the effects of family firms’ governance structures on firm performance and value reveals ambiguous findings. Some studies show that family firms have superior performance and value than non-family firms (Demsetz & Lehn, 1985; Mcconnaughy et al., 1998; Anderson & Reeb, 2003). Other studies show that family firm performance is weaker than non-family firm performance (Burkart & Gromb, 1997; Amoako-adu, 1999; Claessens et al., 2000; Schulze et al., 2001; Lins & Lins, 2003; Westhead & Howorth, 2006; Hoopes & Miller, 2006; Perez-Gonzalez, 2006).

Major controlling shareholders control and monitor managers’ business activities in family firms, which eases potential agency problems. The effect of reduced agency cost is greater firm value. However, the management rights of non-family firm managers are affected by short-term achievements. Thus, unlike family firm managers, the managers of non-family firms tend to make decisions based on short-term achievements, which reduces firm value. On the one hand, a lower value of family firms compared to non-family firms can result from controlling shareholder tendency to seek excessive compensation, related-party transactions, or special dividends. The firm characteristics can cause the firm’s value to decrease (Shleifer & Vishny, 1997).

Korean firms exhibit family ownership characteristics. Kim (2006) state that although Korean family firm performance is greater than non-family firm performance in terms of profitability, productivity, and job creation, non-family firm performance is greater in terms of labor absorptive capacity. Choi and Lee (2005) claim that the management performance of Korean owner-managed firms is greater than the performance of CEO-managed firms.

This research analyzes the effects of family ownership on firm value from the perspective to agency theory. The positive or negative effects of family ownership on Korean firm value is a particular focus of this study. If negative effects are evident, this research explores the factors that cause a decrease in firm value.

The sample includes a total of 5,743 companies listed on the Korea Exchange from 2002 to 2012. The data used in this research temporally connect the cross-section data, forming unbalanced panel data combining time series and cross-section data.

This paper is organized as follows. Section 2 presents a literature review. Section 3 provides the testable hypotheses, data, and variables. Following that the empirical results are described, and the final section concludes the research.

2. Literature Review

The perspective that family firm performance is superior to non-family firm performance can be explained in terms of agency theory and concentrated ownership. The ownership structure of family firms, in which voting rights are concentrated, eases the agency problem and increases firm value. Fama and Jensen (1980) claim that family firms are more efficient than firms managed by CEOs because they have lower monitoring costs.

Demsetz and Lehn (1985) state that firms with combined ownership and control have controlling shareholders with an incentive to alleviate the manager’s agency problem. This is because the family’s wealth is closely related to the firm’s welfare. McConnaughy et al. (1998) argue that family firms have greater efficiency and value than non-family firms because their characteristics provide the incentive to monitor managers and increase firm performance. Anderson and Reeb (2003) claim that a nonlinear relationship exists between family ownership share and firm performance, and that if the CEO is a family member, the firm performance is superior to a firm managed by an unrelated CEO.

The perspective that family firm performance is inferior to non-family firm performance can be explained by factors such as family selfishness, ownership concentration, and the harmful effects of excessive voting rights. According to Shleifer and Vishny (1997), mature family firms may exhibit inferior performance than non-family firms because a controlling shareholder may influence management activities when their capacity to do so is diminished. Moreover, the authors claim that when the managing family seeks excessive pay, related-party transactions, or special dividends, this can trigger a decrease in family firm value. According to Burkart et al. (1997), maximization of the family’s own utility in a family-owned firm can reduce performance. DeAngelo and DeAngelo (2005) argue that when a family pursues extra dividends, limited capital expansion plans negatively affect firm performance and share price. According to Claessens et al. (2000) and Lins and Lins (2003), ownership concentration has a negative effect on the general value of a firm. The concentrated ownership inherent in family firms lowers value.

3. Data and Methodology

3.1. Selection of sample firms and method of analysis

This research selects a total of 5,743 firms listed on the Korea Exchange from the year 2002 to 2012. The sample is composed of firms from non-financial sectors that officially announced closing accounts after 2002, maintained the same accounting period during the sample period, and have a fiscal year end date of December after which accounting documents can be obtained. Financial firms, such as banks, insurance, and securities firms are excluded from the sample because the regulated financial industry in Korea differs from other sectors in terms of ownership structure and management practice.

3.2. Variables for analysis

The classification of family firms is a significant element of the research. Studies that have defined the family firm present family ownership as voting rights, family involvement in management, family succession, and self-perception as a family firm (Litz, 1997; Gomez-mejia et al., 2003).

Considering the criteria and the actual state of affairs in Korea, this research classifies a family firm as one with at least one of three conditions. The first condition is that a founding family must own more than 50% of a firm’s voting rights. Westhead and Cowling (1998) and the Korean Family Business Research Institute (www.familybiz.or.kr) define a family firm as one with more than 50% family ownership rights. The second condition is that a firm must have a family member involved in management. When the family member is a registered executive or non-registered executive, the firm is classified as a family firm. The third condition is a firm that is an affiliate of a large-scale conglomerate with a controlling shareholder. Even when a hired CEO manages the affiliate of a large-scale conglomerate with a controlling shareholder, it is normal practice in Korea for decisions to be made under the auspices of a controlling shareholder who is a member of the founding family.

The market-to-book value (MB) ratio is used as a proxy variable for firm value (Lindenberg & Ross, 1981; Chung & Pruitt, 1994; Lang & Stulz, 1994; Servaes, 1996). The MB ratio is calculated by a firm’s market value divided by its book value. The firm’s market value is sum of the market capitalization and the debt’s book value. The firm’s book value is calculated as the book value of total assets.

This research sets governance structure-related variables such as ownership by a controlling shareholder (OWN), foreign investor ownership (FOR), size of the board of directors (BDSIZE), outside director ratio (OUTDIR), and ownership-control disparity (OCD).

Ownership by a controlling shareholder (OWN) is the shareholding ratio of the founding family shareholder. Ownership by a controlling shareholder can be a proxy variable reflecting an ownership structure with a potential agency problem in the analysis of family firm characteristics. Ownership by a controlling shareholder and firm value can have a positive (+) or a negative (-) relationship (Morck et al., 1988; Stulz, 1988; Core et al., 2003). According to agency theory an signal theory, greater ownership by a controlling shareholder is associated with reduced agency costs and is an indicator that a firm is financially healthy; this can increase firm value. However, according to the management entrenchment hypothesis, when ownership by a controlling shareholder is above a certain level, and the CEO is the controlling shareholder, the CEO has increased self-oriented motivation that can reduce firm value.

Foreign investors as external shareholders are significant agents with respect to the monitoring of management activities. External shareholders, similar to foreign investors, have large-scale operating funds and relatively large ownership; thus, they have an incentive to monitor firm management. According to the efficient monitoring hypothesis, because greater ownership by foreign investors minimizes manager agency problems, a positive (+) relationship exists between foreign ownership and firm value (Douma et al., 2006). A large board of directors is associated with a greater possibility of the availability of excellent managerial resources, a significantly sized board of directors is linked to efficient control and monitoring of management and a reduced management agency problem. Thus, firm value is expected to increase in parallel with the size of the board of directors (Chol & Lee, 2005). The size of the board of directors (BDSIZE) is calculated by the natural log value of the number of registered directors on the board.

If the outside director ratio increases, the likelihood of a manager agency problem is reduced. Thus, firm value is expected to increase (Fama, 1980; Min & Verhoeven, 2013). The outside director ratio (OUTDIR) is calculated as the share of outside directors belonging to the board of directors.

The ownership-control disparity represents the difference between the share directly or indirectly owned by the controlling shareholder of a certain firm and the share that enables the controlling shareholder to directly or indirectly influence the firm (Lemmon & Lins, 2003; Joh, 2003). It is linked to agency problems. Controlling shareholders can transfer the wealth of minority shareholders from an affiliate in which they have low ownership to another affiliate in which they have high ownership through intergroup loans or setting transfer prices (Chang & Shin, 2007).

If ownership-control disparity increases, the management entrenchment effect that uses a control right will be greater than the interest alignment effect, which occurs because of cash flow rights. Consistent with Lemmon and Lins (2003), this research calculates ownership-control disparity (OCD) as the difference between the voting rights and the cash flow rights of a controlling shareholder of a specific firm. We also use variables such as profitability (ROA), leverage ratio (LEV), firm size (SIZE), growth (GROW), and the number of years since establishment (AGE) as proxy variables reflecting firm characteristics.

[Table 1] Descriptions of all the variables

3.3. Descriptive statistics

The summarized descriptive statistics of each variable for the whole sample are presented in [Table 2]. When each variable is compared by dividing the whole sample into family firms and non-family firms, on average, variables such as the controlling shareholder’s share ratio, size of the board of directors, ownership-control disparity, profitability, leverage ratio, firm size, growth, and the number of years since establishment has significant differences.

[Table 2] Descriptive statistics for each variable

Note: The sample includes 11 years of data for 5,743 non-financial companies listed on the Korean Exchange from 2002 to 2012. ***, **, and * indicate significance of at the 1%, 5%, and 10%, respectively.

An ADF-Fisher unit root test was performed to verify whether the time series panel data for each variable used in this study are stationary. The result showed that unit root did not exist for all variables.

3.4. Methodology

This research establishes the following hypothesis to analyze the effects of Korean family ownership characteristics on firm value. If the CEO-related agency problems can be alleviated when the controlling family controls and monitors the CEO’s management activity, a family business will have a positive effect on firm value. However, it will have a negative effect on firm value when the controlling family pursues personal gain in the form of as excessive wages, extra dividends, or related-party transactions.

To verify this hypothesis, firm value (MB) is set as a dependent variable, and family dummy variables (Family, Family1, Family2, Family3) and control variables that indicate whether it is a family firm are set as explanatory variables, forming an analysis model (Model 1).

[Model 1]

\(\begin{aligned} M B_{i t}=& \alpha+\beta_{1} \text { Famil } y_{i t}(\text { Family1 }, \text { Family2, Family3) }\\ &+\beta_{2} O W N+\beta_{3} F O R+\beta_{4} B D S I Z E+\beta_{5} \text { OUTDIR } \\ &+\beta_{6} O C D+\beta_{7} R O A+\beta_{8} L E V_{i t}+\beta_{9} S I Z E \\ &+\beta_{10} G R O W+\beta_{11} A G E_{i t}+\eta_{i}+\lambda_{t}+e \end{aligned}\)       (1)

[Model 2] adds firm value (MB) as the dependent variable and the variables for the factors that reduce the value of family firms (executive wage, ownership-control disparity, dividend, and transactions with related parties) and control variables.

[Model 2]

\(\begin{aligned} M B_{i t}=& \alpha+\beta_{1} \text { CEOWage}_{i t}+\beta_{2} O C D_{i t}+\beta_{3} D I V_{i t} \text { (payout, rat) } \\ &+\beta_{4} A f f_{i t}(L S, \quad T P, M T)+\beta_{5} O W N_{i t}+\beta_{6} F O R_{i t} \\ &+\beta_{7} B D S I Z E_{i t}+\beta_{8} O U T D I R_{i t}+\beta_{9} R O A_{i t}+\beta_{10} L E V_{i t} \\ &+\beta_{11} S I Z E_{i t}+\beta_{12} G R O W_{i t}+\beta_{13} A G E_{i t}+\eta_{i}+\lambda_{t}+e_{i t} \end{aligned}\)       (2)

4. Results

4.1. The effects of family ownership on firm value

[Table 3] shows the results of the panel data regression analysis of the effects of family ownership on firm value. With respect to [Model 1], the regression coefficient of the family firm variable ( Family ) is -0.084 (t = -2.82) and the variable has significant negative (-) effects with a significance level of 1%. This implies that firm value decreases for a family firm. This relationship is consistent with the results of (1) to (4). In (1) through (3), the regression coefficients of the family firm variable are -0.084 (t = -1.94), -0.059 (t = -2.15), and -0.088 (t = -2.52) confirming that the variable has statistically significant negative (-) effects.

[Table 3] Results of the panel data regression analysis of the effects of family ownership on firm value

Note: This table shows the results of the panel data regression analysis of [Model 1]. The analysis uses unbalanced panel data including 11 years (2002 to 2012) of time series data for each firm. The t value is in parentheses. g is the statistic of a Lagrange multiplier test. m is the statistic of a Hausman test. ***, **, and * indicate significance of at the 1%, 5%, and 10%, respectively.

[Table 3] explores governance structure variables that affect firm value and shows that the controlling shareholder’s share ratio (OWN), the ratio of outside directors (OUTDIR), ownership-control disparity (OCD) have negative (-) effects on firm value, and that the board of directors’ size (BDSIZE) has positive (+) effects on firm value.

Increasing the controlling shareholder’s share ratio leads to an increasing likelihood that the controlling shareholder will satisfy the incentive for private consumption and curtail firm value. The result that the outside director ratio has negative (-) effects on firm value reflects the actual situation of Korean firm external directors. These firm outside directors are unable to secure independence in their relationships with CEOs to monitor and control management activities. Ownership-control disparity has negative (-) effects on firm value. When the disparity between control rights and cash flow rights increases, the effect of management entrenchment, which uses control rights, is greater than the interest alignment effect caused by cash flow rights.

This causes frequent agency problems and affects firm value. The result that a large board of directors is associated with increasing firm value can be explained because a large board of directors increases the likelihood that excellent management resources will be available to effectively monitor and control management, thereby reducing the management agency problem.

4.2. Factors reducing family firm value

[Table 3]  illustrates that it is possible for Korean family firms to experience loss in firm value if a family member is involved in management because of the pursuit of excessive wages, transfer of wealth between affiliates, special dividends, or related-party transactions. To verify this finding, the related variables are divided into family and non-family firms and compared. [Table 4] shows a mean difference with a significance level of 10% between family and non-family firms in the relevant variables. Family firms have higher executive wages (CEOWage), higher dividend rates (Rat), lower ownership-control disparity (OCD), and fewer long-term supply contracts (LS) than non-family firms.

[Table 4] A comparison of the characteristics of family firms and non-family firms

Note: Executive wage (CEOWage) = Ln(average wage of one executive); Ownership-control disparity (OCD) = control rights - cash flow rights (control rights = sum of share ratios of the controlling shareholder, relatives, executives, non-profit organizations, and affiliates; cash flow right = sum of share ratios of the controlling shareholder and relatives); Dividend (DIV) (Payout ratio (Payout) = dividend/net profits during a term, Dividend rate (Rat) = dividend/capital); Transaction with a related party (Aff) (Long-term supply contract (LS) = amount of long-term supply/sales, Transfer of assets and business (TP) = amount of transfer of assets and business/total assets, Money transactions and guaranteed obligations (MT) = money transactions and amount of guaranteed obligations/total assets). ***, **, and * indicate significance of at the 1%, 5%, and 10%, respectively.

A panel data regression analysis with [Model 2] is conducted to identify factors that reduce the value of family firms. [Table 5] presents the results of the panel data regression analysis between the factors that reduce firm value and firm value. In [Table 5], the variables of executive wages (CEOWage) and ownership-control disparity (OCD) have significant negative (-) effects on firm value (MB). Higher executive wages and ownership-control disparity are associated with lower firm value.

[Table 5] Results of panel data regression analysis between the factors curtailing corporate value and firm value with respect to family firms

Note: A panel data regression analysis is conducted to discover the factors that reduce family firm value. This table shows the results of the panel data regression analysis for [Model 2]. Control variables include the controlling shareholder’s share ratio (OWN), foreign investors’share ratio (FOR), the size of the directors’ board (BDSIZE), the ratio of outside directors (ORTDIR), profitability (ROA), leverage ratio (LEV), firm size (SIZE), growth (GROW), and the number of years since establishment (AGE). ****, **, and * indicate significance of at the 1%, 5%, and 10%, respectively.

This result implies that Korean family firm value can decrease because of the management involvement of a controlling family shareholder and the pursuit of excessive wages or agency costs caused by ownership-control disparity. The inference is that expansion of control rights leads the controlling shareholder of a Korean family firm to have greater control rights in the general meeting of shareholders and on the board of directors; in this case, the management entrenchment effect on the pursuit of higher executive wages or private benefits increases, thereby decreasing firm value.

5. Conclusion

This research analyzes the effects of Korean family ownership characteristics on firm value. The positive and negative effects of family ownership on Korean firm value were analyzed. The factors that reduce firm value were identified when family ownership has a negative effect on firm value. Moreover, this study addresses whether the factors that reduce firm value vary when chaebol characteristics, which represent the main characteristics of Korean firms, are combined with family ownership characteristics. The study examined a total of 5,743 companies listed on the Korea Exchange from the period 2002 to 2012 using a panel data regression analysis.

The summarized results of the empirical analysis are as follows. First, Korean family ownership diminishes firm value. Korean family firms have the possibility of curtailing firm value through excessive wages, the transfer of wealth between affiliates, special dividends, or related-party transactions.

Second, according to the results of the analysis of factors that reduce the value of family firms, higher executive wages and ownership-control disparity are associated with lower firm value. This result implies that Korean family firm value has been reduced when controlling shareholders are participated in management and pursue excessive wages, or make the management entrenchment effects associated with ownership-control disparity. When the controlling shareholders of family firms have increasing control rights over the shareholders’ general meeting and the directors’ board, the agency costs associated with seeking increasing executive wages or private benefits reduce firm value.

This study has the following limitations. First, although this study analyzed firms listed on the Korea Exchange from the year 2002 to 2012, the generalizability of results to all family firms in Korea is limited. In the future, sample firms should be extended to KOSDAQ-listed firms and outside auditing firms, and the study period could be expanded. Second, the family firm classification standard by Westhead and Cowling (2006), Kim et al. (2014) and the characteristic of Korean large-scale conglomerates were used to establish a classification standard appropriate for Korean firms. However, subjective factors may have been included in data sampling and classification. Despite such limitations, this study has significance because it reveals the negative effect of family ownership in Korea on firm value. These negative effects can be the result of agency problems from controlling family shareholders seeking excessive wages or ownership-control disparity.