DISCUSSION OF Dominant Owners and Performance of Continental European Firms

Abstract

Krivogorsky and Burton (2012) examine the association between dominant shareholders and firm performance for 1,533 firms from seven Continental European countries using ownership data from 2005 to 2007. The primary analysis in the paper tests the effects of four separate types of dominant owners (institutions, blockholders, banks, and individuals and families) on two measures of accounting performance (return on assets and return on shareholder funds) and a measure of firm value (market-to-book ratio). Supplemental tests examine cross-sectional differences in the effects of each type of dominant owner across individual countries. The main results indicate that banks and individual and family owners have a positive effect on firm performance, while institutions and blockholders have a negative effect on firm performance. The evidence from the within-country tests shows that the relation between particular shareholder types and firm performance varies across different jurisdictions, with dominant owners generally having a positive effect.Dominant shareholders have incentives and the ability to influence the firms in which they own a controlling interest. Dominant owners are motivated to utilize their control to monitor managerial actions because of their claims to the residual profits of the firm (Shleifer and Vishny 1997). Dominant shareholders also have the ability to monitor managerial actions because of their access to inside information and their ability to control internal forces designed to curb managerial actions that are not consistent with maximization of firm value. Thus, monitoring by dominant owners can serve to address the classic agency conflicts between shareholders and investors (Jensen and Meckling 1976), thereby having a positive effect on firm value.In an international context, however, country-level institutions, laws, and other regulatory features can interfere with dominant shareholders' typical incentives and ability to monitor managerial behavior. Depending on a country's institutional environment, dominant shareholders could be motivated by a different set of factors, perhaps leading them to take advantage of their ownership status to extract personal benefits from the firm. This type of situation would result in a negative relation between dominant ownership and firm value. Given the potential for either monitoring or expropriation by controlling shareholders, it is, therefore, an empirical question whether dominant owners have a positive or negative effect on firm value and performance.Krivogorsky and Burton's (2012) study contributes to the literature that analyzes the effects of controlling shareholders by focusing on the context of Continental Europe, an environment in which dominant owners play an important role (La Porta et al. 1999). Although it is commonly accepted that a firm's ownership structure has important implications for corporate performance outcomes (Shleifer and Vishny 1997), evidence on these effects from existing research is mixed (Gedajlovic and Shapiro 1998). A primary contribution of Krivogorsky and Burton (2012) is the finding that separate analysis of different ownership types, as well as segmented investigation of firms from specific countries, affects the observed relation between dominant shareholders and firm performance. These results indicate that perhaps a reason for the inconsistent conclusions from prior research is the fact that the interests of different types of dominant owners, and the incentives of these owners in different country environments, have not been separately examined.My discussion is focused on three areas: (1) the connection between the current study and the related literature, (2) the interpretation of the results, and (3) directions for future research. Overall, I conclude that Krivogorsky and Burton's (2012) study makes an important contribution toward understanding the effects of dominant ownership on firm performance and firm value; however, several unanswered and interesting issues that would benefit from future research remain.The research questions covered by Krivogorsky and Burton (2012) lie at the intersection of the corporate governance and international accounting research areas. From a corporate governance perspective, this study examines how dominant shareholders affect the typical agency conflicts related to ownership and control. In the presence of dominant shareholders, where ownership and control of the firm are not separated, it is not clear ex ante whether the controlling ownership will have a positive or negative effect on firm value and performance. On the one hand, controlling shareholders that maintain a sizable and stable ownership share over a long period of time have incentives to serve a monitoring role to constrain unfavorable managerial discretion, which is likely to lead to enhanced firm performance. On the other hand, controlling shareholders can use their power to divert corporate wealth to themselves rather than sharing it with other investors, leading to diminished firm performance. These two potential outcomes are often referred to in the literature (e.g., Claessens et al. 2002) as the positive incentive effect and the negative entrenchment effect, respectively.From an international accounting perspective, Krivogorsky and Burton (2012) examine how a country's institutional environment affects the relation between corporate ownership and performance outcomes. For firms operating in Continental Europe, where controlling ownership structures are common (La Porta et al. 1999), the agency conflict of primary concern is majority shareholder expropriation of minority shareholders. Investor expropriation, also referred to as self-dealing or tunneling, has received much attention from recent international accounting and finance research (e.g., Claessens et al. 2002; Dyck and Zingales 2004; Djankov et al. 2008). This body of research offers mixed findings with respect to the particular corporate governance processes that affect dominant owners' ability to impart significant influence.Prior studies show that different types of dominant shareholders have distinct business characteristics and objectives (Claessens et al. 2002; Chen et al. 2007), resulting in variation in the costs and benefits arising from shareholders' control of the firm. This highlights the importance of considering the effects of different types of dominant owners separately in the empirical analysis, which is an important aspect of Krivogorsky and Burton's (2012) study. It is not surprising that the evidence reported in Panel A of Table 6 in the paper does not indicate that dominant owners have a significant influence on corporate performance outcomes, as this is likely due to conflicting incentives across different types of controlling shareholders from different countries. In other words, certain shareholders have a positive influence due to the incentive effect and others have a negative influence due to the entrenchment effect, and these distinctions are not observed empirically until the groups are analyzed separately. The results reported in Panels B through E of Table 6 provide support for this conclusion.Variation in costs and benefits of dominant ownership across different institutional environments lends further support for the country-level analyses conducted in the study. While most research in this area has focused on the effect of one type of dominant owner across various countries, or on all types of controlling shareholders within one country in isolation, this study allows for variation in both shareholder type and country of origin. This provides a contextually rich sample for the analysis. Krivogorsky and Burton (2012) provide links between the country-level results reported in Table 7 to recent political and institutional developments in each country, which is another important feature of the study.The inferences that can be drawn from Krivogorsky and Burton's (2012) study are a bit more limited than the written manuscript suggests. The authors claim that the first objective of their paper is to “investigate whether different types of dominant owners, which have the capacity to control, exercise active control.” However, this statement seems to be based on the assumption that evidence of actively exercising control is demonstrated by a positive association between dominant ownership and firm performance, which is not necessarily the case. While a positive relation is consistent with the possibility that the dominant owner exercised control, it does not mean that the owner actually did so. A different explanation for a positive relation is that it results from management's decision-making and does not arise from the dominant shareholder's influence. This might be the case if the dominant shareholder takes a passive approach to their investment, and the firm's managers make value-enhancing decisions despite the lack of close monitoring. Alternatively, the dominant owner could have actually exercised control, but that monitoring effort and interference resulted in poor investment returns. This would result in a negative association between dominant ownership and firm performance, and would not be interpreted by the authors as evidence that control was actually exercised, despite the fact that it was.The importance of considering the level of investor protection in a country arises from the idea that ownership concentration can serve as a substitute for weak investor protection rights (Shleifer and Vishny 1997). The measure of investor protection used in the paper, SIP, is a country-level index of the strength of minority shareholder protection compiled by the World Bank, and is based on the methodology of the analyses employed by Djankov et al. (2008). Djankov et al.'s (2008) measure is focused on investor protection of minority shareholders against self-dealing transactions benefiting controlling shareholders and, thus, seems an appropriate measure for the context of Krivogorsky and Burton's (2012) analysis.However, the results on the influence of SIP are somewhat surprising because they are not consistent with findings reported by related research. Given the inverse relation between ownership concentration and a country's strength of minority investor protection documented in other studies (La Porta et al. 1998), it is notable that the evidence reported in Table 5 does not show a negative correlation between the various ownership concentration variables and SIP. In addition, the regression coefficient on SIP is negative or insignificant in all of the subsample analyses of the effects of different types of controlling shareholders (Table 6, Panels B though E). These results indicate that firm performance and firm value are consistently lower in countries where the strength of investor protection is higher, which contrasts with evidence of a positive relation between investor protection and firm performance and firm value (La Porta et al. 2002; Klapper and Love 2004). These somewhat surprising results may be due to the fact that Krivogorsky and Burton (2012) split the ownership concentrations by shareholder types for the majority of these analyses, rather than using aggregate ownership levels.Due to the lack of detailed information on the characteristics of common shares of the sample firms, it is difficult to discern whether the ownership percentage variable (PERCENT) adequately accounts for the effect on incentives arising from non-proportional voting and cash flow rights. With respect to this issue, the authors state that they have “checked for potential lack of proportionality in voting and cash flow rights” and “adjusted the dominant owner percent” in the raw data where necessary. However, given the considerable proportion of shares issued in the sample countries that do not have equal cash flow and voting rights (Institutional Shareholder Services 2007), additional information on the significance of this effect for the sample would have been useful. Taking the lack of proportionality into consideration empirically is particularly notable, because related literature finds that the probability of minority shareholder expropriation (i.e., the negative entrenchment effect from dominant share ownership) is high if large shareholders hold voting rights in excess of cash flow rights (Faccio et al. 2001).To account for such divergence in proportionality, Andres (2008) used an indicator variable to capture separation of cash flow and voting rights in his study of the effect of dominant shareholders in Germany. The results showed that the divergence in proportionality of the controlling ownership shares had an inverse relation with firm performance. Therefore, depending on the magnitude of the divergence from proportionality for the sample firms used in Krivogorsky and Burton's (2012) study, a similar correction to the regression design may have been appropriate in order to fully disentangle the incentive effect (associated with cash flow rights) from the entrenchment effect (associated with voting rights) in the empirical analysis.Finally, it is not clear whether the inferences from Krivogorsky and Burton's (2012) analyses are affected by excluding the firm's debt level from the empirical specification, as the amount of debt a firm has is likely to be correlated with several variables of interest. A firm's amount of leverage is likely to affect the firm's performance (Seifert et al. 2005), as well as the types of dominant owners drawn to invest in the firm. In addition, because debt holders can also serve as monitoring agents (Harris and Raviv 1991), the amount of debt in the firm's capital structure will affect the dominant owner's incentives to monitor managerial actions, as well.Krivogorsky and Burton (2012) provide corroborating evidence on the theoretical foundations of the agency conflicts involved with dominant ownership. Through a comprehensive analysis of multiple types of controlling shareholders, the results from their study show that, depending on the shareholder's characteristics, dominant owners can have either a positive incentive effect or a negative entrenchment effect on firm performance and firm value. Krivogorsky and Burton's (2012) analysis is ambitious because it focuses on the effect of dominant ownership in an international context, where country-level institutions regulating self-dealing behavior further complicate controlling shareholders' incentives to monitor and/or their ability to extract private benefits.The evidence presented in this study provides a number of additional opportunities for future research. The regulatory changes enacted in 2007 within the European Union (i.e., Directive 2007/36/EC of the European Parliament and of the Council) were designed to bring about more diverse share ownership for publicly traded firms. As the composition of shareholders changes and concentrated ownership is less common, the level of influence that dominant shareholders have in affecting firm performance is likely to change, as well. Smaller and less-stable ownership blocks would be expected to result in decreased incentives for shareholders to monitor managerial actions, as well as diminished opportunities for expropriation. Future research can examine whether the historical pattern of dominant ownership has changed following the enactment of this regulation and, if so, how this has affected dominant owners' incentives to monitor and ultimately affect the performance of firms in Continental Europe.Another potential avenue for further work involves understanding the interactive effects of multiple dominant owners. In the present study, the authors have included variables in the empirical specification to account for non-controlling ownership interests (owners that have greater than 5 percent ownership, but less than the dominant ownership percentage). The results from this analysis suggest that other non-controlling owners have a diminished effect on firm performance, as evidenced by the fact that the coefficient on the non-controlling ownership interests is significantly negative in nearly all of the regression specifications reported in Table 6. However, the allocation of control between multiple large shareholders is not taken into consideration. As modeled by Maury and Pajuste (2005), multiple blockholders can have two different roles in firms. On the one hand, by holding a substantial voting block, a non-controlling owner has the power and the incentives to monitor the largest shareholder and, therefore, the ability to reduce profit diversion. On the other hand, the non-controlling owner can form a controlling coalition with other blockholders and share diverted profits.Given that the presence of several large shareholders with substantial blocks of shares is common in Europe (Faccio and Lang 2002), it would be interesting for future research to examine the dynamics between multiple large shareholders. Empirical evidence on this effect is limited, with most studies (e.g., Volpin 2002; Lehman and Weigand 2000; Faccio et al. 2001) focusing on firms within a single country or the presence of multiple owners, without taking into consideration the characteristics of different types of dominant shareholders. Further analysis of contextually rich datasets that represent ownership interests from multiple countries (similar to the data used by Krivogorsky and Burton [2012]), can provide insight into the interaction between different types of multiple dominant owners in a cross-country setting. This could provide useful insights toward improving our understanding of the role dominant owners play in affecting the performance and valuation of firms around the world.