«Three essays on corporate boards R. Øystein Strøm A dissertation submitted to BI Norwegian School of Management for the degree of Dr.Oecon SERIES ...»
Hermalin, B. E. and M. S. Weisbach (1998, Mar). Endogenously chosen boards of directors and their monitoring of the CEO. American Economic Review 88(1), 96–118.
Hermalin, B. E. and M. S. Weisbach (2003, Apr). Boards of directors as an endogenously determined institution: A survey of the economic literature. Economic Policy Review 9(1), 7–26.
Hermalin, B. E. and M. S. Weisbach (2006, nov). A framework for assessing corporate governance reform. NBER Working Papers 12050, National Bureau of Economic Research, Inc.
Hopt, K. J. (1998). The German two-tier board: Experience, theories, reform. See Hopt et al. (1998), Chapter 4(a), pp. 227–258.
CHAPTER 2. ALIGNED, INFORMED, AND DECISIVEHopt, K. J., H. Kanda, M. J. Roe, E. Wymeersch, and S. Prigge (Eds.) (1998).
Comparative Corporate Governance: The state of the art and emerging research.
Oxford: Oxford University Press.
Hsiao, C. (2003). Analysis of Panel Data (2. ed.). Cambridge: Cambridge University Press.
Klein, A. (1998, Apr). Firm performance and board committee structure.
Journal of Law and Economics 46, 275–303.
La Porta, R., F. L. de Silanes, A. Shleifer, and R. Vishny (2000, Oct/Nov).
Investor protection and corporate governance. Journal of Political Economy 58(1-2), 3–28.
Palia, D. (2001). The endogeneity of managerial compensation in ﬁrm valuation: A solution. Review of Financial Studies 14(3), 735–764.
Perry, T. and U. C. Peyer (forthcoming 2005). Board seat accumulation by executives: A shareholder’s perspective. Journal of Finance.
Raheja, C. G. (2005). Determinants of board size and composition: A theory of corporate boards. Journal of Financial and Quantitative Analysis 40(2), 283–306.
Schmid, F. A. and F. Seger (1998). Arbeitnehmermitbestimmung, Allokation von Entscheidungsrechten und Shareholder Value. Zeitschrift fur Betriebswirtschaft 68(5), 453–473.
Shleifer, A. and R. W. Vishny (1986). Large shareholders and corporate control. Journal of Political Economy 94(3), 461–486.
Shrader, C. B., V. B. Blackburn, and P. Iles (1997). Women in management and ﬁrm ﬁnancial performance: An exploratory study. Journal of Managerial Issues 9(3), 355–372.
Smith, N., V. Smith, and M. Verner (2006). Do women in top management affect ﬁrm performance? A panel study of 2300 Danish ﬁrms. International Journal of Productivity and Performance Management 55(7), 569–593.
Wasserman, S. and K. Faust (1994). Social Network Analysis: Methods and Applications. Cambridge, UK: Cambridge University Press.
Westphal, J. D. and J. W. Fredrickson (2001, Dec). Who directs strategic change? Director experience, the selection of new CEOs, and change in corporate strategy. Strategic Management Journal 22(12), 1113–1137.
BIBLIOGRAPHY 53 Williamson, O. E. (1996). The Mechanisms of Governance. Oxford, UK: Oxford University Press.
Woolridge, J. M. (2002). Econometric Analysis of Cross Section and Panel Data.
Cambridge, Mass.: The MIT Press.
Wymeersch, E. (1998). A status report on corporate governance rules and practices in some continental European states. See Hopt et al. (1998), Chapter 12(d), pp. 1045–1200.
Yermack, D. (1996). Higher market valuation of companies with a small board of directors. Journal of Financial Economics 40, 185–212.
CHAPTER 2. ALIGNED, INFORMED, AND DECISIVE
The table shows descriptive statistics for the board design mechanisms, the control variables, and the performance measure. The board design mechanisms are classiﬁed according to their primary function (interest alignment, information provision, and decisiveness) as discussed in section 2.2. The variables are deﬁned in table 2.1.
CHAPTER 2. ALIGNED, INFORMED, AND DECISIVE
The table shows estimates of the base-case ﬁxed eﬀect regression model in expressions (2) and (3) as estimated with GMM. The dependent variable is Tobin’s Q, which we measure as the market value of the ﬁrm over its book value. Every variable is time demeaned by subtracting a given ﬁrm’s observation in a given year from the ﬁrm’s overall mean across the years.
Instruments are the raw, the time-demeaned, and the squared time-demeaned explanatory variables, the average and standard deviation of ﬁrm-demeaned explanatory variables. The ﬁrst column reports unstandardized (regular) coeﬃcient estimates. The second column shows the estimates based on the standardized variables, which we construct by deducting each observation from its mean value and dividing by its standard deviation. The p-values in the third column are identical for both coeﬃcient types.
2.7. TABLES 57
The table explores mechanism endogeneity by means of two models. In the dynamic performance model in the ﬁrst column of results, ﬁrm performance is the dependent variable, and lagged performance is added as an independent variable to the basic model in (2.2). The second model, termed the integrated mechanisms model, consists of ﬁve equations. The ﬁrst equation is the base-case performance model (2.2) from table 2.3. Each of the ﬁve other equations have a board mechanism as the dependent variable, which is Directors’ holding, Independence, Network, Gender, and Size, respectively. The six equations in the second model are estimated one by one. The instruments correspond to those used table in 2.3. Performance is measured as Tobin’s Q, which we operationalize as the market value of the ﬁrm over its book value. All variables in all models are time-demeaned, which means that for each ﬁrm and each variable, we subtract a given year’s observation from the ﬁrm’s overall mean across the years. The table shows the estimates based on the standardized variables, which we construct by deducting each observation from its mean value and dividing by its standard deviation.
Signiﬁcant results at the 5% (10%) level are marked with ∗∗ ( ∗ ).
CHAPTER 2. ALIGNED, INFORMED, AND DECISIVE
The table shows the results of using alternative operationalizations for board independence in models (1)-(3), director network in model (4), and a combination of the two in model (5).
Model (6) is the base-case model from table 2.3. OD is the average number of outside directorships held by the ﬁrm’s board members, and BT is the board tenure proxy from model (1). The dependent variable is Tobin’s Q, which we measure as the market value of the ﬁrm over its book value. Each variable is time demeaned in the regressions. For each ﬁrm and each variable, we time demean by subtracting a given year’s observation from the ﬁrm’s overall mean. The regressions use the same instrument set as in table 2.3, but with new variable deﬁnitions. The table shows the estimates based on the standardized variables, which we construct by deducting each observation from its mean value and dividing by its standard deviation. Signiﬁcant results at the 5% (10%) level are marked with ∗∗ (∗ ).
2.7. TABLES 59
The table shows estimates of the base-case model from table 2.3 when including (model (A)) and not including (model (B)) employee directors in the deﬁnition of board size, gender diversity, and board independence. The dependent variable is Tobin’s Q, which we measure as the market value of the ﬁrm over its book value. Each variable is time demeaned in the regressions. For each ﬁrm and each variable, we time demean by subtracting a given year’s observation from the ﬁrm’s overall mean. The instrument set is as in table 2.3, except that the new variable deﬁnitions replace the ones from table 2.3. The table shows the estimates based on the standardized variables, which we construct by deducting each observation from its mean value and dividing by its standard deviation.
Better rm performance with employees on the board?
This paper1 explores the relationship between employee directors and ﬁrm performance. Shareholders recognise that employee directors bring more heterogeneity to the board, and try to neutralise their value-reducing effects by adjusting the board characteristics and the leverage. Board characteristics and leverage are then endogenously determined by the presence of employee directors. The lagged ﬁrm performance represents another endogeneity inﬂuencing board characteristics and leverage. The relations are modelled in a simultaneous equations framework, allowing an estimation of the relative importance of the endogeneity effects relative to the direct effects of the variables. Panel data of all non-ﬁnancial Norwegian listed ﬁrms from 1989 to 2002 enter the ﬁxed effects estimation. The negative employee director association with ﬁrm performance comes out very clearly, and shareholders’ compensatory actions to neutralise the employee director inﬂuence are clearly visible. Reverse causation running from the lagged ﬁrm performance to the board characteristics and the leverage is conﬁrmed. Yet the endogeneity effects are economically small compared to the direct effects of board characteristics and leverage. Overall, the paper rejects any beneﬁcial effects of one group of stakeholders – the employees – on company boards.
Keywords: Corporate governance, Employee directors, Board composition, Regulation, Endogeneity JEL classiﬁcation codes: G34, G38 1I have beneﬁted from comments from Øyvind Bøhren, Ole Gjølberg, Roswitha King, Gudbrand Lien, participants at the 7th workshop on Corporate Governance and Investment, Jönköping 2006, and the 2nd International Business Economics Workshop, Majorca 13-14 September 2007. Pål Rydland and Bernt Arne Ødegaard have guided me to data.
3.1 Introduction This paper deals with the impact of co-determination2 upon ﬁrm performance. Two conﬂicting views on the beneﬁts of co-determination exist.
One says that co-determination increases ﬁrm performance, either because employee directors supply outside directors with information they would otherwise not have access to (Freeman and Reed, 1983 and Blair, 1995), or because co-determination is a safeguard against dismissal inducing employees to invest in ﬁrm-speciﬁc human capital (Zingales, 2000 and Becht et al., 2003). The other view is that owners’ and employees’ interests are not aligned, and therefore, allowing employees into the boardroom means that conﬂicting goals are pursued. When decision makers with different objectives share in the board’s decisions, its focus may become unclear (Tirole, 2001), its decision time longer (Mueller, 2003), and its decision quality inferior3. The prediction is that ﬁrm performance will be lower than it could otherwise be.
Even though co-determination is important in many European countries4, few ﬁrm-level studies have been made of its ﬁrm performance impact. This paper is an attempt to bring more academic research to the still under-researched (Goergen, 2007) comparison of ﬁrm performance in shareholder determined companies and co-determined companies. Earlier studies give mixed results, showing a negative impact in German ﬁrms (FitzRoy and Kraft, 1993; Schmid and Seger, 1998; and Gorton and Schmid, 2000), Canadian (Falaye et al., 2006), and Norwegian (Bøhren and Strøm, 2007), but a positive impact in a later German study (Fauver and Fuerst, 2006).
Compared to former literature the simultaneous equation estimation of the relationship between ﬁrm performance and explanatory variables 2 Co-determination is deﬁned as employee board representation (Jensen and Meckling, 1979 and Furubotn, 1988).
3 Tirole (2002, p. 118) argues that these “... (c)onﬂicts of interest among the board generate endless haggling, vote-trading and log-rolling. They also focus managerial attention on the delicate search for compromises that are acceptable to everyone; managers thereby lose a clear sense of mission and become political virtuosos.” In a similar vein, Hansmann (1996, p. 44) states that “... because the participants [i.e. stakeholders] are likely to have radically diverging interests, making everybody an owner threatens to increase the costs of collective decision making enormously.” 4 According to the European Industrial Relations Observatory On-line full employee representation is found in Austria, the Nordic countries, and Germany, while the Netherlands and France have systems closer to a consultative function for employee representatives.
3.1. INTRODUCTION 63is the distinctive feature of this paper. The need for simultaneous modelling arises from the fact that the presence of employee directors may induce shareholders to adjust other governance mechanisms, notably board composition and leverage, in order to neutralize the co-determination effects (Buchanan and Tullock, 1962). Employee directors may have a direct impact upon ﬁrm performance, but also an indirect effect. This also means that board composition is at least partly determined by employee directors. Thus, the paper necessarily also relates to the board endogeneity issue (Hermalin and Weisbach, 2003). And since the data cover several periods, it is possible to test the reverse causation hypothesis that ﬁrm performance determines board composition (Hermalin and Weisbach, 1998).
The simultaneous equation setup allows not only the discovery of endogeneity in governance mechanisms, but also a quantiﬁcation of its importance compared to direct effects. I am unaware of former literature containing a measure of the endogeneity effects.