«Three essays on corporate boards R. Øystein Strøm A dissertation submitted to BI Norwegian School of Management for the degree of Dr.Oecon SERIES ...»
Three essays on
R. Øystein Strøm
A dissertation submitted to BI Norwegian School of
Management for the degree of Dr.Oecon
SERIES DISSERTATIONS 1/2008
BI NORWEGIAN SCHOOL OF MANAGEMENT
DEPARTMENT OF FINANCIAL ECONOMICS
R. Øystein Strøm:
Three essays on corporate boards
c R. Øystein Strøm
Series of Dissertations 1/2008 BI Norwegian School of Management N-0442 Oslo Phone: 4641 0000 06600 www.bi.no Printing: Nordberg BI Norwegian School of Management’s dissertation may be ordered from our website www.bi.no (Research - Research Publications) Contents 1 Overview 1
1.1 Introduction............................ 1 1.1.1 Data and institutional background........... 3
1.2 Aligned, informed, and decisive................ 4
1.3 Co-determination......................... 7
1.4 Board control........................... 11
1.5 Looking back........................... 15 2 Aligned, informed, and decisive 21
2.1 Introduction............................ 22
2.2 Theory, evidence, and methodology.............. 27 2.2.1 Interest alignment.................... 27 2.2.2 Information........................ 30 2.2.3 Decisiveness........................ 32 2.2.4 Endogenous board design mechanisms........ 34
2.3 Descriptive statistics....................... 35
2.4 Statistical tests........................... 39 2.4.1 The basic model...................... 40 2.4.2 Endogeneity........................ 42
2.5 Robustness............................. 45
2.6 Summary and conclusions.................... 47
2.7 Tables................................ 54
4.1 The age of arriving and departing CEO............ 137
4.2 The percentage chairman change relative to CEO change.
The line represents the chairman changes, while bars are the number of observations in a given year (right hand scale).138
4.3 The average number of new shareholder elected directors relative to CEO change. The line represents the director changes, while bars are the number of observations in a given year (right hand scale)................... 139
4.4 Board turbulence in ﬁrms experiencing an improved stock return subsequent to a CEO turnover and ﬁrms with reduced stock return. Board turbulence is drawn as a line, observations as bars........................ 140
3.1 The percentage of ﬁrms with zero or more employee directors by employment size........................... 93
3.2 The percentage of ﬁrms with zero or more employee directors by industry and the percentage with zero by year. The Global Industry Classiﬁcation Standard (GICS) is used............. 94
3.3 The main statistical properties of various board measures..... 95
3.4 Is co-determination associated with negative ﬁrm performance and positive governance mechanisms? Full sample (N = 1135) estimations using systematic and ﬁrm speciﬁc risk............. 96
3.5 Is ﬁrm performance (Tobin’s Q) diﬀerently related to governance mechanisms in co-determined (upper part) and in shareholder determined (lower part) ﬁrms?.................... 97
3.6 Are the employee director direct and indirect (endogenous) effects upheld in all ﬁrms with more than 200 employees and in co-determined ﬁrms with more than 200 employees?........ 98
3.7 The employee director direct and indirect (endogenous) eﬀects upon ﬁrm performance when the presence of and individual board variables are used instead of the board index. N = 1135...... 99
4.7 Higher board turbulence following successful CEO turnover? GMM estimation with board turbulence as dependent variable, and varying with diﬀerent deﬁnitions of CEO turnover and ﬁrm performance. All listed non-ﬁnancial ﬁrms at Oslo Børs 1989 to 2002.
Dummy variables indicate improved or reduced performance with new CEO............................... 146 Overview of the three papers
The three essays in this thesis investigate the role of the board of directors from different angles. The ﬁrst – “Aligned, informed, and decisive.
Characteristics of value-creating boards” – is a broad overview and a test of theories that have been put forward on the relationship between board characteristics and ﬁrm performance. The second essay – “Better ﬁrm performance with employees on the board?” – deals in depth with one aspect, the impact of employee directors upon ﬁrm performance. The third essay – “Board control and departures” – takes up the question of CEO entrenchment as viewed through the interaction of CEO turnover and board changes. Overall, the three essays1 ﬁnd that board characteristics matter;
that government regulation to favour worker participation in the board reduces ﬁrm performance; and that CEO entrenchment is not typical of Norwegian boards. While the co-determination essay explores the relationship between owners and employees, the board control essay investigates the relationship between owners, the board, and the CEO.
Why is the board an interesting object of study, and why should one study Norwegian boards? The board is interesting because of its importance. The board appoints the CEO, decides on the CEO’s reward, larger investments and strategy, and monitors the company. Thus, the functioning board’s decisions should have long-term impact upon the company in setting its future direction and in motivating its management. The Norwegian board is interesting because its institutional conditions allow the testing of aspects of the board’s functions that are not possible elsewhere.
This concerns regulations on co-determination and on the protection of minority shareholders (Porta et al., 1998).
Besides the main question of each essay, they all deal with endogeneFor short: AID, co-determination, and board control
CHAPTER 1. OVERVIEW
ity in some form. In the board literature, endogeneity is usually taken to mean that board characteristics are themselves at least partly determined by former ﬁrm performance (Hermalin and Weisbach, 1998). This may be termed reverse causation. The AID and co-determination essays test for this kind of endogeneity. Furthermore, the AID essay tests for simultaneous endogeneity among board characteristics (Agrawal and Knoeber, 1996). However, endogeneity is often deﬁned for a model of simultaneous system of equations, where endogenous variables are determined jointly within the model as a function of exogenous variables (Davidson and MacKinnon, 1993, p. 622). This kind of endogeneity is explored in the co-determination essay. Here, the variable employee directors is taken to be exogenous together with former ﬁrm performance and control variables, and endogenous variables are board characteristics, leverage and average wage. The last essay on board control essay deals more closely with the Hermalin and Weisbach (1998) proposition that board characteristics are endogenously determined from the CEO’s stronger bargaining position arising from former good ﬁrm performance.
Both Bhagat and Black (1999) and Hermalin and Weisbach (2003) divide board studies into overall studies of ﬁrm performance upon a set of board characteristics on the one hand, and on the other studies of the impact of board characteristics in discernable board actions, for instance a takeover bid. The AID and the co-determination essays fall into the ﬁrst category, the board control essay belongs to the second. The two research strategies reﬂect different conceptions of the board. Bhagat and Black hold that the board’s impact is best studied in a long-time data set, since the board’s actions are general and apply to all value-creating aspects of the ﬁrm. Hermalin and Weisbach think that the board intervenes only in times of exception, thus, only the proper, discernable board actions can truly reveal which board characteristics have an impact upon ﬁrm performance.
Both research strategies beneﬁt from the availability of a panel data set.
Such a data set is used here, unlike most board studies.
1.1.1 Data and institutional background Our sample is all non-ﬁnancial ﬁrms listed on the Oslo Stock Exchange (OSE) at year-end at least once over the period 1989–2002.2 To reduce censoring bias in the tenure measures, I start collecting director data in 1986.
The ownership structure data covers every equity holding by every investor in every sample ﬁrm. The public securities register (VPS) provided the ownership data, accounting and share price data is from the OSE, and board data was collected manually from Kierulf’s Håndbok and a public electronic register. The registers provided the name of the CEO as well as the name, date of birth, and board position of every board member.
In most cases, the records also show whether the director was elected by shareholders or the employees. Since the ﬁrst name is known, the director’s gender is known as well.
The resultant data set is of a panel data nature, that is, repeated observations of the same ﬁrms over a number of years are in the data base.
This makes the data ﬁle large in international comparisons. The panel data requires statistical techniques that are only now beginning to be applied among researchers in the corporate governance ﬁeld.
The institutional framework is important in understanding the papers.
First, board regulations are such that the CEO cannot at the same time be the board’s chairman. Thus, a power split is built into the law. The board is small by international comparisons, numbering about ﬁve shareholder elected members. The number of management members is likewise very low. Background checks on the board reveal, for instance, that the chief ﬁnancial ofﬁcer (CFO) is a member of the board only by exception. The CEO is seldom a member, too. Another law-based feature is the presence of employee representatives on the board. A politically induced effort to increase the fraction of female directors is also evident.
The ownership concentration of Norwegian ﬁrms is low by European standards. Norwegian ﬁrms have a less concentrated ownership structure than in any other European country except the UK. For example, the average largest owner holds close to 50% of voting equity in a continentalEuropean listed ﬁrm, and 15% in the UK. The corresponding US ﬁgure 2 The OSE had an aggregate market capitalization of 68 bill. USD equivalents by yearend 2002, ranking the OSE sixteenth among the twenty–two European stock exchanges for which comparable data is available. During our sample period, the number of ﬁrms listed increased from 129 to 203, market capitalization grew by 8% per annum, and market liquidity, measured as transaction value over market value, increased from 52% in 1989 to 72% in 2002 (sources: www.ose.no and www.fibv.com).
CHAPTER 1. OVERVIEWis 3% (Barca and Becht, 2001). Norway has a civil law regime, which is generally considered less investor–protective than common law. Nevertheless, La Porta et al. (2000) ﬁnd that Norway’s regulatory environment provides better protection of shareholder rights than the average common law country. According to their theory of institutionally determined ownership structures, the strong investor protection is a major reason why Norway’s ownership concentration is so low.
1.2 Aligned, informed, and decisive
The ﬁrst essay pulls together different components of board design in order to explain ﬁrm performance, and to explore possible endogenous relationships. Most of the literature in corporate governance approaches the analysis of board structure in a piecemeal manner (Bhagat and Black, 1999). But to be effective in furthering shareholder value, board composition must solve the three problems of alignment with shareholders’ interests, internal and external information collection, and rational and expedient decision-making. The paper builds upon explanations given in the literature, but also develops new measures to reﬂect this many-sided board nature.
Interest alignment in a board is achieved through ownership and board independence. The higher is the share of ownership in the company, the higher is the owner’s incentives and power to monitor the agent (Shleifer and Vishny, 1986). Two variables measure this effect, that is, the directors’ ownership and the outside ownership concentration. The outside ownership concentration is measured by the Herﬁndahl index3. The higher the board independence from monitored ofﬁcers the better aligned is the board with owners.
The Hermalin and Weisbach (1998) model predicts that the longer the history of good performance under the current CEO, the less independent the current board. Consistent with the model, we introduce a new measure of board’s independence as the difference between the average tenure
of its non-CEO directors and the tenure of the CEO:
1n ∑ non-CEO director tenurei − CEO tenure (1.1) Independence ≡ n i=1 3 The Herﬁndahl index for outside ownership concentration is the sum of squared ownership fractions across all the ﬁrm’s outside owners. Its maximum value is one (a single investor owns every share held by the outsiders), approaching its minimum value of zero as the ownership structure gets increasingly diffuse.