Making Boards Effective: An Empirical Examination of Board Task Performance
| DOI | http://doi.org/10.1111/j.1467-8551.2008.00591.x |
| Published date | 01 March 2009 |
| Author | Fabio Zona,Alessandro Zattoni,Alessandro Minichilli |
| Date | 01 March 2009 |
Making Boards Effective: An Empirical
Examination of Board Task Performance
Alessandro Minichilli, Alessandro Zattoni
1
and Fabio Zona
Department of Management, Bocconi University, Viale Isonzo 23, 20135 Milano, Italy, and
1
Management
Department, Parthenope University, and SDA Bocconi School of Management, Strategic and Entrepreneurial
Management Department, Via Bocconi 8, 20136 Milano, Italy
Corresponding author email: alessandro.minichilli@unibocconi.it
Despite the increasing attention of management scholars to boards of directors, there is
still scant evidence on the antecedents of board task performance. The lack of
significant results seems to be due to some theoretical and methodological choices
followed by scholars, i.e. the almost exclusive reliance on agency theory and the use of
demographic data. Following the call for dismantling the fortresses dominating past
studies, this paper contributes to opening the ‘black box’ of boards of directors,
developing a conceptual model that considers the impact of board members’ diversity,
commitment and critical debate on board task effectiveness in performing its service and
control tasks. We collected primary data through a questionnaire survey, and we tested
the model controlling for board, firm and industry characteristics. Our findings suggest
that (i) the predictors we identified, and particularly the board members’ commitment,
are far more important than board demographics to predict board task performance; (ii)
firm and industry contexts exert a significant influence on board task performance; (iii)
predictors have a different impact on specific sets of tasks. Thus, our findings support
the idea that several board characteristics and contingencies at both industry and firm
level must be acknowledged in board design.
Introduction
Despite the increasing attention of management
scholars to corporate governance and boards of
directors, empirical evidence on the relationship
between board composition and structure, on the
one hand, and boards’ outcomes or corporate
financial performance on the other hand is still
equivocal (Dalton et al., 1998, 1999; Hermalin
and Weisbach, 2003; Johnson, Daily and
Ellstrand, 1996). This lack of significant results
seems to be due to two common characteristics of
past research on boards of directors: (1) the
almost exclusive reliance on agency theory, and
the resulting strong emphasis on board control
tasks; (2) the search for a relationship between
board composition and firm financial perfor-
mance, and the consequent lack of process-
oriented board research (e.g. Daily, Dalton and
Cannella, 2003; Roberts, McNulty and Stiles,
2005; Stiles and Taylor, 2001).
First, since Berle and Means (1932) identified the
problems associated with the separation between
ownership and control, the dominant theoretical
lens used in corporate governance studies has been
agency theory (e.g. Daily, Dalton and Cannella,
2003; Dalton et al., 1998). Agency theory dom-
inates corporate governance studies as it proposed
the first satisfactory explanation of the separation
between ownership and control in large public
companies (Jensen and Meckling, 1976). Further, it
was founded on a well-accepted model of man
(Daily, Dalton and Cannella, 2003). However, the
strong reliance on agency theory led scholars and
practitioners to assign strong emphasis on board
control tasks and to assume board effectiveness to
be a function of its independence from manage-
ment (Roberts, McNulty and Stiles, 2005).
British Journal of Management, Vol. 20, 55–74 (2009)
DOI: 10.1111/j.1467-8551.2008.00591.x
r2008 British Academy of Management. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford
OX4 2DQ, UK and 350 Main Street, Malden, MA, 02148, USA.
With respect to the second point, governance
studies have focused on the relationship between
board demographic variables and board’s or
company’s performance (e.g. Daily, Dalton and
Cannella, 2003; Johnson, Daily and Ellstrand,
1996; Pettigrew, 1992). The implicit assumption
behind these studies is that demographic variables
are superior to behavioural and process variables
because they are directly observable, and hence
their measure is more reliable and valid (Pfeffer,
1983). However, some scholars have recently
argued that the use of demographic variables in
the presence of complex group dynamics, as in
boardroom decision-making, cannot be able to
predict board or firm performance (e.g. Daily,
Dalton and Cannella, 2003; Forbes and Milliken,
1999; Johnson, Daily and Ellstrand, 1996).
A recent special issue of the British Journal of
Management (Vol. 16, 2005) underlined the
importance of analysing board effectiveness
through the empirical examination of board
behaviour. Roberts, McNulty and Stiles (2005)
explored the concept of accountability through a
qualitative survey of directors of FTSE 350
companies. They argued that while board dem-
ography conditions board effectiveness, it is the
real behaviour of board members that determines
board effectiveness. Active directors’ behaviour –
i.e. challenging, questioning, informing, encoura-
ging etc. – is an important driver of board
effectiveness. Pye and Pettigrew (2005) pointed
out the importance of focusing on context,
process and time to understand board dynamics.
They called for the development of a new
theoretical framework in board process research,
with the purpose of avoiding confusions among
cause and effect relationships and input and
output variables across multiple levels of analy-
sis. Huse (2005) provided a framework for
exploring behavioural perspectives on boards
and governance. This framework is characterized
by (1) splitting the link between board composi-
tion and corporate financial performance in
intermediate steps (e.g. Zahra and Pearce,
1989); (2) using a pluralistic approach to board
role theories (e.g. Johnson, Daily and Ellstrand,
1996; Zahra and Pearce, 1989); and (3) applying
theories from group and cognitive psychology to
understand boards as decision-making groups
(e.g. Forbes and Milliken, 1999).
Following the call for dismantling the for-
tresses of past research on boards of directors
(Daily, Dalton and Cannella, 2003), this paper
aims to develop and to empirically test a
theoretical model on the impact of board
characteristics on board task performance. We
considered board task performance as the ability
of the board to perform six tasks related both to
service (advice, networking and strategic partici-
pation) and control (behavioural, output and
strategic control) (Huse, 2005). Building on
previous studies (e.g. Forbes and Milliken,
1999; Huse, 2005; Milliken and Martins, 1996;
Zahra and Pearce, 1989), we identified three
antecedents of board task performance: board
members’ diversity, commitment, and critical
debate. We collected data through a survey
directed to CEOs of large Italian companies. We
tested the model using a multiple ordinary least
squares regression analysis. We controlled for the
most investigated board demographic character-
istics (i.e. board size, CEO duality, CEO and
directors’ shareholding, and proportion of outside
directors) and for firm- and industry-level vari-
ables. Our findings support the idea that (i) the
predictors we identified, and particularly board
members’ commitment, are far more important
than board demographics for predicting board
task performance; (ii) firm and industry contexts
have an influence on board task performance; (iii)
predictors have a different impact on specific sets
of tasks. Consequently, our fin dings support the
idea that several board characteristics and con-
tingencies at both industry and firm level must be
acknowledged in board design.
The paper is divided into four sections. In the
first section, we present the theoretical model
considering the relationship between board
members’ diversity, commitment and critical
debate, on the one hand, and board task
performance, on the other. In the second section
we describe the research method, presenting the
sample, the data collection and the operationa-
lization of dependent and independent variables.
Then, we present the results of the study. In the
final section, we discuss our main results and
their implications for research and practice.
Theoretical background
Board tasks
Board tasks have been categorized in several
ways, but the most common distinction is
56 A. Minichilli, A. Zattoni and F. Zona
r2008 British Academy of Management.
To continue reading
Request your trial