臺大管理論叢 NTU Management Review VOL.28 NO.3

Does CEO Career Horizon Lead to Corporate Misconduct? Evidence of Taiwanese Semiconductor Firms 152 However, larger board size is not always detrimental to corporate decision making. Resource dependence theorists suggest that, compared to smaller boards, larger boards may contain more professional expertise and experiences to obtain and process a great deal of information about their firms (Borch and Huse, 1993; Pfeffer, 1973). In addition, large-sized boards can be divided into several subcommittees that separately perform the different administrative functions (Williams et al., 2005), i.e., firms with larger board size are better able to monitor their executives from various angles. Several lines of evidence demonstrate that larger board size can be used to prevent business failure, for example, Chaganti, Mahajan, and Sharma’s (1985) comparison study of failed and non-failed retailing firms reveals that non-failed firms have bigger board sizes, while Gales and Kesner’s (1994) investigation of firms’ bankruptcy shows that non-bankrupt firms tend to have more directors. Judging from the above, larger board size could increase tendencies of monitoring powers so that the effects of self-interest motivated CEOs would be alleviated. By extension of this reasoning, the unlawful desires of CEOs, driven by their career horizon, would be alleviated; that is, the inverted-U effect of CEO career horizon on corporate misconduct will be more salient when a firm has a smaller board size because the monitoring is provided by fewer board members; on the other hand, the relationship between CEO career horizon and corporate misconduct can be mitigated by more frequent and intensive monitoring. H2. The inverted U-shaped relationship between CEO career horizon and corporate misconduct becomes weaker in firms, when their boards are composited of more members. 2.2.2 Board Functional Diversity Besides the board size, the diversity among board members is also likely to affect a board’s decision making. The diversity of a board can basically be broken down into two types, i.e., diversity in observable attributes (e.g., age, race, gender and others) and diversity in underlying attributes (e.g., functional background, technical skills and others). By comparing the causes and effects of the two types of diversity, Milliken and Martins (1996) argue that diversity in underlying attributes is more job- or task-related and has higher potential to impact the board performance. Among the various underlying attributes of board members, several studies further suggest that the functional diversity of board members is particularly relevant to board governance and mostly used to conduct research at the board level (Jensen and Zajac, 2004; Milliken and Martins, 1996; Minichilli,