Diversification strategy and top management team fit.

Matching managers to diversification strategy has long been a cornerstone of strategy implementation research (Finkelstein and Hambrick, 1996; Guthrie and Datta, 1998; Krishnan et al, 1997; Leontiades, 1982; Michel and Harnbrick, 1992; Pitts, 1977; Reed and Reed, 1989; Song, 1982; Tihanyi et al., 2000). The basic premise underlying this body of research is that different strategies pose different management challenges that, in turn, require systematically different management skills and experiences to be implemented success-fully. Managers with backgrounds and skills matched to the critical task demands of a firm's diversification strategy, therefore, should be reflected in superior financial performance. Despite the logical appeal of these arguments, their empirical support remains limited and uneven. The cumulative findings suggest that managers of multi-business firms are generally matched to the task demands of their firm's diversification strategies (Michel and Hambrick, 1992; Pitts, 1977; Song, 1982), although contrary evidence has been found as well (Reed and Reed, 1989). Further, the support for performance improvements from matching managers to diversification strategies is equally mixed (cf, Michel and Hambrick, 1992; Reed and Reed, 1989). One reason for the inconsistencies and limited support found in previous studies may lie in two simplifying assumptions underlying the research designs used. First, previous research has assumed that diversification mode and relatedness typically covary. As such, researchers have either identified firms based on their mode of diversification and made assumptions about the relatedness of their business portfolios (Pitts, 1977; Reed and Reed, 1989; Song, 1982) or classified firms based on the relatedness of their businesses and derived assertions about their typical mode of expansion (Michel and Hambrick, 1992). Generally, it has been assumed that related diversifiers pursue internal development, while unrelated diversifiers expand by acquisition. Other research (Busija et at, 1997; Simmonds, 1990), however, documents that many related diversifiers typically expand lay acquisition and that at least some unrelated business firms primarily expand by internal development within their unrelated subunits. Thus, if this assumption is faulty, our current understanding of matching managers to diversification strategy based on existing evidence would seem incomplete or inaccurate. For the purposes of this study, diversification is defined as a growth strategy, and refers to both the degree of relatedness between a firm's business units (i.e., related vs. unrelated) and the mode in which diversification is achieved (i.e., internal development vs. acquisition). A second limiting assumption in prior work deals with top management skills and characteristics being equally important in and linearly related to the successful implementation of different strategies, despite at least tentative evidence to the contrary (Michel and Hambrick, 1992). This assumption is reflected in the way prior researchers have assessed the performance effects of matching managers to diversification strategies, typically by using interaction terms calculated by multiplying strategy-related data by management-related data (e.g., Michel and Hambrick, 1992; Reed and Reed, 1989). This precludes the possibility that some TMT features are more important for one or more strategies and less important for others, and thus precludes the possibility of identifying strategy-specific linkages between TMT characteristics, diversification strategy and performance. The purpose of this study is to extend previous research on diversification strategy and TMT fit in two ways. First, theoretical linkages between mode and relatedness of diversification and useful TMT characteristics are developed. Previous work has focused on mode or relatedness and only relatedness has been adequately linked theoretically to TMT characteristics. …