The Genesis of top management team diversity : selective turnover among top management teams in the Dutch newspaper publisher industry (1970-1994)

Abstract

This paper develops and tests a model explaining diversity within top executive management teams from the perspective of selective turnover. We draw on two theoretical perspectives to describe the pull toward team homogenization (low diversity) and the push toward team heterogenization (high diversity). Schneider’s attraction-selection-attrition (ASA) model is used to explain the team’s natural tendency to ‘hire likes and fire unlikes’ (so-called homosocial reproduction), provided it has the power to do so. Rational-economic theory, however, suggests firm-specific countervailing imperatives pulling a team toward heterogeneity in order to cope with the requirements of the environment. We propose that the cycle of homosocial reproduction only gets interrupted when the teams face a compelling need for diversity, particularly when organizational performance is poor, diversification is high and competition is tough, but that the reproduction cycle is maintained if the top executive management team is powerful vis-A -vis the board of directors. We test our hypotheses in a population of top executive teams of the Big Five Dutch publishers over a 25-year period. Interestingly, many of our expectations are not supported. Relating to entry, we find that poor performance and high diversification causes teams to select likes. Moreover, although more powerful teams do tend to select likes, this is even more so when competitive intensity increases. Concerning exit, we find that poor performance increases the overall likelihood of executive exit, and that dissimilar managers tend to leave first. We conclude that homosocial reproduction does occur, particularly when the organizations face conditions that at first glance require more team diversity. Apparently, top management teams tend to close ranks when environmental pressure and complexity increases. Explanations for these findings are discussed, which point the way to new research avenues.