We suggest that the equivocal empirical results of board leadership structure on firm performance have both methodological and conceptual roots. We stress that whether board leadership structure enhances or lowers performance depends on its fit with a firm's internal and external conditions, a point that has not been comprehensively addressed by the extant literature. To guide future research in this field, we develop five testable propositions and offer some suggestions on how these propositions may be empirically tested. Copyright Blackwell Publishing Ltd 2005.