Research summary: We use a variance decomposition methodology to assess the degree to which board chairs may influence their companies' performance. To isolate the board chair effect, we focus on firms in which the CEO and board chair positions are separated. Using a U.S. sample of 6,290 firm‐year observations representing 1,828 board chairs in 308 different industries, our results indicate that the board chair effect is substantial at about nine percent. Drawing on resource dependency theory, we also theorize and show how this board chair effect is contingent on the task environment in which firms operate. Our results add to the literature examining the role and influence of board chairs and the context in which chairs may have a greater impact on performance. Managerial summary: Following institutional and regulatory changes, more firms are separating the CEO and board chair positions. With an increasing number of individuals separate from the CEO serving as board chairs, a critical question becomes: What influence do these separate board chairs have on firm performance? Prior research suggests that separate board chairs can provide important resources—including advice and counsel, legitimacy, information linkages, and preferential access to external commitments and support—to their CEOs, other top managers, and overall firms. In turn, who the board chair is and the individual's ability (or lack thereof) to provide these resources may have a significant impact on firm performance. Offering support for this perspective, we find that separate board chairs explain nine percent of the variance in firm performance. Copyright © 2016 John Wiley & Sons, Ltd.