| Introduction A long tradition of research argues that organizations are reflections of their top managers’ characteristics and values (see, e.g., Hambrick 2007; Hambrick and Mason 1984), leading to a strong heterogeneity in corporate practices (Bertrand and Schoar 2003). Recent financial studies highlight a “CEO effect” (captured by means of various observable characteristics [such as age and gender] and behavioral features [such as overconfidence and narcissism]) on corporate decision-making for listed firms.1 However, the link between CEO characteristics and financial decisions remains an under-examined issue for private small and medium-sized enterprises (SMEs). SMEs’ “uniqueness” creates specific financial management issues that stand in contrast to the vast body of corporate finance theory traditionally dedicated to listed firms (Ang 1991). In contrast to publicly traded firms, most private SMEs are run by a CEO who is also the dominant owner (Brunninge, Nordqvist, and Wiklund 2007) and whose investment in the business represents a significant portion of his/her wealth. Moreover, family-owned SMEs typically feature limited corporate governance, particularly the control exercised by the board (Ford 1988). Given these patterns, it should be expected that CEOs (and, therefore, CEO characteristics) would have a substantial impact on corporate decision-making in SMEs (Ang, Cole, and Lawson 2010). This paper focuses on a particular corporate practice—the production of accounting information—and on the related quality of reported earnings numbers.2 Watts and Zimmerman (1986) argue that corporate managers may exercise discretion over the production of accounting information for their own benefit or to reduce certain political costs. Such “earnings management” misleads the stakeholders’ perception of the firm and necessarily erodes firms’ earnings quality.3 As research on earnings management mainly examines samples of listed firms, the magnitude of this phenomenon in small businesses is not well documented. As emphasized by Burgstahler, Hail, and Leuz (2006), private companies experience agency problems that differ from those of public firms. These problems stem from their peculiar relationships with their main stakeholders, which translate into specific incentives to manage earnings. For instance, SMEs are frequently strongly dependent upon banks and can be subject to substantial financial constraints (see, e.g., Cenni et al. 2015). Another prevalent stakeholder for firms located in code-law countries is the tax administration service because financial reporting is strongly influenced by tax regulation in those countries (Coppens and Peek 2005). Our paper offers the first analysis of the impact of CEO characteristics on private SMEs’ decision to engage in earnings management. In so doing, we focus on two observable demographic traits in CEOs, namely gender and age, that have previously been shown to affect organizational behaviors and performance (for gender, see Davis et al. 2010; Pfeffer 1983; for age, see Wiersema and Bantel 1992). We first contribute to the strand of research that investigates whether CEO characteristics affect corporate financial decisions. The existence of a “CEO effect” on corporate decision-making recently became controversial (Fee, Hadlock, and Pierce 2013). Our paper provides strong empirical support for a “gender effect” and an “age effect” concerning the relationship between CEO demographics and earnings management. Second, we extend a line of emerging research that analyzes the interaction between the gender of top management (Barua et al. 2010; Francis et al. |