Does family identity matter for earnings management? Evidence from private family firms

Abstract

PurposeThe purpose of this paper is to examine the role family identity and reputational concerns plays when private family firms engage in earnings management.Design/methodology/approachThe paper is conducted as an archival study using data from private limited liability firms in Norway over the period from 2002 to 2015. The dataset includes financial accounting data and data on family relationships between shareholders, board members and CEOs, where family relationships are determined through bloodlines, adoption and marriage, tracing back four generations and extending out to third cousins. To investigate the incidence of earnings management, the authors employ a measure of accrual-based earnings management (AEM) (Dechow and Dichev, 2002; McNichols, 2002) and a measure of real earnings management (REM) (Roychowdhury, 2006). They use whether or not the family name is included in the firm name (i.e. family name congruence) as a proxy for family members' identification with the family firm and their sensitivity to reputational concerns.FindingsThe authors’ results show that AEM is lower for family-named family firms. Moreover, their findings also indicate that family-named family firms are more likely to select REM over AEM, compared to nonfamily named family firms. This is even more pronounced when detection risk is high (high quality audit proxied by Big 4).Research limitations/implicationsThe quality of the authors’ findings is limited to the validity of their proxy for family firm identification and reputational concerns (the family name included in the firm name). Even though findings from prior research suggest that family name congruence is a valid proxy for identity and reputational concerns (e.g. Kashmiri and Mahajan, 2010, 2014; Rousseau et al., 2018; Zellweger et al., 2013), future research should investigate the validity of these results using alternative proxies for family firm identification. Future research should also investigate whether the authors’ findings are generalizable to public family firms.Practical implicationsThe authors’ results suggest that the risk of AEM is lower for family-named family firms, whereas the risk of REM is somewhat higher, compared to nonfamily named family firms. These results might be relevant for financial accounting users, auditors and supervisory and monitoring bodies when assessing the risk of earnings management.Originality/valueThe paper is, as far as the authors are aware of, the first to investigate the role of family name congruence and detection risk when private family firms select between AEM and REM.