Performance of family-owned firms: The impact of gender at the management level

• Main Author: David, Daniela Sofia Taniça
• Publication Date: 2019
• Other Authors: Félix, Elisabete Gomes Santana
• Format: Article
• Language: por
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10174/25266
• topic: Age; Business Sector; Company size; Family Firm; Gender; Region; Performance

Purpose The purpose of this article is to analyze the impact of gender (F/M), at the management level, on the family company´s performance. Design/methodology/approach Company size, age, region and business sector were used as control variables in order to confirm the adjustment of our model to the theory. GMM dynamic panel models were used in order to control for: endogeneity; time-invariant characteristics; possible collinearity between independent variables; effects from possible omission of independent variables; elimination of non-observable individual effects; and the correct estimation of the relationship between the dependent variable in the previous and current periods. The study used data from 199 Portuguese family companies, from 2006 to 2014. Findings The results confirm the hypothesis from corporate governance literature which argues that board diversity is potentially positively related to firm performance, showing that the presence of a female element in family firms’ direction has positive impacts on their performance, compared to those with only male elements. Also, the results show that region and sector of activity are factors influencing family firm performance. Finally, the study confirms that company size and age are variables helping to explain these companies’ life-cycle. Originality/value The study contributes to the literature on family firms, regarding the effect of gender on family firm performance. The use of dynamic panel data models will make a strong contribution to this, as the problem of endogeneity is dealt with correctly here through using these models, and the possible collinearity between independent variables and correct estimation of the relationship between the dependent variable in previous and current periods.