Gender diversity and firm performance: evidence from Dutch and Danish boardrooms

Abstract

Drawing on the business case for gender diversity, this article examines whether board gender diversity has a positive effect on firm performance, based on evidence from the Netherlands and Denmark. We use empirical data on 186 listed firms observed in 2007. Almost 40% have at least one woman in the boardroom. Within boards, the average share of women is only 5.4%. To investigate the impact of board gender diversity, two-stage least-squares estimation is applied, using Tobin’s Q as a measure of performance. Our findings indicate that on the basis of this data-set, there is no relation between board diversity and firm performance.

Keywords:

• board diversity
• corporate governance
• female board representation
• firm performance

Acknowledgements

The authors would like to thank Thomas van Huizen for his useful help with and comments on the statistical analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The 40% objective refers to the under-represented sex in non-executive board-member positions in publicly listed companies and could therefore also refer to men.

2. One might argue that the diversity variable should take into account the evenness of the distribution as simply using the share of women implies that values above 50% are considered as indicating more gender diversity. An example of a variable that does take this distribution into account is the Blau index which has a maximum of .5, when boards have an equal number of men and women (see e.g. Cambell & Mínguez-Vera, 2008). For the sake of simplicity, however, and because the maximum share of women in our sample is only 40%, we use the share of women as the variable indicating board diversity.

3. In case where exact data were missing, we applied the following procedure. If compliance with the corporate governance recommendation was stated in companies’ annual reports or website, we considered half of the directors elected by the General Meeting as independent. In the case of non-compliance, we took half the number of total supervisory directors elected by the General Meeting minus one to calculate the share of independent directors.

4. The results of the first stage are included in the Appendix 1. The coefficients of both instrumental variables are significant in case board gender diversity is measured as a dummy variable (p < .01), F-test for joint significance is above 4. While this F-statistic is rather low, this could be expected given the sample size. The instruments are somewhat weaker in case the share of women on boards is used as an indicator of board gender diversity. Sargan statistics show no overidentification.