Is there a gender effect on the cost of bank financing?

Abstract

In this paper, we address the question of whether the gender of a firm’s leader affects the cost of bank funding faced by small and medium enterprises in Europe. Using a large sample of observations of non-financial firms, during the years 2009–2013, we empirically test for the presence of discrimination, comparing female-led and male-led firms. After controlling for a rich set of variables and addressing potential endogeneity, our results show that i) female-led enterprises are more likely to face worse price conditions for bank financing compared to their male-led counterparts and, ii) firms whose leadership changes from female to male are more likely to benefit from an improvement in interest rate levels. This evidence is robust to different model specifications and various methodological approaches. The existence of such bias in the credit markets highlights the need of policy measures addressing female-led businesses, thus reducing their bank financing burdens and enhancing their entrepreneurial opportunities

Gender and Financial Constraints: An Empirical Investigation in Italy

“…Arzu and Mantovani (2016) showed that firms with higher rate of women in the board are more performing than ones with higher rate of men in the board. Nevertheless, the authors found that there are differences in the way credit is allocated relating to the gender of manager (Mascia & Rossi, 2017). Indeed, there are more firms led by women that do not receive credit, even if they deserve it because the presence of women in corporate leadership positions may improve firm performance (Noland & Moran, 2016;Lin, Pereira, Topalova & Turk, 2016).…”

Does Corruption Influence the Self-Restraint Attitude of Women-led SMEs towards Bank Lending?

“…On the supply side, the literature investigates whether women-led firms face lower credit availability and/or worse cost conditions and provides mixed evidence. 9 While some studies find that women-led enterprises have greater difficulties than man-led ones in obtaining bank loans (Marlow and Patton, 2005;Becker-Blease and Sohl, 2007;Muravyev et al, 2009;Bellucci et al, 2010;Kwong et al, 2012;Wu and Chua, 2012;Alsos and Ljunggren, 2016;Mascia and Rossi, 2017), others exclude gender discrimination and attribute the differences in cost conditions to economic and financial factors such as credit history, assets, sales, and years in business (Cavalluzzo et al, 2002;Blanchflower et al, 2003). A few papers find that women-led firms face more unfavourable loan contract terms than male firms, motivated by the fact that the formers are less inclined to grow (Fabowale et al, 1995).…”

“…Indeed, the firm's leadership gender may not be completely exogenous. Either reverse causalitythe level of credit rationing may impact on the firm performance and, thus, on the choice of the leader to be hiredor omitted variables namely, unobservable organizational and managerial skills, or a given corporate culture may push towards a given leader rather than anothercan affect our estimates (see, inter alia, Adams and Ferreira, 2009;Liu et al 2014;Sila et al 2016;Mascia and Rossi, 2017). If this is the case, we cannot argue that our results are showing the existence of a causal relation between the leader's gender and the probability of non-application to bank credit.…”

“…Notably, for identification purposes, we need an instrument that is highly correlated with the Female dummy and is not correlated with the error term. A good candidate, in this regard, can be the share of female employment by sector of activity available at Eurostat (see, for instance, Mascia and Rossi, 2017). Therefore, we regress…”

“…Second, we address the possible endogeneity affecting the relation between the manager's gender and the non-application reasons. This problem may arises because the choice of appointing a female manager could be driven by unobservable factors, such as a particular corporate culture characterizing the firm (Adams and Ferreira, 2009;Liu et al, 2014;Sila et al, 2016;Mascia and Rossi, 2017), rather than being exogenous. Therefore, to solve the potential endogeneity we employ a two-step approach with the use of an instrumental variable similarly to Cumming (2008), and Heger and Tykvová (2009).…”