Are female CEOs associated with lower insolvency risk? Evidence from the US property‐casualty insurance industry

• Published date: 01 December 2023
• Author: Jing Li,Jiang Cheng
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/jori.12439

Received: 8 May 2022

|

Revised: 28 June 2023

|

Accepted: 30 June 2023

DOI: 10.1111/jori.12439

ORIGINAL ARTICLE

Are female CEOs associated with lower

insolvency risk? Evidence from the US

property‐casualty insurance industry

Jing Li

1

|Jiang Cheng

2

1

Department of Finance, School of

Economics, Southwest University of

Political Science & Law, Chongqing,

China

2

Department of Finance and Insurance,

Faculty of Business, Lingnan University,

Tuen Mun, Hong Kong

Correspondence

Jiang Cheng, Department of Finance and

Insurance, Faculty of Business, Lingnan

University, Tuen Mun, Hong Kong.

Email: jiangcheng@ln.edu.hk

Abstract

This paper investigates the relationship between

female CEOs and insolvency risk of US property‐

casualty insurance companies. We show that female

CEOs are associated with lower insurer insolvency

propensity, higher z‐score, and lower standard devia-

tion of return on assets. These findings are robust to

alternative econometric specifications to address

potential endogeneity concerns and self‐selection

issues, including propensity score matching, the

instrumental variable approach, and the difference‐in‐

difference approach. Furthermore, we find that the

impact of female CEOs on insurer insolvency risk is

moderated by firm capitalization, the presence of

female directors, and political conservatism of insurers'

home states.

KEYWORDS

capitalization, female CEO, female directors, insolvency risk,

political conservatism

JEL CLASSIFICATION

G22, G32, G33

Journal of Risk and Insurance. 2023;90:941–973. wileyonlinelibrary.com/journal/JORI

|

941

© 2023 American Risk and Insurance Association.

1|INTRODUCTION

The issue of gender equality has gained considerable attention in recent years, both in practice

and from a research perspective. This trend will likely continue as more women obtain business

training and as countries around the world adopt policies to expand women's representation

among business leaders. The number of female CEOs reached a record high in 2021,

accounting for 8.1% of the Fortune 500 list.

1

Yet it remains unclear whether and how women's

management of firms differs from men's management.

In this study, we investigate the relationship between female CEOs and firm risk‐taking in

the property‐casualty insurance industry. We focus on insurers for several reasons. First, risk

management is particularly important for insurers because their business model is wholly built

on their ability to properly identify and price risk to generate profits. The case of AIG in the

financial crisis of 2008 suggests that excessive risk‐taking and inappropriate risk management

can jeopardize insurer solvency and potentially lead to systematic risk throughout the entire

financial market. Insurers' financial stability has always been the main objective of regulators.

Thus, the insurance context is especially interesting to study because of the negative

externalities that excessive risk‐taking may generate for the financial service industry and

ultimately for the whole economy.

Second, insurance is a highly regulated industry with a large number of private firms. This

setting can enrich our understanding of female leadership beyond publicly traded stocks, which

is the focus of most of the literature. Third, focusing on a single industry provides higher

internal validity by eliminating cross‐industry variation. Consequently, insurance provides an

ideal laboratory to study whether and how female leadership helps reduce firm risk and, in

particular, insolvency propensity.

We construct the dataset by manually collecting CEO and board information for a 17‐year

period (2001–2017 inclusive) for each insurer, obtaining the data from the A.M. Best's

Insurance Reports. Gender is identified through the Securities and Exchange Commission's

EDGAR filing system for publicly traded stocks and several online name databases

2

for private

firms (Faccio et al., 2016; Palvia et al., 2015), and the information is cross‐checked via Google,

LinkedIn, Factiva, and company websites. Financial information is obtained from the National

Association of Insurance Commissioners (NAIC) database. Our full sample comprises 1786

firms associated with 2117 unique CEOs and 20,539 firm‐year observations.

Our empirical results suggest that female CEOs are associated with lower insolvency

probability, higher z‐score, and lower standard deviation of return on assets (ROA). In further

tests, we find that the negative association between female CEOs and insurer risk is mitigated

in insurers with higher capitalization firms, insurers with a higher proportion of female

directors, and insurers domiciled in states with lower political conservatism. Finally, the

negative association between female CEOs and insurer insolvency propensity holds in the

financial crisis period (Owen & Temesvary, 2018; Palvia et al., 2015).

Establishing a causal relationship between female CEOs and insurer risk is challenging.

Endogeneity issues may arise because CEO appointment is not an exogenous random variable.

A potential further concern is that a selection bias may influence our results. The possibility

exists that female CEOs might self‐select to join firms that are less risky. Prior evidence shows

1

https://fortune.com/2021/06/02/female-ceos-fortune-500-2021-women-ceo-list-roz-brewer-walgreens-karen-lynch-cvs-

thasunda-brown-duckett-tiaa/.

2

http://www.genderchecker.com;http://www.babynology.com;http://www.namepedia.org/en/firstname.

942

|

LI and CHENG

that women are more likely to join larger boards and to seek employment at firms that are less

risky, with higher growth and better performance (Liu et al., 2014; Mateos de Cabo et al., 2012).

However, extant literature also suggests that female executives are more likely to be appointed

by firms already facing problems (Ryan & Haslam, 2005,2007). We first use a lagged

independent variable (female CEO dummy, Female CEO) to address the possibility of the

results being driven by reverse causality (Chen et al., 2017; Dittmann et al., 2010; Joecks

et al., 2013). Second, we employ a propensity score matching (PSM) approach. We apply PSM to

identify control firms with male CEOs, which are otherwise indistinguishable from our sample

firms with female CEOs. The results consistently suggest that a significant difference in

insolvency risk exists between the two groups. Insurers with female CEOs have significantly

lower insolvency risks than the matched control group with male counterparts. This finding

confirms our baseline results. We further apply the entropy balancing matching method and

obtain consistent results (Hainmueller, 2012).

Third, the concern about endogeneity is somewhat further tempered by our difference‐in‐

differences (DiD) empirical framework as a robustness test. We construct a sample consisting of

insurers experiencing CEO turnovers during our sample period. In particular, our treatment

group is male‐to‐female transitions and the control group is male‐to‐male transitions, following

Huang and Kisgen (2013). Our DiD analyses confirm the baseline results. Specifically, we

observe a decline in insurer risk when a female CEO replaces a male manager. This finding

holds for the full sample of turnover events as well as a subsample that excludes nonroutine

turnovers in which CEO turnover may coincide with possible confounding factors such as poor

performance or excessive risk‐taking.

We further address endogeneity concerns empirically by employing an instrumental

variable (IV) approach. For the instrument, we use the commonness of female‐led insurers in

the home state where an insurer is located (Palvia et al., 2015). We presume that the

commonness of female‐led insurers should be positively associated with the likelihood of

observing an individual insurer with a female CEO. Furthermore, this state‐level variable is a

suitable instrument because it is uncorrelated with our insolvency risk measures and arguably

has no conceptual relation to insolvency risk measures of individual insurers. As expected, we

find that the instrument is positively and significantly correlated with the appointment of

female CEOs. The validity of the instrument is confirmed via the Cragg–Donald's Wald Fweak‐

instrument test statistic. Importantly, the two‐stage least squares regression results confirm the

aforementioned impact of female CEO on insurer insolvency risk. Therefore, we find consistent

results after mitigating the issue of endogeneity.

This study makes at least three contributions to the literature. First, our study contributes to the

gender literature in an important component of the finance service industry, that is, the insurance

industry. The insurance industry undoubtedly plays a significant role in the US financial sector and

economy. How insurers manage risks is associated with insurers' financial stability and their

obligation to policyholders, yet prior research on female leadership in the financial service industry

almost exclusively focuses on banks and yieldsmixedresults(Adams&Mehran,2012;Berger

et al., 2014; García‐Meca et al., 2015;Owen&Temesvary,2018;Pathan&Faff,2013). Our study is the

first to use a combination of techniques (including PSM and IV regressions based on external

instrument) to examine the role of female CEOs with respect to three different risk proxies in

insurance: the probability of insurer insolvency, z‐score, and the standard deviation of ROA. Our

research advances the understanding of how female leadership shapes firm risk for financial

institutions.

LI and CHENG

|

943