Gender and risk taking

Utilizing literature and concepts about risk appetite and decision making, the paper evaluates results of the Simmons Gender and Risk Survey database of 661 female managers.. Findings –

Gender and risk: women, risk taking and risk aversion

Sylvia Maxfield and Mary Shapiro Simmons College, Boston, Massachusetts, USA

Vipin Gupta California State University, San Bernardino, California, USA, and

Susan Hass Simmons College, Boston, Massachusetts, USA

Abstract

Purpose – Labeling women as risk-averse limits the positive benefits both women and organizations can gain from their risk taking The purpose of this paper is to explore women’s risk taking and reasons for stereotype persistence in order to inform human resource practice and women’s career development.

Design/methodology/approach – The paper draws on literature about gender and organizations

to identify reasons for the persisting stereotype of women’s risk aversion Utilizing literature and concepts about risk appetite and decision making, the paper evaluates results of the Simmons Gender and Risk Survey database of 661 female managers.

Findings – The paper finds evidence of gender neutrality in risk propensity and decision making in specific managerial contexts other than portfolio allocation.

Research limitations/implications – More in-depth research is needed to explore the gender-neutral motivators of risk decision making and to explore risk taking in a more diverse sample population.

Practical implications – The paper explores why women’s risk taking remains invisible even as they take risks and offers suggestions on how women and organizations may benefit from their risk-taking activities.

Originality/value – The paper synthesizes evidence on risk taking and gender, and the evidence of female risk taking is an important antidote to persisting stereotypes The paper outlines reasons for this stereotype persistence and implications for human resource development.

Keywords Risk management, Women, Gender, Mentoring Paper type Research paper

Media reporting on business and the economy is replete with language illustrating common stereotypes about gender and risk Depicting Wall Street’s unchecked and excessive risk taking as the culprit for 2008s global financial meltdown, brokers are

“credit default swap cowboys” (Mortgenson, 2008) and “capitalist buccaneers,” and the entire Street is “the Wild West” with “too much testosterone” (Syed, 2008) On the other hand, speculation abounds whether the financial crisis could have been averted had more women been the decision makers, both domestically and globally A financial services sector chief executive officer (CEO) says, “women [ .] have a greater desire to build firm foundations that will endure” (de Vita, 2008) In October 2008, Iceland, after declaring bankruptcy, turned to two women to rebuild its financial system

“after the banking empire built by its young male business-schooled elite collapsed.”

www.emeraldinsight.com/1754-2413.htm

GM

25,7

586

Received 1 April 2010

Revised 25 May 2010

Accepted 24 June 2010

Gender in Management: An

International Journal

Vol 25 No 7, 2010

pp 586-604

q Emerald Group Publishing Limited

1754-2413

A government official noted: “Now, the women are taking over [ .] to clean it up”

(O’Conner, 2008) A French business school study reinforces this view of women in

finance as risk averse and therefore better able than men to “safely” husband financial

resources (Ferrary, 2009)

Both the gendered language used to describe the current risk-induced financial chaos,

and the claims that a greater presence of women would have prevented the crisis reveal a

pervasive and enduring conflation between risk taking and masculinity: men take risks

and women do not Women are risk averse and, if at all, take only tiny, incremental risks

(Slovic, 1966; Wagner, 2001) While this stereotype is socially constructed, it may also

have a biological basis (White et al., 2007; Coates and Herbert, 2008) Yet, we are

interested in gendered behavior around risk and follow the view that gender is more

complicated than sex because it reflects psychological and social considerations as well

as biological factors (Meier-Pesti and Goetz, 2006) We also focus on social and

psychological aspects of risk decision making because these are amenable to change

Against the public discourse on women, men, and risk taking, our study builds on

research about gender bias in risk propensity in specific managerial contexts other

than household or professional financial portfolio allocation Existing research on risk

and gender covers three different general arenas for risk behavior: finance-investment,

gambling, or insurance; physical health and safety; and managerial contexts such as

entrepreneurial or “intrapreneurial” endeavors This paper uses Simmons 2008

“Gender and Risk” database developed from the responses of 661 professional women

to a unique survey designed to capture women’s approach to risk in specific concrete

workplace circumstances The survey specifically invoked career and business

development situations with which respondents were very familiar

Our primary empirical research goal in this paper is to examine evidence about

gender and risk decision making in concrete managerial contexts Analysis of the survey

database reveals that women take risks (Gupta et al., 2009), challenging the common

stereotype and adding to the evidence of gender neutrality in risk propensity in

professional management contexts other than portfolio allocation (He et al., 2007;

Holt and Laury, 2002) In this paper, we explore evidence of women’s risk taking that

suggests reality is more complex than the common stereotype Our research is important

because the predominant stereotype about gender and risk can subtly reinforce

organizational practices that hinder organizational development and women’s careers

We discuss why women’s risk taking remains invisible even as they take risks, and offer

suggestions on how women and organizations may benefit from their

risk-taking activities

Literature review

Existing empirical research on gender and risk concentrates on behavior in three

arenas: health and physical safety; finance including investment, gambling, and

insurance; and strategic decision making in a professional work context Research

suggests that women are clearly more risk averse than men in the arena of physical

health and safety (Hersch, 1996; Barksy et al., 1997; Pacula, 1997; Harrant and Vaillant,

2008; Harris et al., 2006) Results about gender differences in risk taking in the financial

arena (Croson and Gneezy, 2009; Eckel and Grossman, 2003) are more nuanced

A cumulative view of the research suggests that women’s decision making about

risk in investment and insurance decisions is context sensitive in ways different from

Gender and risk

587

men (Badunenko et al., 2009; Hibbert et al., 2008; Beckman and Menkhoff, 2008; Eckel and Grossman, 2008; Charness and Gneezy, 2007; Harrtell, 2007; Fellner and Maciejovsky, 2007; He et al., 2007; Iqbal et al., 2006; Meier-Pesti and Goetze, 2006; Holt and Laury, 2002; Dohmen et al., 2005; Sunden and Surette, 1998; Powell and Ansic, 1998; Jianakoplos and Bernasek, 1998) A dominant finding is that women are more risk sensitive in the context of losses than gains (He et al., 2007) Evidence suggests that this

is because women estimate the probability of gains and losses differently than men (Fehr-Duda et al., 2006; Olsen and Cox, 2001) If the context is insurance against loss, research suggests that men and women’s decision making is similar (Schubert et al.,

1999, 2000) When the context is investment, there is more evidence of gender difference (Hibbert, 2008; Beckman and Menkhoff, 2008) Gender gaps in risk propensity may be sensitive to national or regional cultures Studies of household investment behavior in Europe find no gender difference (Badunenko et al., 2009; Harrtell, 2007; Zinkhan and Karande, 1991) in contrast with several investigations using US population samples (Hibbert et al., 2008; Jianakoplos and Bernasek, 1998; Riley and Chow, 1992) A study of CEOs that awarded additional options to purchase stock in their companies (Iqbal et al., 2006) finds no gender difference in risk propensity If risk averse, women would diversify by selling some of their previously awarded stock but this is not the case In this context, as CEOs, women are highly familiar with the risk environment and perceive that they have some degree of control over future outcomes

Research focused on risk propensity in specific managerial contexts rather than in allocation of investment portfolios, insurance, or simulated gambling situations finds much less, if any, evidence of gender bias In field studies of entrepreneurs (Corman, 2001; Masters and Meier, 1988) and managerial strategy situations (Castillo and Cross, 2008; Johnson and Powell, 1993; Riley and Chow, 1992), gender does not appear to have

a strong impact on risk propensity Building on this stream of research, we sought to explore two questions First, building on previous literature (Iqbal et al., 2006), we wanted to explore risk propensity in concrete managerial contexts, rather than experimental settings typical of research on risk taking in financial resource allocation Second, we sought to draw on literature about gender in organizations to identify reasons for stereotype persistence in the face of evidence that women’s risk decision making in managerial settings may not be so different from men’s For the analysis of risk taking and risk decision making, we used the Simmons 2008 Gender and Risk Survey and database described below

Simmons Gender and Risk Survey and database Simmons 2008 Gender and Risk database is based on a unique survey completed by managerial women attending a women’s leadership conference in the Northeastern USA (Gupta et al., 2009) The survey built on the finding that levels of gender differential in risk propensities vary between studies based on abstract versus concrete managerial settings (Corman, 2001; Masters and Meier, 1988; Castillo and Cross, 2008; Johnson and Powell, 1993; Riley and Chow, 1992) It followed the technique of cueing specific work contexts in cross-cultural research on organizational behavior (House et al., 2004) Accordingly, the survey specifically invoked career and business development situations with which respondents were very familiar

The 34-question survey was divided into four parts: one hypothetical financial allocation question, five questions asking respondents to reflect generally on risk taking,

GM

25,7

588

13 questions asking them to reflect on one specific risk decision, and 14 demographic

questions The Simmons Gender and Risk database was constructed by hosting this

survey online, ensuring anonymity, accessible on 30 laptops located in a large

conference hall The computers’ proximity to refreshments, sponsor booths, and book

signing tables was selected to assure ease of access when participants took a break from

conference proceedings throughout the day Additionally, at multiple times during the

daylong conference, announcements were made encouraging women to take the survey

The final number of respondents was impacted by the finite number of computers, the

wait time of up to 7 minutes and the limited time allotted to official conference breaks

Of the 2,480 participants attending the conference, 661 fully completed the survey,

yielding a 26.6 percent response rate The conference attendees who completed this

survey were employed full-time (92 percent) and had significant work experience:

36 percent had 11-21 years and 41 percent more than 20 years With 46 percent having

received five or more promotions over their careers, 30 percent were now in supervisory

level positions, 44 percent were employed in middle management (director, middle

manager, and assistant vice president), and 13 percent were in more senior positions

About 61 percent of this well-educated group, where 46 percent held an advanced college

degree, had compensation over $100,000 in 2007 and 88 percent contributed 50 percent or

more of their household’s income Most respondents were married or living in a

committed relationship (67 percent) and 80 percent identified themselves as Caucasian

Additionally, 57 percent were working in organizations employing over 10,000 workers,

which reflected the composition of the organizations sponsoring and sending women to

this conference Another impact of using this conference as a convenience sample was

that a majority of the respondents worked in the major industries of the northeastern

USA, namely finance, technology, and health care Although the sample is not

representative of the US workforce, its composition is comparable with those of other

scholars studying the careers of managerial women (Hewlett and Buck Luce, 2005) and

their risk propensities (Corman, 2001)

Measures

Traditionally, scholars have measured risk taking by asking individuals to allocate

financial resources across options with varying levels of financial returns and probable

success The more money allocated to options with higher possible returns and lower

probability outcomes, the “riskier” the behavior (Prince, 1993; Schubert et al., 1999)

These studies find that women tend to be risk averse and a similar question was

included as a baseline in the gender and risk survey

The survey also asked women about the business/professional opportunities they

had actually taken (Table I) The survey intentionally did not label these opportunities

“risky,” but specified them as situations where the respondent had to invest resources

under conditions of uncertainty Respondents were asked to reflect on

“business/professional opportunities, whose success is not assured, that require

learning by doing, and where (you) have to take personal responsibility for failures on

the way.” In this way, each choice involved the investment of personal capital and

carried an unknown outcome for both the business and the personal/career development

Each opportunity was rated on a four-point Likert scale from “never” to “often” (Table I)

To explore women’s decision process in accepting or rejecting these risks, the survey

also asked the respondents’ questions to illicit information about their motivations

Gender and risk

589

and their patterns of consultation The survey asked respondents to rate their level of agreement on each of 12 potential career goals (Schein, 1990; Waldrop and Butler, 2000; Shapiro et al., 2009) on a five-point Likert scale of “strongly disagree” to “strongly agree.” The survey asked respondents to rate eight professional/non-professional sources of potential input into their decision making on a five-point Likert scale from “not at all significant” to “extremely significant.”

Analyses The Simmons Gender and Risk database was analyzed using SPSS 17.0 Frequencies were initially used to analyze results Thereafter, each of the risk-taking opportunities was regressed on factors associated with risk propensity in psychological and sociological literature, using the ordinal regression method

Results and discussion Women’s risk propensity in hypothetical financial allocations The survey asked a version of the traditional financial portfolio allocation question and generated results that confirm the conventional findings of gender-biased risk aversion (Figure 1) The question, what percentage of a fixed sum of dollars would you invest in a new project with unlimited potential reward but high outcome uncertainty, replicates a common approach to studying risk using hypothetical scenarios Approximately,

1 percent in our sample chose to re-allocate the entire budget and two-thirds re-allocated , 50 percent of the dollars available, earning the moniker of “risk averse.” We included this question as a baseline check that indicates the women in our sample responded as anticipated based on the findings from the hypothetical scenario-based literature on gender and risk

Women’s risk propensity in specific managerial contexts When asked to reflect on how frequently they had actually taken on certain business/professional opportunities, women in the survey reported embracing risk:

80 percent reported “sometimes” or “often” pursuing a major change initiative,

79 percent reported “sometimes” or “often” pursuing a major new program, 77 percent reported “sometimes” or “often” pursuing a new job, and 56 percent reported

Types of opportunities

Percentage of respondents indicating they generally take on this type of opportunity – indicating

sometimes or oftena

Major investment decision for business 40 Major business development opportunity 56

Major business relationship 62

Note:an ¼ 641-647

Table I.

Types of opportunities

pursued by respondents

GM

25,7

590

“sometimes” or “often” pursuing a major business development opportunity (Table I).

These frequencies suggest higher levels of risk taking in broader managerial contexts

than in the narrower financial arenas which dominant scholarship on gender and risk

Factors impacting risk taking

Analyses confirmed that several factors included in the survey, such as power,

self-efficacy, and professional networks, impact women’s risk-taking behavior (Table II)

Power A pragmatic approach to the definition and use of power has become

pervasive in current American literature aimed at both male and female managers

(Haass, 1994; Kotter, 1985) Many scholars agree that the need for power arises from the

scarcity of resources, competing demands, and interdependence (Bolman and Deal, 1991,

1997, 2003; Zaleznik and Kets de Vries, 1975; Pfeffer, 1992) While power is neutrally

defined as “a desire to make an impact” (McClelland and Burnham, 2003, p 120;

Merrill-Sands et al., 2005), it is conflated with masculinity Groshev (2002, p 6) conducted

an extensive multi-disciplinary, cross-cultural literature review on the definition

of power and concluded: “the discourse that produced the various definitions of power

[ .] has clearly been masculinized, so the terms are more relevant to men.” For the last

30 years, feminist scholars have sought to examine power from a woman’s experience

Rejecting the classic male definition of “power over,” feminists’ conceptions include

“power to” which is about empowerment (Kitzinger, 1991), and specifically about

sharing power and enhancing “other’s feelings of competence” (Miller and Cummins,

1992, pp 416-17), and “power with” (Fletcher, 2004; Guinier, 1998)

The analysis demonstrated that (Table II) the desire for influence, defined in the

survey as “the power to make an impact,” strongly motivates women to take risks

Figure 1 Percentage of a fixed sum budget that the women will allocate to a new project that may yield up

to ten times the normal returns and has a zero probability of losing the capital investment

% of budget allocated to the new project

0 1–24 25–49 50–74 75–99 100

40

30

20

10

0

40%

6%

1%

1%

Gender and risk

591

Professional networks

2 (%)

* p,

Table II.

Ordinal regression of

risk-taking opportunities

on gender-neutral factors

GM

25,7

592

The stronger the desire for power, the more likely a woman will take on the risky

opportunities This behavior is consistent with sex-blind studies where power was

found to be an organizational necessity (Gandz and Murray, 1980) and that the desire

for power correlates with risk taking (Anderson and Galinsky, 2006; Gaubatz, 2008)

Self-efficacy Self-efficacy, a person’s belief in her or his ability to succeed, is a

concept integral to the dominant social cognitive theory framework used for research

and practice in education, communication, and related fields (Bandura, 1986)

Self-efficacy belief involves an affirmation of capability and strength of belief

(Bandura, 1997) Organizational research finds that self-efficacy is related to individual

performance in organizational settings (Gist, 1987; Lee and Gillen, 1989)

Research also shows that self-efficacy is positively correlated with aspects of

individual performance in organizations that are closely associated with risk taking as

defined in our study High self-efficacy predicts innovative role orientations taken up by

individuals in their first months working in a new organization ( Jones, 1986)

In organizational settings, self-efficacy also determines goal levels individuals set for

themselves and their career success (Bandura, 1982; Ballout, 2009) These studies draw

on mixed-sex populations but are not explicitly designed to test for gender differences

Research oriented toward exploring possible gender differences in the impact of

self-efficacy on individual choices and performance in new venture settings and within

established organizations finds little evidence of gender difference (Mueller and Conway

Dato-On, 2008; Dietz et al., 2003; Chung, 2002)

The analysis reveals that self-efficacy strongly predicts risk taking by women

(Table II) We believe that this shows that women’s decision making about risk is

consistent with “typical” patterns evident sex-blind studies showing the strong impact

of self-efficacy on risk taking (Llewellyn et al., 2008)

Networks Consistent with research on gender and moral decision making (Gilligan,

1981), research on risk decision making suggests that women may contextualize more or

in different ways than men The more women are familiar with the decision context, the

more risk loving they will be (Levin et al., 1988; Johnson and Powell, 1994) This result

may also reflect another claim in the existing literature on gender and risk that women

tend to focus on future consequences while men focus on probabilities (Yates and Stone,

1992) As part of that contextualization, we anticipated that women would report

consulting about risk taking at work in social and informational support networks

A recent study of gender differences among firm-hopping financial managers

(Groysberg, 2008) illustrates women’s context-sensitivity The study finds that women

are more likely to succeed in their new organizations than men because they engage in

more extensive information gathering about the new firm before jumping

A gendered frame might suggest that women would seek advice and support among

friends and family, perhaps more readily than among professional networks, and that

this might be related to the perception/reality of work-family conflict (Cabrera, 2007)

Research on gender and careers finds that women do not generally engage with

professional networks as much as men (Rothstein and Davey, 1995) Human resources

practitioners and career coaches advocate that women network seek information

support within their professional community as much as possible because research

indicates that mentoring and networking improve career success for women (Linehan,

2001; Gersick et al., 2000; Ibarra, 1992; Seiger and Wiese, 2009)

Gender and risk

593

The gender and risk survey included a question about the networks’ women respondents used in the process of deciding to accept or decline risky professional opportunities The analysis reveals that women seek the advice of both family and professional networks in their decision making about risks (Table II) In analysis not reported here (Gupta et al., 2010), there is evidence that women who take up risky opportunities involving new competencies are more likely to consult professional networks than family networks This finding is consistent with literature on the power

of networking and mentorship for women’s career advancement

Explaining stereotype persistence and implications for practitioners Despite strong support in the survey findings that women embrace risk, women are still viewed as risk averse in the business world In this section, we first explore three explanations for this discrepancy Fundamental to each explanation is either the alignment with or violation of role congruency, specifically, “the extent to which men and women behave in a manner that is consistent with the sex role stereotypes society holds for them and socially prescribed expectations for both genders” (Klenke, 2003,

p 1029) We then suggest what individuals and organizations can do to benefit from women’s risk taking

Why are women still seen as risk averse?

Unrecognized risk taking Regarding invisibility, society may not see women taking risks because American culture does not expect them to take risks As humans, we tend

to see what we look for, as a means of confirming our beliefs (Hitt et al., 2009), and conversely, we miss what we do not look for, through inattentional blindness (Simons and Chabris, 1999) Risk taking may even be invisible to the female risk taker herself (Moreschi, 2005) Additionally, two-role-congruent behaviors around claiming risk success may render risk taking invisible

First, women do not seek the conventional visible measures of successful risk taking found in most organizations, namely promotions to high-status positions (Shapiro et al., 2009) Second, a lack of self-promotion due women’s gendered socialization to “not brag” about themselves may further hide their successful risk taking (Tannen, 1994) Even when a woman does talk about the risks she has taken, she may do so using the word

“we” to refer to her own accomplishments, thus redirecting attribution for the risk taking elsewhere (Tannen, 1995)

Cost mitigation Women may undertake strategies to keep their risks hidden in an effort to mitigate the costs arising from “doing gender” and hyper-visibility West and Zimmerman (2003, p 63) frame gender as “the activity of managing situated conduct in light of normative conceptions [ .] for one’s sex category.” If a woman chooses to violate normative expectations through role incongruence, she may be penalized Without a gender-neutral definition of risk, women are thus left with the classic “double bind:” act like a woman, and be called risk averse, or act like a man, and have one’s character, motives and predispositions called into question Carly Fiorina, in her role as CEO of Hewlett-Packard, is a case in point: she was accused of “acting like a rock star” (Karlgaard, 2005) and being haughty and cocky (Lashinsky, 2002)

Women may also want to manage their hyper-visibility, the phenomena of incurring intense scrutiny as a result of being in the minority (Roberts et al., 2008) With women still accounting for only 3.0 percent of Fortune 500 CEOs and 15.7 percent of corporate

GM

25,7

594

officer seats in 2008 (Catalyst, 2009), any risky decision undertaken by one of those

few women is subject to increased analyses and opinion Without sufficient numbers,

where one’s identity is no longer seen as an anomaly, women’s decisions occur in

a white-hot spotlight

Role congruency Finally, women may engage in additional role-congruent behavior

that leads to the perception that they are risk averse As our respondents tap into their

professional networks and seek advice, we can reasonably assume that they are asking

questions while doing so Observers, looking at this advice-seeking through a gendered

lens, may see asking questions as revealing uncertainty or indecisiveness; it may make

women appear afraid or incapable of making decisions alone (Tannen, 1994) While a

woman may be collecting ideas or testing her own by asking, “What do you think

I should do?” some listeners may see the question as a literal invitation to make the

decision for her She is no longer a risk taker She is risk averse

What can individuals, organizations, and educators do?

While many female role-congruent concerns and behaviors may obfuscate women’s risk

taking, increasing research shows that organizations benefit from those behaviors

Female role-congruent behaviors, such as inclusive decision making and collaboration,

are cited as the reasons women are often asked to take on the high-risk situations of

turning around organizational crises and dysfunction Iceland was not unique in tapping

women for help in a crisis: Anne Mulcahy was tapped to be CEO of xerox when the

company was facing bankruptcy, saddled with $17 billion in debt, an Securities and

Exchange Commission scandal, and five consecutive quarters of losses (George, 2008;

Tsiantar, 2006) Brenda Barnes took over the top spot at Sara Lee in time to divest

multiple companies to make a smaller and more profitable company (Warner, 2005)

Termed the “glass cliff” in recognition of the high-risk nature of the opportunity,

Ryan and Haslam (2005, 2007) found that in times of positive performance, organizations

“think manager, think men.” Yet in times of high risk, organizations “think crisis, think

women” and often put women into top leadership, relying on their relational skills to turn

organizations around While the organization benefits from her relational leadership, the

woman herself faces a “glass cliff” in that she assumes leadership right at the time she is

at greatest risk of failure due to the calamity of the situation As evidence of the

vulnerability of precarious leadership positions, Hewlett (2008) points to the demotions

and dismissals of Erin Callan of Lehman Brothers, Zoe Cruz of Morgan Stanley, and

Sallie Krawcheck of Citi Positive and negative association with role-congruent

behaviors can result in an “under-gendered” risk stereotype where women are denied

risky opportunities, or the glass cliff phenomenon results in an “over-gendered” risk

stereotype where women are expected to have special powers to navigate crises

The challenge for women might not be how to take risks, as they already do so

Instead, it may be making those risks visible and capturing the credit for risk taking in

ways that signal their success to those around them Why? Being acknowledged as a

risk taker is essential for a woman’s professional growth Kouzes and Posner (2002)

link risk taking and leadership In their study of thousands of professionals, they

identified 30 behaviors essential in the exercise of leadership, including taking risks,

encouraging others to take risks, and learning from failed risks Indeed, Johnson and

Powell (1994) propose that taking less risks may be a major cause of “glass ceilings” in

corporate promotion ladders Women may want to enhance their career potential by:

Gender and risk

595