Men and women respond differently to financial risk depending on their levels of stress and happiness.
How we behave when it comes to risk is widely believed to differ based on gender. Women, they say, are less likely to take a risk, while men are great risk-takers.
But are they?
A study from Monash Business School suggests that that risk-taking for both genders is much more complicated and has a stronger emotional element than researchers have previously acknowledged.
Professor Philip Grossman and co-authors Anwesha Bandyopadhyay (University of Lincoln, United Kingdom) and Lutfunnahar Begum (Ministry of Finance Bangladesh) looked at whether risk-taking behaviour changes over time.
In other words, over a longer period of time, are women still likely to be more risk-averse?
“Risk attitude across stakes, time, and context has been looked at in a number of studies,” Professor Grossman says.
“Less studied is whether or not there is a gender difference in the stability of risk attitudes.”
Understanding risk and gender
A better understanding of the differences between men’s and women’s risk attitudes can be important for many reasons.
“Women, stereotyped as more risk-averse, may be less likely to be hired into positions in which a willingness to make risky decisions is a desirable attribute,” Professor Grossman says.
And, even if hired, may be less likely to be given promotions.
“Women may face more aggressive negotiations and receive less generous salary offers than men,” he says.
Risk experiment shows gender differences over time
To address the question of whether a gender difference in the stability of risk preferences exists, the researchers designed a unique experiment asking participants to play an online ‘lottery’ made up of two low-stakes lottery choices and one high-stakes lottery choice.
This was followed up with a weekly survey for 12 weeks.
According to Professor Grossman, in the study, the expected winnings in the two low-stakes lotteries were $20-$30. The expected winnings in the higher-stakes lottery were ten times this amount – $200 – $300.
“Our study incorporates self-reported measures of the states of mind of the participants – specifically sources and levels of stress and happiness, as well as financial well-being – to explain the weekly lottery choices of males and females,” Professor Grossman says.
“Understanding their state of mind and levels of financial well-being and happiness was an important part of the process,” Professor Grossman says.
“We wanted to understand if lottery choices vary over time and if choices are correlated with participants’ levels of stress, happiness and financial well-being,”
“At the aggregate level, we did not observe a significant gender difference in average risk attitudes either within or across stakes.”
For both genders, risk tolerance for low-stakes lotteries increases over time; for high-stakes lotteries, only females became more risk-tolerant.
Risk attitudes are not stable
The results from their study have important implications.
“Economists assume that risk attitude is an inherent behavioural trait. And while it doesn’t change a lot, there are important differences, particularly between the genders,” he says.
The results suggest that assessing risk preferences from just one or two decisions in an unfamiliar environment and not knowing the financial and emotional states of the decision-makers is problematic.
“Evidence from week 1, when the environment is new, indicates a gender difference in average risk attitudes,” Professor Grossman says.
“However, over the full 12 weeks, we found, at most, weak evidence of a difference.”
He explains that while week by week, males on average typically make riskier choices than females for all three lotteries, the difference was seldom significant.
What do we mean by stability?
The genders did respond differently over time, particularly in relation to stress.
For males, there was a decrease in the instability (variability) of choices (at both stake levels) as time passes.
Professor Grossman explains that this is consistent with the participants better understanding or becoming more comfortable with the choices they must make.
“It is important to clarify exactly what is meant by stability,” he says.
Holding the stakes constant, stability of risk preferences could mean a constant level of risk tolerance, unchanging regardless of lived experiences.
Alternatively, he says, it could indicate a constant underlying level of risk tolerance, but this tolerance may increase or decrease, depending on what the person is currently experiencing.
And, as the experience fades with time, risk tolerance reverts to the underlying level; a reversion to the mean type process.
“Risk tolerance can vary with a subject’s emotional and financial state and that stress affects the two genders differently.
At both stake levels, males are positively correlated with family-sourced stress.
“For females, experiencing university-sourced stress and relationship-sourced happiness are correlated with increases in risk aversion for the high stakes task.,” Professor Grossman says.
The results suggest that for males, stress is something to conquer, while for females, stress is something to avoid.
They also found that, for females, a better state of financial well-being is positively correlated with increased risk tolerance for the high-stakes task.
Does financial risk taking depend on the size of the stakes?
Comparing risk tolerance across stakes, stability might imply risk preferences that are independent of stake sizes, or it could imply that risk tolerance is constant, but stake-dependent.
Again, with respect to the stability of risk preferences across stake levels, the researchers found weak evidence of a gender difference.
“Week by week, our female (male) subjects were, on average, less risk-tolerant at higher stakes but more risk-tolerant at lower stakes, but insignificantly so,” Professor Grossman says.
“Focusing on individual choices, we find a high degree of stake independence stability. Roughly a third of our sample, exhibited either risk tolerance that is decreasing, constant, or increasing with stakes, regardless of gender.”
Over time, the risk tolerance of males for high stakes relative to low stakes decreases; females exhibit no significant change in relative risk tolerance.
This suggests that males, out of excess confidence, may initially take on too much high-stakes risk, but with time they come to better understand what is at risk and start to scale back their risk-taking.
“Being more cautious, females may initially more carefully consider the risks and have less need to subsequently adjust their choices,” Professor Grossman says.