The Effect of gender and marital status on financial risk tolerance [abstract]
Metadata[+] Show full item record
Abstract only. Abstract: This study investigates the effects of marital status and gender on financial risk tolerance and provides implications for financial advice and education. Ho, Milevsky, and Robinson (1994) suggested that women should have riskier portfolios than men because they live longer. However, most studies analyzing financial risk tolerance by gender have found that women are less risk tolerant than men. Risk tolerance is important because it affects a household's portfolio decisions, which ultimately affect a household's wealth accumulation. Inappropriate levels of risk tolerance might lead to problems. If people behave according to a rational economic model and there are no systematic differences in risk aversion for men and women, then women should tolerate more financial risk because they live longer. Also, because married couples have longer life expectancies than unmarried people of the same age, married couples should tolerate more investment risk than unmarried people. Six Survey of Consumer Finances (SCF) datasets representing 18 years (1983-2001) were combined. Same sex couples and same sex partners that live together were excluded from this research. The total sample size used in the analyses was 24,037. A cross-tabulation of risk tolerance levels and gender/marital status provided percent distributions across the gender/marital status categories. A cumulative logistic method was used to test the hypotheses. The repeated-imputation inference (RII) method was used for statistical analyses. The SCF risk tolerance question includes four choices: substantial financial risk, above average financial risk, average financial risk, and no risk. The three dependent variables for the cumulative logistic model are: substantial risk, high risk (substantial and above average risk) and some risk (substantial, above average and average risk). Household types are: married/partnered female respondents, married/partnered male respondents, divorced females, divorced males, widows, widowers, never married females, and never married males. Never married males are most likely (70%) to be willing to take some risk, followed closely by married males (66%), then divorced males (61%), married females (55%), never married females (50%), divorced females (45%), widowers (38%) and widows (27%). Never married males are the most likely to be willing to take substantial risk (8%), and also to be willing to take high risk (29%), and widows are the least likely to be willing to take substantial risk (2%), and also to be willing to take high risk (5%). Of course, other factors such as age, income, and education might account for some of these differences. When controlling for the effect of other variables in the cumulative logistic model, female respondents are less likely to be willing to take some risk, high risk, and substantial risk than males of the same marital status. Never married males are about twice as likely as otherwise similar never married females to be willing to take some risk. Married male respondents are about 1.6 times as likely as otherwise similar married female respondents to be willing to take some risk, and there are similar differences in willingness to take some risk between male and female divorced respondents, and between male and female widowed respondents. Never married males are more likely to be willing to take some risk than other types of respondents, though the difference from divorced males is not significant. Widowers are more likely to be willing to take high risk than other types of respondents, though the differences from never married men and divorced me were not significant. Divorced males are more likely to be willing to take substantial risk than other types of respondents, though the differences from never married men and widowers were not significant. A divorced male respondent is 1.7 times as likely as an otherwise similar married male respondent to be willing to take substantial risk. No previous publication has analyzed differences in the SCF risk tolerance variable by the gender of the respondent in couple households, so previous studies have compared unmarried males, unmarried females, and married couples for which the sex of the respondent was not identified. This study is the first to differentiate married females and married males in terms of financial risk tolerance. Further research is needed to ascertain whether there are differences in preferences or understanding that cause the differences in the willingness to take financial risk. The male-female difference in the willingness to take some risk are problematic for women, given that normative portfolio analysis implies that they should be willing to take at least as much risk as men, even if they 124 are really more risk averse than men. It is plausible that the differences found in this research are related to women not understanding the nature of investment risk.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.