Results
Table 3 shows the results for financial literacy. Men tended to be more financially literate than women (β: -0.586, standard error [SE]: 0.095); however, the tendencies between men and women were similar: age was associated with an increase in financial literacy (Men, β: 0.249, SE: 0.030; Women, 0.354, SE: 0.026) but the growth rate turned downward among the oldest respondents (Men, β: -0.002, SE: 0.000; Women, -0.003, SE: 0.000).
Table. Estimation results for financial literacy by OLS. (See the original article for the full table.)
Table 4 presents the results of the self-rated financial literacy. While we found similar results to those for financial literacy, among men, the association with age at the tail end moved in the opposite direction (Age, β: -0.021, SE: 0.009, Age2, β: 0.000, SE: 0.000), namely, self-rated financial knowledge increased the most among the older male respondents.
Table. Estimation results for self-rated financial knowledge by ordered logistic regression. (See the original article for the full table.)
Fig 1 represents the association of age and financial literacy and age and self-rated financial knowledge using the previous estimations. The financial literacy trend shows an increase with age, which peaks when both men and women are in their early 60s. However, self-rated financial literacy, particularly for men, appears the lowest around age 70 and then begins to increase among those older than that, in contrast to the trend for the measured financial literacy.
Figure. Age, financial literacy, and self-rated financial knowledge.This figure presents the relationship of age with financial literacy and self-rated financial knowledge based on the estimation results in Table 2 for financial literacy. For self-rated financial literacy, we obtain marginal effects of age and the square of age from ordered logistic regression with the outcomes of self-rated financial knowledge in a three-point Likert scale (very low/low, fair, and high/very high) for presentation simplicity of the results. The lines show point estimates, with shaded areas representing 95% confidence intervals, estimated by robust standard errors. (See the original article for the figure.)
Gender gaps in financial literacy
The results indicate that men are more financially literate than women, with the coefficient of the female dummy variable significant and negative (Table 1). Thus, the hypothesis that there are gender gaps in financial literacy is supported.
Table 5 lists the result for the decomposition analysis of the gender gaps in financial literacy. The observed gender gaps are explained by the differences in the distribution of the explanatory variables (64.1%), the responses for the explanatory variables (53.7%), and their interactions (-17.8%). Among endowment effects, which account for the largest contribution among the three different effects, women’s lower financial literacy is explained mainly by the smaller number of women who graduated with a university degree or higher (85.1%), followed by age, occupation as a part-time worker, receiving financial education, and income.
Table. Decomposition of gender differences in financial literacy. (See the original article for the full table.)
Financial behaviours and attitudes
Table 6 shows the results for financial behaviour. Women tend to reflect more preferable financially behaviour than men (β: 0.329, standard error [SE]: 0.040); however, the tendencies between men and women are similar: age is associated with a decrease in preferable financial behaviour (Men, β: -0.045, SE: 0.013; Women, -0.025, SE: 0.012) but the growth rate increases among the older respondents (Men, β: 0.001, SE: 0.000; Women, 0.000, SE: 0.000).
Table. Estimation results for financial behaviour by OLS. (See the original article for the full table.)
Table 7 presents the results for financial attitude. Women are less likely than men to place a larger value on present benefits (β: 0.639, standard error [SE]: 0.039). The association with age is found only in women (Age, β: 0.035, SE: 0.012; Age2, -0.001, SE: 0.000).
Table. Estimation results for financial attitude by OLS. (See the original article for the full table.)
Fig 2 represents the association of age with financial behaviour and attitude using the previous estimations. In contrast to financial literacy, the preferable financial behaviour trend shows a U-shaped curve with the lowest value around age 45 and higher values among women. For financial attitude, it is associated with age only in women, and shows a decrease with age after around age 35.
Figure. Age, financial behaviour, and financial attitude.This figure presents the relationship of age with financial behaviour and attitude based on the estimation results in Tables 6 and 7. The lines show point estimates, with shaded areas representing 95% confidence intervals estimated by robust standard errors. (See the original article for the figure.)
Gender gaps in financial behaviours and attitudes
As shown in the results in Tables 6 and 7, we find that financial behaviour and attitude vary between men and women: In contrast to financial literacy measured by financial knowledge, women tend to reflect more preferable financial behaviours and attitudes than men. Consequently, we conducted decomposition analyses for these variables.
Table 8 lists the result for the decomposition analysis of the gender gaps in financial behaviour. The observed gender gaps are explained by the differences in the responses for the explanatory variables (157.4%), and their interactions (-34.6%) while the overall endowment effects are not significant. Therefore, the unexplained parts are the major reasons for the gender difference. Among endowments effects, financial knowledge and employment status largely account for the gender difference.
Table. Decomposition of gender differences in financial behaviour. (See the original article for the full table.)
Table 9 lists the result for the decomposition analysis of the gender gaps in financial literacy. In line with financial attitude, the unexplained parts (i.e. coefficients effects and interaction effects) largely account for the gender difference.
Table. Decomposition of gender differences in financial attitude. (See the original article for the full table.)
Robustness check
To check the robustness of our findings, we conduct additional analyses using a different measurement of financial literacy. Here, we define financial literacy by the ’Big 3’ questions [21] (i.e. No. 8, 10, and 14 in Table 1), which address numeracy, understanding of inflation, and knowledge about stocks and risk diversification. Although we find similar results to those from our more detailed measurements, the contribution of the differences in the distribution of the explanatory variables to gender gap in financial literacy is smaller, which may suggest that a small variation in the measurement based on only three questions may generate a measurement error (S1 and S2 Tables).
CC BY 4.0. "Age, gender, and financial literacy in Japan" by Okamoto S, Komamura K, PloS one (2021). https://doi.org/10.1371/journal.pone.0259393. Licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Peer-reviewed and open access; full text from PubMed Central (PMC8598074). Headings and layout adapted for reading; figures and tables are in the original.