![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
The authors of this study investigate the likely influence of required or voluntary financial education on the saving behaviors of U.S. adults. They compare the results for three groups defined by different life experiences with financial education (required, voluntary, and none). Probit models estimate the effects of financial education on four saving behaviors: having a savings account; having an emergency fund; saving for investing; and saving for retirement. The results show similar positive outcomes for required and voluntary financial education on each saving behavior. No difference based on self-selection into financial education is evident. The findings also indicate that multiple exposures to financial education in different venues (high school, college, or employment) increase the apparent effects on saving behaviors compared with a single exposure.
Notes
1 Throughout the text, we use qualifiers such as “likely” or “apparent” when discussing the effects or influences of financial education on financial behaviors to emphasize that with correlational data, no claim is being made that the findings show a causal relationship. For another example of the use of such terms related to financial literacy, see Walstad and Allgood (Citation2022).
2 The Consumer Financial Protection Bureau (CFPB) reviewed some 19 research studies on youth financial education conducted since 2000 and concluded that the findings show that “financial education can improve financial knowledge and financial behaviors” (CFPB Citation2019, 2).
3 Publicly available data, tables, survey questions, methodology explanations, and reports for the 2009, 2012, 2015, and 2018 surveys can be found at usfinancialcapability.org.
4 Respondents that gave inconsistent or incomplete responses to the survey items on financial education were omitted for several reasons. Some respondents (n = 864) stated that financial education was required, but also stated they did not participate in financial education. Other respondents (n = 258) stated they participated in financial education but did not state when they participated (in high school, college, or employment).
5 The survey asks respondents who had not retired if they have tried to figure out how much money they need for retirement (J8). Those respondents who were retired were asked if they figured out how much money they needed for retirement prior to their retirement (J9). This study combined the responses from retirees and non-retirees.
6 State fixed effects may account for some state difference in financial education. No NFCS data are available, however, to directly connect a respondent with state policies for financial education in public schools that correspond to when they resided in the state. It also is not known if they attend a public school rather than a private school.
7 An alternative approach to investigate selection would be IV estimation, but a valid instrument for financial education was not available for the NFCS data. Even if available, IV instruments have problems when weak (Murray Citation2006).
8 The difference in the savings account result compared with the other results is perhaps due to more learning from life experience and less influence from financial education. In the A and C groups, the great majority have a savings account (78 percent). Far fewer report having an emergency fund (53 percent), figured retirement (46 percent), or have non-retirement investments (35 percent), indicating that they are less likely to have learned from life experiences.
9 The findings reported here and in the next subsection assume that the required and voluntary financial education received were similar. Although the NFCS data does not contain information for a full investigation of the assumption, data from four items provide some support. The required and voluntary groups show no major differences in these averages: hours of financial education (2.6 vs. 2.4 on a 1–3 scale) [M41]; perceived quality of financial education (5.5 vs. 5.4 on a 1–7 scale) [M42]; self-rating of financial knowledge (5.6 vs. 5.6 on a 1–7 scale) [M4]; and financial literacy score (3.2 vs 3.4 on a 0–5 scale [M6 to M10]. Also, content guidelines that influence instruction make no distinction between required or voluntary financial education (e.g., CEE Citation2021). Nevertheless, the lack of data to assess this assumption fully is a valid concern.
10 The item also includes the military as a response category, but few adults gave that response, so it is included in the employment category, as being in the military is a type of employment.
11 For subgroup A members with multiple exposures, at least one of the multiple exposures was required as a yes response given to M40 (“were you ever required to take financial education?”). It is not known if an additional exposure also was required given the limitations of the survey. That ambiguity, however, does not apply to subgroup B with multiple exposures as all members gave a no reply to M40 that would apply to all cited exposures.
12 In the single exposure A and B comparison, only having an emergency fund was significantly different (5.2 percentage points). This difference may well be an artifact of the reduced sample sizes of the A and B subgroups or differences in the instructional content received.
13 Although this section focuses on alternative estimation, it should be remembered that the inclusion of four saving behaviors as outcomes also serves as a robustness check on the consistency of the results across different types of saving behaviors.
14 The tables reporting the estimations without demographics and estimations with LPM are available upon request from the authors.
15 Splitting the financial education sample into subgroups for further analysis, as was done in the previous section, does not appear to affect these basis results.