Financial Security Among Workers Eligible for Automated Savings Programs
Feelings of security are tied to household circumstances, but not to participation in Illinois Secure Choice
![](/-/media/post-launch-images/2024/05/gettyimages1229089836jpgmaster/16x9_m.jpg?la=de&h=1024&w=1820&hash=9C7FFD14E6DA53BF2E0BD1C2AE4677C9)
Overview
Illinois launched an automated savings program, Illinois Secure Choice (ILSC), in 2018 to help workers in the state save for retirement. Employees without access to a workplace retirement plan are automatically enrolled in the program and contribute a preset percentage of their wages or salaries;1 they can opt out of the program, change the contribution percentage, or withdraw prior contributions at any time.2 Employees eligible for ILSC, and similar automated savings programs in other states, often earn low to moderate incomes, raising the question on how these programs affect perceptions of overall financial security.
To help The Pew Charitable Trusts learn more about workers’ experiences with ILSC, the Rand Corporation surveyed 1,100 adults eligible to participate in the program—workers enrolled in the program as well as those who opted out. The longitudinal study was conducted in three waves at roughly six-month intervals, with the first wave fielded in the spring of 2020, the second wave in the fall of 2020, and the third wave in the spring of 2021.3 This brief looks at Illinois workers’ perceptions of financial security in early survey waves and whether these perceptions changed across waves. It also examines whether financial security was linked to income and expense patterns, race and ethnicity, marital status, and other personal and household factors. Among the key findings:
- Many workers surveyed said they felt financially insecure, although their feelings of financial insecurity decreased during the course of the survey.
- The survey coincided with the onset of the COVID-19 pandemic. Over half—55%—of all workers eligible for ILSC, whether they participated or opted out, reported feeling financially insecure at the beginning of the pandemic in the spring of 2020, when many workers were laid off or furloughed.
- This figure dropped to 41% by the spring of 2021, coinciding with the U.S. economy’s rebound and the introduction of government support programs tailored to the pandemic: stimulus checks, expanded and enhanced unemployment benefits, student loan repayment relief, and supports to businesses.
- Among all workers eligible for ILSC, whether they participated or opted out, more Blacks (62%) than Whites (56%) or Hispanics (50%) felt financially insecure in spring 2020. Financial insecurity was also disproportionate among those without a college degree, and among workers who had never married or were separated, divorced, or widowed.
- More than half—57%—of participants in the program said they felt financially insecure in spring 2020, compared with 48% of workers who opted out of the program. But the difference in financial insecurity between participants and opt-outs narrowed through the survey’s three waves and had already been statistically eliminated by fall 2020.
- The greater initial sense of financial precarity on the part of program participants may lie, in part, in a greater income-and-expense volatility compared with those who opted out of participation.
- Only slightly more participants (27%) than nonparticipants (25%) said in spring 2020 that they were fairly or very confident of having enough retirement savings by the time they retired, although this comparison wasn’t statistically significant in subsequent waves.
- One possibility for this slight difference between participants and opt-outs may be that having an ILSC account increased participants’ retirement confidence.
- Importantly, participating in the Illinois program, with automated deductions that reduce a worker’s take-home paycheck, was not associated with whether survey respondents reported feelings of financial insecurity. Statistical analyses found instead that household financial characteristics—in particular, household expense variability—were clearly linked to feelings of financial insecurity.
Financial insecurity among all workers eligible for ILSC
The survey asked workers who were eligible for ILSC—program participants and opt-outs alike—about their financial picture, with questions targeted to overall financial well-being, income and expenditure stability, and receipt of public assistance, among other factors. In particular, the survey asked, “Thinking about your household today, how financially secure do you feel?”4
Many workers eligible for ILSC said they felt somewhat or very financially insecure, although financial insecurity among all workers eligible for ILSC decreased during the 1½ years of the survey from 55% reporting that they felt somewhat or very financially insecure in spring 2020 to 49% in fall of that year and 41% in the spring of 2021. The decline in financial insecurity over the three survey waves tracks the national economy’s performance over the same period. In particular, the pandemic’s severe impact on economic growth was felt most strongly in the second quarter of 2020 and followed by economic recovery. Although the figures aren’t directly comparable, in a separate survey conducted by the FINRA Foundation between June and October 2021, 56% of U.S. adults surveyed nationwide reported feeling anxious about their personal financial condition (data is unavailable for later months).5
More Blacks (62%) than Whites (56%) or Hispanics (50%) surveyed by Rand said they felt financially insecure in spring 2020, whether they participated in ILSC or not. Financial insecurity was also higher among those without a college degree, and among single workers who had never married or were separated, divorced, or widowed.
Workers are eligible for ILSC if they don’t have a retirement plan at work,6 a characteristic that may reflect other aspects of job quality, such as job and income stability or having a predictable work schedule. In the spring of 2020, about half (49%) of all workers eligible for ILSC said their income varied somewhat or a lot from month to month, and 47% said their expenses varied somewhat or a lot from month to month.
Financial insecurity is linked to variability in income and expenses. In spring 2020, financial insecurity was 39% among households with a steady income compared with 61% among those with income that varied from month to month (Figure 1). On the expense side, financial insecurity similarly rose from 46% among households with steady expenses to 54% among households with variable expenses.
Financial insecurity among participants compared with opt-outs
Workers who participated in ILSC initially felt more financially insecure than their peers who chose to opt out (Figure 2). In the spring of 2020, 57% of participants said they felt financially insecure, compared with 48% of opt-outs. But the difference between participants and opt-outs narrowed through the survey’s three waves and had already been eliminated, in terms of statistical significance, by fall 2020.7
Participants reported facing greater income and expense volatility than opt-outs, which may help explain participants’ higher financial insecurity. In spring 2020, 50% of participants said their income varied somewhat or a lot from month to month, compared with 43% of opt-outs (Figure 3). Similarly, 47% of participants said in spring 2020 that their monthly expenses varied somewhat or a lot, compared with 42% of opt-outs. Over time, more participants (27%) than opt-outs (21%) reported in all three survey waves that their income was variable; the results for expense variability across the same period were not statistically significant (not shown in Figure 3).
Participants also show other indicators of income or expense precarity that may lead to more financial insecurity than opt-outs. At the survey’s start in March 2020, a somewhat higher share of participants (8%) said they were seasonal employees than did opt-outs (5%). ILSC participants were more likely to rely on public assistance (16%) than opt-outs (9%), suggesting that participants had lower household incomes. Among participants, 43% said they skipped a bill other than their rent or mortgage payment at least once during the previous six months, and a third (33%) said they had overdrawn their checking account at least once during that period, again compared with lower figures for opt-outs.
Other savings and debt
In addition to a subjective self-assessment, financial security may also be measured by whether survey respondents had savings or investments besides an ILSC account. The survey found no significant difference between participants and opt-outs in terms of having another retirement savings account outside of the ILSC at the time the survey was launched. More opt-outs (14%) than participants (9%) said they own mutual funds, stocks, or bonds outside of a retirement plan, although the survey didn’t ask—and therefore could not compare—their stock and bond balances; in any event, for both groups the percentage of respondents reporting these other investments is small. There were no significant differences in homeownership rates between participants and nonparticipants.
Debt ownership and debt balances are another important measure of financial security; servicing debt can affect a household’s ability to save for retirement and other goals. Additionally, some workers may enter retirement with significant debt balances. Yet the survey found no difference between participants and opt-outs in terms of having debt outside a mortgage in any of the three waves.
Retirement savings confidence
The survey asked not only about financial security in the present, but also how confident a person was that they would have enough retirement savings in the future. Slightly more program participants (27%) than nonparticipants (25%) said in spring 2020 that they were fairly or very confident they would have enough retirement savings by the time they retired, although differences were not statistically significant in subsequent waves. Because there were no differences between participants and opt-ins in terms of having other retirement or nonretirement savings, or nonmortgage debt, that could potentially affect retirement confidence—and because few in either group held stocks and bonds—it’s possible that joining ILSC boosted participants’ perception of retirement preparedness, at least initially.
Not surprisingly, confidence about having sufficient retirement savings in the future is linked to a respondent’s current overall financial security at a high level of statistical significance. Among workers who felt fairly or very confident they’d have enough retirement savings, 60% of participants, and 79% of opt-outs, said they felt financially secure in spring 2020. (Among workers who were not confident of having sufficient retirement savings, there was no statistical difference between those who felt currently financially secure and those who did not.) Including those who were somewhat confident about reaching retirement with sufficient savings—in addition to those who were fairly or very confident—changes the figures; 54% of ILSC participants and 68% of nonparticipants who felt somewhat, fairly, or very financially secure in the present said they were somewhat, fairly, or very confident that they’d reach retirement with sufficient savings.
Other factors related to retirement confidence, such as whether the worker had recently (or had ever) tried to figure out their retirement needs, did not vary significantly between participants and opt-outs.
Conclusion
A key issue with state automated retirement savings programs such as ILSC is how they affect workers’ financial security and perceptions of retirement preparedness. Many workers eligible for the programs earn low to moderate incomes, so that saving for retirement under any circumstances may be a challenge.
Many workers eligible for ILSC said they felt financially insecure. But whether they participated or not, their financial insecurity decreased between spring 2020 (when the pandemic’s economic impact was felt most sharply) and spring of the following year. Financial insecurity is linked to a worker’s economic circumstances, including volatility in income and expenses, receiving public assistance, or having to skip paying bills. And financial insecurity was disproportionately reported by Blacks; workers without a college degree; and those who had never married or were separated, divorced, or widowed.
While workers who participated in ILSC felt less financially secure in the spring of 2020 than their peers who didn’t participate, this gap had already been eliminated in a statistical sense by fall 2020, just six months later.
Importantly, the additional financial contributions by participants in the Illinois program were not associated with increasing their financial insecurity. Instead, underlying household characteristics, such as expenses that varied from month to month, were more strongly associated with whether they felt financial insecurity. Furthermore, the statistical analysis suggests that more participants than opt-outs demonstrated underlying traits of financial insecurity, such as skipping bill payments, receiving public assistance, or overdrawing their checking accounts.
Appendix A. Descriptive Statistics
Categorical variables, spring 2020
No | Yes | Don't know | n | |
---|---|---|---|---|
Participates in ILSC | ||||
All | 39% | 38% | 22% | 1,100 |
No | Yes | n | |
---|---|---|---|
Reported household income variability in all waves | |||
All | 75% | 25% | 1,100 |
Participants | 73% | 27% | 497 |
Opt-Outs | 79% | 21% | 382 |
Reported household expense variability in all waves | |||
All | 79% | 21% | 1,091 |
Participants | 80% | 20% | 495 |
Opt-Outs | 80% | 20% | 382 |
Receives public assistance | |||
All | 88% | 12% | 1,091 |
Participants | 84% | 16% | 495 |
Opt-Outs | 91% | 9% | 375 |
Overdrew checking account | |||
All | 71% | 29% | 1,097 |
Participants | 67% | 33% | 497 |
Opt-Outs | 76% | 24% | 380 |
Skipped bill other than rent or mortgage | |||
All | 62% | 38% | 1,099 |
Participants | 57% | 43% | 497 |
Opt-Outs | 70% | 30% | 381 |
Has no other retirement plan | |||
All | 33% | 67% | 866 |
Participants | 33% | 67% | 493 |
Opt-Outs | 34% | 66% | 373 |
Has nonmortgage debt | |||
All | 71% | 29% | 1,090 |
Participants | 71% | 29% | 495 |
Opt-Outs | 75% | 25% | 374 |
Seasonal worker | |||
All | 92% | 8% | 960 |
Participants | 92% | 8% | 443 |
Opt-Outs | 95% | 5% | 347 |
Has stocks, mutual funds or bonds | |||
All | 90% | 10% | 1,086 |
Participants | 91% | 9% | 493 |
Opt-Outs | 86% | 14% | 373 |
Has credit card debt | |||
All | 57% | 43% | 1,090 |
Participants | 56% | 44% | 495 |
Opt-Outs | 54% | 46% | 374 |
Ever tried to figure out retirement needs | |||
All | 74% | 26% | 1,095 |
Participants | 71% | 29% | 495 |
Opt-Outs | 74% | 26% | 379 |
Male | Female | Other | n | |
---|---|---|---|---|
Gender | ||||
All | 45% | 54% | 1% | 1,093 |
Participants | 43% | 56% | 1% | 495 |
Opt-Outs | 44% | 55% | 1% | 379 |
White | Black | Hispanic | Other | n | |
---|---|---|---|---|---|
Race | |||||
All | 37% | 21% | 36% | 6% | 1,093 |
Participants | 39% | 29% | 27% | 6% | 494 |
Opt-Outs | 36% | 13% | 43% | 7% | 378 |
Single, never married | Married | Widowed, divorced, or separated | n | |
---|---|---|---|---|
Marital status | ||||
All | 52% | 33% | 15% | 1,091 |
Participants | 57% | 27% | 15% | 495 |
Opt-Outs | 44% | 41% | 15% | 379 |
High school or less | Some college or above | n | |
---|---|---|---|
Educational attainment | |||
All | 43% | 57% | 1,094 |
Participants | 44% | 56% | 495 |
Opt-Outs | 40% | 60% | 377 |
Continuous variable, spring 2020
Mean | Median | Standard deviation | n | |
---|---|---|---|---|
Age in spring 2020 | ||||
All | 36.9 | 34 | 14 | 1,094 |
Participants | 36.9 | 34 | 13 | 495 |
Opt-Outs | 38.4 | 36 | 14 | 375 |
Appendix B. Logistic Regression Analysis
The research for this issue brief used statistical analysis in the form of logistic regression to supplement the cross-tabulations (Table B1). The key variable in the regression analysis is the individual’s self-reported financial security (the dependent variable). Independent variables, which help to explain financial security, include household income variability, household expense variability, whether the worker is receiving public assistance, whether they’ve bounced a check or skipped a bill other than their rent or mortgage payment in the previous six months, whether they have another retirement plan besides ILSC, and whether they have nonmortgage debt. Additionally, the model includes demographic characteristics such as age, educational attainment, gender, race and ethnicity, and marital status. The regression permits analysis of the relationship between financial security and each independent variable, while holding other independent variables constant.
Odds ratios are interpreted in terms of probability, with odds ratios greater than 1 indicating higher odds than the reference case and odds ratios less than 1 indicating lower odds than the reference case. For the dependent variable of financial security, the reference or base category is feeling somewhat or very financially secure, compared with a second response option of feeling somewhat or very financially insecure. Thus, if a worker had variable household expenses in all three waves of the survey, his or her odds of feeling financially insecure were 1.8 times greater than a worker with relatively constant household expenses, holding all other variables constant. The constant variable estimates baseline odds. These interpretations contain no information about the magnitude of the implied change in probability.
Household financial characteristics are linked to financial insecurity, with workers whose household expenses varied somewhat or a lot from month to month 1.8 times more likely than workers with stable monthly expenses to feel financially insecure, and those who had not skipped a bill payment other than their rent or mortgage less likely to feel financially insecure than those who had skipped such a bill. But most demographic factors, such as race, age, gender, and educational attainment, were not significantly associated with financial security. The exception is marital status: Those who had never married were 1.7 times more likely to feel financially insecure than married workers, and those who were divorced, separated, or widowed were a little more than twice (2.1 times) as likely to feel financially insecure than married workers.
Notably, participation in ILSC is not linked to financial security either negatively or positively in the regression when all other factors are considered. Although many participants made regular, automated contributions to ILSC, doing so did not affect their feelings of financial security relative to those who opted out of the program. As noted in other research based on this survey, 38% of ILSC participants said that they felt more financial security by virtue of participating in the program versus only 14% who said they felt more financially insecure.8
Table B1. Modeling Financial Security Among ILSC Participants and Opt-Outs: Logistic Regression
Odds ratio | P>|t| | ||
---|---|---|---|
Participating | |||
Participating | (reference) | ||
Not participating | 0.8793476 | 0.436 | |
Household income variable in all waves | |||
No | (reference) | ||
Yes | 1.387903 | 0.112 | |
Household expenses variable in all waves | |||
No | (reference) | ||
Yes | 1.796292 | 0.01 ** | |
Receives public assistance | |||
Yes | (reference) | ||
No | 0.7190632 | 0.222 | |
Overdrew checking account | |||
Yes | (reference) | ||
No | 0.674814 | 0.057 | |
Skipped bill other than rent or mortgage | |||
Yes | (reference) | ||
No | 0.2834768 | 0.000 *** | |
Has no other retirement plan | |||
Has other retirement plan | (reference) | ||
Has no other retirement plan | 0.9996375 | 0.999 | |
Has nonmortgage debt | |||
No | (reference) | ||
Yes | 0.8150663 | 0.297 | |
Age, continuous | |||
1.000639 | 0.931 | ||
Educational attainment | |||
High school or less | (reference) | ||
Some college or above | 0.9184244 | 0.638 | |
Gender | |||
Male | (reference) | ||
Female | 0.8345159 | 0.284 | |
Other | 3.453904 | 0.129 | |
Race | |||
White | (reference) | ||
Black | 0.7727983 | 0.25 | |
Hispanic | 0.6770328 | 0.059 | |
Other | 0.6748299 | 0.218 | |
Marital status | |||
Married | (reference) | ||
Single, never married | 1.723076 | 0.015* | |
Widowed, divorced or separated | 2.095139 | 0.005** | |
Constant | |||
5.521281 | 0.012 * |
Legend: * p<0.05; ** p<0.01; *** p<0.001 F=7.02; Prob > F = 0.0000 Number of observations = 865 Note: The number of observations fell because only 866 workers answered the survey question about having a retirement plan besides ILSC. When this variable is dropped from the regression, the odds ratios remain relatively stable, although two variables—household income variable in all three waves and Hispanic ethnicity—become significant at p<0.05%. Source: Pew survey of Illinois Secure Choice eligible participants, spring 2020, fall 2020, and spring 2021 |
Endnotes
- As ILSC was originally enacted, firms with 25 or more employees that did not sponsor a retirement plan were required to automatically enroll their workers, and smaller firms were allowed to join. Illinois phased in the program, initially requiring firms with 500 or more employees to sign up. The employee threshold had dropped to 25 when this survey was fielded. Illinois subsequently amended the program to require firms with five or more workers to enroll.
- ILSC account owners may withdraw any amounts they’ve contributed without taxes or penalties. However, earnings on ILSC balances may be subject to taxes or penalties unless they meet two requirements. First, a five-year holding period must have passed between the first contribution and the withdrawal of earnings. Second, a distribution must be made for one of four qualifying reasons: (1) the account owner has reached 59½ years of age; (2) the account owner has a qualifying disability; (3) the funds are used to pay firsttime homebuyer expenses; or (4) the account owner’s beneficiaries are taking a distribution following the owner’s death. See Illinois Secure Choice Retirement Savings Program, “Withdrawals,” accessed Sept. 12, 2023, https://saver.ilsecurechoice.com/home/savers/withdrawals.html.
- Attrition reduced survey participation by 44% between the baseline and first survey waves. Between wave 1 and wave 2, the completed sample size was reduced by an additional 7.5%, for a total attrition rate between the baseline and second wave of 51.8%. The proportion of respondents who completed all three surveys across Secure Choice opt-out (42%) and remain-in (58%) groups were close to the fielded sample (48% and 52%, respectively) and reasonably well balanced by race and ethnicity. For more information on Pew’s survey of workers eligible for ILSC, see Rand Corporation, “Pew Illinois Secure Choice Survey—Methodology” (n.d.), https://www.pewtrusts.org/-/media/assets/2022/04/methodology_pew_secure-choice.pdf.
- This survey question was presented on a four-point scale with choices including (1) very financially secure, (2) somewhat financially secure, (3) somewhat financially insecure, and (4) very financially insecure.
- FINRA Investor Education Foundation, “Financial Capability in the United States: Highlights From the FINRA Foundation National Financial Capability Study” (2022), https://www.finrafoundation.org/sites/finrafoundation/files/NFCS-Report-Fifth-Edition-July-2022.pdf.
- Employers with five or more employees are required to enroll their employees in ILSC if they don’t already offer a retirement plan. But the program is also available to anyone who wants to enroll and start saving. See “Illinois Secure Choice Retirement Savings Program,” Illinois Secure Choice, accessed Sept. 13, 2023, https://www.ilsecurechoice.com/.
- The difference between participants and opt-outs in spring 2021 was not statistically significant.
- J. Scott and M. Hines, “Many in Illinois Retirement Savings Program Feel Their Financial Security Is Improving,” The Pew Charitable Trusts, April 18, 2022, https://www.pewtrusts.org/en/research-and-analysis/articles/2022/04/18/many-in-illinois-retirement-savings-program-feel-their-financial-security-is-improving.