Skip to main content

A Community-Based Randomized Controlled Trial of an Educational Intervention to Promote Retirement Saving Among Hispanics


We conducted a community-based randomized controlled trial (RCT) to evaluate the impact of an intervention to promote retirement saving among low- and moderate-income, predominantly Spanish-speaking Hispanics, who do not have access to an employer-sponsored retirement account. Our educational intervention provided participants with key information in Spanish related to financial planning for retirement and made use of “behavioral nudges” to encourage participants to open a government-sponsored retirement saving account, my Retirement Account (myRA). Among 142 participants (70 and 72 in control and treatment groups, respectively), we found a significant Difference in Differences (DD) on the proportion of those who opened a myRA. In the treatment group 14% of participants opened myRA, while none in the control group did. Results from a DD regression showed that after controlling for observable and unobservable characteristics through our design as an RCT, the effect of our intervention was 12%. When we restricted our sample to those participants who had a bank account at some point in the study, we found that the effect of our intervention was higher (17%). We also found that our intervention was effective in increasing self-reported knowledge related to retirement saving and preparedness.

This is a preview of subscription content, access via your institution.

Fig. 1


  1. The compound interest question used when measuring financial knowledge/literacy is the following: Suppose you had $100 in a savings account and the interest rate is 20% per year and you never withdraw money or interest payments. After 5 years, how much would you have on this account in total? 1. More than $200; 2. Exactly $200; 3. Less than $200; 4. Don’t know; 5. Refuse to answer.

  2. This study had a caveat since the sample was small and not representative of the population of employees.

  3. The Video “Una Nueva Manera de Ahorrar en La Salle Cleaners” (A new way to save in La Salle Cleaners) was accessed in YouTube and played during the workshop. The video was all in Spanish (or with Spanish subtitles when there was something said in English). This video is no longer available in YouTube.

  4. For example, through a Facebook experiment, Blanco and Rodriguez (2018) find that the message “Many Hispanic women like you have a plan for retirement” is more effective that a standard message and one focus on the family to get Hispanic women interested in learning more about retirement savings. Thus, the more effective message was used in our educational intervention. See message in the selected slides of the workshop presentation included in Online Appendix 1.

  5. We sent a text message at 5 months that reminded participants specifically to open a retirement account before the 6 month deadline.

  6. All complete materials related to our educational intervention are available upon request.

  7. All oral presentations given during our workshops were given in Spanish by a native Spanish speaker, where all participants previously expressed that they were proficient with oral communication in Spanish. All workshop documents were available in Spanish and English, where those participants that felt more comfortable reading and writing in English received the English version per their request.

  8. Note that this website was updated in the Spring of 2018 to take out information on myRA since this program is no longer available. We also changed some of the links of available resources since some of the links we included in the website initially were no longer available. This website was used for another related project conducted by Blanco and Rodriguez (2018).

  9. Our block randomization approach is similar to the one used by Duru et al. (2010), who conduct a randomized controlled trial of a faith-based physical activity program. We use Sealed Envelope available online to create a randomization list for the different groups. Information on groups randomization was not shared with community partners or participants (

  10. Only 4 participants in each group expressed at the second workshop that they contributed to a retirement saving plan different than myRA in a monthly basis (6% of our sample). As a robustness check, we conducted our analysis restricting the sample to those who were not saving in another retirement saving plan during the time of the study and found the same results.

  11. It is important to note that in our main analysis we did not count people who opened a myRA during the second workshop in the data analyses since it will bias the results.

  12. Please refer to Dew and Xiao (2011) for further discussion on the development and validation of this scale. See Table A2 in Online Appendix 1 for the questions used in this scale.

  13. Follow-up surveys are very similar to baseline surveys and are available upon request.

  14. We describe next the control variables we use in our estimations: age (30 to 64), gender (male = 1, female = 0), nativity status (foreign born = 1, native born = 0), educational attainment (no primary education = 0, primary education completed = 1, high school completed = 2, college or more completed = 3), Spanish native speaker (equal to 1, zero otherwise), speaks no English or poorly (equal to 1, zero otherwise), Mexican origin (equal to 1, zero otherwise), employment status (employed part-time, full-time, or self-employed = 1, unemployed, incapacitated, homemaker, retired, student = 0). We use the financial strain indicator developed and validated by Hilton and Devall (1997). Refer to Online Appendices 2 and 3 for the survey questions in English and Spanish from which we construct these variables.

  15. While a Probit model will be more appropriate when using a dichotomous variable as dependent variable, we used OLS since we are including fixed effects, and fixed effect estimators of non-linear models can be biased (see Greene 2002; Arellano and Hahn 2007).

  16. For power calculations we use STATA command “power twoprop” for the dichotomous variable (open myRA, and “power twomeans” for the continuous variables (RKI and M-RKI).

  17. Note that in this part of the analysis, where we show the DD in Table 4, we are assuming a value of zero for those participants we were unable to reach via phone at the one week and one and three months follow-up and who did not attended the second workshop[ (completers and non-completers are considered). For the DD of the RKI we are using data for participants who attended both workshops (completers only).

  18. In this part of the analysis, where we estimate the DD regression, we only consider those participants that attended both workshops (only completers).

  19. Our last workshop for the collection of baseline data and deliver of intervention among treatment participants took place on March 11 of 2017, which gives us more than three months before the announcement on the termination of the myRA program. The announcement came out in July of 2017, and at that time you could no longer open a myRA account, but you could continue to contribute to the account.

  20. For more information on this refer to the California State Treasurer page


Download references


We thank our community partners, the Mexican American Opportunity Foundation (MAOF) and New Economics for Women (NEW), for all their support during the design, data collection, and implementation stages of the project. We thank specially Isaias Hernandez and Raul De la O from MAOF, and Leticia Andueza and Magdalena Cervantes from NEW. We also thank Central City Neighborhood Partners (CCNP) for letting us use their facilities for this study. We also thank myRA Outreach Office for their support to this project, specially to Ingrid Taylor. We acknowledge Ana Francisca Silva, who was the project coordinator, for her excellent work. We thank Michael Ong and Thomas Rice for their useful comments to our manuscript, and to William Humphrey for his research assistance. Dr. Luisa R. Blanco received support for this project from NIA-funded Grant 3P30AG021684-15:S1, 2016–17 Research Supplement to Promote Diversity in Health-Related Research through the University of California, Los Angeles (UCLA) and Charles Drew University (CDU), Resource Centers for Minority Aging Research Center for Health Improvement of Minority Elderly (RCMAR/CHIME). Pepperdine’s Office of Provost also provided funding for this study. The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. All errors are our own.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Luisa R. Blanco.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 23859 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Blanco, L.R., Duru, O.K. & Mangione, C.M. A Community-Based Randomized Controlled Trial of an Educational Intervention to Promote Retirement Saving Among Hispanics. J Fam Econ Iss 41, 300–315 (2020).

Download citation

  • Published:

  • Issue Date:

  • DOI:


  • Aging
  • Behavioral nudges
  • Financial knowledge