Financial technology and behavioral barriers to saving for retirement

How to cite

Haufler, Fabio; Daminato, Claudio; Filippini, Massimo. 2021. "Financial technology and behavioral barriers to saving for retirement." AEA RCT Registry. AEARCTR-0008316. https://doi.org/10.1257/rct.8316-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Fabio Haufler* , Claudio Daminato , Massimo Filippini

* Primary investigator

Registry ID
AEARCTR-0008316
DOI
https://doi.org/10.1257/rct.8316-1.0
First registered
Last updated
Status
In development
Trial dates
2021-10-04 – 2022-08-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/8316

Abstract

A large Swiss pension fund is introducing a new digital pension application to help its insureds plan for retirement. The platform allows individuals to access personal information related to their occupational retirement account. Moreover, the digital pension application allows the insureds to initiate voluntary contributions to their retirement saving plan. In collaboration with the pension fund, we randomize different invitation letters across insureds to test (i) which aspects of the new platform are most important for the insureds and (ii) which behavioral barriers are relevant in the context of adopting and utelizing the digital pension application. We explore whether the different invitation letters affect the probability that individuals access the online platform, its usage, and retirement contribution behavior.

Experimental design

Individuals insured with a pension fund in Switzerland will gain access for the first time to a newly introduced digital pension application. The pension fund sends different invitation letters (via e-mail) to all of its insureds in several consecutive waves. Each individual receives the invitation in its preferred language (German, French, Italian or English). The treatment status (corresponding to different letters) is randomized at the individual level for the six largest firms / institutions insured with the pension fund and at the firm level for the remaining firms / institutions. The design allows to analyse potential information spill-over effects across individuals within the same firm / institution. Individuals working in the same firm receive the invitation letter at the same time. We randomize the timing of the intervention across firms / institutions. The empirical analysis will be carried out using registratoin data from the digital application and administrative pension fund data.

Design details

Randomization method
Randomization done by a computer with the program STATA
Randomization unit
Individual level randomization: Individuals working for one of the 6 largest firms in the sample Group level randomization: Remaining 152 firms in the sample
Clusters
Clusters: 152 distinct firms / institutions
Observations
Individual assignment: 32'803 individuals plus 2'724 individuals in last wave Clustered assignment: 26'264 individuals in 152 clusters (firms) plus 1'664 individuals in last wave
Arms
10 treatment arms with randomization on the individual level: treatment 1: 3210, treatment 2: 3336, treatment 3: 3335, treatment 4: 3372, treatment 5: 3214, treatment 6: 3315, treatment 7: 3388, treatment 8: 3330, treatment 9: 3438, treatment 10 (only women): 2865 4 treatment arms with clustered randomization on firm level: each treatment arm includes 38 firms / institutions.
Primary outcomes
Key outcome variables are the insureds' registration status on the portal, user behavior on the portal, and voluntary retirement contribution behavior.

Cited by

No citation data available for this registration yet.

Keywords