Back to History Current Version

How Do Retirement Income Tools Affect Saving Decisions? Evidence from a Field Experiment

Last registered on August 31, 2018

Pre-Trial

Trial Information

General Information

Title
How Do Retirement Income Tools Affect Saving Decisions? Evidence from a Field Experiment
RCT ID
AEARCTR-0002129
Initial registration date
March 28, 2017

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 30, 2017, 3:04 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
August 31, 2018, 4:20 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Primary Investigator

Affiliation
Stanford University

Other Primary Investigator(s)

PI Affiliation
Stanford University & NBER
PI Affiliation
University of Minnesota
PI Affiliation
London School of Economics
PI Affiliation
University of Minnesota & IZA
PI Affiliation
Claremont Graduate University

Additional Trial Information

Status
In development
Start date
2017-03-29
End date
2019-06-01
Secondary IDs
Abstract
Theory predicts that exponential-growth bias, the tendency to neglect compounding interest, will lead to suboptimal saving decisions. In previous work, we show that this bias is prevalent and explains a meaningful amount of heterogeneity in retirement wealth accumulation. In this project, we aim to investigate an intervention meant to target this bias via a field experiment at a large employer. Our results will provide evidence as to whether this intervention affects retirement savings through the hypothesized mechanism by analyzing the heterogeneity in responses with respect to underlying measures of exponential-growth bias. These interventions could serve as an important proof of concept for future interventions that lead individuals to make better-informed retirement saving decisions.
External Link(s)

Registration Citation

Citation
Goda, Gopi et al. 2018. "How Do Retirement Income Tools Affect Saving Decisions? Evidence from a Field Experiment." AEA RCT Registry. August 31. https://doi.org/10.1257/rct.2129-5.0
Former Citation
Goda, Gopi et al. 2018. "How Do Retirement Income Tools Affect Saving Decisions? Evidence from a Field Experiment." AEA RCT Registry. August 31. https://www.socialscienceregistry.org/trials/2129/history/197704
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
1. Group I: Goals only in time 1, goals + treatment in time 2
a. Group I-A:
i. Time 1: Receive link to goals only tool, receive reminder at week 1 and week 2
ii. Time 2: receive link to goals + projections tools, reminder at week 1 and week 2

b. Group I-B:
i. Time 1: Receive link to goals only tool, receive reminder at week 1 and week 2
ii. Time 2: receive link to goals + projections + cost of delay tool, reminder at week 1 and week 2

2. Group II: goals + treatment in time 1
a. Group II-A
i. Time 1: Receive link to goals + projections tools, reminder at week 1 and week 2

b. Group II-B
i. Time 1: Receive links to goals + projections + cost of delay tool, reminder at week 1 and week 2
Intervention (Hidden)
Intervention Start Date
2017-05-01
Intervention End Date
2018-07-01

Primary Outcomes

Primary Outcomes (end points)
retirement plan contributions, absolute value of change in retirement plan contributions, binary indicator of accessing retirement tool, binary indicator of making any change in retirement plan contributions, difference between Goal and Plan (measure of "intended" change), binary indicator of visiting last page in tool, total time spent on tool, exit tool to Employee Express, clicked on "Print" on tool
Primary Outcomes (explanation)
retirement plan contributions will be measured by both percent of salary and total dollars; all outcomes taken from administrative data on a bi-monthly basis

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We will merge together administrative data on employee retirement savings plan contributions, HR records, and survey data. The survey will ask subjects about basic demographics that are missing from the administrative data. In addition, the survey will elicit employees’ EGB and time preferences using un-incentivized questions. The survey will take approximately 15 minutes. Our administrative data on retirement contributions will span several months before and after our interventions, and will be collected on a high frequency basis (monthly prior to the intervention, and every pay period thereafter).

The experiment will have four groups, following a staggered rollout design. Groups I-A and I-B receive the Goals Only treatment as a placebo at Time 1, and will get either the Goals + Projections (I-A) or the Goals + Projections + Cost of Delay treatment at Time 2, approximately two months later. Groups II-A and II-B receive the Goals + Projections (II-A) or the Goals + Projections + Cost of Delay (II-B) treatment in period 1.

The Goals + Projections + Cost of Delay tool includes a prompt for individuals to think about their retirement lifestyle goals, an interactive calculator that projects a person’s retirement income beginning at an assumed retirement age (projection), and presents calculations of the costs associated with delaying changes (cost of delay). The Goals Only and Goals + Projections tools will offer a subset of those capabilities.
Experimental Design Details
Randomization Method
Block randomization; see randomization file included for more information.
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
6000
Sample size (or number of clusters) by treatment arms
2000 individuals group 1 (1/3; split into 1/6 and 1/6), 2000 individuals group 2 (1/3), 2000 individuals group 3 (1/3)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Minimum Detectable Effect Size for Main Outcomes
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
Claremont Graduate University Institutional Review Board (IRB)
IRB Approval Date
2016-10-10
IRB Approval Number
2813
IRB Name
University of Minnesota HRPP
IRB Approval Date
2016-07-12
IRB Approval Number
1607E90001
IRB Name
Stanford University IRB
IRB Approval Date
2016-09-12
IRB Approval Number
38948
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Abstract
We conduct a randomized controlled trial to understand how a web-based retirement saving calculator affects workers' retirement-savings decisions. In both the treatment and active control conditions, the calculator projects workers' retirement income goal. In the treatment condition only, it also projects retirement income based on defined-contribution savings, prominently displays the gap between projected goal and actual retirement income, and allows users to interactively explore how alternative, future contribution choices would affect the gap. The treatment increased average annual retirement contributions by $174 (2.3 percent). However, effects were larger for those with higher measures of financial knowledge, suggesting this type of tool complements, rather than substitutes for, underlying financial capability.
Citation
@article{GODA2023561, title = {Are retirement planning tools substitutes or complements to financial capability?}, journal = {Journal of Economic Behavior & Organization}, volume = {214}, pages = {561-573}, year = {2023}, issn = {0167-2681}, doi = {https://doi.org/10.1016/j.jebo.2023.08.001}, url = {https://www.sciencedirect.com/science/article/pii/S0167268123002834}, author = {Gopi Shah Goda and Matthew R. Levy and Colleen {Flaherty Manchester} and Aaron Sojourner and Joshua Tasoff and Jiusi Xiao}, keywords = {Retirement planning, Retirement saving, Exponential-growth bias, Present bias, Financial literacy, Financial capability}, abstract = {We conduct a randomized controlled trial to understand how a web-based retirement saving calculator affects workers' retirement-savings decisions. In both the treatment and active control conditions, the calculator projects workers' retirement income goal. In the treatment condition only, it also projects retirement income based on defined-contribution savings, prominently displays the gap between projected goal and actual retirement income, and allows users to interactively explore how alternative, future contribution choices would affect the gap. The treatment increased average annual retirement contributions by $174 (2.3 percent). However, effects were larger for those with higher measures of financial knowledge, suggesting this type of tool complements, rather than substitutes for, underlying financial capability.} }

Reports & Other Materials