Effective Pension Communication

How to cite

Eberhardt, Inka; Smeets, Paul; Bauer, Rob. 2015. "Effective Pension Communication." AEA RCT Registry. AEARCTR-0000987. https://doi.org/10.1257/rct.987-1.0

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

Inka Eberhardt* , Paul Smeets , Rob Bauer

* Primary investigator

Registry ID
AEARCTR-0000987
DOI
https://doi.org/10.1257/rct.987-1.0
First registered
Last updated
Status
In development
Trial dates
2015-12-23 – 2016-01-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/987

Abstract

With the increasing individualization and insecurity of the Dutch pension system, the responsibility for a sufficient pension lies more and more with the pension fund participant. In order to make a decision on whether future pension benefits are sufficient or whether to take additional measures to save for retirement, participants need to have accurate information on the status quo. However, interest in pension communication has been found to be low. Through an experiment where the content of letters is slightly changed per treatment group, we will analyse which manner of communication is most effective in increasing pension awareness. Our main dependent variable is whether participants visit the new website of the pension fund and log in. Furthermore, we are able to see whether different groups of participants react differently to treatments and whether the behavior on the website changes per treatment. Younger participants might react more often to the lottery treatment than older participants. The lottery treatment might lead to more people logging in, while it also might have the consequence that people spend less time to inform themselves.

Experimental design

The letters are sent via a quarterly magazine published by the pension fund and invite the participant to his/her personal space on the pension fund’s website. Each treatment adds one sentence at the beginning of the second paragraph. Four treatments are separated on two dimensions: two letters provide contents on what people think about their future pension income, while two other letters provide information on what people think about the adequacy of their retirement savings. The second dimension lays out whether the pension income or the pension savings will be enough according to a large part of the Dutch. The fifth treatment announces a lottery among the participants who log in to the website. On the website, there is the opportunity to fill in a quiz to test pension knowledge. The quiz contains survey questions on risk and time preferences as well as trust towards the pension fund, current saving behavior and saving intentions. Next to this, we have administrative data on gender, age, income and place of residence.

Design details

Randomization method
Randomization done by a computer
Randomization unit
Pension fund participants
Clusters
0
Observations
~250,000 participants
Arms
~250,000 participants
Primary outcomes
The main variable is the percentage of participants logging in to the website of the pension fund. Additional outcome variables are time spent on the website, whether the participant checks the pension planner and whether the participant takes the pension quiz.

Cited by

No citation data available for this registration yet.

Keywords