Document Type
Working Paper
Date of this Version
9-25-2019
Abstract
Prior research in economics and psychology has documented that individuals exhibit time-inconsistent preferences when faced with the opportunity to take an action that involves immediate costs in return for future benefits – the notion of implementing such an action now is unappealing, but the notion of implementing the same action later is attractive. Because increasing contributions to a retirement savings plan requires a reduction in current consumption (an immediate cost) in order to increase consumption in old age (a future benefit), individuals may be more likely to agree to a contribution rate increase if they have the option to have the increase implemented at a delay. We conducted a field experiment with several universities to test whether the option to choose a delayed contribution rate increase boosts savings. Relative to employees who are offered a convenient mechanism for increasing their contribution rates immediately, employees who are offered a convenient mechanism for increasing their contribution rates immediately or at a delay are no more likely to agree to an increase. In fact, the latter group exhibits lower savings rates over the coming months, as the delayed option attracts some employees. However, when the delayed option is framed as being implemented after a psychologically meaningful moment, such as an employee’s next birthday, the negative effect of offering a delayed option is undone.
Working Paper Number
WP2015-12
Copyright/Permission Statement
The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of NIA, SSA, any agency of the Federal Government, the TIAA-CREF Institute, or the NBER. Benartzi is the Chief Behavioral Economist at the Allianz Global Investors Center for Behavioral Finance. See the authors’ websites for a complete list of outside activities.
Acknowledgements
We thank our partners at the universities and at the retirement savings plan administrator for implementing the experiment, and we thank Hae Nim Lee for excellent research assistance. We acknowledge financial support from the National Institute on Aging (grants P01AG005842 and P30AG034532), the Social Security Administration (grant RRC08098400-06-00 to the National Bureau of Economic Research as part of the SSA Retirement Research Consortium), and the TIAA-CREF Institute.
Date Posted: 12 March 2019