How Do Behavioral Approaches to Increase Savings Compare? Evidence from Multiple Interventions in the U.S. Army
Management Science · 2026 · Richard W. Patterson、William Skimmyhorn
中文摘要
如何比较不同的提升退休储蓄的方法?信息干预、主动选择框架与自动加入均能提高退休储蓄,但支持这些方法的证据来自差异极大的制度环境、人群和时期。在本研究中,我们在一个恒定的环境下,利用2016—2018年间美国军方的干预所提供的实验与准实验变异来比较这些方法。我们发现这些方法的效力存在巨大差异:主动选择与自动加入项目分别比信息干预有效约10倍和100倍。我们还收集了关于项目成本的新数据,并对每个项目的成本效益进行了估计。我们发现,主动选择与自动加入在成本效益上一贯优于任何信息干预;对于小企业,主动选择项目的成本效益最高;而对于较大企业,自动加入项目的成本效益最高。本文由行为经济学与决策分析领域的Jack Soll接收。资助:作者感谢TIAA研究所、沃顿商学院养老金研究委员会与Boettner中心、社会保障署以及NBER退休与残疾研究中心提供的资金支持。本文所述研究依据美国社会保障署的资助[SSA资助编号RDR18000003]完成,该资助为退休与残疾研究联合体项目的一部分。文中表达的观点和结论仅为作者个人观点,不代表社会保障署、联邦政府任何机构、NBER、TIAA研究所或沃顿商学院养老金研究委员会/Boettner中心的观点或政策。美国政府及其任何机构或其任何雇员均不对本研究内容的准确性、完整性或实用性作出任何明示或默示的保证,亦不承担任何法律责任或责任。文中以商品名、商标、制造商或其他方式提及任何特定商业产品、流程或服务,并不必然构成或暗示美国政府的认可、推荐或偏袒。补充材料:在线附录与数据文件可在 https://doi.org/10.1287/mnsc.2024.04939 获取。
Abstract
How can we compare approaches to increase retirement savings? Messaging interventions, active choice frameworks, and automatic enrollment all increase retirement savings, but evidence supporting these approaches spans widely different institutional settings, populations, and time periods. In this study, we use experimental and quasi-experimental variation in these approaches within a constant setting using interventions in the U.S. military during 2016–2018. We find large differences in the efficacy of these approaches, with active choice and automatic enrollment programs being about 10 and 100 times more effective than messaging interventions, respectively. We also collect novel data on program costs and generate estimates of cost-effectiveness for each program. We find that active choice and automatic enrollment are consistently more cost effective than any messaging interventions, active choice programs are the most cost-effective for small firms, and automatic enrollment is the most cost-effective program for larger firms. This paper was accepted by Jack Soll, behavioral economics and decision analysis. Funding: The authors thank the TIAA Institute, Wharton School’s Pension Research Council and Boettner Center, the Social Security Administration, and the NBER Retirement and Disability Research Center for financial support. The research reported herein was performed pursuant to a grant from the U.S. Social Security Administration [SSA Grant Number RDR18000003] funded as part of the Retirement and Disability Research Consortium. The opinions and conclusions expressed are solely those of the author(s) and do not represent the opinions or policy of the Social Security Administration, any agency of the Federal Government, the NBER, the TIAA Institute, or Wharton School’s Pension Research Council/Boettner Center. Neither the U.S. Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this research. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the U.S. Government or any agency thereof. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.04939 .