JOURNAL ARTICLE
Should the Government Be Paying Investment Fees on $3 Trillion of Tax-Deferred Retirement Assets?
Published In: Review of Financial Studies, 2025, v. 38, n. 4. P. 1014 1 of 3
Database: Business Source Ultimate 2 of 3
Authored By: Landoni, Mattia; Zeldes, Stephen P 3 of 3
Abstract
This article analyzes the economic and fiscal implications of tax-deferred ("Traditional") versus front-loaded ("Roth") retirement savings accounts, focusing on the role of asset management fees. While standard models suggest individuals and the government are indifferent between the two systems absent fees, the authors show that investment fees break this equivalence: under Traditional accounts, the government effectively pays higher dollar fees on an implicit portfolio of deferred taxes, estimated at $23.4 billion annually on $3.8 trillion in assets. A supply-side model demonstrates that asset managers optimally charge similar percentage fees under both systems, resulting in greater profits and a larger asset management industry under Traditional. Embedding these findings in a general equilibrium framework, the study finds that Traditional accounts lead to higher taxes, greater resource allocation to asset management, and lower social welfare compared to Roth accounts, with welfare losses amounting to roughly one-quarter to one-third of the government's tax expenditure on retirement savings. The paper also discusses real-world considerations such as progressive taxation, behavioral biases, and political economy factors, concluding that these are unlikely to overturn the core result that Roth systems are more efficient when accounting for asset management fees.
Additional Information
- Source:Review of Financial Studies. 2025/04, Vol. 38, Issue 4, p1014
- Document Type:Article
- Subject Area:Business and Management
- Publication Date:2025
- ISSN:0893-9454
- DOI:10.1093/rfs/hhae070
- Accession Number:184348169
- Copyright Statement:Copyright of Review of Financial Studies is the property of Oxford University Press / USA and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Looking to go deeper into this topic? Look for more articles on EBSCOhost.