Delayed Retirement Preferences: Evidence from a Discrete Choice Experiment in China

How to cite

Wang, Yijie; Zhang, Ping; Tu, Mengjuan. 2026. "Delayed Retirement Preferences: Evidence from a Discrete Choice Experiment in China." AEA RCT Registry. AEARCTR-0018259. https://doi.org/10.1257/rct.18259-1.0

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

Yijie Wang* , Ping Zhang , Mengjuan Tu

* Primary investigator

Registry ID
AEARCTR-0018259
DOI
https://doi.org/10.1257/rct.18259-1.0
First registered
Last updated
Status
On going
Trial dates
2026-03-23 – 2026-05-10
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/18259

Abstract

This study examines individuals’ preferences over delayed retirement policies in China using a discrete choice experiment (DCE). A total of 1,200 participants aged 45–59 will be recruited through an online platform between March 23, 2026 and May 10, 2026. Participants will first read an informational description of the current delayed retirement policy reform in China. They will then complete a series of pairwise choices between hypothetical retirement schemes that vary along multiple policy dimensions, including wage level, retirement age, pension incentives, flexible work arrangements, and employment sector type. The study aims to identify the relative importance of these policy attributes and to evaluate how policy design can influence individuals’ willingness to delay retirement.

Intervention

Participants are presented with multiple pairs of hypothetical retirement scenarios and are asked to choose their preferred option in each pair.

Experimental design

This study employs an online discrete choice experiment (DCE) to investigate individuals’ preferences over delayed retirement policy designs in China. Participants complete multiple pairwise choice tasks, in which they are asked to select their preferred option between two hypothetical retirement schemes. Each task includes two sets of scenarios that differ exogenously in wage levels. Within each set, the scenarios vary along several key dimensions, including retirement age, pension increment rules (e.g., percentage increase per additional working year), availability of flexible work arrangements after age 60, and employment sector (public vs. non-public sector). The attributes are systematically varied across tasks, and the order of presentation is randomized. By observing repeated choices, the design allows for the identification of preferences over policy attributes and trade-offs between them.

Design details

Randomization method
Attributes of retirement schemes are randomly generated across choice sets. The order of choice tasks is randomized across participants. No clustering-based treatment assignment; randomization occurs at the individual level within the DCE design.
Randomization unit
The unit of randomization is the individual participant. Within each participant, the attributes of the hypothetical retirement schemes are randomly varied across choice tasks. The order of the choice tasks is also randomized at the individual level.
Clusters
1,200 individuals
Observations
1,200 individuals
Arms
Not applicable (no treatment arms; attributes are randomized within individuals in a discrete choice experiment).
Primary outcomes
The primary outcome is the participant’s choice in each discrete choice task, indicating the preferred retirement scheme between two alternatives. This binary choice outcome reflects individuals’ preferences over delayed retirement policies. Based on these choices, we will estimate the marginal utilities of key policy attributes, including retirement age, pension increment rules, availability of flexible work arrangements after age 60, employment sector (public vs. non-public sector).

Cited by

No citation data available for this registration yet.

Keywords