Job attributes and the gender retirement income gap
How to cite
Kesternich, Iris; Schouwer, Thimo De; Siflinger, Bettina. 2021. "Job attributes and the gender retirement income gap." AEA RCT Registry. AEARCTR-0007767. https://doi.org/10.1257/rct.7767-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Recent evidence suggests that workers are willing to sacrifice a significant portion of their wages to work jobs with better job attributes (e.g., flexibility in scheduling or the ability to work from home). In our project, we will introduce a hypothetical choice experiment into a Dutch Panel survey (LISS) in order to investigate differences in the valuation of job attributes. We are particularly interested in gender differences. In a second step, we will analyze how preferences for non-wage attributes relate to retirement income. Trading off wages for job attributes has an immediate effect on wages, but this implies important consequences for eventual retirement income. To investigate whether individuals take into account the trade-off between current non-wage attributes and future retirement income, we introduce a treatment variation in a discrete choice experiment that makes the long-term consequences salient.
Experimental design
We use a discrete choice experiment to elicit survey respondents' stated preferences for job attributes. To analyze whether individuals consider the long-term effects of their job choices, we introduce a treatment variation before the start of the discrete choice experiment. This variation consists of two parts. First, we ask our treatment group (25% of our 55-70 year old respondents, randomly selected) to state how big they believe the gender gap in pension income is. This way, they are forced to think about retirement income and its distributional aspects. Second, we provide them with the OECD estimate of the gender pension gap. This informs our respondents about the actual magnitude of the gap. Keeping this information in mind, the treated group is then asked to answer the same job-tradeoff questions as the control group, who are not reminded about retirement income gaps.
To investigate whether the valuation of job attributes is different between the two groups, we estimate treatment – and control-group specific preferences and test whether they are equal. We analyze gender heterogeneity in the valuation and responses to treatment to gain additional insights.
Design details
Cited by
No citation data available for this registration yet.