Men’s Reactions to the Long-Term Implications of Part-Time Work

To test whether learning about the long-term financial consequences of having a reduced workload impacts men’s behaviors and decision-making, we randomly assign an informational video containing three main dimensions of objective information about these long-term consequences:
a. Information on the total earnings lost in the long-term
b. Information on the financial consequences for pensions
c. Information on the financial risk in case of adverse events (such as divorce) for the reducing partner.
The video also puts the life-time financial losses in perspective to the short-term childcare costs. To illustrate these concepts, the video presents the decision-making process of a female teacher regarding her workload together with her partner.

The control group also receives a video but on an unrelated topic. We use a video created by the national public television as part of their regular programming (https://www.srf.ch/). In particular, the video discusses research on the impact of digital media, particularly screens like smartphones and tablets, on children's development, language acquisition, and family relationships.

2023-11-13

2024-01-01

We have three main research questions:

1) How do men adjust their and their partners’ (desired) workload intentions after being exposed to information about the long-term consequences of part-time work? Would this adjustment depend on their partner's career prospects?
2) Do men recommend similar workload adjustments for men and women working part-time?
3) Are men willing to discuss the consequences of part-time work with their partners?

To answer the first research question, we will measure workload intentions for upcoming years, and workload next year from administrative data.

To answer the second research question, we design a vignette to measure if recommended workload adjustments and the factors considered depend on the gender of the person with reduced workload portrayed in the vignette. Half of the sample (within each treatment and control group) sees this vignette portraying a female, while the other half sees a version that reverses the gender.

To answer the third research question, we measure participants willingness to share the survey and video with their partner, by offering a raffle of an additional voucher worth 500 CHF. We also directly ask participants if they intend to discuss any aspect of the video with their partner (and other people), and which specific topics they plan to address.

The field experiment will be implemented by inviting male teachers via email and physical letters.

In Region A, we will generate individualized survey links for every school employee, and the regional authorities will oversee the distribution of survey invitations through letters. In Region B, the regional authorities will establish contact with school principals by email, providing them with the survey link to our study. In region C, we will contact the school principals by email, providing them with the survey link to the study. Subsequently, the school principals will forward this invitation to the teachers within their respective schools, also through email.

In terms of timeline, we time receipt of the letter or email during the second half of November 2023, before teachers start conversations with school principals about their desired workload for the upcoming academic year, which usually takes place beginning of December.

We randomize participants entering the survey into treatment and control at the individual level. We will collect survey data and we will also merge it with administrative data on workload provided by employers.

Not available

Individual-level randomization performed with Qualtrics.

We randomize participants entering the survey into treatment and control at the individual
level.

No

At least 400 individuals
We do not have a predetermined number of observations planned, as invitations to the study
will be sent by the regional authorities and the school principals, and it will ultimately depend
on the response rate.

At least 400

200 observations control, 200 observations treatment