Abstract
Recent work has used surveys and interviews to attempt to understand the importance of various constraints to stock market participation (Choi and Robertson, 2021 JF, Chopra and Haaland, 2024 and Duraj et al. 2025). While these papers have uncovered some new barriers to participation and provided much needed structure to the literature on constraints to participation, the survey and interview-based approaches have some natural limits. For example, they might not be able to accurately measure the impact of deep underlying attitudes towards the stock market that constrain participation.
In this project, we conduct a revealed-preferences test of constraints to stock market participation. We do this in the context of sending out invitations to a financial literacy training programme (focused specifically on training people about investing) that we plan to run in cooperation with a German bank. We randomize the content of the invitations to make one reason for non-participation more salient - for instance, one invitation might specifically address the concern that the respondent considers the stock market too risky while another would address the concern that the respondent considers the stock market too difficult. We then track which of our invitations lead to the most enrolments in our training program.