Connectors and Influencers
We study a setting in which personal acquisition of information is costly and maintaining personal contacts to share information socially is also costly. The theory predicts star networks, with small groups sustaining an active hub and large groups allowing in addition for an inactive hub equilibrium. Experiments with small (8 subjects) and large (50 subjects) groups show star-like networks throughout. In small groups, behavior matches the active hub prediction. In large groups, hubs are inactive or engage in activity far in excess of the active hub equilibrium depending on how much information subjects are given on their own and others’ earnings. The hub’s activity level shapes relative earnings of subjects and the overall performance of the group. This paper was accepted by Roberto Weber, behavioral economics and decision analysis. Funding: This work was supported by H2020 Future and Emerging Technologies [IBSEN Contract 662725]; the Keynes Fund, University of Cambridge; and the Creative-Pioneering Researchers Program, Seoul National University. S. Goyal acknowledges financial support from the Center of Behavioural Institutional Design and Tamkeen [New York University Abu Dhabi Research Institute Award CG005]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.07308 .
Authors
- Frédéric Moisan (ORCID: https://orcid.org/0000-0003-1522-0463)
- Syngjoo Choi (ORCID: https://orcid.org/0000-0003-4006-0377)
- Sanjeev Goyal (ORCID: https://orcid.org/0000-0002-9827-5998)
Institutions
- Centre National de la Recherche Scientifique (FR)
- Seoul National University (KR)
- Université Lumière Lyon 2 (FR)
- University of Cambridge (GB)
- Groupe d'Analyse et de Théorie Economique Lyon St Etienne (FR)
- Université Jean Monnet (FR)
- École de management de Lyon (FR)
- Université Jean Monnet (FR)
Publication Details
- Journal
- Management Science
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1287/mnsc.2024.07308
- Primary Topic
- Game Theory and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00