Who You Know, Who You Recommend: Referrals and Occupational Gender Segregation

Livia Alfonsi and Pedro Ferreira

No 21964, CEPR Discussion Papers from Centre for Economic Policy Research

Abstract: Employee referrals may reinforce occupational gender segregation, but gendered referrals could simply reflect the composition of workers’ networks. We test whether incumbents also make gendered choices among available candidates in a two-part incentivized field experiment with 1, 294 skilled workers across 12 manufacturing and service occupations in Uganda. In Part I, respondents choose whom to refer for real six-week internships at their workplace from profiles in which candidate gender and post-training experience are cross-randomized. Because candidates enter respondents’ choice sets by design, availability and qualifications are held fixed. Gender non- congruent (GNC) candidates—whose gender does not match the occupation’s majority—are 23.4 percentage points less likely to be referred than gender-congruent candidates. In Part II, respondents nominate candidates from their own networks. These networks are gender-segregated, confirming that availability matters; yet among respondents naming contacts of both genders, GNC contacts are 26.2 percentage points less likely to be referred. Candidate evaluations suggest that the penalty reflects incumbents’ own beliefs and preferences, not only anticipated employer behavior. GNC candidates are rated lower as potential coworkers and less likely to be retained; within expected retention, the largest gap is in perceived task ability, exceeding those in anticipated employer valuation and candidate job enjoyment. With additional experience, evaluations of men entering female-dominated occupations reach or exceed a less-experienced GC baseline, while experienced women entering male-dominated occupations remain heavily penalized on task ability and enjoyment. Across calibrations, a dynamic network model incorporating the experimental referral weights implies that the choice penalty meaningfully amplifies occupational segregation. Expanding cross-gender networks alone may therefore leave an important barrier in place.

Date: 2026-09
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cepr.org/publications/DP21964 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cpr:ceprdp:21964

Ordering information: This working paper can be ordered from
https://cepr.org/publications/DP21964

Access Statistics for this paper

More papers in CEPR Discussion Papers from Centre for Economic Policy Research 33 Great Sutton Street, London EC1V 0DX, UK.
Bibliographic data for series maintained by CEPR ().