工资补贴促进女性就业：来自巴基斯坦的证据

Wage Subsidies to Promote Female Hiring Evidence from Pakistan Maurizio BussoloJean LeeMahreen MahmudNayantara SarmaAnaise Williams South Asia RegionOffice of the Chief EconomistFebruary 2026 A verified reproducibility package for this paper isavailable athttp://reproducibility.worldbank.org,clickherefor direct access. Policy Research Working Paper11317 Abstract Can employer-side wage subsidies increase hiring womenin low female labor force participation settings? This papertests this using a randomized experiment with 1,227 Paki-stani firms on a national jobs platform. Treatment firmswere offered a six-month wage subsidy determined viathe Becker–DeGroot–Marschak mechanism. They were11 percentage points more likely to hire a woman, with larger effects for male-only firms. After 18 months, thetreatment effect on employing a woman persisted, althoughthe firm-wide share of female employees did not change.Additionally, administrative data show the treated firmsreduced male-preference language in job postings, consis-tent with emerging demand-side shifts. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about developmentissues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry thenames of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely thoseof the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank andits affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Wage Subsidies to Promote Female Hiring: Evidence from Pakistan∗ Maurizio Bussolo, Jean Lee, Mahreen Mahmud, Nayantara Sarma, Anaise Williams 1Introduction Globally, female labor force participation (FLFP) averages around 50 percent – over 20percentage points lower than that of men – and is substantially lower in regions such asthe Middle East and North Africa and South Asia (International Labour Organization,2025; World Economic Forum, 2020). The supply-side correlates of women’s participation –education and skills, norms and attitudes, transport and childcare access, and safety – arewell documented (Heath et al., 2024; Borker, 2025; Jayachandran, 2021; Groh et al., 2016;Compton and Pollak, 2014; Garlick et al., 2025; McKelway, 2025; Dean and Jayachandran,2019; Afridi et al., 2022). By contrast, less is known about demand-side frictions and policiesthat relax them, even though audit and field evidence documents gender gaps in hiring(Gentile et al., 2023; Jensen, 2012). Motivated by the limited experimental evidence on demand-side interventions and thepersistence of gender gaps in hiring, we evaluate an employer-targeted wage subsidy thatlowers the cost of hiring women.We run the experiment in Pakistan, where FLFP isapproximately 24 percent, one of the lowest rates globally.Our sample consists of 1,227small, medium, and large firms, recruited via the country’s largest online job portal.Ina randomized controlled trial, firms with an active job opening were randomly offered asix-month subsidy for hiring a woman.1 This design allows us to estimate the effect of atemporary women’s wage subsidy on initial hiring and on women’s longer-run employmentafter the subsidy expires, and examine whether it shifts revealed gender preferences in futurejob advertisements and hiring. The sampling frame comprised all vacancies posted on Rozee.pk, our partner job platform,between April and November 2022 that stated no gender preference.Due to budgetconstraints, we restricted our sample to ads offering monthly salaries below PKR 80,000(about USD 1,700 PPP). Vacancies largely involved technical and professional roles and,given the salary cap, were predominantly entry-level. A novel feature of our design is the use of the Becker–DeGroot–Marschak (BDM)mechanism (Becker et al., 1964) to elicit firms’ willingness to hire a woman.2In the treatment group, each firm states the minimum six-month subsidy it would accept to hire a woman; asubsidy is offered only if the stated amount does not exceed a computer-generated randomdraw. Our implementation yields a revealed-preference measure of the subsidy required toequalize hiring choices and lets us quantify how far a targeted subsidy can offset male-leaninghiring preferences.On average, treatment-group firms report needing about a 15 percentsubsidy to hire a woman, well below the subsidy magnitudes tested in prior wage-subsidyexperiments on gender (McKenzie, 2017). Our key finding, based on pre-specified intent-to-treat estimates, is a 10.7 percentagepoints increase in hiring of a woman for the advertised position. This is robust to Lee (2009)bounds for differential attrition. Within the treated group, we implemented the BDM: eachmanager reported the minimum