Women's mobility and labor supply: Experimental evidence from Pakistan

Field, Erica Marie; Vyborny, Kate Working Paper Women's mobility and labor supply: Experimental evidence from Pakistan ADB Economics Working Paper Series, No. 655 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: Field, Erica Marie; Vyborny, Kate (2022) : Women's mobility and labor supply: Experimental evidence from Pakistan, ADB Economics Working Paper Series, No. 655, Asian Development Bank (ADB), Manila, https://doi.org/10.22617/WPS220166-2 This Version is available at: https://hdl.handle.net/10419/259493 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ ASIAN DEVELOPMENT BANK ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org Women’s Mobility and Labor Supply Experimental Evidence from Pakistan In this study, we experimentally vary access to a transport service in Lahore, Pakistan, to quantify the overall impact of transport to work on men and women, and the differential impact of transport exclusively for women. The findings show that reducing physical mobility constraints has a large impact on job searching for women, including those who are not searching at baseline. Women’s response is driven by a women-only transport treatment arm, suggesting that safety and social acceptability, rather than simply cost, are key constraints. About the Asian Development Bank ADB is committed to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific, while sustaining its efforts to eradicate extreme poverty. Established in 1966, it is owned by 68 members —49 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. WOMEN’S MOBILITY AND LABOR SUPPLY EXPERIMENTAL EVIDENCE FROM PAKISTAN Erica Field and Kate Vyborny ADB ECONOMICS WORKING PAPER SERIES NO. 655 April 2022 ASIAN DEVELOPMENT BANK ADB Economics Working Paper Series Women’s Mobility and Labor Supply: Experimental Evidence from Pakistan Erica Field and Kate Vyborny No. 655 | April 2022 The ADB Economics Working Paper Series presents research in progress to elicit comments and encourage debate on development issues in Asia and the Pacific. The views expressed are those of the authors and do not necessarily reflect the views and policies of ADB or its Board of Governors or the governments they represent. Erica Field (field.[email protected]) is a professor of economics and Kate Vyborny ([email protected]) is associate director at the DevLab, Duke University. This study was approved by Duke IRB protocol number 2018-0526. Analysis was pre-registered with the AEA registry (AEARCTR-0002410). We are grateful for useful feedback from participants in workshops at Duke, Harvard, IZA, the University of Nottingham, the Asian Development Bank, the IDRC-Urban Institute, the American Economic Association, and the Lahore School of Economics. We thank Fizzah Sajjad, Syed Uzair Junaid, Alieha Shahid, Huzaifa Akhtar, Noor Anwar Ali, Harmalah Khan, and the entire Job Talash team at Centre for Economic Research in Pakistan for outstanding research assistance and programmatic support. We gratefully acknowledge funding from the Asian Development Bank, DFID-IZA GLM-LIC, 3IE, the International Growth Centre, Gates/JPAL Urban Services Initiative, DFID-PEDL, and the National Science Foundation (1629317). Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) © 2022 Asian Development Bank 6 ADB Avenue, Mandaluyong City, 1550 Metro Manila, Philippines Tel +63 2 8632 4444; Fax +63 2 8636 2444 www.adb.org Some rights reserved. Published in 2022. ISSN 2313-6537 (print), 2313-6545 (electronic) Publication Stock No. WPS220166-2 DOI: http://dx.doi.org/10.22617/WPS220166-2 The views expressed in this publication are those of the authors and do not necessarily reflect the views and policies ofthe Asian Development Bank (ADB) or its Board of Governors or the governments they represent. ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The mention of specific companies or products of manufacturers does not imply that they are endorsed or recommended by ADB in preference to others of a similar nature that are not mentioned. By making any designation of or reference to a particular territory or geographic area, or by using the term “country” inthis document, ADB does not intend to make any judgments as to the legal or other status of any territory or area. This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) https://creativecommons.org/licenses/by/3.0/igo/. By using the content of this publication, you agree to be bound bytheterms of this license. For attribution, translations, adaptations, and permissions, please read the provisions andterms of use at https://www.adb.org/terms-use#openaccess. This CC license does not apply to non-ADB copyright materials in this publication. If the material is attributed toanother source, please contact the copyright owner or publisher of that source for permission to reproduce it. ADB cannot be held liable for any claims that arise as a result of your use of the material. Please contact [email protected] if you have questions or comments with respect to content, or if you wish toobtain copyright permission for your intended use that does not fall within these terms, or for permission to use theADB logo. Corrigenda to ADB publications may be found at http://www.adb.org/publications/corrigenda. Notes: In this publication, “$” refers to United States dollars and “PRs” refers to Pakistan rupees. The ADB Economics Working Paper Series presents data, information, and/or findings from ongoing research and studies to encourage exchange of ideas and to elicit comment and feedback about development issues in Asia and the Pacific. Since papers in this series are intended for quick and easy dissemination, the content may or may not be fully edited and may later be modified for final publication. CONTENTS TABLES AND FIGURES iv ABSTRACT vi I. INTRODUCTION 1 II. SETTING AND DESCRIPTIVES 2 A. Enrolling Respondents 3 B. Enrolling Firms 10 C. Job Matching 17 III. EXPERIMENTAL DESIGN 17 IV. ESTIMATION 21 V. RESULTS 22 VI. DISCUSSION 25 REFERENCES 28 TABLES AND FIGURES TABLES 1 Descriptive Statistics and Balance: Female Subscribers 7 2 Descriptive Statistics and Balance: Male Subscribers 8 3 Descriptive Statistics and Balance: Job Ads 15 4 First stage: Effect of Random Assignment of Transport Treatment 22 on Transport Offer Delivered to Subscriber i with Job Ad j 5 Effect of Transport Offer on Job Applications: Women and Men 23 FIGURES 1 Household Listing Enumeration Blocks and Women’s Initial Interest 3 in Signup for Job Talash 2 Job Talash Flyer 4 3 Stated Anticipated Effect of a Transport Offer on Probability of Taking a Job 5 Among Female Latent Jobseekers in Household Survey 4 Education Composition of Job Talash Subscribers 5 5 Any Step Taken to Search for A Job at Baseline – Respondents Who Sign Up for Job Talash 6 6 Gender Differences in Transport Options Among Jobseekers Who Join Job Talash 9 7 Women’s Willingness to Commute Using Various Transport Modes 9 8 Stated Anticipated Effect of a Transport Offer on Probability of Taking a Job 10 Among Female Latent Jobseekers in Household Survey 9 Firm Density Across Firm Listing Enumeration Blocks 11 10 Size of Firms Signing up to Job Talash Ratios 12 11 Industry Classification of Firms Posting Job Ads on Job Talash 13 12 Share of Female Employees By Firm Size 13 13 Comparison of Job Ads Posted on Job Talash Open to Men and Women Between 14 Firms With No Female Employees and Those With at Least One 14 Salaries of Job Ads Posted on Job Talash Open to Men and Women 14 15 Transport Modes Used by Firms’ Current Employees 16 16 Percentage of Firms Offering Pick-and-Drop Service to Employees 16 17 Pick-and-Drop Van Offered to Jobseekers 18 18 Treatment Assignment of Jobseekers by Residential Neighborhood 19 19 Treatment Assignment of Firms by Neighborhood 20 20 Effect of Transport Offer on Women’s Job Application Rates—Extensive Margin 24 21 Effect of Transport Offer on Women’s Job Application Rates, by Discount Level 24 22 Base Price of Transport Service as a Proportion of Posted Salary in Job Ad 26 ABSTRACT In cities with conservative norms or high crime, female workers may face greater restrictions on their physical mobility. This limits women’s labor market opportunities and the pool of workers that firms can attract. In this study, we experimentally vary access to a transport service in Lahore, Pakistan, to quantify the overall impact of transport to work on men, women, and the differential impact of transport exclusively for women. We show that reducing physical mobility constraints has a large impact on job searching for women, including women who are not searching at baseline. Women’s response is driven by a women-only transport treatment arm, suggesting that safety and social acceptability, rather than simply cost, are key constraints. Keywords: transport, mobility, gender, female labor force participation JEL codes: J16, J22, J28, L91 I. INTRODUCTION Urban labor markets in low-income countries are home to a rising number of unemployed, underemployed, and economically inactive people (International Labour Organization 2017). In South Asian cities, women are less likely than men to participate in the labor market, even at relatively high levels of education (Field and Vyborny 2016; Fletcher, Pande, and Moore 2018). Although some of these women are deeply detached and unlikely to participate in response to policy changes, survey data suggest that a large number of women may be “latent jobseekers” who are at the margin of participation if specific barriers can be addressed (Donovan, Lu, and Schoellman 2020): While one quarter of women in Pakistan’s Demographic and Health Surveys work outside the home, another quarter say they are willing to work—suggesting that female labor force participation could double if these women found employment (Field and Vyborny 2016). One of the reasons women may not search for employment is that they have limited physical mobility, due to pervasive stigma, harassment, or violence on public transport or in the public space (ADB 2015, FIA Foundation 2016, Sajjad et al. 2018, Qutub and Anjum 2015). This may limit women’s ability to commute safely. In this study, we investigate the constraints on women’s labor force participation due to limited safe mobility for the daily commute to and from work. We study a representative sample of male and female residents of Lahore, Pakistan, including the employed, unemployed, and economically inactive. We emphasize the importance of studying the latter group, economically inactive “latent jobseekers,” who represent the population with the largest potential benefit from reductions in labor market barriers. This group is particularly important for understanding gender differences, because in settings such as Pakistan a high fraction of the female population are latent workers. Hence, policies that successfully encourage participation may have a particularly large impact on labor markets and female economic empowerment. However, this population is particularly difficult to identify, reach with interventions, and observe, precisely because they are inactive in the labor market, and female “latent” jobseekers often lack independent access to sources of information such as telephones and internet (half the women in our sample share a phone with another household member, typically a male). To fill this gap, we developed a new job matching platform, Job Talash, and enrolled a population including both active and “latent” female and male jobseekers as well as a large sample of firms. This incorporated representative listings of thousands of households and thousands of firms; development, piloting, and refinement of a high-frequency job matching service that lists jobs and delivers information to respondents about them through short messaging service (SMS); and a call center. Job Talash allows us to both observe job search behavior and deliver interventions. This platform generates high-frequency, detailed data on both the supply and demand sides of the labor market for hundreds of thousands of potential job matches between firm and respondent. Because we provide information to both sides, we observe exactly the same information as both sides of the market up to the point of an interview. In the first phase of scale-up of our transport randomized controlled trial (RCT), we show that (1) reducing physical mobility constraints has a large impact on job search for women and not for men; (2) women’s response is driven by a women-only transport treatment arm, suggesting that safety and social acceptability, rather than simply cost, are key constraints for women’s mobility; and (3) women who are inactive at baseline—neither working nor searching—respond to a women-only safe transport 8 ADB Economics Working Paper Series No. 655 Table 2: Descriptive Statistics and Balance: Male Subscribers (1 ) Mixed Transport (2) Control t-test Difference Variable N / [Clusters] Mean / [SE] N / [Clusters] Mean / [SE] (1)-(2) Proportion applied baseline 2,369 0.048 3,479 0.034 0.014 [120] [0.009] [180] [0.008] Working at baseline 2,368 0.590 3,475 0.616 –0.026 [119] [0.012] [176] [0.011] High women’s mobility household 2,360 0.278 3,457 0.263 0.015 [119] [0.013] [176] [0.009] Years o f experience 2,368 9.977 3,479 10.496 –0.519* [120] [0.212] [180] [0.178] Any work experience 2,369 0.916 3,479 0.924 –0.008 [120] [0.007] [180] [0.005] Few female relatives work 2,353 0.620 3,437 0.621 –0.001 [119] [0.015] [176] [0.011] High school or higher education 2,368 0.352 3,475 0.322 0.030 [119] [0.016] [176] [0.017] Age 2,368 29.241 3,475 29.689 –0.448* [119] [0.191] [176] [0.158] * = p < 0.10, N = sample size, SE = standard error. Source: Authors’ calculation using data from Job Talash. Among women who signed up for Job Talash, most reported that they did not have their own mode of transport to get to work (Figure 6). Women’s stated willingness to commute also varied across modes: only about half the female subscribers were willing to take a public wagon (the most crowded mode of travel, with limited separation between women and men); just under 80% were comfortable traveling by rickshaw (Figure 7). More than 90% of subscribers were willing to take a pick-and-drop to work if provided, suggesting the likely safety and acceptability of the transport treatment we offered. Indeed, at baseline, the vast majority of women who expressed some interest in taking a job reported that an offer of a pick-and-drop service would increase their likelihood of taking a job (Figure 8). Women’s Mobility and Labor Supply 9 Figure 6: Gender Differences in Transport Options Among Jobseekers Who Join Job Talash HS = high school. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash platform. Figure 7: Women’s Willingness to Commute Using Various Transport Modes 0 20 40 60 80 100 % of female JT subscribers Public wagon Public bus Rickshaw Pick−and−drop JT = Job Talash. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash platform. 0 10 20 30 40 % of respondents who do not have own transport to work Male Female Less than HS HS or more 10 ADB Economics Working Paper Series No. 655 Figure 8: Stated Anticipated Effect of a Transport Offer on Probability of Taking a Job Among Female Latent Jobseekers in Household Survey Note: The sample size is the 11,138 female respondent who were not currently working and who initially answered 'definitely willing to work' or 'willing to work' when asked 'would you be willing to work if you find a suitable job?' Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash platform. B. Enrolling Firms We listed a representative sample of 10,000 firms across the metropolitan area, using a similar approach as described above for individual respondents, i.e., a cluster-randomized selection of enumeration blocks followed by a listing of all firms in each selected block (Figure 9). A team of enumerators presented the Job Talash service to firms, offering them the opportunity to enroll to list vacancies immediately or later. Firms who signed up through this mechanism were referred to as Sample A. Most firms responding had never advertised jobs on any public platform, and usually recruited through networks. In a second sample, Sample B, we also promoted the service publicly and included firms who self-selected to sign up. Approximately 1,200 firms signed up across these two samples, of which approximately half had posted jobs as of July 2021. These firms are re-approached on a regular basis approximately once every 1–2 months, and approximately 10% of firms post jobs with the service on each round of outreach. Any firm can also call Job Talash to post a job at any time. 0 20 40 60 % Definitely Increase Increase Maybe No difference Definitely no difference EB= enu m Source: A u Fig u m eration blocks. u thors’ calculatio n u re 9: Firm D n s using data fro m D ensity Acr o m Centre for Econ o o ss Firm List o mic Research in Women ’ ing Enumer a Pakistan Wome n ’ s Mobility an d a tion Block s n ’s Mobility Proje c d Labor Suppl y s c t firm survey. y 11 12 ADB Economics Working Paper Series No. 655 Firms of all sectors signed up for the Job Talash platform. Most had fewer than five employees (Figure 10). The majority were in service sectors (Figure 11); this reflects the large proportion of firms in the representative sample in the service sector, as signup rates were similar in this sector to other sectors. The vast majority of employers had no female employees; larger firms were more likely to have female employees (Figure 12). However, a substantial fraction of employers with no female employees who posted jobs opened them to female applicants, although this is not required by law or by the platform (Figure 13). Firms posted ads for jobseekers across education and experience levels. Most job ads offered a listed monthly salary of PRs10,000–PRs30,000 (corresponding to approximately $60–200 per month), as shown in Figure 14; the salary ranges were similar for jobs open to men and women. The average job in the study sample required two years of experience (Table 3). Figure 10: Size of Firms Signing Up to Job Talash Survey question: How many employees work at this location? Please include yourself or anyone who receives a salary or wage, even if you hired someone from your family or someone you know. Note: The sample size of 841 includes firms that signed up for Job Talash. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. 38.3 25.9 18.5 17.2 0 10 20 30 40 % of firms 0−2 3−4 5−9 >=10 Women’s Mobility and Labor Supply 13 Figure 11: Industry Classification of Firms Posting Job Ads on Job Talash Survey question: Based only on your observation from outside the firm, which industry classification best describes this organization? Note: The sample size of 319 includes only firms from the representative sample that listed ads on Job Talash. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. Figure 12: Share of Female Employees By Firm Size Survey question: Out of the employees who work at this location, how many are female? Note: The sample size is 840 firms. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. 0.2 26.5 3.2 6.1 5.2 2.6 1.3 1.0 7.7 2.3 34.5 1.3 6.5 010 20 30 40 % Others Other service activities Human health, social work Education Scientific, technical Real estate Financial, insurance Information, communication Accommodation, food service Transportation, storage Wholesale, retail trade, repair Construction Manufacturing 0 20 40 60 80 100 % 0−2 3−4 5−9 >=10 0% <25% 25−50% 50−75% 75−100% 100% 14 ADB Economics Working Paper Series No. 655 Figure 13: Comparison of Job Ads Posted on Job Talash Open to Men and Women Between Firms With No Female Employees and Those With at Least One Note: Firm survey representative sample: no. of ads = 616; no. of firms = 316. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. Figure 14: Salaries of Job Ads Posted on Job Talash Open to Men and Women Note: Jobs open to both women and men are included in both groups. Minimum salary range for each job is shown. The sample size of 616 includes ads only from the representative sample of firms. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash platform. 0 20 40 60 80 100 % Number of female employees At least 1 female employee Female only Male only Female or male 0.5 1.0 0.5 1.0 2.2 3.1 6.5 9.4 19.1 20.3 28.5 24.5 29.2 27.6 18.1 17.7 01020 30 % > 100,000 75,000−100,000 PRs 50,000−75,000 30,000−50,000 20,000−30,000 15,000−20,000 10,000−15,000 Under 10,000 Female Male Women’s Mobility and Labor Supply 15 Table 3: Descriptive Statistics and Balance: Job Ads t-test t-test t-test (1) (2) (3) Difference Difference Difference Variable N / [Clusters] Mean / [SE] N / [Clusters] Mean / [SE] N / [Clusters] Mean / [SE] (1)-(2) (1)-(3) (2)-(3) Ln(Job posting salary) 92 7.915 188 8.512 96 8.367 –0.597 –0.452 0.145 [16] [0.693] [34] [0.240] [16] [0.250] Job salary not posted 92 0.163 188 0.101 96 0.115 0.062 0.048 –0.014 [16] [0.070] [34] [0.020] [16] [0.027] Years experience required –5% winsorized 88 1.932 182 2.280 95 2.011 –0.348 –0.079 0.270 [15] [0.195] [33] [0.158] [15] [0.240] Ln = natural log, N = sample size, SE = standard error. Note: Years of experience required have been winsorized at 5%, meaning that the top (bottom) 5% of data points were replaced by the maximum (minimum) values at the threshold to minimize the influence of outliers. Source: Authors’ calculation using data from Job Talash, 2019. 16 ADB Economics Working Paper Series No. 655 The status quo modes of commute by firms’ employees differed dramatically by gender. Far more employers reported that more of their male employees traveled to work on a car or motorbike than their female employees, while far more of their female employees traveled to work by rickshaw or public bus than male employees (Figure 15). Very few employers provided transport services for employees (Figure 16). Figure 15: Transport Modes Used by Firms’ Current Employees Survey question: Do employees use the metrobus/use public buses or wagons/rickshaws or qingqis/cars of motorbikes/employer provided pickand-drop/commute on foot at your organization? Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. Figure 16: Percentage of Firms O f fering Pick-and-Drop Service to Employees (In-Person Survey) Survey question: Does your organization have a transport or pick-and-drop service? Note: The sample size is 2,509 firms. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project firm survey. 51.0 51.4 44.4 14.1 41.3 68.2 24.5 9.5 0 20 40 60 80 % of firms On foot Rickshaw Car/motorbike Public bus Female employees Male employees 0 20 40 60 80 100 % of firms 0−2 3−4 5−9 >=10 Yes, for all employees Yes, for some employees No Women’s Mobility and Labor Supply 17 C. Job Matching At the enrollment stage, each firm was invited to list any current vacancies. Firms were also contacted periodically to submit new vacancies. For each new vacancy listed on Job Talash, the team identified matching applicants based on required educational level and years of experience. Job Talash informed applicants about vacancies to which they were matched by SMS and/or phone call. The Job Talash team sent a packet of CVs of interested applicants to the firm, who then chose whether or not to call them for interviews. After several days, the Job Talash team called the firm to gather follow-up data on the applicants and interviews. Approximately 8,500 of the subscribers in the sample received a job ad for at least one job over the period we study in this paper. Because we provided information to both sides, we observed exactly the same information as both sides of the market up to the point of an interview. This allowed us to experiment easily with this information and reduced the scope for analysis to be distorted by unobserved information. Each observation in our study was one ad sent to one respondent who met the basic qualifications for that job. Our key outcome variable of interest to study labor supply responses was an indicator for whether respondent i applied to job ad j. The key advantage of this outcome variable, collected on the Job Talash platform, is that it isolates the initial labor supply decision, separate from decisions that the employer may make to select applicants based on gender, distance, physical mobility/transport options, or the interaction of these factors. In the control group, approximately 14% of respondents applied to any job; conditional on applying, a respondent applied on average to approximately 8.6% of job ads sent. The overall per-job application rate was thus approximately 1%. This per-job application rate was expected to be this low for two reasons: first, each jobseeker received multiple job ads; second, we deliberately recruited economically inactive individuals who were neither working nor searching for jobs at baseline. III. EXPERIMENTAL DESIGN Using the Job Talash platform, we tested the overall impact of an offer of transport to work on men, women, and the differential impact for women of transport exclusively for women. We experimentally varied access to a subscription service for daily commuting and studied how this affected job search. With each job advertised, all respondents received information on the salary, the distance to the job, and the cost of a rickshaw ride from their home to the office—the status quo commute mode for many women. Treatment group respondents also received an offer of a pick-and-drop service available to reach the specific job in the ad, i.e., a door-to-door subscription 12-seat wagon that would take them to and from the workplace every day (Figure 17). The transport was for commuting daily on the basis of a monthly subscription, not transport to apply or interview for the job. This kind of “pick-and-drop” service is a familiar concept in this context, as a select number of large firms offer it as a benefit or a paid subscription to their employees. The limited use of this option by firms could reflect a combination of factors including coordination costs across small firms, and the difficulty of upfront investment in a transport service when firms are uncertain about how many female employees might join and when. Effectively, our treatment overcame those coordination costs to make the service available to a randomized sample of jobseekers for all jobs in treated areas. 24 ADB Economics Working Paper Series No. 655 Figure 20: Effect of Transport Offer on Women’s Job Application Rates—Extensive Margin Note: Standard errors two-way clustered by jobseeker cluster and firm cluster. Confidence intervals shown are 90%. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash job search platform. Figure 21: Effect of Transport Offer on Women’s Job Application Rates, by Discount Level Note: Standard errors two-way clustered by jobseeker and firm cluster. Confidence intervals shown are 90%. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash job search platform. Proportion of matches treated − female only transport Prop. matches treated − mixed transport −.2 0 .2 .4 .6 .8 Change in proportion who apply to any job Full sample Economically inactive at baseline Offered female only transport at base price Offered female only transport at medium discount (50%) Offered female only transport at large discount (75%) Offered mixed transport at base price Offered mixed transport at medium discount (50%) Offered mixed transport at large discount (75%) −.05 0 .05 Change in proportion who apply to job j Women’s Mobility and Labor Supply 25 There is a clear gradient in the point estimates over the discount level. At the base price, the estimated effect is almost exactly zero. Participants who received a transport offer with a 50% discount applied at three times the rate of the control group. Those who received an offer with a 75% discount applied at four times the rate of the control group. Although we cannot reject that the coefficients are equal with the data collected in this phase, the pattern of results demonstrates that women’s significant overall response to the women-only transport offer shown in Table 5 is driven by the subsidized arms. The estimated coefficient for female-only transport is more than twice the estimate for the mixed transport offer, although we cannot reject that the two effects are equal. The pattern of results is suggestive that not only cost and convenience, but also safety and stigma play an important role in mobility constraints on female labor supply. Finally, we examine whether the transport offer led economically inactive “latent jobseekers” who were not working or searching at baseline, but signed up for Job Talash, to apply for jobs. Figure 20 shows the results. The estimates are imprecise; however, the point estimate suggests a large potential effect on the extensive margin for female-only transport offer, and a zero effect for the mixed-gender transport offer. The point estimate for the full sample is 0.2, suggesting that if half the job ads the respondent received were accompanied by a transport offer, her probability of applying for any job on the platform increased by 10 percentage points. The point estimate is slightly larger for respondents who reported that they were neither working nor searching at baseline. While these results are preliminary, the pattern suggests that reducing safety and acceptability constraints on women’s mobility could ultimately bring more women into the labor force. VI. DISCUSSION These results show a large and gendered effect of constraints to physical mobility on women’s labor supply. The pattern of results, driven by responses to the female-only transport offer, is consistent with safety and social norms being key constraints. Given the apparent importance of these constraints, why do private markets not already provide a service such as the one provided in our experiment? As shown in Figure 21, the job application response in our experiment was driven by the discounted treatments. These were priced at below the break-even point for the service, i.e., the price that would cover the cost of vehicle rental, driver salary, and fuel when 10 of the 12 seats on the vehicle are booked. Figure 22 shows the base price of the transport service, which is approximately 80% of the cost of a rickshaw fare, as a proportion of the posted salary. The X axis breaks down the number of job matches respondents received by quintiles of distance from their homes. All the jobs advertised were within the city of Lahore. However, the cost for the service to break even amounts to more than half the posted salary for more than three-quarters of the job ads respondents received. 26 ADB Economics Working Paper Series No. 655 Figure 22: Base Price of Transport Service as a Proportion of Posted Salary in Job Ad Q = quartile. Source: Authors’ calculations using data from Centre for Economic Research in Pakistan Women’s Mobility Project Job Talash platform. Clearly, paying the full cost of such a transport service would make most potential job matches infeasible for candidates, severely limiting the scope of their job market. Similarly, it would be prohibitively expensive for employers with no female employees who might consider hiring a female candidate; this represents a large percentage of employers in Lahore (Figure 12). In practice, it is only provided by a tiny minority of employers (Figure 16). Grouping employers together so that pick-and-drop routes can be more direct and efficient makes it more cost effective; however, coordination costs are high (e.g., finding people who live and work along a similar route, and work the same hours), which may explain why this practice is uncommon. Ride-hailing services such as Uber and Careem, which are recent entrants to the market, automate this process; but they leave female riders alone with a typically male driver, and thus are often perceived not to address the constraint of safe, acceptable mobility that the group pick-and-drop van provides. However, there is scope for group transport to become more cost effective over time. Given that women candidates are responsive to transport, if the mobility constraints were relaxed in the short term, firms might start to hire more women. Over a longer period, firms may move to a new equilibrium with more female workers, leading to economies of scale in transport. On the worker side, young women who are on the margin of entering the labor force after completing education might find the cost of mobility an important constraint in the early part of their careers. If they get past this hurdle, as they build skills and human capital, transport costs would ultimately become a smaller fraction of their wages. 0 20 40 60 80 Transport service price as % of salary in job ad Distance Q1 Distance Q2 Distance Q3 Distance Q4 Women’s Mobility and Labor Supply 27 These factors, in addition to existing literature demonstrating the importance of women’s economic empowerment for their wellbeing and for the wellbeing of members of their families, suggest that policies that support or subsidize women’s mobility may be called for. The intervention that has the most pronounced effects of what we studied is a reserved space only for women. Thus, our results are relevant to the ongoing policy debate on reserved spaces for women. Our results are consistent with Kondylis et al. (2020), who show that women value these spaces. We further show that offering such an option might help women who wish to do so leave their homes and reach the workplace. However, with our results to date we cannot speak directly to the real concern that they may reinforce attitudes that blame the victim for harassment (Kondylis et al. 2020). More importantly, our experimental results demonstrate the importance of women’s mobility for labor supply. We wish to stress that they do not imply that reserved spaces are the only way to achieve this. We anticipate that these results speak to the economic and labor market importance of a broader set of policies designed to help ensure women’s mobility, including interventions such as high-quality public transportation, streetlights, and gender-sensitive policing, as well as interventions designed to shift norms directly. REFERENCES ADB. 2015. Policy Brief: A Safe Public Transportation Environment for Women and Girls. Manila: Asian Development Bank. Aguilar, Arturo, Emilio Gutiérrez, and Paula Soto Villagrán. 2019. “Benefits and Unintended Consequences of Gender Segregation in Public Transportation: Evidence from Mexico City’s Subway System.” Economic Development and Cultural Change 69 (4): 1379–410. Bernhardt, Arielle, Erica Field, Rohini Pande, Natalia Rigol, Simone Schaner, and Charity Troyer-Moore. 2018. “Male Social Status and Women's Work.” AEA Papers and Proceedings 108: 363–67. Bertrand, Marianne, Emir Kamenica, and Jessica Pan. 2015. “Gender Identity and Relative Income within Households.” The Quarterly Journal of Economics 130 (2): 571–614. Borker, Girija. 2021. “Safety First: Perceived Risk of Street Harassment and Educational Choices of Women.” Policy Research Working Paper No. 9731. World Bank, Washington, DC. Borker, Girija, Gabriel Kreindler, and Dev Patel. 2020. Women’s Urban Mobility Barriers: Evidence from Delhi’s Free Public Transport Policy, IGC Final Report. London: International Growth Centre. Burde, Dana, and Leigh L. Linden. 2013. “Bringing Education to Afghan Girls: A Randomized Controlled Trial of Village-Based Schools.” American Economic Journal: Applied Economics 5 (3): 27–40. Bursztyn, Leonardo, Alessandra L. González, and David Yanagizawa-Drott. 2020. “Misperceived Social Norms: Women Working Outside the Home in Saudi Arabia.” American Economic Review 110 (10): 2997–3029. Cheema, Ali, Asim I Khwaja, Farooq Naseer, and Jacob N Shapiro. 2020. “Glass Walls: Experimental Evidence on Access Constraints Faced by Women.” Unpublished. Chowdhury, Iffat Ara. 2019. “News Reports on Sexual Assault and Women’s Labor Supply: Evidence from Bangladesh.” Mimeo. University of Washington, Seattle, WA. Christensen, Peter, and Adam Osman. 2021. “The Demand for Mobility: Evidence from an Experiment with Uber Riders.” IZA Discussion Paper No. 14179. Institute of Labor Economics, Bonn. Donovan, Kevin, Will Jianyu Lu, and Todd Schoellman. 2020. “Labor Market Dynamics and Development.” Staff Report No. 596. Federal Reserve Bank of Minneapolis, Minneapolis, MN. Fernández, Raquel. 2013. “Cultural Change as Learning: The Evolution of Female Labor Force Participation over a Century.” American Economic Review 103 (1): 472–500. FIA Foundation. 2016. “Safe and Sound: International Research on Women’s Personal Safety on Public Transport.” Research Series Paper 6. London. References 29 Field, Erica, Rohini Pande, Natalia Rigol, Simone Schaner, and Charity Troyer Moore. 2021. “On Her Own Account: How Strengthening Women's Financial Control Impacts Labor Supply and Gender Norms.” American Economic Review 111 (7): 2342–75. Field, Erica and Kate Vyborny. 2019. “Gender, Mobility and Labor Market Outcomes.” The American Economic Association's Registry for Randomized Controlled Trials No. 2410. November 19. ————. 2016. “Female Labor Force Participation in Asia: Pakistan Country Study.” Background report. Asian Development Bank, Manila. Fletcher, Erin K., Rohini Pande, and Charity Troyer Moore. 2018. “Women and Work in India: Descriptive Evidence and a Review of Potential Policies.” HKS Faculty Research Working Paper Series RWP18-004. Harvard Kennedy School, Cambridge, MA. International Labour Organization. 2017. Global Employment Trends for Youth 2017: Paths to a Better Working Future. Geneva. Jacoby, Hanan G., and Ghazala Mansuri. 2015. “Crossing Boundaries: How Social Hierarchy Impedes Economic Mobility.” Journal of Economic Behavior & Organization 117: 135–54. Jayachandran, Seema. 2021. “Social Norms as a Barrier to Women’s Employment in Developing Countries.” IMF Economic Review 69 (3): 576–95. Kondylis, Florence, Arianna Legovini, Kate Vyborny, Astrid Zwager, and Luiza. Andrade. 2020. “Demand for 'Safe Spaces' : Avoiding Harassment and Stigma.” Policy Research Working Paper No. 9269. World Bank, Washington, DC. Muralidharan, Karthik, and Nishith Prakash. 2017. “Cycling to School: Increasing Secondary School Enrollment for Girls in India.” American Economic Journal: Applied Economics 9 (3): 321–50. Qutub, Syed Ayub, and Nomana Anjum. 2015. “Urban Open Spaces for Adolescent Girls: An Assessment for Islamabad and Rawalpindi, Pakistan.” PSSP Working Paper No. 27. International Food Policy Research Institute, Washington, DC. Sajjad, Fizzah, Ghulam Abbas Anjum, Erica Field, and Kate Vyborny. 2018. “Overcoming Barriers to Women‘s Mobility Improving Women‘S Access to Public Transport in Pakistan.” GLM|LIC Policy Brief No. 24. Growth and Labour Markets in Low Income Countries Programme, Institute of Labor Economics, Bonn. Siddique, Zahra. 2018. “Violence and Female Labor Supply.” IZA Discussion Paper No. 11874. Institute of Labor Economics, Bonn. Velasquez, Andrea. 2019. “The Economic Burden of Crime: Evidence from Mexico.” Journal of Human Resources 55: 0716-8072r2. ASIAN DEVELOPMENT BANK ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org Women’s Mobility and Labor Supply Experimental Evidence from Pakistan In this study, we experimentally vary access to a transport service in Lahore, Pakistan, to quantify the overall impact of transport to work on men and women, and the differential impact of transport exclusively for women. The findings show that reducing physical mobility constraints has a large impact on job searching for women, including those who are not searching at baseline. Women’s response is driven by a women-only transport treatment arm, suggesting that safety and social acceptability, rather than simply cost, are key constraints. About the Asian Development Bank ADB is committed to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific, while sustaining its efforts to eradicate extreme poverty. Established in 1966, it is owned by 68 members —49 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. WOMEN’S MOBILITY AND LABOR SUPPLY EXPERIMENTAL EVIDENCE FROM PAKISTAN Erica Field and Kate Vyborny ADB ECONOMICS WORKING PAPER SERIES NO. 655 April 2022