Flexible Work Arrangements and Firm Outcomes
Policy Context
As countries develop, they undergo structural changes in the organization of labor – particularly, a shift from self-employment to employment in firms (Breza and Kaur, 2025; Bandiera et al, 2022). This makes understanding frictions to wage-employment an important topic for both research and policymaking in low-income countries. Recent work attributes the prevalence of high self-employment rates in low-income countries to workers’ preferences for flexibility (Cefala et al, 2025) – this can be a consequence of under-development itself, where the prevalence of health shocks, lack of formal care infrastructure, or dependence on social ties may contribute to such preferences (Breza and Kaur, 2025). However, providing flexibility may be costly for firms, especially when production occurs in teams, such as assembly line manufacturing.
This project studies the trade-off between firm and worker preferences, with the aim to shed light on a new type of labor market friction that may affect firm productivity. I study small-medium factories in India, where workers are hired on piece rates, giving them the flexibility to choose their own work schedules. In the status quo, there is significant dispersion in worker arrival times at the firm, and firm owners agree that such flexibility affects their total output.
Chandra conducts an RCT with 90 small-medium factories and their workers in India to test whether flexibility affects firm output. The RCT has been completed with 60 firms and researchers will be implementing the experiment with an additional sample of 30 firms in August 2025.
Study Design
The research team collect detailed, daily attendance and production data from sample firms for a period of 6-weeks. At the beginning of the study, firms are randomized into a treatment and control group; workers in treatment firms are incentivized to follow fixed-work schedules for a 4 out the 6-weeks. Treatment incentives are designed as a daily bonus of Rs. 100 (15% of average daily earnings), paid conditional on arriving at the firm by the owners’ pre-specified work start time. The incentive thus encourages both attendance and before-time arrival. Workers in control firms receive unconditional incentives for the same period, matched to a randomly chosen treatment firm each week to address concerns about income effects on worker productivity.
Researchers will use daily attendance data and arrival times observed by enumerators to estimate firms’ cost of incentivizing workers to show up for work and arrive on time. Researchers will use data on worker task assignments, daily and weekly output and earnings to estimate the effects of the incentive on worker output. Finally, using information on the task-content of firms’ products, researchers will construct a measure of overall firm output (using total worker output on the last task in the production line) to estimate the effect of incentives on firm-level output. These results will allow the research team to assess both whether flexibility affects firm productivity, and how gains in firm profits compare with payments made to workers.
Finally, using endline survey data on owner time-use in the firm, and worker time-use before/during the work day, researchers will investigate how incentives affect the allocation of (1) production versus management roles between owners and workers, and (2) workers’ time in household work, leisure and economic activity, given both preferences for flexibility, and social norms around household work and caregiving activities.
These results together will shed light on the role of worker preferences on firm productivity and firm organization in developing countries.
Results and Policy Lessons
This pilot is ongoing and findings are forthcoming.
