Menstrual Health and Worker Productivity in the Bangladeshi Garment Sector

We are testing the effectiveness of free Menstrual health classes, and of access to free sanitary pads, in improving MHM management, worker well-being, and work absenteeism.

We are testing three main intervention arms to test their effectiveness in improving MHM management, worker well-being, and work absenteeism.

First, we provide free Menstrual health classes to workers in the factories, provided by an experienced local NGO, that last for about one hour. The MHM classes cover various topics related to MHM, such as the physical basics of menstruation, materials to use to mange menstruation, personal hygiene during menstruation, and how to clean and dispose of menstruation materials. The local NGO has been active in promoting good MHM practice for workers in the garment sector since several years and we use their established information session without further adjustments (except for the added experiment in some sessions - see details below).

Second, we provide access to free sanitary pads, distributed at the factories about twice weekly. Each worker receiving free sanitary pads will receive a voucher for which they can receive a monthly ration of pads over 6 consecutive months. They can redeem the voucher against sanitary pads at the medical room in their factory which will be staffed with outreach workers employed for he project, who distribute the pads on two days a week in a given factory.

Third, we will send some workers regular SMS reminders about the importance of hygienic menstrual health management and the availability of free pads at the workplace.

We further plan to implement two further variations of the treatments. We plan to vary in an exogenous, but for the workers predictable way, the gender of the person distributing the free pads. Each week, there will be two days of distribution of pads in any participating factory (always the same days in the week). One day it will be manned by a male, and the other by a female. The assignment of the week-day to the male and female outreach worker will swap each month. Workers will be told this system at the initial information session when we start the project, and it will be printed on back of the eligibility card they receive (“voucher”). Workers will also get monthly reminders when the order within the week changes.

Second we vary the content of the menstrual health classes, with some workers getting exposed to a short additional module that stresses that pads are more effective in preventing potentially embarrassing leakage, to test whether this stigma ameliorating message is more powerful than a simple health based message in affecting our outcomes. This involves a demonstration of the absortion capacity of menstrual pads vs. standard cloth in an experimental and interactive way.

2018-10-08

2019-06-30

Work Absenteeism (measured through factory HR records at individual worker level)
Adoption of pads (measured both through surveys, through administrative data from pad distribution, and at those factory where available, purchases from factory shop).
Self-reported well-being (surveys).
Willingness to pay for sanitary pads.
Perceived taboos around menstruation.

Self-reported well-being: Individual survey questions and additive index from survey questions.
Willingness to pay for sanitary pads: Through Becker -de Groot Mechanism.
Perceived taboos around menstruation: Elicited through incentivised survey questions (monetary reward of mentioning modal extend of taboo as reported by others in survey). E.g. can a women cook during her period, go to mosque, go to bazaar....

Reported difficulty of working and achieving targets during period (survey based).
Worker Turnover (measured through factory HR records at individual worker level).
Menstrual Health Care Knowledge of Workers.

Menstrual Health Care Knowledge of Workers: E.g. measured through question at baseline and follow-up survey, "Should washed cloth used during menstruation be dried in the sunlight?"

We implement at 2x2 Design of our two basic interventions - menstrual health classes, and access to free pads - with 1000 workers in the initial phase (250 workers in each cell).

The treatments will be assigned at the worker level, randomisation will be stratified at factory level.

Subject to funding, the project will be scaled up to 2000 workers in total.

Phase 1: 1000 Female Workers (Sewing Operators) across 3-5 factories near Dhaka, Bangladesh
(Factories in that area typically have anything between 200 and 10,000 workers. We focus on sewing workers, which typically make up at least half of the work-force, and are usually more than 80% female).

2x2 Design: 250 workers attend one hour long MHM sessions provided by local NGO, 250 get access to free pads over 5-6 months, distributed 2 times weekly at known times at a known place in the factory, 250 get access to both, and 250 to none (control group).
Randomisation stratified at factory level.

Cross-Randomised to basic design:
- Weekly sms reminders on good MHM health to 500 workers
- Every distribution date at each factory will at 50% of days be staffed by male distribution worker, and at other 50% by female distribution worker.
- 50% of MHM sessions will stress ability of sanitary pads to better absorb liquids (blood) compared to traditional remedies (cloth, paper), reducing the danger of "embarrassing" leakage. Other 50% will remain with core-message of health benefits of using sanitary pads.

At computer in office.

Individual workers

No

1000 workers (in Phase 1)
1000 workers (in Phase 2)

1000 workers (in Phase 1) 1000 workers (in Phase 2)

250 workers controls, 250 workers attending MHM classes, 250 worker getting access to free pads, 250 worker getting access to both MHM classes and free pads.

We have good data on worker absenteeism of the same kind we plan to collect during our experiment on our sample worker. We assume that inluding the planned Phase 2 of the project (which will be largely identical to Phase 1), we will eventually have had 2000 workers in combined sample. If we assume we have on average 10 monthly absenteeism observations per worker, and use the empirical mean absenteeism rate (3.2%), its standard deviation (7.6%), and intra-worker correlation (0.23), a sample of 2000 workers will allow to detect a reduction of 18% in absenteeism with power 80%, and 21% with power 90% (two-sided tests).

Menstrual hygiene practices in low-income countries are often limited by lack of finance and information, with potentially adverse consequences for women’s well-being and workplace outcomes. In a randomized controlled trial with around 1,900 female workers from four Bangladeshi garment factories, we relax both constraints individually and jointly by providing free sanitary pads and information on hygienic menstrual practices. Both access to sanitary pads and information improve menstrual practices, either by the adoption of new products, or by knowledge gains and improved use of traditional materials, and both interventions improve health outcomes. However, these positive effects do not translate to better labor outcomes, such as earnings and work
attendance.

Czura, Kristina, Andreas Menzel, Martina Miotto (2023): Improved menstrual health and the workplace: An RCT with female Bangladeshi garment workers. Accepted, Journal of Development Economics

https://www.sciencedirect.com/user/identity/landing?code=WvfrJeyVG5ByKzBbqatcjrsw6EaKgg8r2Vg3V7Xb&state=retryCounter%3D0%26csrfToken%3D3b4c25e9-f4e1-431c-a559-588893875e8e%26idpPolicy%3Durn%253Acom%253Aelsevier%253Aidp%253Apolicy%253Aproduct%253Ainst_assoc%26returnUrl%3D%252Fscience%252Farticle%252Fabs%252Fpii%252FS030438782300130X%26prompt%3Dnone%26cid%3Darp-e86c4e7b-a474-40a9-99c6-13b306fbebbf