Freeing Women's Time, Shifting Household Power

Arm 1 — Solvatten device, 360 households. A free Solvatten household solar water-treatment canister, distributed by World Vision Tanzania, with a short standardised demonstration of use, cleaning and storage at handover. No cash, no employment information, and no gender-related messaging accompanies delivery.

Arm 2 — Cash, 180 households. An unconditional transfer in two equal cells of 90: one calibrated to the local cost of safe drinking water, the other to the value of the Solvatten canister.

Arm 3 — Control, 360 households. Survey and economic games only. No device and no transfer.

Universal end-of-endline module — the dyadic job-offer experiment. At the very end of the endline interview, every sampled household in all three arms receives an identical offer: entry into a lottery for a paid community job for the wife, such as street cleaning, at the going wage. Consent is elicited privately and separately from each spouse, and neither is told the other's answer. Only dual-consenting households enter the lottery, and the selection probability among them is deliberately low so that a household without a job cannot infer who refused.

The job offer is not randomised and is not cross-randomised with device assignment. All identifying variation for the consent outcomes comes from the parent trial's three-arm household randomisation.

2026-09-05

2026-12-10

P1. Wife's willingness to accept — binary, 1 if she privately consents to the household entering the job lottery.
P2. Husband's willingness to accept — binary, defined identically.
P3. Consent typology — Both consent / Wife only / Husband only / Neither.

S1–S2. Female labour supply, extensive (any paid work in the last 7 and 30 days; any work outside the homestead) and intensive (hours in the last 7 days; days in the last 30).
S3. Earnings: the wife's cash earnings in the last 30 days; household cash earnings; her share.
S4–S5. Occupation and search: sector and type of main activity and whether it is performed outside the homestead; job search in the past four weeks, hours spent searching in the past week, willingness to work outside the home.
S6. Time use: time spent on water collection, boiling and fuel gathering, illness caregiving, domestic work, and income generation, for each spouse.
S7. Bargaining power: the wife's say over purchases, her earnings, her employment and her mobility; an attitudes-to-female-work module administered separately to each spouse.
S8. Beliefs: spouses' beliefs about each other's labor market opportunities.
S9. Water and health (first stage): whether the household treats its water, illness among household members in the past four weeks, days of illness care given by the wife.

A three-arm, household-level randomised controlled trial in six villages of Uyui District, Tabora Region — Kalangale, Kizengi, Lutona, Misole, Mwakadala and Nyankombe — with a universal elicitation module at endline.

The six villages were selected because they contain unimproved dug wells without treatment and broken or missing pumps, so publicly provided safe water is limited and roughly comparable ex ante. The sampling frame was built from building footprints inside village boundaries; 150 main-sample households were drawn per village, giving 900. Enumerators screen each geo-located building for an eligible couple and interview two adults per household, yielding approximately 1,800 respondents per round. Kizengi is a pure-control village receiving no devices, which bounds spillovers.

The endline round, roughly eleven weeks after distribution, closes with the job-offer module. Identifying variation for the primary outcomes is the parent trial's household-level arm assignment; the job offer is a constant. The lottery among consenting households makes the offer credible and creates deniability.

Not available

Assignment of the Solvatten canisters is at the household level, stratified by village, with a fixed recorded seed, performed in R before any field activity. Kizengi receives no devices; cash and control households there are allocated in the same proportions as elsewhere, scaled to fill the 150-household quota. The employment lottery is conducted via similar software at the village level.

The household, stratified by village. The offer to enter the job lottery is not randomised; it goes to every sampled household. The subsequent employment lottery is randomised at the household level within each village, conditional on dual consent.

Yes

900 households, across six villages in Uyui District.

1,800 (2 adults x 900 households) per round (baseline, endline)

Arm 1 — Solvatten device: 360 households
Arm 2 — Cash: 180 households
Arm 3 — Control: 360 households