Digital financial services are widely promoted as tools to expand women’s economic empowerment. Mobile money, in particular, has been linked to higher savings, improved resilience, and greater financial agency in many low-income settings (Riley, 2024, Heath and Riley 2024, Aker et al., 2016, Suri and Jack, 2016, Dizon et al. 2020). Yet a fundamental question remains: how does mobile money actually empower women?
A common explanation is privacy. If money can be stored and transferred digitally, women may be better able to keep resources hidden from others in the household. Another possibility is that digital payments introduce transaction frictions that make funds harder to appropriate. A third potential mechanism is earmarking: mobile money may change how funds are perceived and claimed within households, strengthening women’s sense of ownership even when privacy is limited.
We study these mechanisms using a lab-in-the-field experiment with 235 married women in Mwanza, Tanzania, recruited from current and former clients of BRAC Tanzania’s microfinance program. Participants made real, incentivized financial decisions, allowing us to observe behavior rather than rely on self-reports. In our games, financial control and empowerment are defined behaviorally through the choices women make. In some situations, women face a tradeoff between maximizing income and retaining control over resources – for example, choosing a less profitable option if it allows them to better claim or manage the money. We interpret empowerment as the ability to choose the income-maximizing option regardless of whether payments are public or private, or whether money is received in cash or digitally. In this sense, empowerment reflects practical control over resources, revealed through decisions rather than self-reported attitudes.
The experiment consisted of two simple games designed to isolate privacy, earmarking, and transaction costs. In the first game—the mobile money preference game—women made a series of straightforward choices. Each time, they chose between receiving a fixed amount of cash or a gradually increasing amount of money sent to their mobile phone. For every choice, we simply asked: Would you rather receive cash, or mobile money? By seeing how large the mobile money payment had to be before a woman switched away from cash, we could measure how much she valued mobile money relative to cash. To understand the role of privacy, we randomly varied whether these choices were private or public. Some women were told that their husband would be informed about both the amount they received and whether it was paid in cash or mobile money. Other women were told that this information would remain completely private. If privacy is a key benefit of mobile money, we would expect women to value mobile money less when their choice is public—because it no longer offers any privacy advantage over cash.
In the second game—the investment game—women were given a sum of money (endowment) and asked to decide how to divide it between two options. One option was safe and guaranteed. The other was a risky investment that could generate much higher returns, but might also pay out nothing. A key feature of this game is that husbands were not told how much money the woman received at the start. This created an opportunity for concealment: if a woman received a large endowment, she could choose to invest only part of it and keep the rest hidden. We again varied whether these decisions were made in public or in private. In addition, we changed how any earnings were paid out. Some women were paid in loose cash, others received cash in a clearly labeled envelope (to signal that the money was meant for the woman), and others were paid via mobile money. The labeled cash allows us to separate the effect of digital payments from a simpler explanation: that clearly marking money as personally owned—rather than the technology itself— changes how women use it.
Across both games, privacy plays a surprisingly limited role. Making women’s choices public does not significantly reduce how they value mobile money relative to cash in the first game, nor does it substantially change their investment behavior in the second game.
Figure 1 shows what we learn from the investment game. The horizontal axis measures how likely women are to hide income, and each dot shows the effect of a different condition compared to receiving loose cash. First, simply making decisions public—so that husbands can observe them— does not change behavior very much. On its own, public observability has little effect on whether women hide income. This mirrors what we saw in the first game.
These findings suggest that women’s empowerment through digital finance does not depend on privacy. Even when spouses know about an account, mobile money can strengthen women’s financial control by clarifying which resources are “hers” rather than part of a shared household pool.
For policymakers and practitioners, this highlights the importance of product design not just access. Features such as labeled wallets, goal-based subaccounts, or clearly framed transfers may be powerful tools for strengthening women’s financial agency. Behavioral design that reinforces ownership and purpose may be as important as expanding access or protecting secrecy.
In short, mobile money appears to empower women not primarily by helping them hide money, but by changing how money is perceived, labeled, and claimed within households.
Aker, J. C., Boumnijel, R., McClelland, A., & Tierney, N. (2016). Payment mechanisms and antipoverty programs: Evidence from a mobile money cash transfer experiment in Niger. Economic Development and Cultural Change, 65(1), 1-37
Dizon, F., Gong, E., & Jones, K. (2020). The effect of promoting savings on informal risk sharing: experimental evidence from vulnerable women in Kenya. Journal of Human Resources, 55(3), 963-998
Heath, R. & Riley, M. (2024). Digital Financial Services and Women’s Empowerment: Experimental Evidence from Tanzania
Riley, E. (2024). Resisting social pressure in the household using mobile money: Experimental evidence on microenterprise investment in Uganda. American Economic Review, 114(5), 1415-1447
Suri, T., & Jack, W. (2016). The long-run poverty and gender impacts of mobile money. Science, 354(6317), 1288-129
