Are Mobile Savings the Silver Bullet to Help Women Grow Their Businesses?

The Africa Gender Innovation Lab (AFRGIL) and the East Asia Pacific Gender Innovation Lab (EAPGIL) carry out impact evaluations and inferential research to generate evidence on how to close the gender gap in earnings, productivity, assets and agency. The impact objective of the GILs is to increase take-up of effective policies by governments, development organizations, and the private sector in order to address the underlying causes of gender inequality. The labs aim to do this by producing and delivering a new body of evidence and developing a compelling narrative, geared towards policymakers, on what works and what does not work in promoting gender equality. Policy Brief Issue 29 P ub lic D is cl os ur e A ut ho riz ed

• In both countries, the training enhanced the impact of promoting mobile savings. In Indonesia it led women to save more overall, including a nascent use of mobile accounts, and report greater decision making power within the household. In Tanzania, it led to substantially higher mobile savings, new businesses and products, more capital investment, labor effort, and better business practices.
• However, these short-term impacts have yet to translate into higher business profits. In Indonesia, we observe increased household welfare, but no discernible effects on business outcomes shortly after the training ended. In Tanzania, the increased business investments were not accompanied by greater profitability.
Savings can help businesses expand, by enabling them to finance lumpy investments and absorb unexpected shocks. However, several barriers stand in the way of women firm owners in developing countries who want to increase their savings. For instance: social pressures women might face to share their incomes with friends and family members, the high transaction costs of traditional savings products, and implicit or explicit biases from financial service providers against women as potential good customers. If despite these barriers women manage to set aside some savings, any effective use of these funds towards business expansion may take time to materialize, compete with other, more urgent uses, or be hindered by human capital limitations such as limited entrepreneurial ability.

HERE'S WHAT WE DID
We designed two randomized control trials: one in Tanzania, the other in Indonesia.
In both countries, we tested the impact of two interventions, separately and together, in alleviating some of the constraints impeding savings and business growth among women microentrepreneurs. The first intervention ("mobile savings" intervention) expanded access to mobile savings bank accounts. The second intervention ("business training" intervention) provided training to improve business skills and financial literacy. The specifics of each intervention differ by country, as follows.

Tanzania
In Tanzania, the mobile savings intervention promoted M-Pawa: a mobile banking product linked to M-Pesa that allows customers to save money on an interestbearing mobile savings account. It also gives customers access to uncollateralized instant microloans. The intervention invited women microentrepreneurs for a training session on M-Pawa. At the end of the session, we helped women create an account with M-Pawa. We also offered them the option to set personal savings goals and receive weekly SMS savings reminders. This latter feature attempted to relax possible psychological barriers related to limited attention that could lead to under-adoption of the product.
The business training, designed and implemented by TechnoServe, lasted for 12 weekly group sessions. The training was designed by TechnoServe and it was explicitly adapted to women business owners.
The goal was to teach and motivate women to set business goals, identify business opportunities, put ideas into practice, and mobilize finance (with a focus on savings). The curriculum also emphasized the importance of fostering personal effectiveness and perseverance, as well as the benefits of a more gender equal society. Additionally, it covered standard good business practices. The face-toface group training was supplemented with an interactive mobile phone learning platform.
The research design in Tanzania allows us to identify the impacts of the mobile savings intervention, with and without the business training. To do so, we randomly assigned about 4,000 women microentrepreneurs (operating in Dodoma and Mbeya) into three groups: (i) 1,000 women were invited to the mobile savings intervention only, (ii) 2,000 women were invited to both the mobile savings and the business training interventions, and (iii) 1,000 women served as a control group. All women were first surveyed in June-July 2016. The interventions run from August-November 2016. Follow-up surveys took place at one year and at 18 months after the baseline survey. For the analysis we also exploit administrative data on M-Pawa transactions made available by Vodacom for all women in the study.

Indonesia
In Indonesia, the mobile savings intervention also promoted access to a new branchless banking mobile savings product, but that was dependent on the successful operation of a new mobile banking network. Unlike in Tanzania where we worked directly with women business owners, in Indonesia the intervention targeted local bank agents and included men and women entrepreneurs. It provided agents with enhanced financial incentives for signing up new customers, as well as training on the product and the importance of targeting female customers. The intervention thus targeted the supply side of the market for mobile savings, whereas in Tanzania we directly targeted the demand side.
The business training, designed and implemented by Mercy Corps Indonesia, invited microentrepreneurs for an average 3 hours-long training session on financial literacy with a focus on mobile savings. This group session was followed by three group mentoring sessions to reinforce the training with practice.
The research design in Indonesia allows us to identify the impacts of the two interventions, separately and together. To do so, we followed a two-stage randomization procedure. The first stage randomized the mobile savings intervention across about 400 villages (in East Java): the villages were either paired with bank agents receiving high (Rp. 10,000, around $0.77) or low (Rp. 2,000, or $0.15) monetary incentives for signing up new customers. The second stage randomized women microentrepreneurs within both types of villages into two groups: about 1,600 women were invited to the business training intervention, and about 1,200 women served as a comparison group. Microentrepreneurs were surveyed in two waves, with the first wave surveyed in January-February 2017 and the second in July-November 2017. The results below are restricted to half the sample, where a majority of the women (77%), were re-interviewed in early 2018, a month or less after the last group mentoring session (as delays resulted from operational challenges in setting up the network, now resolved).

HERE'S WHAT WE FOUND Tanzania
The mobile savings intervention led to firms saving more on M-Pawa, and the business training enhanced this impact. Transaction-level data shows women in the mobile savings only group saved two times more money weekly on M-Pawa than women in the control group, while those in the mobile savings + business training group saved almost four times more.
Borrowing from M-Pawa also increased. The mobile savings intervention led to a 14pp statistically significant increase in the likelihood that women received a mobile loan. This impact is slightly higher among women who received the additional business training intervention, though not significantly different.

The increase in mobile savings and loans appear to crowd out other types of savings and credit.
Survey data show that there are no systematic effects on overall savings or loans. Firm owners appear to have shifted both savings and loans to the mobile platform from other sources, including cash savings at home. So although savings balances haven't increased, they appear to be more secure.
The mobile savings intervention led to firms opening new businesses. Women in the mobile savings only intervention are 4pp more likely to grow their business activities by operating a second business. This impact is higher among women who receive the additional business training intervention, though not significantly different from the mobile-savings only group.

The mobile savings intervention also led to an increase in women's intra-household decisionmaking power vis-à-vis their husbands.
Selfreported data from women who were encouraged to open mobile-savings accounts indicates that they have greater decision-making power over business and household expenditure allocations compared to women in the control group. This evidence is consistent with mobile savings accounts offering greater privacy and security.
The business training further led to better business practices (such as record keeping and financial planning), more labor effort and capital investment. The marginal impact of adding business training to the mobile savings intervention was a 4pp increase in the number of good business practices , a 2-hour increase in time the woman devotes to work per week, and a 23% increase in capital investment.

Indonesia
The business training increased women's knowledge of mobile money and mobile savings and their likelihood of opening a mobile savings account.
The business training in combination with the mobile savings intervention led women to increase total savings and mobile savings. These women increased total savings balances by about 24%, including a statistically significant 4% increase in mobile savings The researchers would like to acknowledge the ExxonMobil Foundation's pivotal role in the creation of "She Counts"-a global platform that seeks to put financial services in the hands of women business owners. The researchers are privileged to provide the evidence that feeds into this partnership.
Photo credit: Marijo Silva and the "She Counts" global platform balance. Women who only received training and women who were only exposed to better paid agents also increased total savings (11 and 7%, respectively), but these increases were not statistically significant.
Women who received training and were served by high-incentive agents reported being more empowered, reflected in an index of household decision-making. This empowerment effect seems to have been mediated by the increase in savings and mobile savings, providing privacy and enhancing women's economic self-reliance.
The combination of the two interventions also had positive economic welfare effects and increased women's self-confidence, but there were no discernible effects of either intervention alone or combined on women's business outcomes. Both interventions independently increased household assets -women who received training reported a 20% increase in household consumer durables (including smartphones and TV connection) and women served by high-incentive agents reported a 24% increase in these durables. These women likely invested at least part of their increased savings in durable household assets. An outstanding question is whether increased savings will also be invested in the business over time.

WHAT HAVE WE LEARNED SO FAR?
Our results suggest that easing supply-side constraints increases the uptake of mobile savings accounts by women microentrepreneurs in developing countries, and has beneficial economic empowerment effects. However, the evidence also shows that without relaxing complementary demand-sided constraints that women might face, access to mobile savings accounts per se is no silver bullet. Providing women with complementary business training bolsters the effectiveness of mobile savings accounts, although this is still not enough to improve business profits in the short run. We plan to conduct additional follow-up surveys for both the Tanzania and Indonesia projects to track longer run effects on business performance.
The first draft of this policy brief was released in September 2018.