The Financial Inclusion Program (FIP) provides technical and financial support to rigourous evaluations and pilot projects related to financial service design, digital finance, and financial capability. The Program’s projects, which range in scale from pilots to multi-country randomized evaluations, are implemented across developing and advanced economies and focus on innovations that are informed by behavioral insights, are cost-effective, and present a promising business case for scale-up. FIP identifies new research projects and promising partnerships through open calls for proposals and periodic matchmaking and training events, and disseminates recent results through conferences, webinars, and publications.
Theoretically, weather-index insurance is an effective risk reduction option for small-scale farmers in low-income countries. Renewed policy and donor emphasis on bridging gender gaps in development also emphasizes the potential social safety net benefits that weather-index insurance could bring to women farmers who are disproportionately vulnerable to climate change risk and have low adaptive capacity. To date, no quantitative studies have experimentally explored weather-index insurance preferences through a gender lens, and little information exists regarding gender-specific preferences for (and constraints to) smallholder investment in agricultural weather-index insurance. This study responds to this gap, and advances the understanding of preference heterogeneity for weather-index insurance by analysing data collected from 433 male and female farmers living on a climate change vulnerable coastal island in Bangladesh, where an increasing number of farmers are adopting maize as a potentially remunerative, but high-risk cash crop. We implemented a choice experiment designed to investigate farmers’ valuations for, and trade-offs among, the key attributes of a hypothetical maize crop weatherindex insurance program that offered different options for bundling insurance with financial saving mechanisms. Our results reveal significant insurance aversion among female farmers, irrespective of the attributes of the insurance scheme. Heterogeneity in insurance choices could however not be explained by differences in men’s and women’s risk and time preferences, or agency in making agriculturally related decisions. Rather, gendered differences in farmers’ level of trust in insurance institutions and financial literacy were the key factors driving the heterogeneous preferences observed between men and women. Efforts to fulfill gender equity mandates in climate-smart agricultural development programs that rely on weather-index insurance as a risk-abatement tool are therefore likely to require a strengthening of institutional credibility, while coupling such interventions with financial literacy programs for female farmers.
Individuals across the world use high-transaction-cost savings devices, even when lower-cost technologies are available. High costs may help savers protect resources from the demands of others. I investigate this hypothesis by randomly assigning ATM cards to 1,100 newly-opened bank accounts in rural Kenya. These cards reduced withdrawal fees by 50 percent. While the cards increased overall account use, the positive treatment effect is entirely driven by joint and male-owned accounts. I also find that individuals with low levels of household bargaining power save less when accounts have ATM cards, while individuals with high levels of household bargaining power save more.
Through a field experiment in Afghanistan, we show that default enrollment in payroll deductions increases rates of savings by 40 percentage points, and that this increase is driven by present-biased preferences. Working with Afghanistan’s primary mobile phone operator, we designed and deployed a new mobile phone-based automatic payroll deduction system. Each of 967 employees at the country’s largest firm was randomly assigned a default contribution rate (either 0% or 5%) as well as a matching incentive rate (0%, 25%, or 50%). We find that employees initially assigned a default contribution rate of 5% are 40 percentage points more likely to contribute to the account 6 months later than individuals assigned to a default contribution rate of zero; to achieve this effect through financial incentives alone would require a 50% match from the employer. We also find evidence of habit formation: default enrollment increases the likelihood that employees continue to save after the trial ended, and increases employees’ self-reported interest in saving and sense of financial security. To understand why default enrollment increases participation, we conducted several interventions designed to induce employees to make a non-default election, and separately measured employee time preferences. Ruling out several competing explanations, we find evidence that the default effect is driven largely by present-biased preferences that cause the employee to procrastinate in making a non-default election.
Weather index insurance protects farmers against losses from extreme weather and facilitates investment in their farms, but randomized evaluations in South Asia and sub-Saharan Africa have shown low demand for these products at market prices, suggesting the need for alternative approaches.
Without substantial subsidies, take-up of insurance was low. Large discounts increased take-up substantially, and interventions designed to increase financial literacy or reduce basis risk also had positive effects. However, at market prices, take-up was in the range of 6–18 percent, which cannot sustain unsubsidized markets.
Insured farmers were more likely to plant riskier but higher-yielding crops. In the three studies that measured changes in farmer behavior, farmers who felt protected against weather risks shifted production toward crops that were more sensitive to weather but more profitable on average.
While self-sustaining markets for weather index insurance have not emerged, finding ways to address weather risk remains a priority for agricultural development. Some possibilities are improving index quality, providing subsidized insurance, selling insurance to institutions, and exploring other risk-mitigating technologies, such as irrigation and stress-tolerant crops.
Loans and business management training helped men grow their small business profits, but women did not experience any impacts on their businesses as a result of loans, training, or grants.
With the support of the Citi Foundation, the Financial Capability Initiative at IPA incubates, develops, and rigorously evaluates products and programs that improve the ability of the poor to make informed financial decisions and adopt healthy financial behaviors. The Initiative conducts tests and evaluations of innovative, product linked financial education interventions and financial products that aim to improve financial capability.
I use a field experiment in rural Kenya to study how temporary incentives to save impact long-run economic outcomes. Study participants who were randomly selected to receive large temporary interest rates on an individual bank account had signifi- cantly more income and assets 2.5 years after the interest rates expired. These changes are much larger than the short-run impacts on experimental bank account use and almost entirely driven by increased rates of entrepreneurship. Temporary interest rates directed to joint bank accounts had no detectable long-run impacts on entrepreneurship or income, but increased investment in household public goods and led to greater spousal consensus over financial matters. The short-run effects of modest unconditional cash payments were similar to those of the interest rates, but the cash payments had no apparent long-run impact on economic outcomes.
Anti-poverty programs in developing countries are often difficult to implement; in particular, many governments lack the capacity to deliver payments securely to targeted beneficiaries. We evaluate the impact of biometrically-authenticated payments infrastructure (“Smartcards”) on beneficiaries of employment (NREGS) and pension (SSP) programs in the Indian state of Andhra Pradesh, using a large-scale experiment that randomized the rollout of Smartcards over 157 subdistricts and 19 million people. We find that, while incompletely implemented, the new system delivered a faster, more predictable, and less corrupt NREGS payments process without adversely affecting program access. For each of these outcomes, treatment group distributions first-order stochastically dominated those of the control group. The investment was cost-effective, as time savings to NREGS beneficiaries alone were equal to the cost of the intervention, and there was also a significant reduction in the “leakage” of funds between the government and beneficiaries in both NREGS and SSP programs. Beneficiaries overwhelmingly preferred the new system for both programs. Overall, our results suggest that investing in secure payments infrastructure can significantly enhance “state capacity” to implement welfare programs in developing countries.
The pricing and advertising of tied add-ons and overages have come under increasing scrutiny. Working with a large Turkish bank to test SMS direct marketing promotions to 108,000 existing holders of “free” checking accounts, we find that promoting a large discount on the 60% APR charged for overdrafts reduces overdraft usage. In contrast, messages mentioning overdraft availability without mentioning price increase usage. Neither change persists long after messages stop, suggesting that induced overdrafting is not habit-forming. We discuss implications for interventions to promote transparency in pricing and advertising, and for models of shrouded equilibria, limited attention, and salience.
This paper tests whether uncertainty about future rainfall affects farmers’ decision-making through cognitive load. Behavioral theories predict that rainfall risk could impose a psychological tax on farmers, leading to material consequences at all times and across all states of nature, even within decisions unrelated to consumption smoothing, and even when negative rainfall shocks do not materialize down the line. Using a novel technology to run lab experiments in the field, we combine recent rainfall shocks and survey experiments to test the effects of rainfall risk on farmers’ cognition, and find that it decreases farmers’ attention, memory and impulse control, and increases their susceptibility to a variety of behavioral biases. In theory, insurance could mitigate those effects by alleviating the material consequences of rainfall risk. To test this hypothesis, we randomly assign offers of an index insurance product, and find that it does not affect farmers’ cognitive load. These results suggest that farmers’ anxiety might be relatively difficult to alleviate.
We implemented a randomized intervention among Malawian farmers aimed at facilitating formal savings for agricultural inputs. Treated farmers were offered the opportunity to have their cash crop harvest proceeds deposited directly into new bank accounts in their own names, while farmers in the control group were paid harvest proceeds in cash (the status quo). The treatment led to higher savings in the months immediately prior to the next agricultural planting season, and raised agricultural input usage in that season. We also find positive treatment effects on subsequent crop sale proceeds and household expenditures. Because the treatment effect on savings was only a small fraction of the treatment effect on the value of agricultural inputs, mechanisms other than alleviation of savings constraints per se are needed to explain the treatment’s impact on input utilization. We discuss other possible mechanisms through which treatment effects may have operated.
Financial products and transfer schemes are often designed to help individuals follow through intertemporal plans by not giving into “temptation”. We implement an artefactual field experiment in Malawi to test the ability of households to manage a cash windfall and learn whether savings defaults and delayed payments change consumption patterns in ways that are likely to be welfare improving. We vary whether 600 households receive a windfall payment in cash or through direct deposit into pre-established accounts at a local bank, and whether payments are made immediately, with one day’s delay, or with eight days’ delay. Defaulting the payment into a savings account leads to higher bank savings, an effect that persists for a number of weeks afterwards. However, neither savings defaults nor payment delays affect the amount or composition of spending, suggesting that households manage cash effectively without the use of formal financial products.
Experimental tests of microcredit programs have consistently failed to find effects on business and household income. Does the current microfinance model and targeting of clients miss important effects from finance? I present results of a randomized experiment with microenterprise owners in Uganda that sought to expand access to finance for men and women who generally did not qualify for finance under normal circumstances with the goal of inceasing business profits and employment. Participants were offered either capital with repayment (subsidized loans) or without (grants) and were randomly chosen to receive or not receive business skills training in conjunction with the capital. Consistent with existing literature, I find no effect for female enterprises from either form of capital or the training. However, I find large effects for men with access to loans combined with training. There is no effect for men or women from the grants, suggesting repayment requirements can increase the likelihood of productive investment. I also find little evidence that investing capital and training in a few enterprises crowds out other businesses. The results indicate that cash-constrained male-owned enterprises—a sample that is not well targeted by microcredit organizations or researchers—can benefit from subsidized finance, and that this may have larger, positive income and employment growth effects for an economy.
In recent years, the influx of available consumer data has presented corporate firms, non-profit organizations, and governments alike with an opportunity to increase the efficacy and targeting of their products and services. The key to identifying what works is to build experimentation through randomized controlled trials (RCTs) into the process of designing new products and services. Running RCTs, however, is not always straightforward: there are a multitude of technical, analytical, and logistical hurdles that arise during the course of designing and implementing an RCT.
To this end, the US Finance Initiative at Innovations for Poverty Action has compiled best practices gleaned from years of experience running RCTs in the finance sector into a toolkit. The toolkit assumes a certain amount of technical knowledge and is intended for researchers, but details the often-neglected “softer” skills of managing an RCT, including the logistics of implementation and the interaction between the researcher and the partner institution, that are equally central to the success of the experiment. This guide focuses specifically on using RCTs to develop and test new financial products and services for consumers in the United States, but is applicable to RCTs in other disciplines.
Please send any feedback, questions, or comments to firstname.lastname@example.org.