Beyond Policy: How Gendered Interactions on the Ground Shape Development

National governments and international bodies often implement policies to promote more gender-equitable social and economic development. For example, initiatives like the UN’s Gender Equality Sustainable Development Goal highlight gender inequities across the globe and set benchmarks for progress towards equality. Similarly, USAID highlights gender equality and women’s empowerment as one of its main areas of focus, aiming to “unlock human potential on a transformational scale” by investing in women’s skills and advancement. These ambitious, large-scale policies and goals function as powerful tools for orienting global attention, and channeling funds toward initiatives that promote gender equality.

In our own research and in observations of practice, we found that although such policies and goals provide an important starting point for women’s economic development, the deeper impact of the policies come through more mundane, day-to-day practices. Indeed, the actual changes associated with gendered policies occur through interpersonal interactions. Women’s economic and social development happens through on-the-ground exchanges among ordinary citizens, low-level government officials, NGO workers, and others. As individuals meet, converse, argue, teach, establish relationships, and exchange resources, the real outcomes associated with development policies come to life.

Consider the experience of introducing clean cook stoves into countries where people tend to cook over open fires. These improved cook stoves have the potential to improve the lives of families through enhanced health, reduced labor for collecting fuels, and reduced air pollution. Yet despite the obvious potential benefits, not all cook stove projects have been successful. Developers of new cook stoves have often struggled to gain adoption. The Global Alliance for Clean Cook Stoves realized that to have a more significant impact, they needed to take a gender lens to all of the interactions associated with designing, developing, selling, distributing, and using the cook stoves. Interactions at each stage of the process could determine whether the project would be successful. Many cook stoves designed in North America, for example, simply didn’t meet the needs of the communities for whom they were designing. People ended up using new cook stoves for storage, because the stoves did not cook food fast enough. Distributors did not understand how using a new cook stove would involve a lifestyle change for the women using them—for example, they would no longer have the opportunity to socialize with one another while collecting wood. By paying attention to gender-based interactions with users at all stages, distributors have been able to promote greater uptake of clean stoves. In one study, researchers found that, with sales training, women distributors outsold men 3:1. When women sold the improved cook stoves, adoptees were more likely to report that the stove was easy to use, safe, and better than a traditional stove, and that they used it more frequently.

As in any interpersonal communication, gendered biases shape interactions related to social and economic development and, as a result, can yield positive or negative results. For example, research on microfinance demonstrates how gendered expectations in interactions can lead to favorable or unfavorable outcomes for women.

On the one hand, studies have shown that poor women who borrow in mutual accountability groups tend to miss fewer payments than male borrowers (although, we recognize that researchers debate the overall merits of women-focused lending). In the context of lending circles, women’s high repayment rates reflect gendered expectations around responsibility (women are often viewed as more responsible with household finances than men), as well as expectations of female collaboration and cooperation. These gendered behaviors and expectations are useful for microfinance institutions and female borrowers: They can facilitate better repayment rates and allow low-income women to access credit.

On the other hand, gendered expectations can also undermine women in the context of microfinance. Our own research shows that microfinance borrowers (both male and female) miss more payments if their loan officer is female than if their loan officer is male. These trends occur even when accounting for non-gendered characteristics that might affect repayment, such as officers’ experience and borrowers’ incomes. As in many other social contexts, individuals who take out microloans are less likely to follow the directives of women in positions of authority (in this case, their loan officers). This has implications for the careers of the women bankers as well as for the financial status of the borrowers.

These examples demonstrate how gendered biases and expectations can result in divergent outcomes within even one type of development initiative: microfinance. Through on-the-ground social interactions, these biases can amplify the advantages of a gender-focused development policy, such as better repayment rates among women in lending circles. But they can also lead to unexpected disadvantages for women, such as more missed payments to female loan officers.

When we view social and economic development as a series of interpersonal interactions, we can understand why development outcomes sometimes diverge from policy goals, and how gendered interactions shape the degree to which they achieve social and economic development. To incorporate an interaction perspective into our understanding of gendered impacts of development, we propose the following strategies:

1. Solicit feedback from targeted beneficiaries and project coordinators when designing development policies and goals. These individuals are closest to the on-the-ground realities in which projects will unfold and are best equipped to anticipate the points at which interpersonal interactions may unexpectedly shape development outcomes. Recent examples illustrate what happens when policymakers neglect consultation with targeted beneficiaries. In India, a campaign to encourage citizens to use toilets resulted in an unanticipated gender imbalance. Although women have readily adopted the new toilets, men appear less inclined to use them, undermining the public health goals of installing them. One woman explained the difference along gendered lines: “Men can go out to the open fields, but for women who wear veils all day, a toilet in the home is a good idea.” Gendered beliefs about appropriate uses of private spaces can lead to differential uses of public goods and, in this case, affect the public health outcomes associated with large-scale development policies.
2. Measure the actual effects of development programs rather than assuming that programs have the impact on women that you anticipate. Based on a needs assessment, measure outcomes that matter to the beneficiaries themselves. For example, while many randomized controlled trial (RCT) studies of microfinance look at growth of microenterprises or changes in consumption as outcomes, research suggests that many of the empowerment benefits might be less financially tangible. For example, a field study of Avon representatives in South Africa showed that a primary outcome for the women who became “Avon Ladies” was the increased prestige they experienced in their communities, an outcome not typically measured in quantitative RCTs. This study also demonstrated that the financing models associated with being a trader for Avon are much easier to manage and less costly for women than traditional microfinance, thus opening up the possibility that such sales models are indeed more effective forms of economic development. As this example suggests, it’s important to analyze the metrics that matter to beneficiaries, not to policymakers.  
3. Modify practices (not just policies) if your anticipated objectives are not achieved. Given the scope and power of gendered expectations and biases in social interactions, it is likely that some aspect of your program may unfold in unexpected ways. Rather than glossing over such divergent outcomes, anticipate that they will occur. I will be useful to build steps to “accommodate the unexpected” into your program architecture. For instance, you might schedule mid- and end-point discussions with beneficiaries, coordinators, and other community members to identify the interaction points that drive unexpected effects. By adapting your program—from the interaction level on up—to accommodate these insights, you can more effectively realize development initiatives in the real-world, gendered contexts in which they unfold.

Read more stories by Sarah Kaplan & Laura Doering.