For the past several years, a number of co-authors and I have been working on a series of projects that have explored the relationship between the psychology of poverty and economic development. I thought it might be worth a blog post to summarize what we’ve learned so far, mainly for the sake of practitioners who might want to incorporate behavioral economics-based interventions into their work, but also for other academics interested in this area. My co-authors deserve much if not most of the credit for this work, and they include Travis Lybbert and Irvin Rojas at UC Davis, my colleague Alessandra Cassar at the University of San Francisco, Paul Glewwe at the University of Minnesota, Phillip Ross at Boston University, and Daniel Prudencio at Rice University.
Traditionally development economics has focused on relieving “external constraints” for people in poverty. In layman’s terms essentially what this means is we try to help poor people by providing them stuff. Stuff may include resources devoted to education, finance, infrastructure, even health. And even when we don’t actually give these things away, resources are used to provide people access to them. The point is to relieve some kind of tangible economic constraint in the hopes that once this external constraint is relieved, a virtuous cycle begins in which at long last people earn enough income to save, to educate their children, and to plan in productive waves about the future, rather than having to focus on meeting the needs of today.
But what if poverty isn’t all about external constraints? What if some of the most important aspects of poverty relate to internal constraints like feeling worthless, hopeless, and without aspirations for the future? If these things are important, then perhaps no amount of external aid and development will have much of an effect. And perhaps where hope, self-efficacy, and aspirations are strong, it allows people to overcome their own external constraints without outsiders having to do it for them. So essentially the questions we ask is whether interventions that are focused around the development of goals, aspirations, and even hope itself can make a difference to the poor.
My work with Travis is based around a paper forthcoming in Economic Development and Cultural Change that lays out a theoretical framework for thinking about the relationship between internal and external constraints. More than the analytical model presented in the paper, practitioners may appreciate the effort of the paper to understand the nature of hope and its role in upward economic mobility. What we argue that may be useful to practitioners is that unless there are pre-existing aspirations for an intervention (such as microfinance or myriad programs related to education and other types of empowerment), the relief of an external constraint in a particular area is likely to meet with poor results. But if aspirations (for, say, entrepreneurialism or education) are pre-existent, or perhaps if they can be effectively developed as a complement to a tangible intervention, then the intervention is much more likely to realize significant and positive impacts.
In work with Paul, Phillip, and Daniel, we find an example of this type of intervention with studies we carried out in Kenya, Indonesia, and Mexico on children who are currently sponsored through Compassion International. Unlike many development NGOs, Compassion places a substantial emphasis on the relief of internal constraints of children rather than solely focusing on paying for their schooling and relieving other tangible economic constraints, although they do this too. In Paul’s and my initial study with Laine Rutledge, we found in a sample of over 10,000 individuals in six countries that by adulthood, child sponsorship increased secondary school completion by 45%, white collar work by 35%, and income by about 20%. Our hypothesis was that Compassion’s heavy emphasis on nurturing the socio-emotional and spiritual development of children, their dreams, and their aspirations may have had something to do with these big impacts. What we found in this new three-country study, not on adults who were sponsored as children as before, but on currently sponsored children was that being a sponsored child caused a child to be a qualitatively “different kind of kid” from the inside out. Our results from a sample of 2,022 currently-sponsored children in Kenya, Indonesia, and Mexico find that sponsorship increased indices of self-esteem and optimism about the future, each by a quarter of a standard deviation (a modestly big impact), and increased educational aspirations by nearly half a year. Of the three countries, results are by far strongest in Kenya, although positive everywhere. These results complement those from a study forthcoming in the Journal of Human Resources, where in Indonesia we used a digital analysis of children’s self-portraits to find that Compassion’s international sponsorship significantly raised levels of hope by two-thirds of a standard deviation, happiness by nearly half a standard deviation, and self-efficacy by nearly one-third of a standard deviation.
Interestingly, even the relatively small increases in aspirations among children may yield big effects in the future. Our currently sponsored children only aspire to about half a year more schooling than the non-treated kids, whereas the study on adults found an impact closer to 1.5 years. But the point is once they complete high school, their realized level of education may later lead to greater aspirations for more down the road. Aspirations breed higher aspirations.
While it is possible that this kind of holistic child sponsorship approach might directly affect adult life outcomes in a way that is independent from the dramatic changes it causes on children’s psychology, we view this as highly unlikely. What is far more likely is that the aspects of Compassion’s program that relieve internal and external constraints act as strong complements to one another, which causes the program outcomes (in for example, education) to exceed that of secular and more uni-dimensional programs such as conditional cash transfer programs that relieve external constraints without addressing internal ones.
Aspirations may be vital to development outcomes, and we find in a paper now in resubmission to the Journal of Economic Behavior and Organization that even an intervention that provides the most rudimentary support around goal-setting may yield substantial results. Three of Alessandra’s and my M.S. students, Paulina Aguinaga, Jennifer Graham, and Lauren Skora, carried out an experiment in Colombia that replicated the components of the Family Independence Initiative (FII) among entrepreneurs in Medellin. What we found was that the mere act of establishing a goal plays a large and significant role in outcomes, where subjects in any of the treatments where a goal is established achieved 35%-38% of the possible goals compared to rates of 15%-25% in control groups. However, the FII combination of providing a small cash incentive for the realization of the goal and a small network of peer support to root each other on for achieving the goal (and getting the prize) was the most powerful. A similar experiment in Colombia run by a different set of students finds even bigger impacts using the FII approach for educational goals among teenagers shows. What I like about the FII approach is that it hits on all cylinders, combining psychological, social, and economic motivations to propel individuals forward in their aspirations.
Travis, Irvin and I carried out a randomized controlled trial among 732 indigenous women with microfinance loans in Oaxaca, what we call the Oaxaca Hope Project. More details on this experiment are available in a previous blog post, but I will summarize them here. Our intervention was created within the framework of the EDCC paper, and builds upon classic work in positive psychology by Synder (1994) which breaks down aspirational hope into goals, agency, and pathways. As such, some film-school students from Sacramento State University created a documentary on the four most successful women involved in the community banks of Fuentes Libres, a faith-based nonprofit operating through the Evangelical Covenant Church (Iglesia Evangelica del Pacto) in Oaxaca. We also created very snazzy looking refrigerator magnets which contained inspirational Bible verses on the importance of goals, agency, and pathways (e.g. I can do all things through Him who gives me strength.) And on these refrigerator magnets, women penned in sales and savings goals for their micro-enterprise. Program personnel carried out a year-long curriculum on the importance of goal setting, realization of talents and capabilities, and helping women to conceptualize pathways out of poverty (by altering product, sales location, approach with customers, etc.) but this intervention eschewed any kind of formal business training, where the focus was entirely on the development of women’s spiritual lives and socio-emotional skills.
What we found 12 months after the intervention was that aspirations among women in treated community banks were about one-sixth of a standard deviation higher than women in the control group and similarly was an index of business performance. Only some of our psychological and business measures were individually statistically significant, but every one of them was positive. Profits and sales were higher, but not significantly different for treated women. Most of the impact on the business performance index related to new employees, where there were small but significant gains in employees for women in the treated groups and zero in the control groups. And among treated women, there were much higher plans for new employees in the future.
We are obviously still at work here, but there are a few central themes that stand out to me from our research so far. First, internal constraints matter, and probably matter more than mainstream development economics ever thought they did, although certainly more research needs to be done to verify this claim. Second, some of the most effective interventions that help alleviate internal constraints may be very inexpensive, such as goalsetting. However, the most effective interventions are more expensive because they involve the long-term nurture of individuals. And these are probably more effective when carried out with children. For example, the Compassion results on children are stronger than our results on adult women. The FII intervention is stronger on teenagers than on adult entrepreneurs. With adults, old patterns of thinking die hard. With children, there are no old patterns of thinking.
Finally, the integrated, holistic model of human development is not just some pleasant-sounding, wishy-washy approach adopted by faith-based NGOs. Integrated development has teeth to it, to the extent that both faith-based and secular practitioners simultaneously addressing internal and external constraints are showing impacts are in many cases realizing impacts that are significantly greater than interventions focused on external constraints alone. I would love to see more productive collaborations between NGOs who are implementing, or who would like to implement, holistic, integrated development approaches and academics who would like to rigorously evaluate them.