The challenge of building sustainable livelihoods for the poor
Policymakers worldwide search for strategies to reduce extreme poverty. This issue is particularly pressing in the Sahel, where over half of rural households live in poverty, and many rely on subsistence agriculture in environments exposed to frequent climatic shocks.
Evidence shows that cash transfer programs positively impact welfare and human capital, and, over the last decade, governments across Africa have developed social protection systems. Niger, the poorest country in the Sahel in GDP per capita terms, started with a government program providing monthly cash transfers to poor households.
However, it was recognized that more may be needed to address the root causes of extreme poverty and open pathways toward sustainable livelihoods. Therefore, policymakers and program implementers across the Sahel looked to complement their cash transfer programs with additional interventions.
Another reason for this keen interest is that while evidence has consistently documented positive impacts of regular cash transfers on productive activities, many policymakers still harbor lingering concerns around social assistance. For example, several years ago, a high-ranking official told us that, in his mind, there were two categories of spending: “expenditures” and “investments.” He believed social protection programs fell in the first, less appealing, category.
There is no simple policy solution to sustainably improving women's economic activities in extremely poor households because many families face multiple barriers to starting or diversifying their income-generating activities. Therefore, programs likely need to jointly address several constraints, as indicated by earlier findings on multifaceted programs implemented by NGOs around the world.
With this context in mind, our first challenge was to identify and prioritize the main constraints to address. In a multi-country effort involving government representatives, NGOs, and technical experts, we engaged in a diagnostic phase combining qualitative work and quantitative data analysis across several Sahel countries. We also examined the evidence of effective interventions for each key constraint, and considered how they could be packaged together and added to existing safety net programs.
At this stage, psychosocial constraints stood out alongside the lack of capital or skills. For example, in the context of very low levels of education and high gender inequality, women often lack self-esteem and mental bandwidth while frequently suffering from depression. Social and gender norms further undermine their capacity to control resources, go to the market, and run small businesses. To explore how to tackle these psychosocial constraints, we reviewed international examples that had documented impacts, including video-based interventions at the community level and socioemotional skills trainings.
As a result, the policy and research teams identified the following interventions for inclusion in an integrated package:
- A core set of measures consisting of coaching, group savings facilitation, entrepreneurship training, and improved access to markets.
- A lump-sum cash grant.
- Psychosocial interventions in the form of life skills training and a community sensitization on aspirations and social norms (delivered via community video screenings), which highlighted how the objectives of the program aligned with positive values and traditions of the community at large (such as cohesiveness and adaptation) and aimed to foster community support for women.
Implementation through government systems
How could governments with limited capacity deliver and scale such an integrated package of interventions for some of their hardest-to-reach population?
Government program managers made some critical design decisions. For example, activities would be group-based to foster positive social dynamics and minimize implementation costs; training modules would be adapted to a non-literate audience and focus on fundamental cross-cutting business practices, rather than trade-specific technical skills; and the asset transfer would be cash-based (rather than in-kind) to facilitate delivery at scale and prevent adverse impacts on prices of specific assets. A delivery apparatus was then organized involving safety net workers, private training providers, local NGOs, and community volunteers.
Priority (policy) research questions
But there were still many questions on how to get the most out of this integrated package while making it cost-effective. For example, was the prioritization of interventions done right? Did the package address the most binding constraints that drive poverty persistence?
Our experiment was designed to provide answers to these questions. Specifically, the leading questions revolved around the effectiveness of both the lump-sum cash grant (one of the main cost drivers) and the psychosocial interventions (which were the most innovative). There were debates among the policy teams, with some arguing that these additional components may not be necessary.
Therefore, we assessed the impacts and cost-effectiveness of three packages: a capital package that included the cash grant in addition to the core components (components 1 and 2, above), a psychosocial package that included the psychosocial components in addition to the core components (components 1 and 3, above), and a full package that included all components.
Impacts on consumption and revenues were strong
Fast-forward a few intense years of program implementation and research activities: what did we find in Niger?
Eighteen months after the intervention ended, all three packages improved household consumption and decreased food insecurity. The impact of the psychosocial package was initially lower six months after the intervention but grew over time––the psychosocial interventions helped women strengthen social relationships with their community and their partner, and build social capital.
We also observed large increases in revenues. For example, revenues from beneficiaries’ off-farm businesses increased by 62–107 percent 18 months post-intervention. Revenues from livestock and agriculture also increased. Across all three packages, households developed new income sources.
Beyond the economic impacts, we saw widespread improvements in psychological and social well-being.
The program was highly cost-effective, especially when including psychosocial interventions
Costs were kept low by using the delivery mechanisms of the government-led national cash transfer program. Specifically, costs were $263 per beneficiary for the psychosocial package, $482 for the capital package, and $584 for the full package. For the latter two packages, the cash grant ($321) was the main cost driver.
As a result of the large observed impacts and low costs, the program was highly cost-effective. The benefits measured on consumption alone were so large that they already exceeded costs 18 months after the intervention ended for the psychosocial and full packages. The capital package was also cost-effective under weak assumptions. Yet the full and psychosocial packages had higher benefit-cost ratios than the capital package, which shows the value of integrating psychosocial interventions.
Assuming that impacts dissipate by 50 percent per year, the benefits were larger than costs by a ratio of 1.25 for the capital package, 2.08 for the full package, and 2.98 for the psychosocial package. This implies internal rates of returns of 15%, respectively 44% and 66%.
These numbers are rather impressive. While multi-faceted interventions had already shown to be effective when implemented by NGOs, leveraging government systems enhances scalability and cost-effectiveness. Psychosocial interventions carefully tailored to the needs of the beneficiaries may also be critical to unlock their potential.
In the end, our evidence demonstrates that government-led, multi-faceted social protection interventions can indeed be high-return investments in the welfare of extremely poor families.