Building stable livelihoods for low-income households

Multifaceted livelihood programs caused broad and lasting economic impacts. The bundle of goods increased basic entrepreneurial activities, primarily concentrated on livestock and trade. Recent studies measuring outcomes eighteen months to three years later have found beneficiaries of the program increased their incomes from 7–65 percent, while consumption­—a common measure of well-being—increased by 11­–30 percent [1][4][6][8][14][19][21]. These households experienced similar improvements in food security, asset holdings, and savings.

Two studies found persistent effects ten years later. In West Bengal, India, an evaluation with long-term follow-up surveys of three, seven, and ten years found that the program increased consumption by 22 percent and income by 38 percent at year ten. Additionally, it enabled recipient households to migrate 25 days longer (a 20 percent increase) to farther urban centers where they improved their wages for the same work by US$89 in year seven and US$51 in year ten [6]. In Bangladesh, a four-year follow-up found Graduation recipients’ consumption increased by 11 percent and earnings by 21 percent, and they were 14 percent less likely to fall below the poverty line[4]. Further, an eleven-year follow-up found evidence that the persistent effects of the program could lead to increases in the accumulation of assets and the total labor supply of working-age women. Specifically, the results suggest that the size of the asset transfer must be sufficiently large to lift beneficiaries out of the poverty trap by allowing them to pursue more lucrative jobs or build upon the assets originally transferred [4]. Future research aimed at assessing whether the program continues to deliver benefits beyond ten years in regions outside South Asia would bolster the evidence base.

Graduation led to some improvements in psychosocial well-being and women’s empowerment. Psychosocial well-being and mental health measures, such as anxiety, happiness, and stress, and some physical health and political engagement measures improved for participants. Specifically, five out of nine studies that measured these outcomes found improvements in psychosocial well-being by the end of the evaluation [4][5][8][12][22]. In Afghanistan, participants experienced substantial improvements in life satisfaction, happiness, depression, self-esteem, and stress[8]. On the other hand, four studies did not find changes to participants’ mental health [1][7][11][18], pointing to the need for additional research to disentangle effects by programmatic components.

Targeting resources toward women to promote household decision-making ability, ownership of productive assets, and control over income improved women’s intrahousehold welfare, gender equity, and economic engagement in eight [1][5][8][12][17][18][19][22] out of eleven [3][7][9] studies where this was measured. In Nepal, bundled intervention beneficiaries were 24 percentage points from a base of 60 percent more likely to have control over decisions regarding livestock sales (a 40 percent increase) and 14 percentage points from a base of 40 percent more likely to control the income earned from livestock (a 35 percent increase) [18]. In Niger, recipients had positive sustained effects on an index of women’s control over their own earnings, productive activities, social support, and social standing[12]. A six-country comparative analysis of Graduation programs found pooled short-term effects but these did not persist over time[5].

Long-run benefits of economic inclusion programs outweigh the up-front costs. Eight studies across thirteen countries compared the program’s net economic benefits to its net costs. In the six-country comparative study, there was a positive rate of return three years after the asset transfer in all contexts except Honduras (likely due to the death of livestock from illness), ranging from 133 to 433 percent[5]. Similarly, six studies in Afghanistan, Bangladesh, the Democratic Republic of the Congo, Ghana, Nepal, and Niger found a benefit-cost ratio between 121 to 379 percent and internal rates of return between 16 to 66 percent. This highlights that in addition to being effective at helping households rise out of poverty, the Graduation approach's economic benefits exceed its costs[1][4][6][7][8][12][18][22]. In West Bengal, India, the program cost the equivalent of US$2,163, broke even by year four, and had a benefit-cost ratio of 379 percent by year ten (paying for itself 2.8 times over the course of ten years). However, another study found the intervention breaking even by eighteen months [12]. These figures do not account for any non-financial social benefits produced, such as improved mental well-being, acquired life skills through coaching, or the empowerment of women [8][12]. To policymakers, the substantial evidence from diverse contexts underscores the Graduation approaches enduring effectiveness at yielding economic benefits that surpass investment.

Simplifying the standard program by removing key components tends to reduce overall impact. Two studies demonstrate how, in the pursuit of a lower-cost model, unbundling can still generate positive effects, albeit with reduced impact [1][12]. In Niger, a multi-faceted program laid on top of a government cash transfer removed the asset but complimented the program with psychosocial supports that included life skills and social norms training. This model was highly cost-effective while still improving economic, psychosocial, and women’s empowerment measures [12].  In two studies, the productive asset alone did not attain the same impact relative to the full suite [7][22]. For example, in Ghana, an evaluation examined whether individual components could achieve the same impacts relative to the full suite of services. Neither access to savings accounts with deposit collection services, nor simply receiving the productive asset (a goat) was sufficient to achieve the full program’s impact over time[7]. Four studies found that removing individual components such as livestock, savings support, or coaching reduced the impact of the program on outcomes such as income, consumption, and savings [7][9][12][22].

Table of studies that test the impacts of individual components of the Graduation program relative to the full suite.

Further long-term evaluations, which test innovative complementary features and explore ways to reduce costs, could help to identify and scale a more robust and less expensive model while retaining sustained impact. For instance, a study conducted in Bangladesh suggested that extending credit to slightly wealthier participants could be as beneficial as providing them with cash grants. This variation in the funding model has the potential to enhance the sustainability of the program. By supplying productive assets through credit to the higher-wealth subset of beneficiaries, a more scalable and equally effective program could be implemented [21].

Altering the traditional Graduation approach, such as group microenterprises and the type of productive asset, has generally not had the same results. Tweaking the original model whereby individuals were targeted with a productive asset has had varying results. For example, a small-group business development grant for skilled trade in Uganda may have acted as a kick-start for recipients rather than a sustained solution. There were large, short- and medium-term economic gains, but recipients experienced a convergence of economic well-being measures with the comparison group by year nine[11]. In another evaluation in Uganda, microenterprises run by three households observed gains in key economic welfare indicators, increased entrepreneurial activity, and a reduction in poverty[22]. Yet, in Tanzania, funding for a group enterprise complemented with short-term training and cash transfers did not prove to have strong effects on beneficiary welfare[2]. Policymakers interested in Graduation-style alternatives should consider investing in research of bundled interventions that explore the group business model, relevance in urban markets, and conditions and components that promote persistent impacts.

Policymakers seeking to maximize impact may consider a suite of programs. Often policymakers need to make a decision between implementing different programs to address poverty, like simple cash transfers or other poverty-reduction programs, and cost is a common, limiting factor. Although cash transfers are less costly to implement, they do not fully replicate the positive impacts of multifaceted livelihood programs on recipients [6][20]. It is important to note, however, that cash transfer programs still have their place, as it is not uncommon for recipients to continue living in poverty even after participating in Graduation programs. For policymakers, an optimal strategy may consider a portfolio of anti-poverty programs such that cash transfers are targeted toward the individuals it would impact most, whereas Graduation-style interventions can remove simultaneous socioeconomic barriers while offering sustained gains [16].