Investigating Rural Labor Markets Among the Ultra-Poor in Northern Ghana
- Earnings and income
- Recruitment and hiring
Few subsistence farmers engage in paid labor, even when pursuing these opportunities would enable them to earn more and protect against risk. A variety of factors may influence this trend, including poor nutrition, the complexity of the work, and lack of access to financial services. Researchers designed an employment program in northern Ghana and evaluated participants’ willingness to engage in paid labor and carry out complex tasks on the job.
Self-employed farmers in sub-Saharan Africa may earn more by engaging in wage labor, but few pursue off-farm opportunities, forgoing higher and more predictable incomes. There are a number of possible explanations for this trend, including lack of information about job opportunities or high transportation costs to get to work sites. The choice to engage in wage labor may also be affected by the wage level, the complexity of the work, or an individual’s nutrition status. For instance, farmers may only be willing to pursue wage labor above a certain wage threshold, or they may find the tasks too complex or onerous to complete. Poor nutrition could also make workers less productive and, thereby, more costly to hire. Rigorous research is needed to disentangle the various mechanisms that affect individuals’ choices to engage in wage labor.
Context of the evaluation
Subsistence agriculture is the main economic activity for the vast majority of rural households in Ghana. Researchers introduced an employment program in communities in northern Ghana that were already participating in an ultra-poor graduation program. The two-year Graduation from Ultra Poverty (GUP) program, originally developed and evaluated in Bangladesh, offers a comprehensive package of services (productive assets, consumption support, financial literacy training, and healthcare) to households living in extreme poverty.
A random subset of participants in GUP was assigned to a voluntary savings program called Savings Out of Ultra Poverty (SOUP), in which clients had the opportunity to earn a 50 percent match on their savings.
Researchers partnered with Presbyterian Agricultural Services (PAS) and Innovations for Poverty Action (IPA) to implement GUP and SOUP in 155 communities in northern Ghana.
Details of the intervention
Researchers introduced an employment program in communities in northern Ghana that were already receiving GUP and SOUP or part of the comparison group for the GUP evaluation. Half of all GUP, SOUP, and comparison villages were randomly selected to participate in an 11-month employment program that offered wages for the production of cloth bags. This amounted to 120 villages: 42 comparison villages, 39 GUP villages, and 39 SOUP villages (1,098 households total).
Researchers randomly varied the complexity of the bag design and wage per bag to understand the impact of these factors on individuals’ decisions to engage in paid labor and whether willingness to engage in complex tasks depended on the wage level. Of the 120 villages assigned to the bags program, 60 were assigned to produce a simple bag and 60 were assigned to produce a complex bag, based on the difficulty of the stitching required. Participants were randomly assigned to receive either a high wage of GHC 0.70 (approximately US$0.39) or a low wage of GHC 0.30 (approximately US$0.16), and the wage assignment switched every four weeks.
Communities received all sewing materials and a four-day training on how to produce the bags prior to the start of the employment program. Program facilitators carried out regular quality checks to assess how well the simple and complex bags produced met the pre-determined quality standards. Within each wage category, participants could earn an additional GHC 0.10 more per bag for producing a high quality bag or lose GHC 0.10 per bag for producing a low quality bag.
Results and policy lessons
Banerjee, Abhijit, Dean Karlan, Hannah Trachtman, and Christopher Udry. "Does Poverty Change Labor Supply? Evidence from Multiple Income Effects and 115,579 Bags." Working Paper, May 2020.