EthioChicken Impact Evaluation

Executive Summary

EthioChicken is a private company that breeds high-productivity chickens and sells them to smallholder farmers in Ethiopia. EthioChicken distributes chickens through its network of agents (who in Oromia grow chickens from one day old to 56 days before distributing) and government agricultural extension agents who take orders and deliver chickens to smallholder farmers.

EthioChicken makes two breeds available: Bovans layers and Sasso dual-purpose birds. Both breeds are expected to grow four times faster and produce four times as many eggs as traditional Ethiopian backyard birds,¹ while also being able to grow and develop in the local environment. These higher-productivity birds have the potential to increase household income and provide sources of animal protein for consumption. As women are primarily responsible for raising chickens in Ethiopia, higher productivity chickens could also lead to increased decision-making power for women.

This report presents the results of an impact evaluation of providing access to improved chickens, leveraging EthioChicken’s expansion into the state of Oromia in Ethiopia. It studies the short-term impact of owning improved chickens on smallholders’ income, nutrition, and female decision-making roughly 6-14 months after purchasing improved chickens. To identify the causal effect of owning improved chickens, we match purchasers of improved chickens with households who did not have access to these chickens. The matching algorithm ensures comparability on a number of key baseline characteristics as well as a predicted probability of purchasing chickens (generated using a machine learning model). Our endline data, collected from October 2017-February 2018, was collected 6-14 months after chickens became available in our treatment areas. Since EthioChicken’s chicks are sold at 56 days old in Oromia and ferenj chickens take around 5-6 months to begin laying eggs, the timing of the endline assures that the chickens have had the opportunity to mature and start producing eggs or be sold for meat.

We find that households who purchased improved chickens:²

• Are likely to keep rearing ferenj chickens, as around 65% have improved chickens at endline. Purchasers seem to rear ferenj chickens in addition to rather than in place of habesha chickens. Both the treatment and control groups of this study had an average of 2.5 habesha chickens, while the treatment households also had an average of 2.2 ferenj chickens.
• Produce around 6.6 extra eggs per week (compared to 4.5 eggs/week in the comparison group) and consume an additional 3.0 eggs per week (compared to 3.6 eggs/week in the comparison group). Purchasers of improved chickens consumed eggs on an additional 0.7 days per week (compared to 0.80 days per week in the comparison group).
• Increase sales of eggs, resulting in an increase in average income from egg sales in the prior month of around 14 ETB (.51 USD). This is compared to average income from eggs of around 7 ETB (.26 USD) in the comparison group.
• Increase their income from chicken sales. Revenue from chicken sales in the last 6 months increased by around 88 ETB (3.21 USD) over an average of around 103.5 ETB (3.76 USD) in the comparison group.
• Increase monthly expenditure on chicken rearing by around 28 ETB (1.02 USD), compared to average expenditures of 7 ETB (.29 USD) in the control group. The main expenditure category for owners of improved chickens was feed. The increase in average expenditures seen in treatment households is a consequence of a sizeable minority of households taking on greater expenses; the median treatment household saw a marginal increase in expenses compared to the control group.
• Increase egg consumption. Households have an increased Food Consumption Score driven by increases in egg consumption. Based on 24-hr dietary recall, we find that women are 4 percentage points more likely to consume eggs (compared to an average of 4.9% in the control group), but not more likely to consume meat. We find that children under 5 are not significantly more likely to consume eggs or animal protein.
• Do not see an improvement in women’s decision-making power over how income from chickens is used, but report modest gains on a constructed index of women’s decision-making that also takes into account participation in and input on certain chicken-related activities.

Additionally, we do not find any evidence of significant spillovers onto households who did not purchase chickens but lived near other households who did purchase chickens. This is not surprising given the modest magnitude of direct effects that were hypothesized to have driven spillovers, specifically egg production.

Overall, we find that purchasing improved chickens results in a statistically significant increase in income from egg and chicken sales compared to households that own only local breeds of chicken. However, the impact on households’ well-being is likely to be limited given the small magnitude of the effects found; more transformative impact might require owning larger flocks of improved chickens. Furthermore, households experience higher income related to poultry, but also higher poultry-related expenses. Although we do not see increased average net poultry income in our sample, there is significant heterogeneity in the results and the impact of improved chicken ownership on long-term profit remains unclear.

Lastly, we find rather modest results on nutrition and no improvements on children’s consumption of animal protein. This may be partially explained by issues of power: since the overall increase in egg consumption by families is quite modest, it is hard to tease out increases for individual family members. However, it may also point to the need for complimentary interventions if chicken rearing is going to reach its potential as a nutrition booster for children.

Recommendations for future research:

• Assessing the impact of high productivity chickens on medium to long run outcomes to facilitate a better understanding of nutrition and income effects over the lifecycle of chicken ownership
• Exploring why chicken expenses were higher than expected, and whether productivity gains can be achieved without these extra expenses
• Understanding the evolution of chicken expenses, income, and profits as flock size increases
• Addressing the overall income effects of chicken rearing and whether investing in chickens is in addition to or in substitution of other potential investments
• Determining whether complementary interventions aimed at women’s nutrition, women’s empowerment, and children’s nutrition can boost the effect of high productivity chickens on those outcomes