Paintings by Niko Pirosmani, 1862–1918: (Left) ‘White Cow on a Black Background’ and (right) ‘Goat’, Art Museum of Georgia, Tbilisi, Georgia, via Wikiart.

A new and particularly careful and robust analysis by scientists at Cornell University and the University of Illinois at Urbana-Champaign uses a livestock donation program by Heifer International to examine the impacts of expanded animal agriculture on food security and expenditure in a resource-poor community in Zambia.

[T]he livestock donation program provides an experimental setting in which to examine the effects of expanded animal agriculture in an impoverished rural community, where livestock was not already prevalent. . . .

Results show that in this sample and specific context, livestock development increases income, raises the food security of those holding animals, and alters the food environment to enhance the diets of the recipients’ communities.

Other excerpts and findings from this important new paper follow.

‘Smallholder livestock ownership has potential to enhance food security by raising incomes of the poor and by increasing the availability of nutrient-dense foods. This paper exploits the staggered rollout of livestock distribution by Heifer International in Zambia to identify the effects of livestock using statistically similar treatment and control groups in a balanced panel of households. Results indicate that livestock ownership improves dietary diversity through both direct consumption of animal products produced on farm and through increased consumption expenditures. Further results indicate that expanded livestock ownership alters the local food economy to influence food consumption by households lacking farm animals.

‘Livestock ownership is increasingly promoted in food security strategies because farm animals can provide nutrient-dense foods, regular income, and other benefits. Nonetheless, little empirical work actually demonstrates a causal link between livestock ownership and food security (FAO, 2012). Measuring the effectiveness of livestock in alleviating poverty and food insecurity is undermined by endogeneity: those households that have livestock likely differ systematically from those that do not. . . .

‘In this article, particular features of the Heifer International livestock donation program allow us to identify current and future recipients of farm animals. We use future adopters of livestock as a control group for current adopters and collect baseline data on livestock recipients and comparable non-recipients. We use a difference-in-differences approach controlling for time-invariant household characteristics to compare outcomes for the recipients against those who have selected into the program, but have not yet received animals. This approach follows suggestions from de Janvry, Dustan, and Sadoulet (2011) who note that selection bias can be overcome in part through utilization of staggered rollouts, which can be analyzed similarly to randomized control trials (RCTs) even when they lack explicit randomness. Using households who have selected into the program as controls and applying household fixed effects addresses concerns of endogeneity more completely than existing research (DFID, 2014). We run several placebo tests to rule out systematic pre-treatment differences among the groups and other possible confounding factors that could drive our results.

‘Our analysis uses panel data from Zambia covering 300 households over four rounds, spanning 18 months from the distribution of donated animals. To our knowledge, this work is one of the first studies to use a balanced panel of data to examine the effects of livestock on expenditure and food security. While we expect to see a direct impact of donated livestock through increased availability of animal products, we also expect an indirect effect on food security through increased revenue, which can be used to access a wider variety of foods. Moreover, we anticipate potential spillover effects throughout communities as perishable animal products become more available due to increased local production.

‘We find significant effects on household outcomes for recipients of dairy cattle, meat goats, and draft cattle. The receipt of livestock triggers increases in dietary diversity and in consumption expenditure per capita. In all specifications, expenditure per capita has a positive and significant effect on dietary diversity, implying that livestock ownership affects dietary diversity through two channels: first, through the direct impact of ownership on access to animal sourced foods, and second, indirectly, through the increase in total consumption expenditures allowed by livestock ownership. . . . We further find that non-recipients living in communities with recipients also experience increased milk consumption. In our setting, we find the benefits from livestock, particularly dairy cattle, are large. When we compare the size of effects from the livestock gift to those expected from an equivalent gift of cash, we find that the cash gift would have to yield an annual return of nearly 70% to generate the same effects on dietary diversity as the average gift of livestock.

Thus, our findings imply that livestock can have an impact on food security beyond what would emerge from a cash transfer of similar scale and that livestock development may affect the local food economy to enhance food security of non-recipients as well.

‘. . . Both dairy cow and goat ownership have a positive, direct impact on the expected number of food groups eaten by a household on any given day. This result implies that in our setting, expanded animal agriculture leads to increased food security through direct consumption of home-produced animal products. Expanded livestock activities also contribute indirectly to increased dietary diversity through increased expenditure. Long-term ownership of livestock could lead to increased income from the sale of animals, animal products such as milk, and animal services for hauling and plowing. The survey data show that livestock revenue for dairy cow recipients increases more than 100 fold, from less than ZMK5,000 ($1.00) per household per quarter in Round 1 to over ZMK774,525 ($154.91) in Round 4. Goat and draft cattle recipients see gains as well: going from no revenue in Round 1 to ZMK349,300. . . .

The relative strength of the effect of livestock on dietary diversity suggests that in our setting, animal donations, particularly dairy cattle, change the local food economy, making nutrient dense foods more available and accessible in addition to raising incomes.

‘. . . While the increase in consumption is economically significant, whether livestock development can transform households fundamentally through sustained asset accumulation and scaling can only be revealed over a longer time frame. . . .

Livestock donation both increases income and skews expenditures toward food security.

The results of this analysis are conclusive: livestock development among resource poor smallholders in Zambia’s Copperbelt increases household dietary diversity and total consumption expenditures, with dietary impacts that are substantially greater for animals that produce food products for direct consumption.

The data used in this analysis come from the Copperbelt Rural Livelihoods Enhancement Support Project (CRLESP) in the Copperbelt Province of Zambia, which is implemented by Heifer International with funding from Elanco Animal Health (USA).

Read the paper (paywall) in World Development: Milk in the data: Food security impacts from a livestock field experiment in Zambia, by Margaret Jodlowski, Alex Winter-Nelson, Kathy Baylis and Peter Goldsmith, Vol 77, Jan 2016.