Last updated on August 2, 2015
That’s the title of a new working paper which my PhD student Lindsey Novak and I have just finished working on this past week, and in which we ask whether smallholder participation in agricultural value chains via contract farming leads to better food security.
In case you are not familiar with contract farming, it is the economic institution wherein a processing firm contracts the production of some agricultural commodity out to a grower (in developing countries, growers are typically smallholder households). In the literature, contract farming agreements are also known as grower-processor contracts, or as outgrower schemes, and because it is halfway between spot markets (i.e., growers selling their output at market after harvest) and vertical integration (i.e., processors growing, transporting, processing, and so on), contract farming has been referred to as a vertical coordination contract (i.e., it allows coordinating the actions of agents upstream and downstream).
There is now a good amount of evidence that participation in contract farming increases the income of smallholders who choose to participate as growers. I myself published an article in World Development in 2012 in which I asked whether participation in contract farming improved welfare via increased incomes.
In this paper, we use the same data I used in my 2012 article to look at a longer causal chain and ask whether participation in contract farming leads to improved food security in the form of shorter hungry seasons (i.e., the length of time a household goes without eating three meals a day). Why wouldn’t increased incomes not necessarily lead to shorter hungry seasons? Because the income from participating in contract farming materializes immediately around harvest time, and the hungry season occurs in the months leading to harvest, i.e., almost one year after harvest. Because it is not always possible to save, and because of self-control problems, it is not immediately obvious that a greater income will improve food security almost a year later.
Because we look at duration data–the hungry season is measured in months–and because we adopt a selection-on-observables design thanks to the field experiment that was conducted when collecting the data to elicit people’s willingness to pay (WTP) to participate in contract farming, we can look at the following:
- In a regression context, whether participation in contract farming leads to a shorter hungry season on average. It turns out that participation in contract farming appears to shorten the hungry season by about 10 days on average.
- Also in a regression context, whether participation in contract farming makes it more likely to see one’s hungry season end at any given time during the hungry season. It turns out that participation in contract farming increases the likelihood that one’s hungry season will end by about 20 percent at any given time.
- Also in a regression context, whether participation in contract farming is more beneficial for households with more children, and for households with more children of either gender. It turns out that participation in contract farming has more beneficial effects for households with more children and households with more girls.
- In a matching context, given the selection-on-observables design, what is the effect of participation in contract farming on average (ATE), on those who choose to participate (ATT), and on those who choose not to participate (ATU). As one would expect, in absolute value, ATT > ATE > ATU, and the estimated ATE from matching is very close to the estimated ATE from regression.
Here is the abstract of the paper:
Contract farming has often been associated with an increase in the income of participating households. It is unclear, however, whether contract farming increases other aspects of household welfare. Using data from six regions of Madagascar and a selection-on-observables design in which we control for a household’s marginal utility of participating in contract farming, which we elicited via a contingent valuation experiment, we show that participating in contract farming reduces the duration of a household’s hungry season by about ten days on average, and that it makes participating households about 20 percent more likely to see their hungry season end at any point in time. Further, we find that these effects are more pronounced for households with a larger number of children, and for households with a larger number of girls. This is an important result as children—especially girls—often bear the burden of food insecurity.