Another plausible reason why households often allocate more labor to some plots relative to others
Collective farming remains widely practiced among subsistence farming households in West Africa, despite numerous accounts of the inefficiencies of such a centralized household production system. A concurrent rise in the individualization of farming has led to the coexistence of collective and individual plots within the same household Guirkinger and Platteau (2014). Among these households, labor remains an important productive resource and given that most agricultural technologies among subsistence farmers tend to be labor intensive Barrett et al. (2004) it, it is critical to examine the implications of the simultaneous management of the two types of plot ownership for labor allocation.
In my Job Market paper (JMP), which uses a two-wave panel dataset (2009 and 2012) from Mali, I find that collective plots –managed by household heads of extended families, with output shared amongst all family members– are more intensively farmed with labor by more than 40% relative to individual plots, managed by other male household members. However, an overwhelming majority of the literature finds evidence that collectively managed farms suffer from low effort provision leading to lower overall output level Carter, 1987; Guirkinger and Platteau, 2014)).
As a result, a natural question arises from the current findings: Why do we find that collective plots are more intensively farmed? Particularly, why are my findings different from those obtained by Guirkinger and Platteau (2014) where the authors find the opposite to be the case in another locality in the same country? While it is not possible to offer a definite answer to these questions, I posit that motives for insurance against production risk partly explain these observed labor and yield gaps across the two types of plots management.
How can collective farming reduce production risk despite same covariate shocks?
The argument that collective farming serves as insurance against production risk due to weather variability may seem at first glance counter-intuitive, since rainfall instabilities are perceived as covariate shocks and, consequently, assumed to affect all plots in a given region in a similar manner. Nevertheless, as I describe in my JMP, the setting of the study area makes this arrangement mechanism suitable for ex ante risk mitigation. Specifically, rainfall, which usually occurs from May to September, constitutes the main source of water for plots in the region, as households have very limited access to irrigation. Thus, untimely execution of farming activities during the rainy season can have dire impacts on crop yield performance. At the same time, while labor is an important production factor among these households, they are not able to hire labor either because of liquidity constraints or missing labor markets. As a result, households in the region are at risk of unpredictable labor bottlenecks when there are weather instabilities such that they have to accelerate or repeat the execution of some agricultural activities. Given that households composed by extended family members are usually large and likely to be heterogeneous, by farming collectively, it is possible for them to overcome the adverse impacts of weather instabilities. Precisely:
· Extended households are able to mobilize a larger pool of labor enabling them to apply “brute strength” labor.
· Heterogeneity of the extended households allows them to have workers with different endowments. While younger members are physically more productive in the field, elder members are likely to have more experience in strategic planning in the wake of weather instabilities (Rosenzweig and Wolpin, 1985).
Empirical evidence of motives for risk insurance against ex ante risk
To test the claim that differences in labor allocation and crop yields are due to motives for ex ante risk insurance, I introduce village level historical rainfall variability, represented by the coefficient of variation (CV) of rainfall from 1981 to 2007 in the labor allocation and yield regressions. I find that difference in total labor allocation between collective and individual plots among households in sites with high rainfall variability (defined as places where the CV of rainfall is one standard deviation above the sample mean) is about 0.43 while the corresponding figure for households in villages with low rainfall variability (where the CV is one standard deviation below the sample mean) is only 0.26. With a sample mean equals to 105 person-days per hectare, these differences translate to about 18 person-days per hectare. The differences in labor allocation are substantiated in the yield results where the gap in monetary value is about 34,000 FCFA (i.e., $68). Furthermore, I find that households living in places with high rainfall variability are more likely to participate in collective farming relative to those living in areas with low rainfall variability area by up to six percentage points.
Sample restrictions for internal validity and identification strategy
Since this paper aims to investigate labor allocation and crop yield across collective plots and individual plots, I restrict the sample to households within which collective and individual plots coexist. Similar as in Udry (1996), the estimations are carried out at the plot level controlling for household-year-crop-fixed effects (i.e., for plots grown to the same crops, in the same household, and in the same year). One major additional analysis not yet performed in existing similar work, to my knowledge, is the robustness check of the finding controlling for plot manager year fixed effects.
The findings of higher labor allocation on collective plots relative to individual plots suggest efficiency loss. Since there are decreasing marginal returns to labor, households could increase their productivity by reallocating labor from collective to individual plots. As a result, if motives for ex ante risk insurance are plausible driving factors in the observed labor allocation gaps, increasing farmers’ access to productive safety nets and developing labor markets would increase overall efficiency. Yet, the differences in the empirical literature regarding rural household economies highlight the challenge faced by development practitioners in West Africa to improve aggregate measures of household welfare in contexts where intra-household dynamics can be location-specific. It is therefore often difficult to tailor policies to specific groups of people or locations. This stresses the need for continuous research efforts in understanding household production decision making in these regions.
A shortcoming in the current work is the inclusion of farming management structure as an exogenous variable because of the inability to use the instrumental variable approach, as potential instruments for this variable are most likely to be correlated with the household labor allocation decisions and crop yield performance. Ideally, one would control for plot fixed effects. Unfortunately, such an approach is not feasible because the data are not plot-panel data. In addition, plot management structure is unlikely to change over time.
Another point to highlight here is that the current work only provides suggestive evidence that collective farming constitutes an ex ante risk mitigation strategy for households in the study area. It does not claim to establish a framework to determine a causal correlation between rainfall variability and the household decision to participate in collective farming.
Aissatou Ouedraogo is a PhD student at Michigan State University