Managing resources in erratic environments : an analysis of pastoralist systems in Ethiopia, Niger, and Burkina Faso

In these regions, livestock is generally the predominant production activity, although cropping can also be important in semiarid regions. The ability to move livestock to different pastures is a key strategy for mitigating exposure to erratic rainfall, and reliable access to a wide range of pasture resources has long been essential to the viability and sustainability of such systems. In addition, various types of common-tenure regimes facilitate herd mobility. Although the flexibility inherent in such common-tenure systems enables herders to cope with different rainfall patterns and thus limits their exposure to this risk, one potential cost of such systems may be in terms of the use and management of the natural resource base. As is well known, common resources may be subject to externalities; these externalities open up the possibility that resources will not be well managed. Thus, there may be a trade-off in terms of flexible access to mitigate risk and the use and management of common-pool pastures. In this report, we look at three community-level outcomes that can be affected by both climate variability and by externalities generated when managing the commons is costly. In the second chapter, we develop a theoretical model incorporating variability and costly cooperation. The model generates the following hypotheses: greater variability will lead to lower stock densities, increased herd mobility, and, likely, larger amounts of land allocated to the common pastures. Furthermore, when cooperation is costly, stock densities will be too high, herd mobility will be too low, and pressure will build to encroach on common pasture for private uses. To capture the costs of cooperation, we consider factors that affect cooperative capacity, hypothesizing that greater cooperative capacity reduces the costs of cooperation. Empirical results indicate that there are some general lessons to be drawn. First, greater cooperative capacity does indeed lead to lower stock densities and increased mobility, as we expect from the theoretical model. Cooperative capacity has a more limited impact on land allocated to private uses vs. common pastures, although its impact is particularly strong in Burkina Faso. Also of interest is the great variability in the capacity of communities to manage pastures and allocate land to its best use, both within and across countries. Factors that are generally associated with greater cooperative capacity include relatively small community size, more equal distribution of wealth, and fewer adults migrating for wage work, all of which should reduce negotiation and enforcement costs of undertaking collective action. Other factors affecting cooperation differ across countries. For instance, external pressure to use community resources appears to have a much greater impact on cooperation in Ethiopia than in Burkina Faso or Niger. Higher productivity rangelands and higher effective livestock prices are associated with greater cooperative capacity in Ethiopia, but have no impact in Burkina Faso. Thus, the evidence suggests that more favorable livestock market conditions either increase cooperative capacity or have no impact; there is nothing to suggest that better market conditions erode this capacity. Second, there is little evidence to suggest that livestock owners accumulate larger herds to mitigate vulnerability to rainfall shocks in high variability environments. Our results instead indicate that herd sizes do increase with rainfall variability at relatively low variability, but decrease precisely in the higher variability environments. In other words, we would expect that policies and programs that directly "insure" livestock owners -- through feed subsidies in response to drought, for instance -- would likely lead to larger herds precisely in the environments subject to the greatest variability. We must emphasize that our results are consistent with this hypothesis, but, given the one-period nature of the survey, we did not test the hypothesis directly. This is still a contentious issue, as a wide range of researchers, policymakers -- and indeed, herders themselves -- believe that holding onto more livestock is a strategy to mitigate the impact of such climate shocks as drought. Results presented here imply that policymakers designing crises mitigation strategies -- including those in many governments that are signatories to the UN Convention to Combat Desertification -- must carefully consider insurance and crises mitigation mechanisms that do not lead to dramatic increases in the size of the national herd. Not only do policymakers need to consider the impact of such programs on herd size, but also on herd mobility. Mobility remains an important part of these systems; our results indicate that current rainfall patterns -- and thus, locally available feed resources -- heavily influence the extent of herd mobility. Given the rather complicated patterns of herd mobility into and out of community areas in Ethiopia, we were not able to gather good enough data to include this variable in the statistical analyses. Still, more than 84 percent of the communities surveyed relied on mobility for at least part of the previous year, and in the 12 communities for which data were quite good, herds were mobile for nearly 40 percent of the year. The number of communities where at least some members engaged in herd mobility is lower in Niger and Burkina Faso, but mobility is still practiced in more than 40 percent of communities in both these countries. And, as noted above, better cooperative capacity within communities supports greater herd mobility. Nonetheless, herders" rights to access traditional grazing areas are generally eroding everywhere. Results indicate that communities with more traditional pastoralists do tend to rely more heavily on herd mobility, but the impact is weak and not robust across specifications. Thus, pastoral land tenure and drought mitigation policies will need to take into account the continued reliance on herd mobility -- even by those not considered to be traditionally pastoralist -- as well as factors that are either directly or indirectly limiting mobility. (Author abstract)