|dc.description.abstracteng||Global demand for food and farm commodities continues to grow, while land and other natural resources are becoming increasingly scarce. In Sub-Saharan Africa (SSA), rising population, decreasing per capita arable land, soil degradation and changes in weather patterns are the major challenges affecting productivity and welfare of smallholder farmers. Existing agricultural practices may not be able to meet production needs in the face of these challenges. Sustainable intensification has been proposed as one way of increasing agricultural productivity in a socially and environmentally responsible way. This requires a broad portfolio of technologies, including input-intensive and various natural resource management (NRM) practices. However, while there is consensus that these agricultural technologies play an important role in improving the welfare of smallholder farmers, there are disagreements as to which type of technology is best suited for farmers in SSA. Some support input-intensive technologies while others support NRM technologies. In reality these technologies are not incompatible and there may even be synergies in combining them. We use nationally representative data from maize production systems in Kenya to understand tradeoffs, complementarities and synergies that exist between different input-intensive and NRM technologies. Maize is the most important crop in Kenya, providing much of the daily calorie requirements and is grown by the majority of smallholder farmers in almost all of the country’s agroecological zones.
This dissertation comprises three essays. In the first essay, we analyze adoption as well as tradeoffs and complementarities that exist between different types of input-intensive and NRM technologies practiced by smallholder maize farmers in Kenya. There is a lively debate about which type of technology is the most appropriate to foster sustainable development. In the public debate, the two strategies are often perceived as incompatible. Environmental non-governmental organizations (NGOs) in particular consider NRM practices, which are mostly low-external input strategies as the only sustainable form of agriculture, a view that has considerable influence on policymakers and the international donor community. Most existing adoption studies have either looked at input-intensive technologies or at NRM techniques, using different data and methodologies, so that comparisons were not easily possible. We apply a multivariate probit in the analysis since it allows for correlation between the error terms in the adoption of these technologies. This enables us to analyze adoption of different technologies simultaneously. Specifically, we consider seven technologies; improved seeds, chemical fertilizers (input intensive), terracing, soil bunds, zero tillage, crop residue management and use of animal manure. Results indicate that NRM technologies and strategies that build on external inputs are not incompatible. Interesting complementarities exist, which are not yet sufficiently exploited, because many organizations promote either one type of technology or the other, but rarely a combination of both. NRM technologies are mostly promoted by the public extension service and NGOs, whereas for improved seeds and mineral fertilizer the private sector plays a larger role. While this view is short-sighted, it influences development programs and prevents more widespread implementation of combined approaches that can bring about important synergies.
In the second essay, we analyze income effects of various technologies and technology combinations. Possible synergies in smallholder environments are not yet sufficiently understood since most impact studies focus on the effects of single technologies. We compare income effects of various input-intensive technologies, NRM technologies, and selected combinations. We use propensity score matching to correct for selection bias. When adopted alone, some innovations produce positive effects, while others do not. Effects of certain technology combinations are larger. The largest income gains occur when improved seeds are adopted together with organic manure and zero tillage practices. This clearly underlines that there are important synergies between input-intensive and NRM technologies. However, the number of farmers that have adopted such promising technology combinations is relatively small, again implying that synergies are not yet fully exploited.
In the third essay, we estimate and make a distinction between technical efficiency (TE) and environmental gaps among maize farmers distributed across different agroecological zones (AEZs). Reducing inefficiencies among smallholder farmers is one way of sustainably increasing yields. While there are numerous studies on productivity in SSA, few make a distinction between TE and environmental gaps resulting from climatic differences. Not differentiating can contribute to misinformed policies, for instance when existing environmental gaps are falsely attributed to farmers’ inefficiencies. The AEZs in Kenya differ substantially in their climatic conditions and other factors. We apply the stochastic meta-frontier production function framework, which allows distinguishing between TEs and meta-technology ratios (which capture environmental gaps). We further assess factors explaining inefficiencies among farmers in each of the AEZs as well as those associated with meta-technology ratios (MTRs). We find large deviations between TEs and MTRs across the AEZs. TEs relative to the group frontiers are relatively high; in some zones (mostly the drier zones) they are two to three times higher than the TEs relative to the meta-frontier. This suggests that farmers compare much more favorably with farmers in the same AEZ as opposed to the whole industry. In the same vein, results suggest that environmental gaps contribute more to observed yield differences than farmers’ inefficiencies, implying that farmers have not been able to sufficiently adapt their agricultural practices to the constraints they face. This interpretation is supported by low magnitudes of MTRs coupled with wide variation across the AEZs. Environmental gaps can be explained by the amount of rainfall received, maximum daily temperatures, population pressure, access to infrastructure, incidences of maize lethal necrosis disease and technologies adopted. Efforts to narrow down environmental gaps among farmers are thus more urgent than efforts to reduce technical inefficiencies.
We draw several conclusions from the study. Synergistic income effects exist between input-intensive and NRM technologies. However, these synergistic relationships have not been fully exploited. Furthermore, some technology combinations which were shown to be beneficial in other places are rarely observed among Kenyan farmers. This may be due to the divide among different organizations on the type of technologies they promote. Some tend to focus on promoting specific seed varieties or techniques, rather than more holistic approaches to increasing yields while protecting the environment. To be able to exploit these synergies there is need for more integrated extension by different organizations. In addition, narrowing down environmental gaps among farmers in different AEZs should be a priority for increasing maize yields. This can be achieved through promotion of appropriate technologies necessary to adapt to environmental stresses, such as drought and heat tolerant varieties for the dry regions as well as proper infrastructure. Further research is also necessary to be able to understand more synergistic associations between different types of technologies.||de