Agent‐based modeling for ex ante assessment of tree crop innovations: litchis in northern Thailand

This study uses an agent‐based model for ex ante assessment of agricultural innovations. The model builds on whole farm mathematical programming (MP) and extends the methodology with a spatial representation of the system, the heterogeneity of farm households and landscapes, and the interaction between farm households. We apply the model in a northern Thai watershed to study the potential of four innovations to increase the profitability of litchi orchards. Cost‐benefit analysis shows that each innovation would increase the profitability of litchi growing; however, the results of the agent‐based model show that at current price levels these innovations alone would not be enough to stem the decline in the area under litchis. The model was validated and the sensitivity of the results tested for variations in the irrigated water supply and liquidity. We report on how farmers responded to these results and discuss the implications for other areas in northern Thailand.     

Introduction to the Special Feature: Evaluating the Impact of Agricultural Projects in Developing Countries

This article provides an introduction to the special feature on impact evaluation of agricultural projects in developing countries. The special feature is motivated by an increased interest in impact evaluation both within the economics profession and the development community. The article highlights methodological issues in conducting such evaluations and discusses the current literature that empirically assesses the effectiveness of agricultural projects. A synthesis of the articles presented in the special feature and their contribution to the literature are noted as well as conclusions on the next steps in evaluating agricultural projects in developing countries.     

Impact of cooperatives on smallholders' commercialization behavior: evidence from Ethiopia

This article examines the impact of marketing cooperatives on smallholder commercialization of cereals using detailed household data in rural Ethiopia. We use the strong government role in promoting the establishment of cooperatives to justify the use of propensity score matching to compare households that are cooperative members to similar households in comparable areas without cooperatives. The analysis reveals that although cooperatives obtain higher prices for their members, they are not associated with a significant increase in the overall share of cereal production sold commercially by their members. However, these average results hide considerable heterogeneity across households. In particular, we find that smaller farmers tend to reduce their marketed output as a result of higher prices, whereas the opposite is true for larger farmers.     

Sharecropping efficiency in Ethiopia: threats of eviction and kinship

We tested a theoretical model with the Marshallian inefficiency (H1) and threat of eviction (H2) hypotheses having opposite effects on land productivity on sharecropped plots. The model also assumes that kinship contracts may eliminate or reduce the Marshallian inefficiency (H3) and threat of eviction (H4) effects on land productivity. Our empirical findings were consistent with H2 and H4 being true. We found higher land productivity on sharecropped plots than on share tenants' own plots and higher land productivity on sharecropped plots of nonkin than of kin tenants. The nature of the data allowed controlling for unobservable household characteristics through household fixed effects and for observable plot characteristics. Analyses with and without plot characteristics revealed that these findings were stronger with plot characteristics than without them. Based on the plausible assumption that observable plot characteristics are positively correlated with unobservable plot characteristics this strengthens our conclusion. The results are also supported by first-order stochastic dominance analysis. Sharecropped plots' output value distribution unambiguously dominated the output value distribution from share tenants' own plots. Nonkin sharecropped plots' output value distribution also first-order stochastically dominated the output value distribution from kin sharecrop plots.     

Land use and income diversification: comparing traditional and colonist populations in the Brazilian Amazon

This article compares and analyzes land use and income diversification among two distinct groups of farmers in the Brazilian Amazon: recent colonists in Ouro Preto do Oeste, Rondônia, and traditional long‐term residents along the Tapajós River, Pará. We investigate the hypothesis that farmers who diversify their cash income sources clear less forest on an annual basis, and we compare these livelihood choices across colonist and traditional populations. In particular, we develop a conceptual model based on the household production framework and use econometric models to identify determinants of diversification and forest clearing. We find that diversification of agricultural cash crops is negatively correlated with forest clearing by colonists, providing limited evidence for the hypothesis. Other significant covariates of diversification and forest clearing include cash income levels, stage in family life cycle, cattle ownership, and chemical inputs. Differences in these variables, and differences in household response to these variables, explain variation in diversification and forest clearing across the two populations.     

Investment in soil conservation in northern Ethiopia: the role of land tenure security and public programs

A double hurdle statistical analysis of 250 farms in the Tigray region of Ethiopia reveals different causal factors for soil conservation adoption versus intensity of use. Farmers' reasons for adopting soil conservation measures vary sharply between stone terraces and soil bunds. Long‐term investments in stone terraces were associated with secure land tenure, labour availability, proximity to the farmstead and learning opportunities via the existence of local food‐for‐work (FFW) projects. By contrast, short‐term investments in soil bunds were strongly linked to insecure land tenure and the absence of local food‐for‐work projects. Public conservation campaigns on private plots reduced adoption of both stone terraces and soil bunds. Whereas capacity factors largely influenced the adoption decision, expected returns carried more influence for the intensity of stone terrace adoption (measured as metres of terrace per hectare). More stone terracing was built where fertile but erodible silty soils in higher rainfall areas offered valuable yield benefits. Intensity of terracing was also greater in remote villages where limited off‐farm employment opportunities reduced construction costs. These results highlight the importance of the right kind of public interventions. Direct public involvement in constructing soil conservation structures on private lands appears to undermine incentives for private conservation investments. When done on public lands, however, public conservation activities may encourage private soil conservation by example. Secure land tenure rights clearly reinforce private incentives to make long‐term investments in soil conservation.     

Heterogeneity of commercialization gains in the rural economy

Farmers in low‐income rural economies often fail to switch to cash crops from staple production despite the positive income effects of commercialization expected by policy makers. The literature suggests that market failures prevent households from cash crop adoption but remains inconclusive regarding farmers' motives and the effects of commercialization. This article contributes to the debate by offering an original approach to the analysis of commercialization outcomes and farmers' decisions. It consists of applying a model with essential heterogeneity and a semiparametric estimation technique to analyzing harvest value returns to cash cropping. Using Malawian data, we show considerable heterogeneity in harvest value returns to cash cropping both within and between groups of farmers choosing different crop portfolios. Importantly, the results imply rational choices based on comparative advantage considerations of farming households: farmers self‐select into the activity where they expect higher gains and adopt cash crops when facing weaker market barriers.     

Agricultural production and children's diets: evidence from rural Ethiopia

We study the relationship between pre‐school children's food consumption and household agricultural production. Using a large household survey from rural Ethiopia, we find that increasing household production diversity leads to considerable improvements in children's dietary diversity. However, we also document how this nonseparability of consumption and production does not hold for households that have access to food markets. These findings imply that nutrition‐sensitive agricultural interventions that push for market integration are likely to be more effective in reducing under‐nutrition than those promoting production diversity.     

Tenure security, resource poverty, public programs, and household plot-level conservation investments in the highlands of northern Ethiopia

Land degradation poses a serious problem for the livelihoods of rural producers. Furthermore, there is rarely enough private investment taking place to commensurate the scale of the problem. This article examines the role of tenure insecurity, resource poverty, risk and time preferences, and community‐led land conservation on differentiated patterns of household investment in land conservation in northern Ethiopia. We control for biophysical, household characteristics, market access conditions, and village level factors. Investments in soil bunds and stone terraces are specifically studied so as to capture the link between these various factors and the durability of conservation investments. We introduce the distinction between the determinants of the decision to invest and how much to invest in conservation. Regression results show that publicly led conservation programs seem to significantly stimulate private investment. A host of plot‐level variables and household perceptions of returns on conservation investments, expressed in terms of perceived improvements in land quality and increased crop yields, were found to be critical to the decision to invest and intensify soil conservation. The evidence on the significance of households' attitudes toward risk aversion suggests the important role of risk and the household's risk‐bearing capacity in the decision to intensify conservation measures. At the same time, tenure security indicators and households' resource endowments (resource poverty) had weaker effects in increasing willingness to invest and the level of investment made. The policy implications of these results point to the importance of agricultural research and extension efforts that target technologies which reduce household risk and poverty while enabling sustainable investments in conservation measures by individual households.     

Does human capital raise farm or nonfarm earning more? New insight from a rural Pakistan household panel

This study explores how human capital affects farm household earnings using two tools to refine measurement of human capital effects. First, it employs a two‐sector model to allow the allocation of family labor between farm and nonfarm activities. Second, it accounts for village fixed effects to evaluate whether results from panel data differ meaningfully from a cross‐sectional data analysis with local binary variables. The results show that education has a negligible effect on farm earnings; instead, experience appears to be the principal channel by which human capital affects agricultural performance in a traditional rural setting. Our results also suggest that prior models that fail to separate nonfarm activities spuriously exaggerated the effect of education to the farm sector. In addition, typical cross‐sectional analyses that ignore fixed effects may cause the effects of education on rural household earnings to be significantly overstated. The fact that panel data regressions accounting for village‐level fixed effects found only one instance of education raising earnings—the effect of literacy on nonfarm income—suggests that considerable heterogeneity may have been ignored in cross‐sectional data analyses, especially ones that omitted village‐level effects.     

A revised Tobit procedure for mitigating bias in the presence of non-zero censoring with an application to milk-market participation in the Ethiopian highlands

Fixed transactions costs that prohibit exchange engender bias in supply analysis due to censoring of the sample observations. The associated bias in conventional regression procedures applied to censored data and the construction of robust methods for mitigating bias have been preoccupations of applied economists since Tobin [Econometrica 26 (1958) 24]. This literature assumes that the true point of censoring in the data is zero and, when this is not the case, imparts a bias to parameter estimates of the censored regression model. We conjecture that this bias can be significant; affirm this from experiments; and suggest techniques for mitigating this bias using Bayesian procedures. The bias-mitigating procedures are based on modifications of the key step that facilitates Bayesian estimation of the censored regression model; are easy to implement; work well in both small and large samples; and lead to significantly improved inference in the censored regression model. These findings are important in light of the widespread use of the zero-censored Tobit regression and we investigate their consequences using data on milk-market participation in the Ethiopian highlands.     

The impact of water conservation and intensification technologies: empirical evidence for rice farmers in Ghana

This study employs the propensity score matching model to examine the impact of the adoption of water conservation and intensification technologies on farm output and income among small-scale lowland rice farmers in the Northern Region of Ghana. The matching was conducted based on the Mahalanobis distance combined with the propensity score. Balancing tests by checking the mean standardized absolute bias in the matched sample were conducted as well as a sensitivity analysis was conducted to check for hidden bias due to unobservable selection. The empirical results show a positive and significant impact of the adoption of bunds technology on input demand as well as a positive but insignificant impact on output supply and net returns. Adopters of the dibbling technology were found to have higher output supply, whereas no statistically significant difference was found between the incomes of adopters and nonadopters of dibbling seed method. The results, however, reveal a positive and significant effect on output and net returns when the dibbling method is combined with intensified weeding. The estimates were found to be relatively insensitive to hidden bias.     

Land titling policy and soil conservation in the northern uplands of Vietnam

In Vietnam, a quasi-private property regime has been established in 1993 with the issuance of exchangeable and mortgageable long-term land use right certificates. Using primary qualitative and quantitative data collected in a mountainous district of Northern Vietnam, this paper investigates the role of the land policy in the adoption of soil conservation technologies by farmers. This issue is of crucial importance in the region where population growth and growing market demands have induced farmers to intensify agricultural production. While poverty has been reduced, environmental problems such as soil erosion, landslides, and declining soil fertility have become more severe over the past years. Our findings suggest that despite farmers’ awareness of erosion, soil conservation technologies are perceived as being economically unattractive; therefore, most upland farmers continue to practice the prevailing erosion-prone cultivation system. Focusing on agroforestry as one major soil conservation option, we estimate household and plot-level econometric models to empirically assess the determinants of adoption. We find that the possession of a formal land title positively influences adoption, but that the threat of land reallocations in villages discourages adoption by creating uncertainty and tenure insecurity. The analyses reveal that these two effects interact with each other but are of small magnitude. We conclude that the issuance of land titles is a necessary but not sufficient prerequisite to encouraging the adoption of soil conservation practices. However, current practices remain economically unattractive to farmers. This deficiency needs to be addressed by interdisciplinary research and complemented by strong efforts by local authorities to promote sustainable land use.     

How sustainable is participatory watershed development in India?

Watershed conservation is widely recognized as a major strategy for rural development throughout the developing world. In India, the apparent success of participatory approaches to watershed development resulted in a decentralization of project planning, implementation, and management to local communities at the village scale. We explore the effectiveness of this so‐called community‐based approach in achieving sustainable soil and water conservation in four semi‐arid regions in India, and analyze what factors explain project success. We confirm the result of earlier studies that participatory approaches are more effective in establishing soil and water conservation in the short run. However, our main result is that investments in community organization fail to ensure household commitment to maintenance in the longer term. Without better returns to investment in soil and water conservation and without local institutions to coordinate investment in the long run, the sustainability of participatory watershed management is seriously threatened.     

Supply response in Ethiopia: accounting for technical inefficiency

Few empirical studies of supply response using the profit function have accounted for technical inefficiency. Using farm‐level panel data from Ethiopia, this study examines the effect of incorporating technical inefficiency in estimating the supply response of peasant farmers. Two systems of output supply and input demand equations are estimated and compared: the conventional model in which technical efficiency is assumed and another in which technical inefficiency is explicitly incorporated. The model with technical inefficiency is preferred on grounds of theoretical consistency and improved estimates, although model comparison tests are not conclusive. Incorporation of inefficiency generally increases the magnitudes and the statistical significance of own price elasticities, substantially so in the case of fertilizer and fertilizer‐intensive crops, and alters the priority attached to nonprice factors. An important result is that only the specification with inefficiency reveals a significant effect of access to extension services on output. Only this specification finds that output increases with household size, which one expects as the farms in the sample are largely subsistence and producing for own consumption. Furthermore, the results demonstrate that farmers' response to incentives is considerably restricted by inefficiency, suggesting that the traditional model would overstate response by excluding the efficiency variable.     

Are small‐scale rice farmers in eastern India really inefficient? Examining the effects of microtopography on technical efficiency estimates

The article analyzes how controlling for differences in land types (defined by position on a low‐scale toposequence) affects estimates of farm technical efficiency for rice farms in eastern India. Contrasting previous research, we find that farms are considerably more technically efficient when efficiency estimates are carried out at the plot level and control for plot characteristics rather than at the farm level without such controls. Estimates show farms cultivating modern varieties are technically efficient and plots planted with traditional varieties on less productive lands (upland and midupland) operate close to the production frontier. Significant technical inefficiency is found on more productive lands (medium and lowland plots) planted with traditional rice varieties. The finding that these smallholder rain‐fed rice farms are efficient cultivators on some plots contrasts with previous findings of farm‐level inefficiency (i.e., rejects overarching explanations linked to farm operator ignorance or lack of motivation) and suggests more complex explanations are required to address the inefficiency that is present.     

Diffusion and spillover of new technology: a heterogeneous-agent model for cassava in West Africa

Understanding what determines the geographic spread of innovations can help guide the funding and implementation of research and extension programs. Our approach uses household survey data as model parameters, to simulate behavior across the entire surveyed population and avoid the aggregation bias associated with representative‐farm models. Such a “heterogeneous agent” approach allows us to infer the distribution of a technology's impacts across one set of households, and predict the potential for spreading to another set that shares similar characteristics with respect to natural resource endowments and farming systems. We apply the technique to new cassava varieties in West Africa, finding a strongly poverty‐alleviating impact, with substantial spillover potential from Nigeria to neighboring countries.     

Rural nonfarm activities and agricultural crop production in Nigeria

Although most rural households are involved in the farm sector, the nonfarm sector has grown significantly in recent decades, and its role in rural development has become increasingly important. This article examines the effect of participation in nonfarm activities on crop expenses of farm households in Nigeria. The relationship is modeled using a nonseparable agricultural household model that suggests that participating in nonfarm activities can relax the credit constraints facing farm households and reduce risk thereby helping households improve farm production and smooth consumption over time. The results show that participation in nonfarm activities by Nigerian farmers has a positive and significant effect on crop expenses and in particular on payments for hired labor and inorganic fertilizers. Separate analysis of the six geopolitical zones in Nigeria indicates that it is in the South-South and South-East zones where nonfarm participation appears to induce more hiring of labor. The results support the hypothesis that nonfarm participation helps relax liquidity constraints but suggests how that liquidity is used is zone-specific. In general, the results also indicate that liquidity is used more to pay for inputs into staple production as opposed to cash crops.