| This study examines the technical efficiency of coffee Arabica ' farmers in Ethiopia and also investigates the factors that could affect these levels of efficiency. This study uses a stochastic frontier approach (SFA) to accomplish this task because SFA acknowledges the fact that not all farmers are equally technically and/or allocatively efficient. Therefore, it explicitly incorporates efficiency differences in the analysis.;Coffee Arabica is selected for this study because of its important contribution to Ethiopian economy and to the coffee producing households. Primary data for the study was collected with the use of a questionnaire from 49 sample coffee farmers during 2006-2007, and the secondary data were provided by Coffee and Tea Authority officers.;The Maximum Likelihood Estimation (MLE) methodology is used to estimate the technical efficiency of 49 sample coffee farmers in Ethiopia. Efficiency analysis is an issue of interest given that the overall productivity of an economic system is directly related to the efficiency of production components in the system. The empirical results show that the estimates of technical efficiency vary among farmers, ranging between 18 percent and 96 percent. In addition, the study reveals the mean technical efficiency for the farmers in the sample is 71 percent, implying that a significant proportion of production is lost due to farm-specific inefficiency that result in production taking place below the frontier. The economic implication of the study is that efficiency in coffee production among Ethiopian farmers could be increased by 29 (1-TE) percent through better use of available resources, given the current state of the technology. This can be achieved through policy interventions that contribute to better access to training, encouraging farmers' participation in soil and forest conservation as well as participation in extension program.;The result of the stochastic frontier production function model reveals that the elasticities of output with respect land (0.503) and capital (0.784) are statistically significant at 5 percent level with t-statistics value of (2.821) and (5.145) respectively. Therefore, an increment of land (farm size) by one percent will increase coffee output by 0.503 percent and an increment of capital by one percent will increase coffee output by (0.784). This indicates the need for improving land quality, increasing farm size and improving technology in order to increase the performance of farmers. The elasticity of family labor is positively related with coffee output; however, it is not statistically significant at 5 percent level.;In addition, the inefficiency model reveals that technical efficiency of coffee farmers increases with age, membership, farming experience in other crops, family size (this implies that larger households face fewer labor bottlenecks), and extension farm visit, while it decreases with altitude and coffee farming experience. In sum, we found a positive effect of the age of the farmer, extension agents' farm visit, membership, farmer experience in other crops (proxy for other source of income), and family size on technical efficiency. However, the result reveals that altitude and coffee farming experience have a negative effect on technical efficiency. The negative effect of coffee farming experience is an unexpected outcome, but it could be related to the fact that coffee trees age as farmers age; and the trees become less productive. On the other hand, the negative effect of altitude is as expected because of poor soil conditions as well as biodiversity loss.;The economic implication of this study is that efficiency in coffee production among Ethiopian coffee farmers could be increased by 29 percent through better use of available sources, given the current state of technology. This increase can be achieved by policy intervention that contributes to better access to training, improving soil and protecting biodiversity loss. |
