Growth in total factor productivity in the Ethiopian agriculture sector

Growth accounting and econometric assessments of sources of growth

During the 2003/04–2008/09 period agricultural production in Ethiopia grew annually at 9.3 percent while cultivated area expanded at 4.7 percent. The remaining growth resulted from intensive use of other inputs, increased productivity, increased efficiency, or a combination of these factors. This study applies growth accounting and two econometric approaches on sector and administrative zone level data to investigate the issue.
The baseline three-factor growth accounting specification applied to sector level data covering the 2004/05–2009/2010 period implies an average year-to-year change in total factor productivity (TFP) of 4.5 percent. The five-factor specification that includes the effects of intermediate manufacturing and services inputs implies that out of the 8.4 percent average, annual growth in real value of output TFP growth accounted for 4 percent. Out of the remaining 4.4 percent growth in output, increased application of labor accounted for 2.7 percent, land and capital for 1 percent, and intermediate inputs for 0.7 percent. A modification of the growth accounting model, to take into account factors that indirectly affect TFP, resulted in an average annual growth in TFP of 2.7 percent. I applied the Cobb-Douglas and stochastic production frontiers on zone level data of four agriculturally important regions. The results of the Cobb-Douglas production function (CDPF) imply that TFP grew at an average annual rate of 5.6 percent. Growth in cultivated area and employment of labor accounted for 3.6 and 1.0 percent of the growth in output. Similarly, results of the stochastic production frontier model imply that growth in labor and cultivated area accounted for 0.6 and 3.7, respectively, of the growth in output. Changes in TFP and gains in efficiency accounted for 3.9 and 0.5 percent of the growth in output. A typical farmer in these regions had an average level of efficiency of 0.73, which implies that output can be increased by about 37 percent from its current level without increased application of inputs.
During the same period, efficiency improved as the proportion of literate farmers grew. It also improved with the application of the extension package, and as the number of farmers who receive advisory services grew. The fact that growth in efficiency between 2003/04 and 2005/06 represents about 84 percent of the average annual change in TFP, which mainly occurred between 2003/04 and 2005/06, has an important policy implication. This result, together with the sources of growth in efficiency, implies that the Ethiopian government’s efforts to educate farmers and expand modern production practices—efforts that have been going on for well over a decade—are starting to pay off.

Bachewe, Fantu Nisrane
Published date: 
International Food Policy Research Institute (IFPRI) and Ethiopian Development Research Institute (EDRI)
Series number: 
PDF file: