tural spending, which was equal to 4.8 percent of government spending in 2002 (Figure 4.5). This share comprised 0.2 percent for irrigation and 1.3 percent for R&E; the remaining 3.3 percent was for other areas of agriculture. Because all ex- penditure areas grew at 2 percent per year in the baseline scenario, there was no change in the fi nal composition of total expenditure. However, in the irrigation sce- nario we gradually increase the share of government expenditure on irrigation from 0.2 to 2.7 percent during 2006–15. In the extension scenario we also increase the share of R&E spending by 2.7 percent, so that agricultural spending as a whole is 10 percent of total spending. This scenario is equivalent to meeting the expenditure target identifi ed in the Maputo Declaration. It is important to note that the sce- narios are cumulative, meaning that the extension scenario includes the effects of the irrigation scenario. Therefore, the counterfactual for the extension scenario is the irrigation scenario rather than the baseline scenario. Increasing irrigation’s share of total spending from 0.2 to 2.9 percent is equiva-
lent to increasing the share of irrigable land under irrigation from 5.3 percent to 19.3 percent by 2015. In the baseline scenario, the share of land under irrigation would have risen to 6.9 percent by 2015. So in the irrigation scenario we are more than doubling the amount of irrigated land over a period of 10 years. The impact of increasing irrigation investment is an acceleration of agricultural growth from 3 percent per year in the baseline scenario to 3.8 percent in the irrigation scenario
Figure 4.5—Final agricultural expenditure shares under investment scenarios, 2015
Source: Kenyan dynamic computable general equilibrium model results. Note: Outcomes are cumulative (for example, “Roads” includes the expenditures from “Irrigation” and “Extension”).