Structure of Public Support for National Agricultural Research Systems: A Political Economy Perspective

This paper initiates development of a set of stylized facts concerning the structure of public support for national agricultural research systems (NARS) within a neoclassical political economy framework. The aim is to place public funding of NARS in the broader context of the overall level of direct government assistance to agriculture. Using a newly constructed data set on NARS expenditures over the 1970‐85 period, we observe a growing disparity in agricultural research intensity ratios, which measure the level of public support for NARS in relation to agricultural gross domestic production (Aggdp) between low and high‐income countries. This growing disparity appears to be driven by much larger increases in support for agricultural research by high‐income countries, coupled with a significantly slower growth in the size of their agricultural sector, despite the propensity of low and middle‐income countries to increase real support to agricultural research. As per‐capita incomes rise the public agricultural expenditure ratio, which measures public expenditures on agriculture relative to the size of the agricultural sector, Aggdp, increases substantially. Public expenditures on agriculture were indexed on agricultural and non‐agricultural populations to give a rough indication of the increasing incentives for rural 'distributional coalitions' to seek a redistribution of public expenditures in their favor. A relative research expenditure (rre) ratio is developed, which measures the proportion of total public expenditure on agriculture spent on agricultural research. It provides an indication of the relative importance given to research on agriculture within the constraints imposed by overall public spending on agriculture. In contrast to the agricultural research intensity ratios, the rre ratios suggest that agricultural research appears to command as large a share of the public purse devoted to agriculture in low and middle‐income countries as it does in high‐income countries. Expectations derived from the neoclassical political economy literature that research may have fared relatively better in high compared with low‐income countries were not supported by the data.