The contribution of agricultural input industries to agricultural innovation

This paper presents an analytical framework to measure how farmers (i.e., primary agriculture) in countries at different stages of economic development absorb new technology by buying inputs from other industries. By combining information on the use of purchased inputs in primary agriculture (based on country-specific input±output matrices) with information on R&D intensities in the various supplying industries (based on S&T surveys), an approximation can be made of the R&D investments embodied in those inputs. These "acquired" R&D investments can then be contrasted with local R&D investments that target primary agriculture directly, creating a more complete picture of the sources of innovation in primary agriculture. Data from three case study countries representing low-, middle- , and high-income countries (Colombia, Brazil, and the Netherlands) reveal quite remarkable differences and unexpected results, such as: the poorest of the three countries (Colombia) is relatively most dependent on technology that is imported, privately financed, and nonagricultural, while the richest of the three countries (the Netherlands) scores only highest in absolute terms; most acquired R&D concentrates in just two industrial clusters: the agrochemical industry and the agricultural machinery and transport equipment industry; moving from poor to rich countries, a deepening and diversification of input use in primary agriculture takes place mainly in relatively low-tech industries (e.g., services and energy). As a result, the average R&D intensity of agricultural inputs used in Dutch agriculture is not higher than those used in Colombian or Brazilian agriculture.