Diffusion of new knowledge and technologies in agriculture can offset the sometimes explosive nature of food price increases, especially in developing countries. Closely following the notion of innovative geographic clusters, this thesis examines knowledge flows in the US agriculture industry for evidence of innovative agglomeration. Each agricultural patent granted from 1972 - 2002 was spatially tagged using Geographic Information Systems software. The data indicate that a closer distance between any two patent origins increases the probability that one cites the other as prior art, with subtle interregional variations on the degree to which proximity advances agricultural innovation. Policies that exploit the relative ease of knowledge within localized networks can encourage development of cost-cutting technologies which can in turn lower world food prices.
This thesis focuses on the United States’ “National Innovation System” (NIS) and the role of private and public actors in this system. It is part of a broader literature seeking to identify the key catalysts of innovative activity, and hence economic growth. Although there are many elements that constitute the NIS, this study emphasizes one private institution, the venture capital industry, and one public institution, the Small Business Innovation Research program. The study uses state-level panel data aggregated over a fourteen-year period (2002-2015) and is operationalized by implementing an Ordinary Least Squares fixed-effects model that uses utility patent applications—a proxy for innovation—as the dependent variable. This thesis argues that public-private partnerships and symbiosis between the two sectors is critical to empowering technological innovation and sustainable long-run economic growth.
In this study I apply the theory that changing energy prices induce innovation to producers of energy, specifically the oil and gas industry. Using pricing, production and patent data from 1980 – 2011, I model the share of total patents that are applicable to oil and gas as a function of expected future commodity prices, production of each commodity and previous stock of knowledge. In the building of the model, I develop knowledge stock variables and expected future prices specific to the industry. I find a significant, positive and highly elastic correlation between expected commodity prices and innovation, that is in line with previous work and the induced innovation theory.