1. An Examination of Private (Sectoral and Firm-Level) Investments in U.S. Food and Agricultural R&D (with Philip G. Pardey, Steven P. Dehmer and Steve Miller)
Abstract: The relationship between a firm’s financial position and its spending on research and development (R&D) has been studied for a number of manufacturing sectors. However, few studies have been conducted for the food and agricultural sectors, which is most likely a reflection of the substantive public R&D presence plus a paucity of suitable firm-level data. This paper presents a new set of firm-level data pertaining to firms’ financial position and R&D conducted by food and agricultural-related businesses operating in the United States from 1950 to 2014. Within the food and agricultural sectors, we identify the shifting structure of investments in machinery, agriculture and chemicals, and food and beverage processing R&D. In particular, we emphasize, in detail, the changes in the portfolio of firms conducting this research.
We also econometrically examine the associations between firm-level sales, profit rates, and R&D spending among these firms. We find a large variation in their absolute and relative (to sales) investments in R&D, a variation that becomes more pronounced over time. We also find that firms are likely to invest more in R&D when they anticipate higher future sales. In particular, when firms expect higher sales, they tend to invest more in capital, which lowers the cash flow to capital ratio and eventually increases R&D investment. These estimated effects are larger for food firms than agricultural firms, suggesting that food firms are more sensitive to changes in sales when investing in R&D. These results hold under robustness checks designed to eliminate outliers in R&D growth that might otherwise skew the results.
2. Assessing the Propensity to Collaborate in Life Sciences Research (with Steve Miller and Philip G. Pardey)
Abstract: Collaboration among researchers is increasing, but among all possible pairs of researchers few actually work together. It is thus natural to ask what information signals do researchers use in selecting research partners from among a large pool of potential collaborators. This study has addressed that question for life sciences researchers focusing on three particular types of signals: perceived research productivity, knowledge complementarity, and professional familiarity. We investigate the collaboration consequences of these signals by using academic publication data published by life sciences faculty at the University of Minnesota (UMN) from 1999 to 2014, focusing on papers in which all of the authors are affiliated with UMN. We have found perceived research productivity is positively associated with the initiation of first collaboration while complementary knowledge is not. For continuing collaborations, however, knowledge complementarity becomes significant as well as the recency of first collaboration and the frequency of past collaborations. While complementarity is noted as a key driver of repetitive collaborations, researchers are more likely to collaborate with coauthors who have moderate differences in research profile. This suggests that the factors researchers consider when choosing coauthors vary with respect to the types of collaboration. This study contributes to the empirical literature on research team formation and to university administrators seeking to promote more collaboration among researchers.
3. Patterns of Research Collaboration in the Life Sciences (with Steve Miller and Philip G. Pardey)
Abstract: The perceived rise of collaborative research raises a range of questions about who collaborates, with whom and why, and whether those partnerships are ultimately beneficial. We investigate these questions using a newly constructed panel dataset of 3,305 scientists from three large academic units within the University of Minnesota (UMN). We found that for this sample of researchers the total number of publications nearly doubled during that period, and involved a large increase in the number of outside collaborations. This suggests that researchers increasingly perceived that the benefits from outside collaborations outweighed the additional costs that may arise from collaborating externally. Those benefits apparently do not include increased research (or more specifically publication) productivity at the university level, but rather, it seems, that researchers pursue these external collaborations to boost citation rates. On the other hand, we also find evidence consistent with declining costs of cross-institution collaboration. Re-examining these questions at the college level reveals patterns consistent with heterogeneity in benefits and costs of collaboration across colleges. On close examination, we did find that researchers in the other life-science colleges may seek outside collaborations in search of productivity rather than citation gains.