Productivity, Innovation, and Entrepreneurship
March 21, 2014
Pian Shu, Harvard University
Shu analyzes the characteristics of entry-level financiers and the impact of the recent financial crisis on their career outcomes. She presents three key results using detailed data on MIT bachelor's graduates from 2006 to 2012. First, graduate school and careers in finance, especially at hedge funds and trading positions, both attract students with high raw academic talent. However, financiers are significantly less likely to develop academic skills and more likely to participate in extracurricular activities that develop soft skills. As a result, the two groups and their skill sets do not appear to be substitutable at the time of college graduation. Second, the recent financial crisis caused an exogenous and large decrease in entry-level jobs in finance that was mostly concentrated in the less quantitative side of finance, for example, investment banking. The shock had the largest impact on management and economics students who do not have a second major in science and engineering. Although many of them did not enter finance because of the contraction in the sector, there was no evidence that they pursued a career in science and engineering instead. Third, the author finds no evidence that the crisis has affected students' tendency to major in management and economics. These results suggest that finance does not attract the most productive scientists and engineers from MIT. It is not clear whether limiting entry into finance in the case of the recent financial crisis has improved the overall efficiency in talent allocation.
Walker Hanlon, University of California at Los Angeles and NBER, and Antonio Miscio, Columbia University
Hanlon and Miscio introduce a dynamic approach for estimating the sources of agglomeration economies. Their framework allows them to simultaneously estimate the importance of agglomeration forces attributable to cross-industry spillovers, to within-industry spillovers, and to overall city size in the growth of city-industries. This is done while controlling for fixed locational fundamentals, city-specific shocks, and national industry growth rates. The authors implement the approach using detailed new data describing the industry composition of English cities from 1851 to 1911. They find strong evidence that cross-industry connections can influence industry growth, particularly for industries with small firms and more skilled workers. Within-industry effects are not present for most industries, but may be important in a small number of sectors characterized by large firms and low-skilled workers. Once the authors control for the role of industry composition, they find a strong negative relationship between city size and city-industry growth rates.
Ajay Agrawal, University of Toronto and NBER; Carlos Rosell, Department of Finance, Canada; and Timothy Simcoe, Boston University and NBER
Agrawal, Rosell, and Simcoe exploit a change in eligibility rules for the Canadian Scientific Research and Experimental Development (SRED) tax credit to gain insight on how tax credits impact small-firm R&D expenditures. After a 2004 program change, privately owned firms that became eligible for a 35 percent tax credit (up from a 20 percent rate) on a greater amount of qualifying R&D expenditures increased their R&D spending by an average of 15 percent. Using policy-induced variation in tax rates and R&D tax credits, the authors estimate the after-tax cost elasticity of R&D to be roughly -1.5. They also show that the response to the after-tax cost of R&D is larger for contract R&D expenditures than for the R&D wage bill, and is larger for firms that perform contract R&D services, or that recently made R&D-related capital investments. The authors interpret this heterogeneity as evidence that small firms face fixed adjustment costs that lower their responsiveness to a change in the after-tax cost of R&D.
Pierre Azoulay, MIT and NBER; Joshua Graff Zivin, University of California, San Diego and NBER; Danielle Li, Northwestern University; and Bhaven Sampat, Columbia University and NBER
Azoulay, Graff Zivin, Li, and Sampat measure the impact of public R&D investments on innovation by private sector firms. They quantify the returns to grant spending at the National Institutes of Health (NIH) in terms of the biomedical patents it generates. The paper makes two contributions. First, the authors use newly constructed bibliometric data to develop a method for flexibly measuring the outcomes of basic science investments. Second, they take advantage of the institutional features of NIH peer review to address concerns about the endogeneity of grant funding. The results show that NIH funding generates more private patents than it crowds out. A $10 million increase in NIH support generates 2.8 additional patents; given an average grant award of $1.34 million the authors expect one additional private sector patent to be produced for every three additional NIH grants. Moreover, the authors find substantial cross-disease spillovers in funding for biomedical research; approximately half of all patents generated by NIH funding for one disease area are primarily relevant for a different disease.
Manuel Adelino and Song Ma, Duke University, and David Robinson, Duke University and NBER
Adelino, Ma, and Robinson ask whether startups react more to changing investment opportunities than more mature firms do. They use the fact that a region's pre-existing industrial structure creates exogenous variation in the severity of its exposure to nationwide manufacturing shocks to develop an instrument for changing investment opportunities, and examine employment creation in the non-tradable sector as a response to those opportunities. Startups are much more responsive to changing local economic conditions than older firms are. Moreover, their responsiveness doubles in areas with better access to small business finance, suggesting that financing constraints are an important brake on job creation in the startup sector. Although the authors focus mostly on the non-tradable sector for empirical identification, their results extend to other sectors of the economy, indicating that the mechanisms uncovered are economically pervasive. This suggests that factors like organizational flexibility and innovativeness may be important drivers of job creation among startups.
Shai Bernstein, Stanford University, and Albert Sheen, Harvard University
Do private equity buyouts disrupt company operations to maximize short-term goals? Bernstein and Sheen document significant operational changes in 103 restaurant chain buyouts between 2002 and 2012 using health inspection records for more than 50,000 stores in Florida. Store-level operational practices improve after private equity buyouts as restaurants become cleaner, safer, and better maintained. Supporting a causal interpretation, this effect is stronger in chain-owned stores than in franchised locations-"twin restaurants" over which private equity owners have limited control. Private equity targets also reduce employee headcount, lower menu prices, and experience a lower likelihood of store closuresa proxy for poor financial performance. These changes to store-level operations require monitoring, training, and better alignment of worker incentives, suggesting private equity firms improve management practices throughout the organization.
Achyuta Adhvaryu, University of Michigan; Namrata Kala, Yale University; and Anant Nyshadham, University of Southern California
The impacts of technologies that mitigate climate change are tempered by the willingness of individuals and firms to adopt them. Absent costly sustained taxes or subsidies, this willingness rests on the technologies' private returns, which are generally assumed to be modest. Adhvaryu, Kala, and Nyshadham document a novel productivity co-benefit to the adoption of energy-saving LED lighting. First, using detailed daily production data from garment factories in Bangalore, India, they show that efficiency (realized output over target output) decreases substantially on hotter days. But the introduction of energy-saving LED lighting, which emits less heat than incandescent or fluorescent lighting, attenuates 75 percent of the negative impact of temperature on productivity. The results are robust to a variety of measures of temperature, specifications, and the incorporation of relative humidity. The estimates suggest that the productivity returns to LED adoption are more than four times larger than the energy savings.