Kevin BryanJob Market Candidate, Managerial Economics and Strategy
Fields: Industrial Organization (Innovation), Applied Micro Theory, Economic History
PhD (expected), Managerial Economics and Strategy, Kellogg School of Management, 2014
M.S., Managerial Economics and Strategy, Kellogg School of Management, 2011
M.S., Mathematics, Virginia Commonwealth University, 2008
M.A., Economics, Boston University, 2006
B.A., Intl. Relations and Economics, Boston University, 2006
My CV is here.
I am a fifth-year Ph.D. student in Managerial Economics and Strategy at the Kellogg School of Management; beginning in July 2014, I will move to the University of Toronto as an Assistant Professor of Strategy at the Rotman School of Management. My work primarily consists of applied theoretical analyses of innovation, generally supported by detailed qualitative historical investigation. Right now, I am particularly interested in questions related to which research lines inventors and firms pursue, and in how "directional inefficiency" in these choices can be mitigated. Over the next few years, I am interested in exploring, theoretically, the government's role in encouraging diffusion, especially via testing and standards; anecdotally, this appears to me a more important role than directly funding R&D. As a side project, I enjoy studying history of thought and methodology, and am in the early stages of a project relating developments in modern logics to developments in microeconomics in the middle of the 20th century.
I maintain the research weblog A Fine Theorem devoted to discussions of new research in my areas of interest; despite a technical focus, the site has been visited over a quarter million times and has been discussed by The Economist, Slate, Reuters, Forbes, The Browser and many others.
R&D Policy and the Direction of Innovation
How should the government sponsor R&D? This problem is often posed in terms of inducing firms to perform the optimal amount of R&D, but you may also want policy to encourage firms to work on the "right" projects, particularly when fixed factors of production mean that government-sponsored research crowds out private research. I show that many common policies (R&D tax credits, patents, etc.) which are well-known to solve the problem of suboptimal research effort can have the opposite effect when it comes to solving directional inefficiency. I then apply the theory to qualitative data about U.S. nuclear power policy decisionmaking in the 1950s and 1960s, showing why certain policies led to unexpected outcomes, and why path dependence alone is insufficient to explain the eventual dominance of light water reactors.
The Direction of Innovation (with Jorge Lemus)
Why do firms work on the "wrong" research projects? We construct a tractable theory of invention graphs, where every successful invention potentially opens up new research lines in the future. We show that firms work on socially suboptimal projects for three reasons: they maximize only profits in excess of the continuation value they receive for doing nothing, they put less weight than the planner on future inventions their work makes possible since other firms may be first to invent those, and they effectively discount at a different rate than the planner due to concerns that other firms may invent less valuable yet easier projects first.
The Wright Brothers invent the airplane in the United States in 1903, yet by the start of World War I, the U.S. airplane industry is close to dead. These "industrial reversals of fortune" are surprisingly common the the annals of technology. Are they inevitable, or merely the result of bad policy after the initial invention? I suggest that the U.S. aviation industry withered not because of policy mistakes, but rather because the minimum national technical ability needed to get a plane in the air is not the same as the minimum national technical ability needed to deveop a commercially viable airplane. Using a new database of airplane-related microinventions between 1896 and 1917, I show that the U.S. even in 1903 was behind multiple European countries in areas like metallurgy, theoretical aerodynamics, and light engine design which would prove critical to the commercial viability of aviation. Once Europeans were made aware of the Wrights' ideas on maintaining lateral stability, they quickly grafted these better control methods onto otherwise superior airplanes. The strong Wright patent further limited the ability of U.S. designers to incorporate certain European techniques, as lawsuits limited transatlantic travel and technology demonstration on the part of frontier firms. The moral for modern technology policy is clear: a nation that funds basic research had better ensure either that domestic firms also have the knowledge necessary to transform the results of basic research into commercial products or that domestic firms have the ability to quickly learn that knowledge from abroad.
Open Access Increases Commercialization of Medical Research (with Yasin Ozcan)
[Note: this paper is currently under preparation for a medical journal, hence I cannot post a working paper online. Please contact me if you'd like a copy.]
Expansion of open access mandates has proven controversial, partly because there has been limited quantification of the benefits of open access beyond measures such as academic citation counts. We link every research article in 44 top medical journals since 2005 to every U.S. patent application over that period, a challenging task as patent applications have much less standardized citation practices than granted patents. In the raw data controlling for month of publication and journal name, open access articles are cited 46 percent more often by patents than gated articles. Using the 2008 NIH open access mandate to instrument for selection, we find an open access citation advantage of over 35 percent. A back-of-the-envelope calculation suggests that the benefit of additional invention due to open access exceeds the entire social cost of funding open access journals.
Multigraph Diffusion, with an Application to the One Child Policy
Ideas, policies, inventions and macroeconomic shocks are all subject to diffusion over time. Often, the diffusion occurs simultaneously across multiple networks. Given a single realized diffusion, I show how to back out the relative speed of the diffusion on each network. With those parameters in hand, we can state in a formal way how important each network, and how important the location of the initial shock, was to the eventual diffusion. I apply these ideas to the spread of family size restrictions in China in the 1970s and 1980s, which was hypothesized by William Skinner to have diffused both through spatial distance and through the "urban hierarchy". I further suggest that these techniques may be of use to modeling economic phenomena such as recessions, which are known to spread both via spatial proximity and via input-output chains.
How Does Open Access Increase Follow-Up Invention? (with Yasin Ozcan)
In an earlier paper, Yasin and I suggested that open access to academic research seems to substantially increase the use of that research in inventions. Here, we analyze the mechanisms behind that result in greater depth by matching medical research articles, patent applications, patentee characteristics (including location of the patentee, academic versus commercial employment, experience with the patent system, etc.) and patent characteristics (such as the technology class of the patent). We further investigate how the impact of open access compares to other shifts in lab-to-application knowledge flow patterns since the mid-1970s.
The Evolution of City Population Density in the United States (with Pierre-Daniel Sarte and Brian D. Minton)
Federal Reserve Bank of Richmond Economic Quarterly 93.4 (2007)
American urban areas have become less dense over recent decades. Is this merely due to regional differences between East and West, a recent occurrence, or resulting from the difficulty of measuring the concept of "cities"? We digitize census data from 1930 to 2000 on the area and population of all U.S. places with population above 5000, all urbanized areas and all metropolitan statistical areas. These data show that decreasing density has occurred continuously since at least 1930, in all regions, in all definitions of "city", and both in places founded before the 20th century began as well as newly-founded cities.
(Data and Replication Code)
On the Evolution of Income Inequality in the United States (with Leonardo Martinez)
Federal Reserve Bank of Richmond Economic Quarterly 94.2 (2008)
How has income inequality trended over the past four decades, and to what extent do different data sources and measures of income inequality disagree? We analyze a number of different measures and data surveys, and show (largely for a general policy audience) the robustness of basic qualitative facts - inequality rising since around 1970, particularly stagnant outcomes for men, rising inequality between the middle class and the bottom in the 1980s and between the top and the middle class at other times.
Semiparametric Estimation of Land Price Gradients Using Large Data Sets (with Pierre-Daniel Sarte)
Federal Reserve Bank of Richmond Economic Quarterly 95.1 (2009)
The nature of the land price gradient has played an important role in economic theory going back to von Thünen in 1826. In particular, modern urban models often assume polycentric cities, with multiple "peaks" in the land price density surface. Empirically measuring land prices is often quite challenging, since there simply aren't that many sales of unfinished land. We show how to extract relative land prices using a semiparametric model of housing prices that allows the price of land to vary nonparametrically over space. An algorithm based on nearest neighbor allows quick surface estimation even with hundreds of thousands of observations. We apply this method to the region surrounding Richmond, Virginia, showing evidence of polycentric land price gradients.
The Economic Ideas of Ronald Coase | VoxEU.org, September 2013
Residential Externalities (with Pierre-Daniel Sarte) | Richmond Fed Economic Brief 2009
Jargon Alert: Arbitrage | Richmond Fed Region Focus, Spring 2007
Economic Adviser for "The Future of Taxpayer-Funded Research"
The results of this paper were presented before Congress by Elliot Maxwell on March 19, 2012.