The Rise of American Ingenuity: Innovation and Inventors of the Golden Age
We examine the golden age of U.S. innovation by undertaking a major data collection exercise linking historical U.S. patents to state and county-level aggregates and matching inventors to Federal Censuses between 1880 and 1940. We identify a causal relationship between patented inventions and long-run economic growth and outline a basic framework for analyzing key macro and micro-level determinants. We find a positive relationship between innovation and drivers of regional performance including population density, financial development and geographic connectedness. We also explore the impact of social structure measured by slavery and religion. We then profile the characteristics of inventors and their life cycle finding that inventors were highly educated, positively selected through exit early in their careers, made time allocation decisions such as delayed marriage, and tended to migrate to places that were conducive to innovation. Father's income was positively correlated with becoming an inventor, though not when controlling for the child's education. We show there were strong financial returns to technological development. Finally, we document an inverted-U shaped relationship between inequality and innovation but also show that innovative places tended to be more socially mobile. Our new data help to address important questions related to innovation and long-run growth dynamics.
We would like to thank Philippe Aghion, Harun Alp, Price Fishback, David Galenson, Rick Hornbeck, Ali Hortacsu, Xavier Jaravel, Taylor Jaworski, Ben Jones, Chad Jones, Fatih Karahan, Pete Klenow, Alan Marco, Casey Mulligan, Rob Shimer, Andrei Shleifer, Stefanie Stantcheva, Nancy Stokey, John Van Reenen, and Dietrich Vollrath, and the seminar and conference participants at the NBER Summer Institute Economic Growth (EFJK) Group, Development of the American Economy Group, Firms in Macroeconomics Conference, College de France, the USPTO, the University of Houston, the Fuji Conference in Tokyo, Queen's University, Koc University, and Northwestern for very helpful feedback and discussions. We thank the Minnesota Population Center for access to the Census data and William Kerr, Richard Hornbeck, and Price Fishback for sharing various datasets with us. Akcigit gratefully acknowledges the National Science Foundation, the Alfred P. Sloan Foundation, and the Ewing Marion Kauffman Foundation for financial support. Nicholas received funding from the Division of Research and Faculty Development at Harvard Business School. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.