Academic journal article The American Journal of Economics and Sociology

Do We Still Need Cities? Evidence on Rates of Innovation from Count Data Models of Metropolitan Statistical Area Patents

Academic journal article The American Journal of Economics and Sociology

Do We Still Need Cities? Evidence on Rates of Innovation from Count Data Models of Metropolitan Statistical Area Patents

Article excerpt


In a 1998 Journal of Economic Perspectives symposium on urban agglomeration Edward Glaeser asks, "Are Cities Dying?" This article attempts to take some first steps toward empirically answering this question. The most important reason cities continue to exist, given the dramatic drop in transportation costs for physical goods over the last century, is often argued to be related to the forces of agglomeration as they apply to knowledge creation. Using new data on USA Metropolitan Statistical Areas (MSAs), we analyze the role played by population density in the determination of the rate of innovation. The empirical methodology used in this article accounts for the discreet nature of patent data, the dependent variable in our empirical model. These count data models are extensions of the simple Poisson regression model. The empirical investigation proposed here is an important part of understanding the viability of urban areas in the future.

The ideas of agglomeration and offsetting congestion, common to regional and urban economics, are employed to explain patenting activity across US cities. This article, therefore, provides an empirical link between the "new economic geography" and the "new economic growth theory" (Krugman 1998). Both literatures depend heavily on recent developments in modeling imperfect competition in a general equilibrium setting (see Dixit and Stiglitz 1977). We believe it is safe to say that a full recognition of the importance of space and regional economies toward improving the understanding of macroeconomic growth is not realized by many economists, save for a few notable exceptions (e.g., Krugman 1991; Glaeser et al. 1992, 1995; Ciccone and Hall 1996; Beeson 1987).

The new economic growth theory looks closely at the economics of new knowledge and ideas. A common implication of new growth theory is that economies experience some type of scale effect, where larger economies are at an advantage in innovating as a result of more relaxed resource constraints. While this is a unique implication of these new models of growth, there is little evidence that scale effects actually exist (Helliwell and Chung 1992; Jones 1995a; Pack 1994; Grossman and Helpman 1994). This lack of evidence of scale effects is the impetus behind a number of theoretical extensions of new growth theory with the goal of ridding the models of their scale implication (Lucas 1988; Jones 1995b; Segerstrom 1998; Young 1998).

However, the typical measure of scale does not take spatial aspects of the economy into consideration since the scale variable is typically measured as total population or GDP.

The theory may be getting ahead of itself in these attempts to reformulate new growth models. The lack of evidence of scale effects may be due to a poor specification of the nature of scale in an economy within the typical macroeconomic growth model, which should account for agglomeration economies and, possibly, the offsetting forces of congestion. Sedgley and Elmslie (2004) show evidence of scale effects in innovation in terms of population density at the US state level.

Our empirical analysis of agglomeration and congestion differs from other studies. Previous research focuses on urban agglomeration and congestion in explaining worker productivity measures. These forces are well documented, but little attention has been paid to empirically investigating the role of these forces in the production of new knowledge and ideas. A knowledge production function (KPF) (Jaffe 1986; Griliches 1986; Acs and Audretsch 1989) is a key feature in empirically studying patent statistics. The KPF is equated with the equation of motion for new ideas common to the new growth literature. By incorporating the density of economic activity, agglomeration forces and congestion forces are given an explicit role via the functional form of the KPF.

The next section reviews some salient ideas from regional and urban economics to provide motivation for the empirical exercise. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.