Smart Cities and a Stochastic Frontier Analysis: A Comparison among European Cities

The level of interest in smart cities is growing, and the recent literature on this topic (Holland, 2008; Caragliu et al., 2009, Nijkamp et al., 2011 and Lombardi et al., 2012) identifies a number of factors that characterise a city as smart, such as economic development, environment, human capital, culture and leisure, and e-governance. Thus, the smartness concept is strictly linked to urban efficiency in a multifaceted way. A seminal research for European policy conducted by Giffinger et al. (2007) defines a smart city on the basis of several intangible indicators, such as a smart economy, smart mobility, smart environment, smart people, smart living, and smart governance. These authors’ methodology results in a ranking of 70 European cities in terms of their smartness. Our aim is to verify the robustness of these smartness indicators in explaining the efficiency of the same sample of European cities. Using the concept of output maximising, we built a stochastic frontier function in terms of urban productivity and/or urban efficiency by assessing the economic distance that separates cities from being smart. Moreover, this approach, which distinguishes between inputs and efficiency, allows us to incorporate the smartness indicators into the systematic component within the error term. As a result, our conclusions identify a different ranking of European cities with respect to Giffinger et al. (2007)’s analysis, thereby highlighting the need for a better and more robust definition of these indicators.