Territories Acting for Economic Growth Rome, 4 and 5 December 2014 Second Plenary Session Evidence on medium and small cities performance in the EU Roberto Camagni Politecnico di Milano, ABC Department

Urban Growth in the EU (annual GDP growth - LUZ) London Paris Ruhrgebiet

Urban Growth in the EU (annual GDP growth - LUZ)

About some questionable theoretical shortcuts (Static vs. Dynamic Agglomeration Economies Roberto Camagni, Roberta Capello and Andrea Caragliu (61st Annual North American Meeting of the RSAI, Washington, November 12-15, 2014) The NEG (Fujita et al, 1999) and the World Bank (2009) claim that: - Mega cities show a higher productivity (GDP per capita): TRUE!  They will grow more than smaller cities (NO!)  policies have to support mega cities in order to maximise growth (NO!!) As: - size-derivative (“static” agglomeration economies) ≠ time-derivative (“dynamic” agglomeration economies), - > productivity does not mean > attractiveness: this latter should be measures by “net” benefits (benefits – location costs), not by gross benefits (GDP per head alone)

Results of an econometric analysis on the sources of urban development Dependent variable: changes over time of productivity advantages associated to urban size (operationally: growth of net urban benefits vs ) Highly significant explanatory variables (sources of growth): - not initial urban size! But: - an upgrading and increase in hosted urban functions (innovation) - (and small cities benefit more from this upgrading of urban functions) - an upgrading of functions present in the entire regional urban system - an increase in the demographic size of the regional urban system (the phenomenon of “borrowed size”: Meijers, 2013) - the level of long-distance city-networks (cooperations with other cities). There exist multiple options and strategies for the development of small and medium cities, located both inside and outside large urban regions

The econometric analysis and results Dependent variable: urban productivity increases Model (1)(2)(3)(4)(5)(6)(7)(8) Constant term * ** (0.51)(0.52)(0.06) (0.05)(0.06) City population (0.03) (0.04) Growth of high level urban functions 0.17***0.15***0.15** 0.14***0.15***0.16*** (0.06) (0.07)(0.03)(0.05)(0.06) Growth of borrowed size (0.00) (0.001)(0.00) Growth of borrowed functions **0.46**0.49**0.45*0.46**0.43*0.45* (0.23) (0.24) Growth of networks ** (0.19) (0.00) Urban networks * ---- (0.11) Growth of high level urban functions * City population *** --- (0.05) Growth of borrowed size * City population ** -- (0.001) Growth of borrowed functions * City population (0.31) Growth of networks* City population (0.00) Number of obs. 136 Robust standard errors Yes Method of estimation OLS Pseudo-R Joint F-test test 3.01**3.35**3.09**2.97***14.50***5.52***2.46**2.33**

THANK YOU VERY MUCH FOR YOUR ATTENTION! Roberto Camagni Department ABC - Politecnico di Milano Piazza Leonardo da Vinci MILANO tel: fax: