Un « marché des villes » en économie géographique avec des coûts urbains endogènes

Urban costs are introduced into an economic geography model between two cities. These costs are made up of land rents, the plot size been endogenous, and of commuting costs, a local tax allowing a development of the urban network. Urban costs are the only centrifugal force in the model, which is without farmers. Consumers share the same Cobb-Douglas tastes for housing and manufactured goods, enjoying diversity of theses in a Constant-elasticity-of-substitution subutility function. The industrial sector makes differentiated goods with an economies of scale technology and it is in a monopolistic competition equilibrium (Dixit-Stiglitz). The differenciated goods can be carried between cities with an iceberg transport cost (Samuelson). General equilibria between the two cities are obtained by means of simulation by clearing the good, the labour and the land markets, in the short run and the long run. The main findings are: (i)Without farmers, Krugman?s findings are reversed: population concentrates into a single city if the transport cost of industrial goods is high and scatters if it is weak. (ii)Assuming that the urban network can be developed by three ways: a land tax, an equal household?s tax or a kilometre toll, the property tax is the most effective tool for maximising the welfare and the toll is the worse. (iii)A subtle play of forces can be analysed: on the one hand economic geography forces, i.e. the trade off increasing returns ? transport costs and on the other hand urban economic ones, i.e. the trade off land costs ? commuting costs. They may play in the same way or in opposite ways: (iv)When City One develops its network in a context of weak tax pressure, consumer benefits from the decrease of urban costs and from economies of agglomeration; consumer?s utility increases in the two cities and (in the long run) the population of City One steeply increases; (v) With intermediate levels of tax pressure in City one, urban costs increase but consumer?s utility in City One benefits from a transfer of utility from City Two; in the long run, the population of City One still slowly rises; (vi) With a higher level of taxation in City One, when taxes again increase, City One loses the advantage of been a large market and bears high urban costs: utility declines in the two cities and the population of City One also decreases.