Georg Hirte (p), Hyok-Joo Rhee, Sangyun Yu

Soushi Suzuki

It is well known that the free market equilibrium fails to achieve the maximum welfare in the model with homogeneous households under Benthamite welfare. We construct a model with unobserved heterogeneity and show that the same proposition holds. However, as a result of spatial re-sorting and distributional terms, the conventional rule to achieve a first-best – i.e., Pigouvian tolls plus redistribution to equalize social marginal utilities of income – is not guaranteed to hold. The same problem plagues the self-financing rule and the Henry George Theorem in the model with heterogeneity. We find that at the optimum a city’s transport budget runs a surplus, but that optimum population conditioning on financing fixed costs of infrastructure or the city is smaller than in case of household homogeneity.

Soushi Suzuki (p), Peter Nijkamp

Georg Hirte

This paper aims to provide an advanced dynamic assessment methodology for city performance strategies, based on an extended Data Envelopment Analysis (DEA). The use of this novel efficiency-improving approach based on DEA originates from the earlier developed, so-called Distance Friction Minimisation (DFM) method. To design a feasible and realistic improvement strategy for low-efficiency Decision-Making Units (DMUs), we introduce a Target-Oriented (TO) DFM model on top of a Super-Efficiency model, in order to generate an appropriate efficiency-improving projection model. The standard TO approach specifies a target-efficiency score (TES) for inefficient DMUs. This approach is able to compute an input reduction value and an output increase value in order to achieve a desired TES. To develop a dynamic DEA perspective, we create next a new model from a blend of the TO-DFM approach and a Time-Series (TS) approach which incorporates a multi-temporal time horizon and a stepwise target score to achieve a final target efficiency score so as to generate a more appropriate efficiency-improving DEA projection. This new model is able to incorporate a catch-up effect in the efficiency projection. However, the regular TS approach assumes that the efficiency frontier is fixed over any time period. However, in reality, efficiency frontiers vary from year to year. That is to say, the TS approach is not able to incorporate a frontier shift effect in setting the overall target improvement level. Therefore, it is necessary to develop a more realistic efficiency improvement projection which includes a dynamic system of target-settings to achieve a target improvement level so as to programme more realistic policy initiatives. In the present paper we develop a new multi-period model from a blend of the TO-DFM model and a dynamic TS decision approach. The above-mentioned Dynamic TO-DFM model will be applied to and tested for a multi-dimensional efficiency assessment of several large Japanese cities. In this study, due to comparative data limitations, we consider two inputs (population and city budget) and two outputs (GDP and tax revenues). Based on these items, this study assesses the relative economic performance of 16 Japanese big cities (i.e., “government-ordinance-designated cities”) by means of the above described, extended super-efficient DEA model. Finally, we provide an efficiency improvement programme based on the Dynamic TO-DFM model for inefficient cites.