Base on Krugman¡¦s NEG model, we subsume urban cost in this study. Referring to study procedure of Murata and Thisse, make few changes of the model. Then, we get 3 conclusions of this study: (1) Due to the difference between commuting costs of each region, the symmetric distribution of workers will no longer make the largest total mass of varieties. While a region attend to traffic construction in order to drop people¡¦s commuting cost, agglomeration economic will not certainly make a negative impact on urban cost or effect people¡¦s variety consumption behavior. (2) In the case of an agglomeration, if the transportation costs are sufficiently large, the agglomeration equilibrium will be stable. The larger the transportation costs between regions, the larger cost of consuming other regions¡¦ commodity. Thus, chose to agglomerate in a bigger region can get cheaper and variety manufacturing commodities easily. And the agglomeration equilibrium will be stable. (3) Any change of commuting cost of the other region will not influence the stable of agglomeration equilibrium.

This paper provides a non-technical overview of NEG models dealing with policy issues. Considered policy measures include alternative categories of public expenditure, international tax competition, unilateral actions of protection/liberalisation, and trade agreements. The implications of public intervention in two-region NEG models are discussed by unfolding the impact of policy measures on agglomeration/dispersion forces. Results are described in contrast with those obtained in standard non-NEG theoretical models. The high degree of abstraction limits the applicability of NEG models to real world policy issues. We discuss in some detail two extensions of NEG models to reduce this applicability gap: the cases of multi-regional frameworks and firm heterogeneity.

This paper provides a non-technical overview of NEG models dealing with policy issues. Considered policy measures include alternative categories of public expenditure, international tax competition, unilateral actions of protection/liberalisation, and trade agreements. The implications of public intervention in two-region NEG models are discussed by unfolding the impact of policy measures on agglomeration/dispersion forces. Results are described in contrast with those obtained in standard non-NEG theoretical models. The high degree of abstraction limits the applicability of NEG models to real world policy issues. We discuss in some detail two extensions of NEG models to reduce this applicability gap: the cases of multi-regional frameworks and firm heterogeneity.

The interest in the phenomenon of industrial agglomeration dates back to the late 19th century (particularly, the work of Alfred Marshall). The late 20th century saw a renaissance in this interest, for example, in the ‘new geographical economics’ (e.g. Krugman) and Porter’s framework of competitive forces. Subsequently, clustering has come to be seen as a feature of high technology industries, despite the difficulty to gather empirical evidence to underpin this assertion and to corroborate underlying cluster-related mechanisms that would result in benefits for firms in clustered environments.This thesis extends the discourse on rationales of clustering into the field of the management software industry sector. Its objective is to use the example of the cluster around the German software giant SAP in South-western Germany as a model to explore the strengths and characteristics of clustering in this industry context. A survey-based research design was selected and interview responses were collected from 206 management software firms located throughout Germany. The empirical analysis was based on two complementary statistical approaches. The SAP cluster was compared to two control groups, one consisting of firms from non-clustered regions, the other of firms from other clustered locations. Diversities between environments were analysed using bivariate statistical techniques. The findings of this descriptive analysis substantiate disparities between the SAP cluster and the non-clustered control group. Beneficial effects available to SAP cluster firms can be observed in respect to nearly all factors analysed, i.e. access to specialised human resources, access to investment capital, demand proximity, knowledge spillover externalities, spin-off activity and the accumulation of social capital.This in-depth comparative perspective is complemented through a holistic evaluation employing structural equation modelling. Among the key findings of this analysis are the significant links between location in the SAP cluster and superior growth performance, in which long term professional networks and social capital derived from major software vendors (particularly SAP) act as mediators. Structural equation modelling also highlights a positive effect of being located in the SAP cluster on knowledge spillover externalities, which is mediated by the cultural environment in the SAP cluster. Finally, the SAP cluster environment stimulates spin-off activity, which functions as an important mediator in the accumulation and retention of social capital held with players in the industry. In sum, the empirical analysis suggests that a wide-ranging network of factors exists, in which social capital functions as a catalyst in the mechanisms resulting in benefits available to SAP cluster firms.