要旨：Using the original survey data conducted to small and medium-sized firms in Zhongguancun Science Park (ZSP), we examine the effects of R&D cooperation on firms' intensity of R&D activities, with particular attention to type and proximity to partners like large sized firm, other small and medium-sized firm, and university and research institute. We found that direct effects to enhance employment R&D intensity of ZSP small firms from research cooperation with small and medium-sized firms, and university and research institute that are located nearby. Our results also suggest that indirect effects of research cooperation such that cooperation with nearby large firms and that with far-off small and medium-sized firms enhance further R&D of ZSP small firms when such cooperation yields commercial success.

要旨：This paper describes how differences in costs between downstream firms affect their relative performance in a Hotelling-type location model. I consider three aspects of production costs in discussing the effect of cost advantage on firm performance: (i) costs to differentiate their products, (ii) efficiencies of production, and (iii) endogenously determined input procurement conditions. I show that a less efficient firm can be the market leader if the per-unit production cost of each firm depends on its location (costs to differentiate their products exist) and if the firms procure inputs from outside suppliers. That is, the profit of a less efficient firm can be larger than that of a more efficient firm. In equilibrium, the less efficient firm locates closer to the center than the efficient firm. This equilibrium outcome is quite different from that in which the efficiency difference is given exogenously. Moreover, I investigate how a change of cost structures affects the profits of firms. I show that improving the procurement conditions of a firm does not always increase its profit, whereas improving its production efficiency always does.

要旨：By adding housing congestion to the model of Fujita and Thisse [Does geographical agglomeration foster economic growth? And who gains and who loses from it? Japanese Economic Review 54 (2003), 121-145], we study interactions among growth, agglomeration, and urban congestion. However, unlike Fujita and Thisse (2003), we formulate the model as a one-shot game and take an evolutionary game-theoretic approach for stability analysis. Our approach is simple and enables us to analyze stability of nonstationary equilibria. We show that if both expenditure share for housing and transport cost are small, a stable steady-state equilibrium does not exist. Moreover, in such a case, we show that there can exist a stable nonstationary equilibrium.

【大阪大学経済学研究会と共催】

要旨：Not a few cities around the world have experienced population decrease during the past several decades. The change in population is a key factor of urban development process. For the case of increasing population, Fujita[9], Wheaton[15] and others investigate spatial patterns of residential development, but no previous papers take account of decreasing population. This paper provides a framework to examine urban spatial dynamics on both growing and declining phases of an open monocentric city. Specifically, the paper examines when, where, and how the land is developed, renewed and returned to agricultural land (or abandoned). Conclusively, the paper shows several transition patterns of land use, and clarifies which economic condition brings about such patterns.

要旨：This paper extends the work of Kempf and Rota-Graziosi (Journal of Public Economics, 2010, 94, 768-776), which argues that the sub-game perfect equilibria (SPEs) correspond to two Stackelberg outcomes under capital tax competition. The findings show that their result depends on the form of capital-ownership, that is, absentee ownership. By generalizing the form of capital-ownership, this paper shows that the simultaneous-move outcome prevails as an SPE if the capital is owned by residents in the countries, whereas the Kempf and Rota-Graziosi argument holds if the capital is owned by absentee owners.

本講では分割表の階層モデルの交互作用項に構造を入れることによるモデルの一般化について議論し, hierarchical subspace model(HSM)という新たなモデルのクラスを導入する. これまでもそうした一般化は比較的低次元の分割表モデルにおいては多くなされてきたが, HSMは推論の局所化という視点から, それらのモデルを統一的に扱うための理論的枠組を提供するとともに, 高次元分割表モデルへの自然な一般化を与える.本講演では特に尤度の分解, 最尤推定の局所計算, マルコフ基底の局所計算などについて議論し, マルコフ基底を用いた正確検定の数値例も紹介する.