此博士論文由三篇獨立論文組合而成，每篇論文所討論的議題為區域經濟或產業經濟。第一篇論文為「 Spatial Competition and Flexible Manufacturing」，採用一個二維(two-dimensional)空間模型，假設廠商從事Bertrand價格競爭，探討廠商擁有彈性生產下，廠商最適區位的選擇。此篇文章發現，給定服務全市場的條件成立下，若運輸費率相對於調整費率低時，兩廠商會聚集在市場中點；反之，當運輸費率相對於大時，兩廠商會選擇分離區位。不管運輸費率大小，基礎產品的位置會在1/4與3/4來區位。此外，若運輸費率(調整費率)越大，則兩廠商分離越遠(越相互靠近)。 第二篇論文為「Patent Licensing and Double Marginalization in Vertically Related Markets」，給定垂直相關市場，上下游廠商經由「談判」決定中間財價格，主要探討產業外專利權擁有者的最適授權方式與社會福利的影響。此篇文章發現若上游廠商談判力越強，則專利權擁有者的最適授權方式為固定權利金授權(fixed-fee licensing)；反之，上游廠商談判力越弱，則單位權利金授權(royalty licensing)為最適授權方式。此外，傳統文獻指出，若消除完全雙重邊際化會增進社會福利。本篇文章卻發現，當上游廠商談判力越強，最適授權方式由單位權利金授權改變為固定權利金授權，進而使社會福利上升。此一結果與傳統文獻有明顯的不同。 第三篇論文為「Patent Licensing in a Leadership Structure: Comment」，探討Kabiraj (2004)的結果不同於Kamien and Tauman (1986)的結論。此篇文章依循Kabiraj (2004)的模型架構，但採用Kamien and Tauman (1986)固定權利金的定義。此篇文章發現，當創新程度相對小時，且專利權擁有者採用固定權利金授權，其最適授權家數為兩家廠商。此外，當專利擁有者的授權策略為固定權利金、單位權利金與拍賣時，單位權利金決對不會被選擇。 This dissertation consists of three independent papers. Each of them explores issues involved in either regional or industrial economics. The first paper is entitled “Spatial Competition and Flexible Manufacturing”. This paper constructs a two-dimensional framework to explore firms’ equilibrium locations by taking into account flexible manufacturing. It shows that the two firms will agglomerate at the center of the location line and the optimal attributes of the two basic products are located at the first and third quartiles of the attribute line, as the transport rate relative to the marginal modification rate is lower, while will get apart on the location line and the optimal attributes of the two basic products are located at the first and third quartiles of the attribute line, as the transport rate is relatively higher. Secondly, the two firms would locate at first and third quartiles of the location line and the attribute addresses of the basic products would agglomerate at the center of the attribute line as the relative transport rate is further higher so that market-serving condition is violated. Lastly, two firms would locate at the opposite endpoints of the line segment and the optimal attributes of the two basic products would be located at the first and third quartiles of the attribute line under mill pricing with quadratic transportation costs. The title of the second paper is “Patent Licensing and Double Marginalization in Vertically Related Markets”. The conventional wisdom indicates that the elimination of the double marginalization unambiguously increases social welfare. This paper develops a three-stage model, in which the input price is determined by a bargaining process between the upstream and the downstream firms in vertically related markets. It shows that the outsider patentee prefers royalty (fixed-fee) licensing to fixed-fee (royalty) licensing, as the bargaining power of the upstream firm is small (large) irrespective of the innovation size. Moreover, it also proves that the social welfare may get improved by switching from a royalty licensing to a fixed-fee licensing selected by the outsider patentee, as the bargaining power of the upstream firm is large enough. The third paper is “Patent Licensing in a Leadership Structure: Comment” and it has been accepted by the Manchester School. This paper explains why Kabiraj (2004) has obtained so different results compared with those of Kamien and Tauman (1986) and utilizes a different definition from that in Kamien and Tauman (1986) and Kamien et al. (1992). It shows that a transfer to both firms under fee licensing is optimal when the innovation size is small. It also proves that a royalty contract will never be an optimal contract when fee, royalty and auction contracts are available.