【作者】 徐杨；

【导师】 安虎森；

【作者基本信息】 南开大学 ， 区域经济学， 2012， 博士

【摘要】 我国自改革开放以来，整体经济发展迅速，总体实力不断增强，然而同时国内的经济发展并不平衡，长期以来大量的工业部门和生产要素高度聚集在东部沿海地区，使其得到了突飞猛进地发展，而广大中西部内陆地区的经济增长与发展速度相对缓慢，与东部地区的差距也越来越大。对此党和国家政府确立了实现区域经济协调发展，逐步缩小地区发展差距的指导方针，提出要“促进区域间生产要素合理流动和产业有序转移，在中西部地区培育新的区域经济增长极，增强区域发展的协调性”。当前在中国希冀通过产业在区域之间合理、有序地转移，促进产业空间布局结构的优化，已经成为国家对未来有效改善区域发展差距问题、实现区域经济协调发展的一种美好愿景。本文尝试采用新经济地理学视角对上述问题进行研究。在对区域经济差异理论及其在我国的创新发展、产业转移的相关理论与研究进展进行回顾梳理与分析比较之后，对近年来逐渐兴起的新经济地理学理论的发展过程、理论基础及其核心思想作了评述，在此基础上以新经济地理学理论为分析框架，研究了产业的区际转移、聚集从而造成区域发展差距问题背后的内在机制与影响因素。之后以中国为案例进行了实证分析，并在此基础上提出了合理引导产业转移改善区域发展差距的政策含义与参考。本文在新经济地理学的垂直联系核心—边缘模型的理论框架下，引入了资源禀赋系数这一变量，并放松了原模型关于劳动力不能跨区域流动的限制，讨论了在存在不同地区之间的初始资源禀赋差异、工业部门之间的投入产出联系和劳动力的跨部门跨地区自由流动的情况下，由于产业在地理空间上的转移与聚集而形成的产业区域分布模式的内在机制和影响因素。引入的资源禀赋系数通过影响对称均衡状态被打破时对应的贸易自由度水平（即突破点B），从而对稳定的均衡状态具有重要的影响作用，这是因为一个地区所拥有的资源禀赋越丰裕，区内企业的生产能够节省更多的贸易成本，其总的成本支出就越低，从而会增强该地区的聚集力。劳动力不仅可以在同一地区的不同部门之间自由流动，也可以在不同的区域之间进行流动，在部门之间和区域之间名义工资水平差异的驱使下，其具体的流动情况可以通过在达到均衡状态时某地区工业部门所需要的劳动力数目与该地区所拥有的劳动力禀赋之比较来考察判定。在迪克希特—斯蒂格利茨的垄断竞争和规模收益递增的框架下，不连续的“块状”区域经济中存在着本地市场效应、价格指数效应和市场拥挤效应。前两种效应通过需求关联（后向联系）和成本关联（前向联系）的循环累积因果链形成了聚集力，它促进产业向某一地区集中，第三种效应则因企业之间竞争的存在形成了发散力，它使得企业为避开激烈竞争而向外迁移。聚集力与分散力的合力决定了最终的产业分布模式。在不均匀的产业空间分布模式下，产业活动密集的地区成为发达地区，位于该地区的企业和生产要素能够享受到区域“聚集租金”带来的好处，从而稳固强化了聚集的自我强化机制，即既有的区域经济发展差距会抑制产业从发达地区转移至欠发达地区，从而阻碍对自身差距程度的改善。不过聚集的这种自我强化机制会受到区域贸易自由度、区域市场规模、初始的资源禀赋等变量影响，本文的实证部分在此基础上进行了深入细致的讨论分析。在我国当前东部地区第二产业高度聚集，全国的区域工业发展差距正在逐年加大，并且东、中、西部三大地区之间的差距已经远远超过各自地区内部的差距。在理论模型的基础上，通过计量回归检验发现：我国工业的区域聚集程度与企业的迁移变化存在明显的相关关系；自然层面的贸易自由度水平越高越有利于区域产业聚集，但制度层面的贸易自由度对其影响还要视当地市场规模的大小而定；区域市场规模的大小对区域产业的聚集程度具有正向影响，并且目前在我国源自企业对中间投入品的支出购买力所产生的市场规模对区域的工业聚集具有更重要的影响；区域的资源禀赋丰裕程度对产业聚集的影响作用需要有当地政府层面的保护才能得到更好地发挥；一个地区对工业产品的支出份额对该地区的产业聚集程度有正向影响；市场规模较小地区的国有经济的适度发展更有利于该地区的产业聚集。在区域之间存在着既有的发展差距的情况下，由于发达地区的区位聚集租金优势、区位的黏性特征以及政策的门槛效应的存在，要想打破极度不均衡的区域产业分布模式、合理引导发达地区的部分产业向欠发达地区转移从而改善区域发展差距，就要加大政府区域经济政策的调控力度，使其越过政策的门槛值，跳出原有的发展路径。制订政策时需要从区域贸易自由度、区域市场规模、产业部门之间的投入产出联系等多角度出发，如加强欠发达地区内部交通通讯等基础设施与物流体系建设；适度发挥地方保护的作用，并因地制宜实行差别化的区域经济政策；建立完善的产业配套体系并积极发展工业园区经济；改革完善收入分配制度以缩小居民收入差距；在不发达地区合理适度地发展国有经济等。更多还原

【Abstract】 China’s economy has developed rapidly and the comprehensive national strengthhas been reinforced increasingly since reform and opening up. However, the domesticeconomy doesn’t developed evenly at the same time. A great deal of industrial sectorsand factors of production have highly aggregated in the eastern coastal areas so thatthese regions have improved by leaps an bounds. In contrast, the pace of the inlandregions’ economic growth and development is so slow that the the regional economicdevelopment disparities are enlarging. The Chinese Communist Party and the stategovernment have established the guidelines of coordinated and balanced developmentand reducing regional economic disparities step by step. It has been put forward thatit necessary to “advance the factors of production flowing reasonably and industrialtransferring ordely between rigions; foster new regional economic growth pole in thecentral and western regions so as to enhance the coordination of regional economicdevelopment”. At present the whole country hope to resolve the regionaldevelopment inequalities and achieve coordinated and balanced developmnet, via therational and ordered regional industrial transfer so as to optimize the industrial spatiallayout.This dissertation attempts to study on the above issues from the perspective ofnew economic geography. It reviews the theoies of regional economic differences andtheir innovation and development in China, along with the theoies and researchprogress of industrial transfer. Then we introduce the development progress,theoretical basis and core ideas of new economic geography theory which has beengradually on the rise in recent years. On the theoretical framework of new economicgeography theory, we try to study the internal mechanism and influencing factorshiding behind the regional industrial transfer and agglomeration that cause theregional development inequalities. After that, we take China’s economic developmentas the case for empirical analysis. Based on the theoretical and empirical analysis, thisdissertation proposes some policy implications and elicitation. Grounded on the theoretical framework of new economic geography, thisdissertation introduces the coefficient of resource endowment to the core-peripheryvertical-linkage model (CPVL model for short), and loosens the limitation that laborscannot flow among different regions. It discusses the internal mechanism andinfluencing factors of regional industrial distribution pattern as a result of thegeographical spatial industrial transfer and agglomeration, in the case that differentregions own dissimilar initial resource endowments, input-output linkages in thepresence of industrial inter-sectoral, and labors are mobile among not only industrialsectors but also regions. The coefficient of resource endowment affects the breakpoint (B) where the symmetric equilibrium switches from stable to unstable,therefore it plays an important role in the stable equilibrium of the extensive model.This is because the more abundant resource endowment a region owns, the more tradecost the firms located there can save, and the total production cost is lower so that thisregion’s agglomeration forces could be reinforced. Labors can flow freely not onlyamony various sectors in the same region, but also amony different regions. Drivenby the nominal wage gap, the specific flow can be reviewed and determined throughthe comparison between the labor demanded by some region’s industrial sectors andthe labor endowment that the region initially possesses.Upon the Dixit-Stiglitz monopolistic competition and increasing returns frametheory, the economy is a very lumpy place where exists three effects: the ‘homemarket effect’, the ‘price index effect’ and the ‘market crowding effect’. The ‘homemarket effect’and the ‘price index effect’ shape into agglomeration forces thatstimulate industries to aggregate into some region, through the potential for circularcausality which is also knowns as cumulative causality, or backward and forwardlinkages. In contrast, the ‘market crowding effect’ shapes into dispersion force, and itcauses imperfectly competitive firms to move outwards in order to avoid fiercecopetition. It is the resultant force of agglomeration forces and dispersion force thatdicides the final industrial distribution pattern. Under the unequal spatial industrialdistribution pattern, the areas that own intensive industrial activities becomedeveloped regions. Firms and factors of production located in these regions can enjoythe benefit of agglomeration rents so that the self-reforcing mechanism can be strengthened. That is to say, the existing regional economic development disparitieswill restrain industries transferring from the developed areas to underdeveloped areas,and impede the narrowness of the regional disparities itself. Nevertheless, thisself-reinforcing cycle of agglomeration will be affected by regional free-ness of trade,regional market size, initial resource endowment and so on. According to thetheorectical analysis, the empirical part of this dissertation discusses and analyzesparticularly.In China the secondary industry highly aggregated in the eastern regions and thewhole country’s industrial development disparity is increasing year by year. What’smore, disparities among the eastern, central and western regions have exceeded thatinside the three regions respectively. Based upon the theoretical model and by theregression models, the empirical analysis observes some relevant conclusions: there isnotable correlation betwween the regional industrial agglomeration degree and themove of firms; the high natural free-ness of trade contributes to regional industrialagglomeration, while the effect of the institutional free-ness of trade also depends onthe local market size; market size has a positive impact on the regional industrialagglomeration degree, and at present the market size derived from the purchasingpower on intermediate inputs plays a more inportant role; the impact of abundantdegree of regional resource endowment on industrial agglomeration needs theprotection from the local government; the share of expenditure of industrial producthas a positive effect on the regional industrial agglomeration; the appropriatedevelopment of state-owned economy contributes to the industrial agglomerationwhere the market size is mall.On the condition of the existing regional development disparities, since thedeveloped areas owns agglomeration rents, the existence of locational hysteresis andthe threshold effects of policy, it is necessary to enhance the control potency ofregional economic policies and make it overleap the threshold so as to change theoriginal development path. The formulation of policies needs to consider from theperspective of regional free-ness of trade, regional market size, input-output linkagesand so on. For example, it is necessary to enhance the infrastructure including thetraffic transportation, communication and logistics system inside the underdeveloped regions; making the local protection play a appropriate role, and formulatingdifferential regional economic policies; establishing complete coordinated industrialsystem and devoloping industrial park economy; reforming the income distributionsystem economy in order to narrow the income gap; developing the state-ownedeconomy in an appropriate degree in the underdeveloped regions and so on.更多还原