中文摘要：

本文通过理论模型揭示了制造业就业与服务业就业之间存在交互乘数及空间溢出效应。利用中国2003—2013年285个地级及以上城市的数据，建立联立方程模型，基于不同空间权重矩阵，使用GS3SLS方法进行事实检验。研究发现：制造业就业与服务业就业之间存在双向促进作用机制，城市制造业就业每增加1％，会带来服务业就业0．45oA～0．47％的提升，其中，生产性服务业的受益最大；服务业就业每增加1％，会带来制造业就业1．15％～1．16％的提升。城市之间制造业就业和服务业就业均存在正向的空间溢出效应，而二者的空间交互溢出效应为负，表现为发达地区拥有高端制造业的同时，会吸引周边地区从事服务业的劳动力迁入。进一步的分析表明：随着时间的推移，服务业就业对制造业就业的乘数效应占主导，制造业的发展愈发依靠服务业的不断完善，二者的空间交互溢出效应逐渐减弱。政策上，促进劳动力的流动、深化市场化改革、充分发挥大城市的规模经济、完善城市结构体系、加强交通基础设施建设和公共服务有助于提升第二、三产业之间相互创造就业岗位的能力。

英文摘要：

This paper reveals the existence of multiplier interaction and spatial spillover effect between manufacturing employment and service employment through theoretical model. Based on the data of 285 prefecture-level cities in China from 2003 to 2013, and different spatial weight matrix, we establish a simultaneous equation model with GS3SLS method. Empirical results show that there is a mutual promoting effect mechanism between manufacturing employment and service employment, urban manufacturing employment increased by 1%, will lead to a rise of 0.45%-0.47$ in service employment, while service employment increased by 1% will bring a promotion of 1.15 % - 1.16% in manufacturing employment. There is a positive spatial spillover effect between manufacturing employment and service employment, but the spatial interaction spillover effect is negative, performance for the developed regions own high-end manufacturing, at the same time, will attract the labor migration from surrounding area. Further analysis shows that with the passage of time, the multiplier effect of service employment on manufacturing employment is dominant, the development of manufacturing industry relies increasingly on continuous improvement of service industry, and the interactive space spillover effect between them gradually become weakened. We argue that it needed to promote labor flow, deepen the market-oriented reforms, give full play to the big city economies of scale, improve the structure of urban system, strengthen the traffic infrastructure construction and public service to enhance the employment multiplier effect.