【作者】 李文星；

【导师】 徐长生；

【作者基本信息】 华中科技大学 ， 西方经济学， 2008， 博士

【摘要】 居民消费率过低是中国经济当前面临的一个长期结构性问题。本文从理论和实证的角度分析中国人口年龄结构变化对居民消费产生的影响,首先对中国人口年龄结构状况进行了介绍,然后讨论了人口年龄结构对居民消费的影响渠道,最后总结了新中国人口年龄结构与居民消费的经验关系。新中国建国以来中国人口年龄结构的特点体现为时间和空间两个维度的差异。新中国人口年龄结构的历史演变过程可分为年轻型(1949-1964年)、成年型(1965-1999年)和老年型(2000年以来)三个阶段。中国人口年龄结构在地域上存在显著的差异,东部发达地区老龄化进程一般更快,中部地区次之,西部和少数民族地区尚未步入老龄社会。人口年龄结构的不同发展阶段对居民消费具有不同的影响渠道。短期内,人口生育率冲击会增加人口总量,并提高抚养系数。长期内,这种冲击还会传导到就业和老龄化。就业高峰冲击一方面由于劳动人口的增加而提高产出,并且就业高峰意味着较低的总抚养系数,这有助于提高居民消费水平;另一方面,就业高峰冲击又会降低人均资本存量,并且就业竞争会压低长期均衡工资水平,这又会降低居民消费。短期内,由于老龄人口只花费过去的积累,老龄化冲击会增加消费,但它会减少劳动人口,降低未来的产出和居民消费水平。但如果老龄化诱致的技术进步足够抵消老龄化导致的劳动人口的减少,未来的产出和消费则不一定下降。日本、亚洲“四小龙”和中国经济起飞时都正好处于“人口红利”时期,经济增长带动了居民消费不同程度的上升。但中国居民消费率的事实表明,“人口红利”与居民消费之间并未显示出显著的正相关。本文首先分析了新中国人口年龄结构变化的第一个阶段——人口生育率高峰冲击对居民消费的影响。使用中国1952-2004年的时间序列数据和两种不同的方法,对中国实际居民消费与实际GDP、人口自然增长率三个变量进行协整检验和回归后发现,无论是在长期还是短期,中国人口自然增长率与居民消费之间存在显著的正相关,表明中国人口自然增长率的下降是中国居民消费率下降的重要原因。鉴于目前中国人口自然增长率下降已接近谷底,因此,如果维持现行计划生育政策,可以推断未来中国人口自然增长率对中国居民实际消费率的影响将是有限的。人口生育率高峰冲击之后,是中国人口年龄结构变化的第二阶段——人口就业高峰冲击。本文也同样使用了中国1952-2004年的时间序列数据和两种不同的协整回归方法来分析就业高峰冲击对居民消费的影响。这部分在上述三变量的基础上,将就业人口比率引入计量模型,以分析就业高峰冲击对居民消费的影响。对中国实际居民消费与实际GDP、人口自然增长率、就业人口比率的协整检验和回归结果表明,中国人口自然增长率与居民消费之间仍然存在显著的正相关。而无论是在长期还是短期,衡量就业高峰冲击的就业人口比率与中国实际居民消费之间并没有显著的相关性。在分析中国人口年龄变化的第三阶段——人口老龄化冲击对居民消费的影响,本文使用了中国1989-2004年的省际宏观面板数据和动态面板GMM估计方法,并区分了儿童抚养系数和老年抚养系数对居民消费的不同影响。结果发现,中国儿童抚养系数对居民消费率具有弱显著的负影响,但这种影响在数量并不大;而中国老年抚养系数与居民消费率之间并没有显著的相关性。对中国人口年龄结构三个不同发展阶段的实证结论基本是一致的,中国人口年龄结构与居民消费之间并不存在稳定的关系,因此,不能认为人口年龄结构对中国居民消费产生了显著的影响;而中国人口自然增长率的长期下降是中国居民消费率下降的重要原因。更多还原

【Abstract】 China has experienced a long-run decline in the ratio of household consumption to GDP in the last decades, This dissertation aims to investigate whether or not the population age structure(PAS)is an important factor that leads to the low ratio of household consumption to GDP. Firstly,it summarizes the changes in PAS in China. Secondly, it examines the various transimission mechanisms how PAS affects household consumption. Finally, it tests the empirical relationships between PAS and the household consumption.Descriptions of differences in both time and space are given about the characteristics of changes in PAS since the foundation of P. R. China. Firstly, the evolutionary stages of PAS in China are divided into three segments which are Young Stage(from the year of 1949-1964), Mature Stage(from the year of 1965-1999), and Old Stage(since the year of 2000). Secondly, the diversity of the development of PAS among China is presented using the old dependency ratio during the period of 1989-2005.Basically, the developed regions have entered a period of faster ageing than the less-developed regions. While other regions still remain in the young stage of PAS.PAS affects household consumption through different mechanisms. In the short run, the fertility shock will increase population and makes the young dependency rate higher. In the long run, it will cause the employment shock and the ageing shock. For the employment shock, on the one hand, it keeps the household consumption higher because of the low dependency rate, and it also leads to more aggregate output due to more labors; on the other hand, the employment shock makes the household consumption lower in the long run because of the lower per capita capital-stock and lower real equilibrium wages in the labor market. For the ageing shock, it will increase the household consumption in the short run, but it will also imply less labors which make aggregate output and the household consumption lower in the long run. If the ageing shock triggers advances in technologies, both aggregate output and the household consumption are less likely to decrease. Some Asian countries or regions like Japan, the Asian“Four Small Dragons”, and China’s mainland, enjoyed a quick economic growth during their demograhic windows of opportunity. So economic growth causes the household consumption to increase at differet paces in different countries or regions. It seems that demographic dividend does not have a significant and positive correlation with the household consumption if we consider the low ratio of the household consumption to GDP in China.About the effects of the first development stage of PAS, that is the fertility shock, on household consumption, this dissertation uses time series data for the period of 1952-2004 in China and conducts cointegration regressions with three variables, which are real household consumption, real GDP, and natural population growth rate. The results show that natural population growth rate plays a significant and positive impacts on the household consumption both in the long run and in the short run. This suggests that the decline in natural population growth rate led partly to the decline in the ratio of household consumption to GDP in China.Because natural population growth rate has nearly reached the bottom of its stage, it is expected that it would play a limited impact on the household consumption in the near future if the One-Child Policy is still maintained.With regard to the effects of the second development stage of PAS, that is the employment shock, on the household consumption, this dissertation also uses the time series data for the period of 1952-2004 in China and makes cointegration regressions with four variables, which are real household consumption, real GDP, natural population growth rate, and emloyment rate which is measured by employment population to whole population. The difference between the fourth and fifth chapter is a new addition of employment rate in the regression equation in the fifth chapter. This will help test the effects of employment rate on the household consumption. The results show that there is a significant and positive relationship between the natural population growth rate and the household consumption, but there is not a significant relationshiop bewteen employment rate and the household consumption both in the long run and in the short run.About the effects of the third development stage of PAS, that is the ageing shock, on the household consumption, this dissertation employs the provincial panel data and conducts a dynamic panel data GMM regression. Especially it separates the impact of the young dependency rate on household consumption from that of the old dependency ratio. The results show that the youth dependency ratio has a negative but small impact on the household consumption, while the old dependency ratio does not have a significant impact on the household consumption. Hence changes in PAS can not explain the low ratio of the household consumption to GDP in China.These empirical results above are consistent. They suggest that there is not a stable relationship between PAS and the household consumption in China. It is not accepted that changes in PAS contribute to the decline in the ratio of the household consumption to GDP. But the natural population growth rate is an important factor which leads to the decline in the ratio.更多还原