老龄化对货币需求影响的研究-基于省级面板数据
摘要
人口老龄化是一个国家或地区经济社会发展到某个阶段下出现的社会现象，
经济社会快速发展叠加独特的计划生育政策导致我国面临“未富先老”局面。人
口老龄化不断加剧是我国的国情，研究和应对老龄化带来的各项社会经济方面的
冲击和影响，是学者普遍关心的议题。人口结构方面的经验研究主要集中于储蓄
率和经济主体风险偏好上。有关我国货币需求的文献较为丰富，可是鲜有从人口
结构的角度出发的。有针对性地研究老龄化对货币需求影响的文献现在较为空白。
本文的创新之处在于，基于国内的省级面板数据，研究老龄化对货币需求的影响，
有助于补充有关我国老龄化对货币需求影响的相关研究，有助于解释进入老龄化
社会后我国的高货币化现象。
本文的经验逻辑基础是双向固定效应模型，根据 LSDV法，认为模型存在
个体效应。在混合回归和随机效应之间，LM检验表明结果选择随机效应。依据
Hausman检验选择随机效应还是固定效应。实证分析之前，本文对面板数据做了
单位根检验，并通过协整检验。确定基础模型之后，本文处理了自相关、异方差
和截面相关问题，并使用系统 GMM方法处理可能存在的内生性问题。为了研究
老龄化对货币需求的异质性现象，本文运用了分组回归模型。其中分组的依据是
我国东、中和西部地区的划分。老龄化对货币需求影响存在扩散现象或者说边际
效率递增现象，原因可能是经济发展水平的提升。本文运用了门槛模型，并选择
实际 GDP作为门槛变量。为了验证研究结论的稳健性，本文把被解释变量城乡
储蓄存款替换为银行存款总和，更换了欧元区样本，两次验证了结果的稳健性。
实证结果表明老龄化对货币需求存在正向影响关系。老龄化对不同地区货币
需求的影响具有异质性。人口老龄化对经济发达的东部地区的货币需求有正影响；
对经济发展程度中等和较低的中西部地区的货币需求有负影响。门槛模型的回归
结果表明，门槛效应存在，老龄化对货币需求的影响随着实际 GDP提高而显著
增加。最后，本文基于老龄化背景和相关研究结论提出了有关政策建议。
关键词：老龄化；资产配置；货币需求；异质性
I

老龄化对货币需求影响的研究-基于省级面板数据
Abstract
The Population aging is a social phenomenon that occurs when the economic and
social development of a country or region reaches a certain stage. The rapid economic
and social development combined with my country's unique family planning policy
has caused my country to face the situation of "getting old before getting rich". The
increasing aging of the population is the national condition of our country, and
studying and responding to the various social and economic impacts and impacts
brought about by aging are topics of general concern to scholars. The empirical
research on population structure mainly focuses on the savings rate and the preference
of economic entities. The literature on my country's currency demand is relatively rich,
but there are few starting from the perspective of population structure. The literature
on the targeted study of the impact of aging on the demand for money is now
relatively blank. The innovation of this article is that, based on the domestic
provincial panel data, studying the impact of aging on currency demand will help
supplement the relevant research on the impact of my country’s aging on currency
demand, and help explain my country’s entry into an aging society. The phenomenon
of high monetization.
The empirical logic basis of this article is the two-way fixed-effect model.
According to the LSDV method, it is believed that the model has individual effects.
Between mixed regression and random effects, the LM test indicates that the results
choose random effects. Choose random effect or fixed effect according to Hausman
test. Before the empirical analysis, this paper does a unit root test on panel data and
passes the cointegration test. After determining the basic model, this paper deals with
autocorrelation, heteroscedasticity and cross-section related issues, and uses the
systematic GMM method to deal with possible endogeneity issues. In order to study
the heterogeneity of aging demand for money, this paper uses a group regression
model. The grouping is based on the division of my country's eastern, central and
western regions. The impact of aging on currency demand has a diffusion
II

老龄化对货币需求影响的研究-基于省级面板数据
phenomenon or a phenomenon of increasing marginal efficiency. The reason may be
the increase in the level of economic development. This article uses a threshold model
and chooses actual GDP as the threshold variable. In order to verify the robustness of
the research conclusions, this article replaced the explanatory variable urban and rural
savings deposits with the sum of bank deposits, replaced the euro area sample, and
verified the robustness of the results twice.
The empirical results show that aging has a positive influence on money demand.
The impact of aging on currency demand in different regions is heterogeneous. The
aging of the population has a positive impact on the currency demand in the
economically developed eastern region; it has a negative impact on the currency
demand in the central and western regions with medium and low economic
development. The regression results of the threshold model show that the threshold
effect exists, and the impact of aging on money demand increases significantly with
the increase of real GDP. Finally, this article puts forward relevant policy
recommendations based on the aging background and relevant research conclusions.
Keywords: aging; asset allocation; money demand; heterogeneity
III
。。。以下略