The impact of population changes on economic development in OECD countries

WANG Jinying1 LI Tianran1

(1.School of Economics, Hebei University of China)

【Abstract】Based on the data of 17 OECD countries from 1960 to 2016, this paper examined the trajectories of population size, population quality, age structure and distribution. And it explored the impact of population changes on economic development, industrial structure and development quality by using panel-corrected standard errors and feasible generalized least squares. The result shows that, given the population size and quality fixed, the population growth rate has negative impacts on economic growth and the quality of economic development. The migration rate contributes positively to economic growth and the quality of economic development, although it restrains the change of economic structure. An increase in working-age population has a positive impact on economic growth, economic structure adjustment, and development quality. Likewise, population aging promotes the adjustment of industrial structure. The improvement of education has a significantly positive impact on economic growth and development quality. The increase of life expectancy promotes the vigorous development of service industry. In addition, population agglomeration and migration accelerate the economic growth, enhance the quality of economic development, and promote the change of economic development mode in general. The paper points out that the impact of population on economic development results from all demographic aspects.

【Keywords】 population changes; human capital; industrial structure; economic development; development mode;


【Funds】 Major Project of National Social Science Fund of China (12ARK001)

Download this article


    [1]. ① GDP data are converted from China Yuan to US dollar using the official exchange rate in 2010. For a few countries where the official exchange rate does not reflect the effective exchange rate used in actual foreign exchange transactions, an alternative conversion factor is used. [^Back]

    [2]. ② This paper calculates total factor productivity based on the total factor productivity index (100 in 2010), using the data of real GDP, real net capital stock and total employment with 2010 as the base in the AMECO database. [^Back]

    [3]. ③ This paper uses Barro-Lee’s database of educational attainment, which is based on the average years of schooling with an interval of five years, and uses the linear interpolation method to calculate. Among them, the average years of schooling from 2014 to 2016 are predicted data in the database. [^Back]

    [4]. ① The OECD countries selected in this paper include Austria, Australia, Portugal, Greece, the Netherlands, Japan, Finland, Italy, Norway, Belgium, the UK, Canada, Spain, Sweden, the United States, France and Denmark. [^Back]

    [5]. ② [^Back]

    [6]. ③ Since the sample data in this paper are long panel data, LLC, IPS, Fisher-ADF and Fisher-PP test methods are adopted according to the precondition of asymptotic theory test. However, these four tests have certain deficiencies and differences. Theoretically, each method has its own characteristics, but it cannot solve all the problems. For example, LLC test assumes that each panel unit has the same autoregressive coefficient, that is, a common root, and requires panel data to be balanced panels. However, the other three tests allow panels to have different autoregressive coefficients and allow panel data to be unbalanced panels. Therefore, four test methods are usually used. When passing the test at the same time, it can indicate the unnecessary deviation that the regression of the panel data used will not cause. [^Back]

    [7]. ④ Similarly, when the tests are conducted on whether there is a long-term equilibrium relationship between variables in panel data, Kao test, Pedroni test and Westerlund test have their respective advantages and disadvantages and can complement each other. Therefore, they are usually used at the same time in order to eliminate some interference factors as much as possible. [^Back]

    [8]. ① Since the panel data in this paper are unbalanced panel data, it is only assumed that the disturbance terms of different individuals are allowed to have heteroscedasticity when the full FGLS method is adopted. [^Back]

    [9]. ② Since FGLS is more effective than PCSE, this paper mainly analyzes the estimation results of FGLS. [^Back]

    [10]. ① Due to the large time span of this paper, the use of LSDV method to introduce both individual and time effect variables will lose more sample size, so the time trend t is chosen as the time effect variable of the double fixed effect model; and t was 1 in 1960, 2 in 1961 and so on. [^Back]


    1. Cai, F. et al. Economic Research Journal (经济研究), (9) (2009).

    2. Liang, H. & Xu, C. Northwest Population Journal (西北人口), (2) (2016).

    3. Liu, C. & Zhang, X. Journal of Xiangtan University (Philosophy and Social Sciences) 湘潭大学学报(哲学社会科学版), (3) (2016).

    4. Sun, S. et al. Economic Research Journal (经济研究), (1) (2014).

    5. Wang, J. & Fu, X. Population Research (人口研究), (1) (2006).

    6. Wang, J. & Li, J. Chinese Journal of Population Science (中国人口科学), (3) (2016).

    7. Wang, Y. & Ni, C. China Population, Resources and Environment (中国人口·资源与环境), (5) (2013).

    8. Wang, Z. & Liang, C. China Population, Resources and Environment (中国人口·资源与环境), (10) (2012).

    9. Xiao, W. & Yang, Y. Population Research (人口研究), (4) (2017).

    10. Yang, X. 新兴古典经济学和超边际分析. Beijing: China Renmin University Press, (2000).

    11. Yao, Y. Journal of Finance and Economics (财经研究), (5) (2017).

    12. Yu, H. Economic Research Journal (经济研究), (10) (2015).

    13. Zhang, T. Economic Science (经济科学), (5) (2016).

    14. Aoki, M. and Yoshikawa H. (2002), Demand Saturation-creation And Economic Growth. Journal of Economic Behavior and Organization. 48(2): 127–154.

    15. Berliant, M. and Konishi H. (2000), The Endogenous Formation of a City: Population Agglomeration and Marketplaces in a Location-specific Production Economy. Regional Science and Urban Economics. 30(3): 289–324.

    16. Brunner, J. K. and Zweimüller, J. (2005), Innovation and Growth with Rich and Poor Consumers. Metroeconomica. 56(2): 233–262.

    17. Desmet, K. and Parente, S. L. (2010), Bigger is Better: Market Size, Demand Elasticity and Innovation. International Economic Review. 51(2): 319–333.

    18. Fu, Y. and Gabriel, S. A. (2012), Labor Migration, Human Capital Agglomeration and Regional Development in China. Regional Science and Urban Economics. 42(3): 473–484.

    19. Murphy, K. M., Shleifer, A. and Vishny, R. (1989), Income Distribution, Market Size, and Industrialization. The Quarterly Journal of Economics. 104(3): 537–564.

    20. Smith, R. G. (1972), Optical Power Handling Capacity of Low Loss Optical Fibers as Determined by Stimulated Raman and Brillouin Scattering. Applied Optics. 11(11): 2489–2494.

This Article


CN: 11-1043/C

Vol , No. 06, Pages 2-16+126

December 2018


Article Outline



  • 1 Introduction
  • 2 Model setting, variable selection and data sources
  • 3 Test methods of model regression
  • 4 Empirical results
  • 5 Conclusion and implication
  • Footnote