![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
The aim of this paper is to empirically examine the regional determinants of the fertility rate in Russia using panel data for the period of 2005–2015. The estimation results of a system GMM dynamic model revealed that economic growth, employment opportunity, favourable local business conditions, educational opportunity, quality of social infrastructure, and housing supply serve to increase the fertility rate in Russian regions, while the presence of a Slavic population, migration inflow, poverty and ecological risks tend to suppress it. Furthermore, we found that combinations of factors that strongly affect the reproductive behaviour of Russian women vary greatly among age groups and regions. To mitigate the declining trend of fertility in Russia, it is necessary to implement policies that take generational differences and regional heterogeneity into serious consideration.
Acknowledgments
In the study meeting of the Institute of Economic Research at Hitotsubashi University on 25 July 2018, we received numerous comments on the earlier version of the paper. In this regard, we would like to express particular gratitude to Naohito Abe, Reiko Goto, Ryo Jinnai, Yukinobu Kitamura, Takashi Kurosaki, Hodaka Morita, Yuka Takeda, Masahiko Tsutsumi, Tsuyoshi Tsuru, and Emiko Usui. We also wish to thank Charles M. Becker, Richard Connolly (the editor), Róbert I. Gál, and two anonymous reviewers of the journal for their helpful advices, Keiko Suganuma for her research assistance, and Dawn Brandon for her editorial assistance.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. According to Russian labour statistics, ‘working-age population’ is defined as males aged 16–59 years and females aged 16–54 years. The typical age of eligibility to receive a pension differs for men and women, being 60 years of age for the former and 55 years of age for the latter.
2. See the website of the Federal State Statistics Service (hereinafter, ‘Rosstat’) at http://www.gks.ru/.
3. These acts denote the Federal Law of the Amendment of Several Federal Laws and Regulations Concerning State Assistance for Citizens with Children dated 5 December 2006, and the Federal Law of Additional State Support for Families with Children dated 29 December 2006, respectively.
4. The amount of ‘mother’s capital’ was 429 thousand rubles in 2014. The necessary amount to buy a standard apartment with an average floor space for one person in Moscow city in the year was 2,570 thousand rubles, and this meant that the mother’s capital could compensate less than 17% of this cost. Even in the Russian Far East, in the peripheral but comparatively urbanised areas, the amount of mother’s capital is less than 29% of the price for a standard apartment for one person on average. Note that the nominal amount of mother’s capital does not differ from region to region: hence, the meaning of mother’s capital must be very limited in Moscow or in other urban areas. On the other hand, if one calculates in the same way for the Republic of Kalmykia, where more than 50% of people are rural residents and the average price of apartments is the cheapest in Russia, the mother’s capital could compensate more than 69% of the cost for a standard apartment for one person (all the data are available at the Rosstat Website [http://www.gks.ru/] and Rosstat, Citation2017). Then the scheme of mother’s capital may have some effects on fertility in rural areas. In Russia, however, the urbanisation ratio is well above 70%; therefore the significance of this policy may not be nation-wide one. This is the reason why the authors pay special attention to the regional variation of living conditions in determining fertility rates. We acknowledge a debt of gratitude to Charles M. Becker for his suggestions.
5. National Centre for Health Statistics, https://www.cdc.gov/nchs/nvss/births.htm, accessed on 24 August 2019; United States Census Bureau, https://www.census.gov/, accessed on 24 August 2019.
6. Research in Japan has produced similar findings, with Kamata and Iwasawa (Citation2009) using prefectural data to perform empirical research. The findings indicate a significant negative correlation between the unemployment rate and the fertility rate; further, the rate of employment has a significant and positive effect on the fertility rate.
7. In 2016 and 2017, the TFR in Russia decreased, possibly due to the social tension caused by conflict with Ukraine and economic sanctions introduced by the United States and many European countries.
8. A ‘federal district’ is a regional zone under the control of a plenipotentiary representative of the president that is established for the purpose of overseeing and supervising regional administration. Federal districts differ from ‘federal entities,’ which are general administrative divisions and comprise federal cities, republics, krais, oblasts, autonomous oblasts, and autonomous okrugs. Federal districts comprise federal entities with a great deal in common geographically, historically, and ethnically. Federal districts, therefore, constitute administrative zones useful for identifying and comparing general characteristics of Russian regions.
9. White indicates a region for which data was not available. Rosstat has not published fertility rates for six autonomous districts since 2010.
10. Although the age range of women subject to the calculation of the total specific fertility rate is 15–49, Rosstat does not publish life-stage fertility rates for women aged 35 years or older, so it is impossible to perform analyses of these higher age groups.
11. The mean (median) of the TFR for each federal entity in the same year is 1.836 (1.815).
12. Here, attention needs to be paid to the 15–19 age group. During the period of 2005–2015, the proportion of children born outside of wedlock in Russia as a whole declined fairly steadily from more than 29% to less than 22%, but for the 15–19 age group, it remained in the 47–49% range, which is far higher than the figures for the 20–34 age group, which did not exceed 30% at any point during the period. In other words, the proportion of unmarried mothers is extremely high among the younger age groups, and it is highly likely that this factor could result in reproductive behaviour that differs from that seen in other age groups. However, the focus of the subsequent analysis in this paper is to show that fertility determinants for each age group differ, so the issue of why fertility determinants differ is beyond this paper’s scope. To answer that question, analysis involving micro-data will be required.
13. Table A1 reports a correlation matrix of the dependent variables and correlation coefficients between the dependent variables and independent variables.
14. As the table shows, the Arellano–Bond test for AR(2) rejects the null hypothesis of no autocorrelation at the 10% significance level for two of the nine models only, indicating that the assumption of the disturbance term is adequately satisfied for the system GMM estimation. Although we have omitted the test results, the test for AR(1) rejected the null hypothesis for all of the models reported in this paper at the 5% significance level or lower. On the other hand, the Sargan test rejects the null hypothesis that overidentifying restrictions are valid in all cases. The estimation results of the dynamic model, therefore, have some room for improvement in model specification.
15. If regional differences in age structure are attributable to the determinants of fertility rates, direct control for the female population in each age group in each region could be adopted for regression analysis. In Russia, however, fertility-rate data for each five-year age bracket is not available. Furthermore, estimates to extrapolate the population in each bracket are not made between each census. This is the reason we employed this approach for our study.
17. The results are reported in Table A6.
18. In 2016, the nationwide average was 359,000 rubles per year, while in the Nenets Autonomous Okrug, it was 699,000 rubles per year.