As a direct result of several influencing factors, including those that are economic, political, scientific, technological, cultural, and historical, the fertility of China’s population presented a trend of fluctuating decreases. This study explored the impact of economy and education on population fertility using existing literature and analysis of longitudinal data at the provincial level. This study found that the economy does play a role in improving the fertility rate although inverse development trends of gross domestic product (GDP) and crude birth rate (CBR) were found. Furthermore, schooling years per capita (PEDU) also had an inverse relationship with fertility, but the proportion of higher education population (HEDU) was positively associated with fertility. Due to urgency concerning the continuously low fertility rate in the post-demographic transition period in China, this study provides some evidence for the formulation of fertility policy in China. The results also suggest that promoting economic development and advancing the popularization of higher education are important paths to enhance female fertility willingness and fertility rates.
Data used in this study were obtained from the China Statistical Yearbooks from National Bureau of Statistics of China and from provincial-level bureaus of statistics. Based on the CBR (from 1949 to 2020) and GDP (from 1952 to 2020) of China, this study analyzed the correlation between fertility and economic activity nationwide. The GDP and CBR trends at the regional level (the eastern, central, western, and northeastern region, divided by the overall national levels provided by the National Bureau of Statistics of China①) were further described. Then, a panel database of 31 provincial-level administrative divisions (PLADs) of China from 2002 to 2019 was constructed with indicators of CBR (%), GDP (100 billion CNY), PEDU (year), and HEDU (%), sex ratio (female=100), urban population density (100 persons/km2), urbanization rate (%), urban registered unemployment rate (%), old-age dependency ratio (%), labor proportion (total population=1), and average family household size (persons).
Based on the panel data, a fixed-effects model was used for regression analysis with CBR as the predicted variable, GDP, and PEDU and HEDU as explanatory variables, and other indicators mentioned above as control variables. Stata 13.0 (developed by StataCorp LLC, Texas, USA) was used for all the data analysis.
Figure 1 presents long-term changes in the economy and fertility. At the national level, Chinese CBR decreased overall from 1949 to 2020, which is inverse to the overall increase in GDP. From a regional perspective (Figure 2), the CBR of the four regions also presented an inverse-economic distribution pattern since 1949. However, the northeastern region displayed a both decreased economic activity and decreased fertility over this time period since 2014.Figure 1. Changes of CBR of China (1949–2020) and GDP of China (1952–2020). Source: the National Bureau of Statistics of China. Abbreviations: CBR=crude birth rate, GDP=gross domestic product.Figure 2. Changes of CBR of 4 regions in China (1949–2019) and GDP of 4 regions in China (1952–2020). Source: the National Bureau of Statistics of China. Abbreviations: CBR=crude birth rate, GDP=gross domestic product.
Table 1 shows the results of the association of fertility with economic activity and education from the regression analysis. Based on the fixed-effects model, it was found that after controlling for multiple covariates, a significant positive impact of GDP on CBR was found (coefficient=0.045, P<0.001). In addition, we found a negative impact of PEDU on CBR (coefficient=–0.945, P<0.001), but a significant positive impact of HEDU on CBR (coefficient=0.085, P=0.004). The Hausman test was conducted and the results suggested using the fixed-effects model. The multi-collinearity of the 10 variables in the regression was also tested and found that the variance inflation factors of them were all less than 10 (Supplementary Table S1), meaning that there was no multi-collinearity among the variables in the model.
Variables Coef Std. Err t P 95% CI GDP 0.045 0.006 6.94 <0.001 0.032 to 0.057 PEDU −0.945 0.253 −3.74 <0.001 −1.442 to −0.449 HEDU 0.085 0.029 2.91 0.004 0.027 to 0.142 Sex ratio −0.037 0.015 −2.49 0.013 −0.066 to −0.008 Old-age dependency ratio −0.083 0.030 −2.81 0.005 −0.141 to −0.025 Labor proportion 0.387 3.411 0.11 0.910 −6.315 to 7.089 Average household size 0.795 0.310 2.56 0.011 0.185 to 1.405 Urban population density −0.013 0.006 −2.32 0.021 −0.024 to −0.002 Urbanization rate 0.016 0.013 1.26 0.207 −0.009 to 0.042 Urban registered
0.175 0.124 1.41 0.160 −0.069 to 0.419 Abbreviations: PLADs=provincial-level administrative divisions, CBR=crude birth rate, GDP=gross domestic product, PEDU=schooling years per capita, HEDU=proportion of higher education population, Coef=regression coefficient, Std. Err=standard error, CI=confidence interval.
Table 1. The association between economy and education with CBR: results of fixed-effects model (n=533) in China, 2002–2019.
|①||The National Bureau of Statistics of China divides 31 PLADs in mainland China into 4 major regions: the eastern, central, western, and northeastern regions. The eastern region includes Beijing, Tianjin, Hebei, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan; the central region includes Shanxi, Anhui, Jiangxi, Henan, Hubei, and Hunan; the western region includes Neimenggu, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Xizang, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang; the northeastern region includes Liaoning, Jilin, and Heilongjiang.|