Page 117 - IJPS-11-5

International Journal of
Population Studies Endowment insurance and family consumption in China

influence of endowment insurance on younger individuals many are exclusively involved in each type versus those
while suggesting potential downsides for older recipients. who might have multiple types of coverage. Of the 35872
households in the sample, 14225 participate in endowment
3.5. Robustness check insurance, with the majority engaged in a single type.

3.5.1. The impact of participating in different types The results shown in Table 10 reveal that pensions from
of endowment insurance on household consumption previous institutions do not have a statistically significant
levels
effect on household consumption. In contrast, basic
Table 9 illustrates the distribution of participants across endowment insurance for urban employees is associated
various types of endowment insurance, showing how with an 11.1% increase in consumption, indicating
a strong positive impact. Similarly, supplementary
Table 6. Regression results of consumption of households enterprise endowment insurance leads to a 6.7% increase
with different household registration in consumption, underscoring its role in enhancing
household spending.
Independent variable Models
(1) (2) Conversely, commercial endowment insurance appears
Urban household Rural household to have a slight negative effect, although this result
registration registration is not statistically significant, suggesting that further
Dependent variable: Household investigation may be necessary to understand this trend
consumption more comprehensively. Old rural endowment insurance,
Participation in 0.116*** (4.891) 0.127*** (10.835) new rural social pension insurance, and urban and
endowment insurance rural residents endowment insurance are all positively
Pension (log) 0.002 (0.731) 0.008** (2.438) and statistically significantly associated with increased
Constant 9.893*** (94.387) 9.763*** (190.600) household consumption, with coefficients of 0.141, 0.106,
Control variables Join Join and 0.088, respectively (p < 0.01).
Household fixed effects Yes Yes 3.5.2. Breakpoint regression
Year fixed effects Yes Yes The breakpoint regression analysis examines the impact
Fixed effects F test 2.980 (p=0.000) 2.620 (p=0.000) of pension receipt on household consumption, with
Model significance test 46.350 (p=0.0000) 86.750 (p=0.0000) particular attention to retirement age eligibility in China.
Within R² 0.162 0.101 Specifically, men are eligible for pensions at age 60, while
Sample size 7224 23008 women qualify at ages 50 and 55. In rural areas, individuals
Notes: *p<0.1; **p<0.05; ***p<0.01. become eligible for pensions at age 60. To ensure

Table 7. Regression results of household consumption in different development regions
Independent variable Models
(1) (2) (3) (4)
Eastern Region Central Region Western Region Northeast Region
Dependent variable: Household consumption
Participation in endowment insurance 0.104*** (5.790) 0.183*** (9.502) 0.107*** (6.030) 0.079*** (3.222)
Pension (log) 0.004 (1.252) 0.008** (1.976) 0.003 (0.774) 0.005 (1.555)
Constant 9.841*** (113.538) 9.781*** (108.041) 9.730*** (144.461) 10.037*** (86.392)
Control variables Join Join Join Join
Household fixed effects Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes
Fixed effects F test 3.080 (p=0.000) 2.780 (p=0.000) 2.330 (p=0.000) 3.040 (p=0.000)
Model significance test 52.220 (p=0.000) 53.560 (p=0.000) 35.750 (p=0.000) 15.420 (p=0.000)
Within R² 0.127 0.154 0.097 0.075
Sample size 11328 8872 9944 5728
Notes: *p<0.1; **p<0.05; ***p<0.01.

Volume 11 Issue 5 (2025) 111 https://doi.org/10.36922/ijps.4857