Page 111 - IJPS-11-5

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

consumption levels. In addition, Tian & Huang (2024) Survey at Peking University, China. The CFPS surveys were
found that pension insurance reduces income uncertainty, conducted in 2012, 2014, 2016, and 2018, and the 2010
encouraging households, particularly in rural areas, to survey was excluded due to missing variable information.
diversify their asset holdings. To address the issue of missing data, we employed multiple
Research has also explored pension insurance’s role imputations, a method that allows for the creation of several
in shaping housing wealth and asset allocation. Ertuğrul plausible datasets based on observed values, ensuring
& Gebeşoğlu (2020) found that Turkey’s private pension more accurate and unbiased estimates. Winsorization was
system significantly contributed to domestic savings and conducted on continuous variables, adjusting them to the
st
th
economic growth. Unnikrishnan & Imai (2020) showed that 1 and 99 percentiles. Economic variables were adjusted
India’s pension scheme enhanced consumption and asset for price using the national Consumer Price Index with
accumulation, particularly among women. However, Madeira 2012 as the base year. In regression analysis, the logarithm
(2022) & Kang et al. (2022) noted that large-scale pension of adjusted quantities was used for smoother results.
withdrawals, such as those during the COVID-19 pandemic A balanced panel dataset of 35,872 observations from
in Chile and South Korea, temporarily boosted consumption 8,968 households across the 4-year period formed the basis
but ultimately led to long-term reductions in savings. for analysis and findings.
In this investigation, we aimed to better understand how 2.2. Variables
endowment systems influence household consumption The explanatory variables related to endowment insurance
based on the participation and payment stages in the context included a participation dummy variable, a variable
of China’s pension landscape. The primary contribution of denoting the type of endowment insurance, and the
this article lies in three key areas. One of the most innovative amount of pension income received. Control variables
aspects is its comparative approach to understanding the were incorporated to account for a range of demographic
impact of pension insurance on household consumption. and socioeconomic characteristics. At the individual
The research distinguishes between different stages of level, these characteristics include age, total income,
pension insurance—participation and payment—and gender, household registration status, marital status,
analyzes their effects on consumption at each stage. By education level, health status, and employment status.
examining how these two aspects influence household Household-level variables encompassed the number of
behavior, the study offers crucial insights into how pension household members, demographic composition, financial
systems can be structured to optimize consumption assets, property values, liabilities, and expenditures on
patterns. The second contribution lies in its macroeconomic
analysis of the heterogeneous effects of pension insurance commercial insurance. In addition, a dummy variable was
policies. Rather than assuming a uniform impact across employed to indicate the presence of risky financial assets.
households, the research provides a nuanced understanding A comprehensive overview of these variables is provided
of how pension insurance influences various consumption in Table 1.
categories—food, clothing, housing, healthcare, and more. 2.3. Model
This detailed breakdown allows for a deeper exploration of
the consumption dynamics within different segments of We use a two-way fixed effects model with panel data to
society. The third aspect focuses on the equity implications study the effect of endowment insurance on household
of pension insurance in promoting inclusive economic consumption behavior. The Hausman test indicated
development and social welfare. By examining how that the fixed effect model was more appropriate for our
pension insurance affects household consumption across data, as it suggested a significant correlation between
different income and age groups, this study highlights the the individual effects and the explanatory variables. The
role of social security policies in addressing inequalities. function is specified as follows:
Taken together, the findings from this study are crucial for ln(consumptio ) = α + β insured + β ln(payment )
optimizing pension systems to reduce disparities and for i,t 1 i,t 2 i,t (I)
+ X γ + λ + δ + ε
ensuring that benefits are distributed in ways that foster i,t t t i,t
broader economic stability and welfare. Where ln(consumptio ) represents household
i,t
consumption (log), using total and disaggregated
2. Methods variables in the regressions, respectively; insured i,t
represents insured status, using 0 – 1 dummy variables for
2.1. Data participation or not and multicategory dummy variables
In this study, we utilized data from the China Family Panel for type of endowment insurance in the regressions;
Studies (CFPS) conducted by the Institute of Social Science ln(payment ) represents pension (log); X includes a series
i,t
i,t
Volume 11 Issue 5 (2025) 105 https://doi.org/10.36922/ijps.4857