The Impact of Multi-level Pension Insurance Participation on the Upgrading of Household Consumption Structure: Evidence from China

The Impact of Multi-level Pension Insurance Participation on the Upgrading of Household Consumption Structure: Evidence from China | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Impact of Multi-level Pension Insurance Participation on the Upgrading of Household Consumption Structure: Evidence from China Zhiqi Zhang, Xingyu Xu, Huidan Xu, Qiye Meng This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9194470/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract During China’s economic transition from high-speed growth to high-quality development, stimulating domestic demand and optimizing the consumption structure have emerged as critical economic priorities. Drawing on the life-cycle hypothesis, social comparison theory, and Maslow’s hierarchy of needs, this study employs panel data from the China Household Finance Survey (CHFS) to investigate the impact of multi-level pension insurance participation on the upgrading of household consumption structure. The findings reveal that participation in multi-level pension insurance significantly facilitates a shift in household consumption structure from subsistence-oriented to development- and enjoyment-oriented categories. Pension insurance enhances household consumption confidence both by increasing stable absolute expected future income and by improving households’ relative economic standing within their communities. Mechanism analysis further indicates that pension insurance promotes consumption upgrading by alleviating residents’ sense of relative income deprivation. This paper provides empirical evidence supporting the refinement of the multi-level pension insurance system in China, and offers policy insights for mitigating residents’ financial insecurity to foster sustainable consumption upgrading. Multi-level Pension Insurance Household Consumption Structure Upgrading Chinese Households 1. Introduction In recent years, China has advanced consumption transformation alongside economic recovery, and the upgrading of household consumption has become the core focus of this study. At present, China’s per capita GDP exceeds 13,000 U.S. dollars, and household consumption has shifted from a survival-oriented model to a development- and enjoyment-oriented one. Consumption is gradually shifting from commodity consumption-led to service consumption-led (Chen & Guo, 2023 ), but future income and old age uncertainty still inhibit households’ willingness to upgrade their consumption (Hong, Wang,& Tian. 2022). Pension insurance as social security can reduce residents’ uncertainty about future income and other factors, increasing current consumption and thus supporting consumption upgrading (Wang & He, 2024 ). According to the National Bureau of Statistics of China, as of 2025, 15.9% of the population was aged 65 and above, and China has entered a deeply aged society by international standards. Population aging poses a challenge to economic development, while the silver economy presents an opportunity for consumption growth (Wei & Chen, 2024 ). Pension insurance can reduce pressure of old age and promote economic growth at the same time (Diamond,1965). China has established a “three-level” pension insurance system jointly developed by governments, employers, and individuals: the first level is government-led basic pension insurance, the second is employer- and market-led supplementary pension insurance, and the third is voluntary individual pension savings and commercial pension insurance. Existing studies generally confirm that pension insurance exerts a significant positive effect on household consumption(Huang, 2021 ;Cao et al.,2022). However, existing studies on the relationship between pension insurance and household consumption mainly focus on either its impact on a single type of consumption(Yang,Cai, Zhang,& Li,2025) or its effect on the overall scale of household consumption. There are studies on multi-level pension insurance and superimposed effect on household consumption(Cai, 2025 ;Harenberg et al.,2015), but few studies focus on the relationship between pension insurance and consumption upgrading. Consumption transformation and consumption upgrading are prerequisites for China's high-quality development (Zhang, & Wang, 2025), focusing on multi-level pension insurance on household consumption structure can theoretically lead to reasonable policies to boost consumption. Multi-level pension insurance and superimposed effect on household consumption needs to be studied further. Additionally, relevant foreign studies offer cross-regional empirical support for the consumption-enhancing effects of pension insurance. At the level of developing countries, Cuong's (2013) research utilizing Vietnamese data demonstrates that both old-age insurance and social subsidies significantly increase household food and non-food consumption, with the consumption-boosting effect of old-age insurance being more pronounced, while also effectively reducing the poverty rate. At the level of developed countries in East Asia, Kang et al.2022) found that Korea's Basic Old Age Pension (BOAP) significantly influences the total consumption expenditure of impoverished and disadvantaged groups, underscoring the importance of consumption protection for vulnerable elderly individuals. In Europe and the United States, Larsen et al.'s ( 2025 ) examination of the Danish mandatory pension system reveals a dual effect: 85% of workers can sustain 90% of their pre-retirement consumption levels through the system; however, mandatory contributions may reduce disposable consumption during the working years and limit non-retirement savings, illustrating a combination of "work-period crowding out" and "retirement protection." This situation presents a policy trade-off between "crowding out at work" and "protection at retirement." These cross-country studies affirm the universal impact of pension insurance on household consumption, providing critical references and a framework for this paper to investigate the role of pension insurance in enhancing household consumption and consumption upgrading through the mechanisms of alleviating future uncertainty and facilitating social comparison within the Chinese context. According to the life-cycle hypothesis (Modigliani & Brumberg, 1954 ), pension insurance can alleviate residents’ pressure around post-retirement life, provide stable pension income, and reduce uncertainty about future income, thereby boosting consumer confidence and current consumption (Feldstein, 1974 ;Zhang et al.,2025). Maslow’s hierarchy of needs demonstrates that once basic survival needs are secured by pension insurance, individuals will be more willing to upgrade their consumption and allocate more resources to higher-level needs for development and self-actualization. This is reflected in consumption as an increase in the proportion of expenditure on development and enjoyment(Maslow, 1943 ). Given the social nature of humans (Agarwal, Qian, & Zou,2021), household consumption behavior is often profoundly influenced by the communities to which households belong. Building on the operating mechanism of pension insurance, this paper introduces the social comparison theory (Hogg, 2000 ;Stutzer, 2004 ), which posits that pension insurance can alleviate residents’social comparison pressure by narrowing the income gap among residents within the same community. This, in turn, boosts their consumption confidence and enables them to spend more on meeting needs other than basic survival. Panel data from 2015, 2017 and 2019 China Household Finance Surveys (CHFS) are used to analyze the effect of multi-level pension insurance and its superposition on household consumption structure by building the analysis of “protection portfolio-consumption structure” and further exploring mechanisms and heterogeneity of the effect. The main contributions of this study are as follows: First, based on data from the China Household Finance Survey, We integrated the life cycle hypothesis, social comparison theory and Maslow's hierarchy of needs theory to construct a theoretical framework of "pension insurance portfolio - household income differences - consumption structure". we explore the upgrading of Chinese households’ consumption structure from the perspective of multi-level pension insurance coordination, providing theoretical support for the formulation and optimization of pension insurance policies. Second, we analyze the regional and urban-rural heterogeneity of the effects of pension insurance, and provide constructive suggestions for China’s balanced development and common prosperity. Finally, we integrate the social comparison theory and construct a mediation effect model, using the inter-household income gap as a mediator to explore the intrinsic mechanism of pension insurance’s impact on household consumption structure from a psychological perspective. 2. Theoretical hypotheses 2.1 Impact of single-level pension insurance on household consumption structure As a core part of the social security system, pension insurance can provide residents with stable long-term expectations of future income and reduce disincentives to consumption caused by old-age uncertainty. According to the life cycle hypothesis, rational consumers will smooth their consumption over their lifetime based on expected lifetime income. By providing a stable stream of future income, pension insurance alleviates residents’ concerns about old age and increases their confidence to raise current consumption. Existing studies show that pension insurance has a boosting effect on household consumption. (Zheng, & Zhong,2016)found that the expansion of social security coverage significantly increased the consumption level of rural households. (Zheng et al.,2022) demonstrated that the individual pension system exerts a positive influence on the consumption structure optimization of urban households. These studies affirm the positive role of pension insurance in stimulating household consumption. Based on Maslow's hierarchy of needs theory, when an individual's survival needs are basically guaranteed, the center of gravity of consumption will gradually shift to the needs for self-development and enjoyment. Pension insurance provides basic protection for old age and ensures the bottom line of residents' basic survival needs. As a result, families are able to allocate more resources in their consumption budgets from survival necessities such as food and clothing to education, health, culture and entertainment that can improve human capital and quality of life, thus promoting the upgrading of the consumption structure from survival to development and enjoyment. Accordingly, this paper proposes research hypothesis H1: Participation in any single level of pension insurance can significantly promote the upgrading of consumption structure. 2.2 Impact of multi-level pension insurance on consumption structure China's three-level pension insurance system, each complementing the other, may have interactions between different levels of pension insurance. In terms of their institutional positioning, the attributes and functions of each level of China's multi-level system are distinct. The first level basic pension insurance and the second level enterprise annuities and occupational pensions are institutionalized, supply-led forms of public protection. In contrast, the third level, commercial pension insurance is a market-oriented, demand-led form of protection. This type of protection is highly autonomous and citizens can select it according to their needs. The superposition of multi-level protection may have two effects: one is the enhancement effect between the first and the third level of the pension insurance. Since there exists an association between the two types of insurance where the first level provides basic protection and the third level offers supplementary coverage, they should complement each other in their roles, synergizing and enhancing each other to produce the effect “1 + 1>2”. The second effect is the substitution effect or role which is independent of each other, between the first and the second level of the pension insurance. Since these two types of pension insurance are similar in purpose and function, when they are combined, the diminishing marginal effect of their roles may lead to either a substitution effect or a situation where the two effects are independent of each other, such that their combined effect is merely a simple summation of the two. Accordingly, this paper puts forward the research hypothesis H2: There are synergistic, substitution or independent effects of the impact of multi-level pension insurance on the upgrading of household consumption structure. Specifically, the superposition of the first level and the third level of pension insurance will produce a significant synergistic enhancement effect; the superposition of the first level and the second level of pension insurance will produce a substitution effect or the two roles are independent of each other. 3. Model setting and data 3.1 Data sources and sampling This study uses panel data from the China Household Finance Survey (CHFS) 2015, 2017, and 2019. The CHFS is a large-scale national micro-survey organized and implemented by the SURVEY AND RESEARCH CENTER FOR CHINA HOUSEHOLD FINANCE, which adopts a scientific sampling design covering the whole country, except Tibet, Xinjiang, Hong Kong, Macao and Taiwan, and covers 29 provinces (autonomous regions and municipalities directly under the central government), with national and provincial representativeness. The survey systematically collected detailed information on household assets, liabilities, income, consumption, social security participation and other dimensions, providing a high-quality data base for studying the relationship between pension insurance and household consumption structure. In order to construct a data sample suitable for this study, this paper carried out the following processes. First, considering the primary role of the head of household in household decision making (Bernard et al.,2020), this paper excluded all non-head of household samples. Second, this paper vertically merged the household head data samples to construct an unbalanced panel dataset in a long format. For sample screening, this paper removed all samples with missing key variables to ensure that the observations that end up in the model are complete and consistent. After processing, the valid samples cover households in the eastern, central, and western regions of the country, resulting in a dataset containing 55,857 households with a total of 96,765 observations. 3.2 Definition of variables 3.2.1 Dependent Variable: Household Consumption Structure With reference to Maslow's hierarchy of needs theory, this paper has categorized total household consumption expenditure into three types: survival, development and enjoyment. Survival consumption includes expenditures to satisfy basic physiological and safety needs such as food, clothing, housing, water, electricity, and fuel; developmental consumption includes expenditures to improve human capital and health such as education, training, and health care; and enjoyment consumption includes expenditures to improve the quality of life and to satisfy physical and mental pleasures such as culture and recreation, tourism, and beauty care. The dependent variable in this study is the household consumption structure, defined as the share of total consumption expenditures accounted for by the sum of developmental and enjoyment consumption expenditures. The higher the ratio, the more optimized the household consumption structure and the deeper the degree of consumption upgrading. 3.2.2 Core explanatory variables: multi-level pension insurance According to the framework for the construction of China's pension insurance system, China's pension insurance system includes the first level of government-led basic pension insurance, the second level of unit-supplemented enterprise annuities and occupational pensions, and the third level of individual voluntary commercial pension insurance. Since the CHFS dataset does not include direct measures of commercial pension insurance, this study uses commercial life insurance as a proxy for commercial pension insurance. Therefore, this paper constructs three dummy variables for whether residents participate in the first level of basic pension insurance, the second level of enterprise annuities and occupational pensions, and the third level of commercial life insurance. If participation in a certain level of pension insurance takes the value of 1 for that level, and 0 otherwise. 3.2.3 Control variables To control for other factors that may affect the structure of household consumption, this paper includes total household income (taken in logarithmic form), household size (number of household members). age and its squared term, gender, education, work status, and type of hukou (rural or urban hukou) as control variables in the model. 3.2.4 Descriptive statistics Table 1 shows the descriptive statistics of the main variables in this paper. The data show that the mean value of the dependent variable, the proportion of household development and enjoyment consumption, is 0.401, indicating that a considerable portion of the expenditure of the sample households has been used for higher-level needs such as education and recreation, and that there is a great deal of room for consumption upgrading, and its standard deviation is 0.253, indicating that there are significant differences in consumption upgrading among different households. The participation in multi-level pension insurance is in line with the reality that "one level is dominant, while the development of the second and third levels is weak" in China. Table 1 Descriptive statistics of variables Variable Symbol Measurement Obs Mean SD Min Max Explained variable con_it Share of development and enjoyment-oriented consumption 96765 0.401 0.253 0 1 Core explanatory variables ins_1st 1 = participated; 0 = did not participate 96765 0.816 0.387 0 1 ins_2nd 1 = participated; 0 = did not participate 96765 0.027 0.162 0 1 ins_3rd 1 = participated; 0 = did not participate 96765 0.045 0.206 0 1 Control variables ln_finc ln_total income 96765 10.681 1.430 -2 16 hk 1 = urban; 0 = rural 96765 0.524 0.499 0 1 gender 1 = male; 0 = female 96765 0.787 0.409 0 1 age Age (years) 96765 55.553 13.845 18 117 age2 Age squared 96765 3277.790 1549.067 324 13689 edu 1 = below bachelor’s degree; 2 = bachelor’s degree or above 96765 1.076 0.264 1 2 work 1 = employed; 0 = not employed 96765 0.637 0.481 0 1 size Household size (persons) 96765 3.177 1.555 1 15 3.3 Empirical models 3.3.1 Benchmark regression model First, this paper uses a two-way fixed-effects model to estimate the independent effects of each level of pension insurance, with models (1)-(4) set up as follows: $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_1st+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(1\right)\end{array}$$ $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_2nd+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(2\right)\end{array}$$ $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_3rd+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(3\right)\end{array}$$ $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_1st+{\beta\:}_{2}ins\_2nd+{\beta\:}_{3}ins\_3rd+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(4\right)\end{array}$$ In the above model, Con it is a dependent Variable indicating the proportion of development and enjoyment consumption. ins_1st, ins_2nd, and ins_3rd indicate whether or not one participates in the first, second, and third level of pension insurance, respectively. X it is the set of control variables mentioned above. µ i is an individual fixed effect, and θ t is a time fixed effect. ε it is a random error term. After determining the effect of a single level of pension insurance, in order to test whether there is a synergistic or substitutive superposition effect between different levels of pension insurance, this paper adds the interaction term of each level of insurance in the benchmark model. It should be noted that China's pension insurance system is based on basic pension insurance, and employees employed in enterprises and public institutions participating in the second level of pension insurance industry are also bound to participate in the first level, so there is no such thing as only participating in the second level and not the first level. This is also reflected in the data, the sample size of only participating in the second level without participating in the basic pension is zero. Models (5)-(7) are set up as follows: $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_1st+{\beta\:}_{2}ins\_2nd+{\delta\:}_{1}(ins\_1st\times\:ins\_2nd)+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(5\right)\end{array}$$ $$\:\begin{array}{c}Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_1st+{\beta\:}_{2}ins\_3rd+{\delta\:}_{1}(ins\_1st\times\:ins\_3rd)+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(6\right)\end{array}$$ $$\:Co{n}_{it}=\alpha\:+{\beta\:}_{1}ins\_1st+{\beta\:}_{2}ins\_2nd+{\beta\:}_{3}ins\_3rd+{\delta\:}_{1}(ins\_1st\times\:ins\_2nd)+{\delta\:}_{2}(ins\_1st\times\:ins\_3rd)$$ $$\:\begin{array}{c}+{\delta\:}_{3}(ins\_1st\times\:ins\_2nd\times\:ins\_3rd)+\lambda\:{X}_{it}+{\mu\:}_{i}+{\theta\:}_{t}+{\epsilon\:}_{it} \left(7\right)\end{array}$$ In the above model, Ins_1st×Ins_2nd denotes the interaction effect of the first level pension insurance with the second level pension insurance, Ins_1st×Ins_3rd denotes the interaction effect of the first level pension insurance with the third level pension insurance, and Ins_1st×Ins_2nd×Ins_3rd denotes the interaction effect of the first level pension insurance, the second level pension insurance and the third level pension insurance, and other variables are consistent with the previous section. If 'δ > 0' and significant, it indicates that there is a synergistic effect between the two types of protection, and the effect of joint participation is greater than the sum of the effects of separate participation; otherwise, it indicates that there is a substitution effect. 3.3.2 Endogenous treatment: instrumental variables approach To address potential endogeneity issues caused by reverse causality and omitted variables between pension insurance participation and household consumption structure, this study uses the instrumental variable (IV) method combined with two-stage least squares (2SLS) for more rigorous causal identification. Drawing on the research idea of community cohort effect(Angrist, 2014 ), this paper chooses "the participation rate of community (village/residence) in the same level of pension insurance" as the instrumental variable. This instrumental variable belongs to the macro-level community characteristics, which only affects the consumption structure indirectly by influencing the participation decision of households, which meets the requirement of exogeneity; meanwhile, the community participation rate is highly correlated with the participation behavior of households, which meets the requirement of correlation(Zhang,& Zhang,2023). Since CHFS does not provide direct data on community participation rates, this paper uses the ratio of the sample size of a level of participation in a surveyed community to the sample size of the surveyed community as the estimation of the participation rate, and excludes communities with a sample size of less than 5 to avoid the interference of extreme values. 4. Empirical results 4.1 Benchmark regression results 4.1.1 Impact of single-level pension insurance on household consumption structure Table 2 presents the results of the benchmark regression on the impact of single level pension insurance on household consumption structure. Columns (1) through (3) examine the independent impact of the first, second, and third levels of pension insurance, respectively. The results show that participation in any level of pension insurance significantly increases the share of household development and enjoyment consumption, and the strength of the impact increases with the level of pension insurance participation. Column (4) confirms the independence of the promotion effect of each level of protection by including all three in the model at the same time, and the coefficients remain significant. Column (5) further incorporates all control variables and the core findings remain robust. Among the control variables, the positive effect of household size on consumption structure may be mainly reflected in the increase of developmental expenditures such as education. The above results provide support for research hypothesis H1. Table 2 Single-level pension insurance and household consumption structure ins_1st (1) Model 1 (2) Model 2 (3) Model 3 (4) Model 4 (5) Model 5 0.0158*** (3.8434) 0.0157*** (3.8065) 0.0173*** (4.1852) ins_2nd 0.0189** (2.0057) 0.0179* (1.9036) 0.0176* (1.8895) ins_3rd 0.0394*** (5.6110) 0.0393*** (5.6109) 0.0396*** (5.6528) ln_finc 0.0067*** (4.6345) hk 0.0113** (2.4828) gender -0.0021 (-0.4102) age -0.0110*** (-8.8298) age 2 0.0001*** (8.3916) edu 0.0108 (0.9553) work 0.0024 (0.6211) size 0.0242*** (13.2000) Constant 0.3144*** (84.3822) 0.3264*** (176.9837) 0.3250*** (174.0005) 0.3123*** (83.4194) 0.4405*** (11.3037) Time FE Yes Yes Yes Yes Yes Individual FE Yes Yes Yes Yes Yes Observations 96765 96765 96765 96765 96765 R 2 0.052 0.052 0.052 0.053 0.058 Adjusted R 2 0.052 0.052 0.052 0.053 0.058 Notes: t statistics in parentheses, * p < 0.10, ** p < 0.05, *** p < 0.01 4.1.2 Overlapping effect of multi-level pension insurance on household consumption structure Table 3 shows the test results of the superimposed effect of multi level pension insurance. The results show that the coefficient of the interaction term ins_1st×_3rd is significantly positive (0.0368, p < 0.05), indicating that there is a significant synergistic effect between the two, specifically, the households that participate in basic pension insurance and commercial life insurance at the same time have a consumption structure upgrade that is significantly larger than the sum of the independent effects of the two types of insurance. The consumption upgrading effect of commercial pension insurance is enhanced when basic pension insurance is owned. In contrast, after controlling for this interaction term, the independent effect of the third-level insurance becomes statistically insignificant. This suggests that the effect of the third level pension insurance in promoting consumption upgrading depends on the first level basic pension insurance.This also elucidates the fundamental reason for the sluggish development of China's third level: a segment of the population has not yet achieved comprehensive coverage under the first level and therefore lacks both the foundation and the incentive to invest in commercial pensions. For the interaction term ins_1st×_2nd between the first and second levels, no valid regression coefficient is generated. Since there is no sample that only participates in the second level without the first level, the interaction term is completely collinear with the main effect and automatically omitted by the model, indicating that the effects of the first and second levels on consumption structure upgrading are independent of each other, with no synergistic or substitution effects. This arises from the significant overlap in their institutional attributes and the weak complementarity of their functions. Additionally, the second level serves as an extension of the first level, ensuring that the first level’s basic protection role remains intact. Consequently, both levels can be developed concurrently. In addition, the coefficient of the triple interaction is also insignificant, suggesting that having all three types of coverage at the same time did not generate additional nonlinear gains. These findings partially support research hypothesis H2 and highlight the complementarity that exists between basic and market supplemental protection, providing empirical evidence to optimize the structural design of the multilevel pension insurance system. Table 3 Multi-level pension insurance superposition and household consumption structure upgrading ins_1st (1) Model 6 (2) Model 7 (3) Model 8 0.0174*** (4.1940) 0.0161*** (3.8361) 0.0159*** (3.7902) ins_2nd 0.0177* (1.8975) 0.0183* (1.8877) ins_3rd 0.0091 (0.5815) 0.0092 (0.5890) ins_1st_2nd 0.0000 (.) 0.0000 (.) ins_1st_3rd 0.0368** (2.1905) 0.0371** (2.1978) ins_1st_2nd_3rd -0.0074 (-0.2922) ln_finc 0.0068*** (4.6916) 0.0067*** (4.6358) 0.0067*** (4.6399) hk 0.0111** (2.4442) 0.0113** (2.4793) 0.0113** (2.4823) gender -0.0023 (-0.4668) -0.0020 (-0.3961) -0.0021 (-0.4216) age -0.0109*** (-8.7944) -0.0109*** (-8.8023) -0.0110*** (-8.8234) age2 0.0001*** (8.3252) 0.0001*** (8.3628) 0.0001*** (8.3915) edu 0.0108 (0.9566) 0.0106 (0.9293) 0.0105 (0.9242) work 0.0025 (0.6372) 0.0025 (0.6497) 0.0025 (0.6284) size 0.0241*** (13.1451) 0.0242*** (13.2009) 0.0242*** (13.1998) Constant 0.4419*** (11.3408) 0.4413*** (11.3150) 0.4416*** (11.3266) Time FE Yes Yes Yes Individual FE Yes Yes Yes Observations 96765 96765 96765 R 2 0.058 0.058 0.059 4.2 Robustness Tests In order to verify the robustness of the benchmark regression results, this study examines the robustness of the benchmark regression results in turn by three methods: shrinking the tails, excluding the samples over 80 years of age, and replacing the core explanatory variables with the core explanatory variables lagged by one period. The results are shown in Table 4 . After shrinking the tails at the 1% level for continuous variables (Column 1) and excluding samples where the head of household is over 80 years old (Column 2), the promotional effect of pension insurance at all levels remains significant, consistent with the previous section. However, the results of the lagged one-period model (Column 3) show a change in the direction and significance of the effects of the core explanatory variables. This signal suggests that there may be a problem of reverse causality or endogeneity due to omitted variables between pension insurance participation and household consumption structure, which may not be fully overcome by the baseline fixed effects model. Therefore, a more rigorous causal identification using an instrumental variables approach is presented below. Table 4 Robustness test results ins_1st (1) shrinking the tails (2) excluding elderly samples (3) Lagging period 0.0157*** (3.8294) 0.0170*** (4.0676) ins_2nd 0.0178* (1.8968) 0.0160* (1.6971) ins_3rd 0.0394*** (5.6279) 0.0389*** (5.5353) ins_1st_lag -0.0154** (-2.5738) ins_2nd_lag 0.0111 (0.8289) ins_3rd_lag -0.0142 (-1.4295) Constant 0.3114*** (83.4812) 0.3160*** (83.1211) 0.4167*** (85.1091) Time FE Yes Yes Yes Individual FE Yes Yes Yes Observations 96765 93580 40899 R 2 0.054 0.050 0.037 4.3 Endogeneity test This paper first tests the endogeneity of the core explanatory variables through the Durbin-Wu-Hausman (DWH) test, and the results are shown in Table 5 . The results show that the p-value of the first level of pension insurance is 0.24, which cannot reject the original hypothesis of "no endogeneity" at the 5% significance level; The p-values of the second level and the third level of pension insurance are 4.88888916260e-07 and 3.52073428047e-19, respectively, which strongly reject the original hypothesis at the 1% level. This indicates that there is a significant endogeneity problem in the participation of the second and third level of pension insurance, therefore, this study used instrumental variable method to correct it. Table 5 Results of endogeneity test (DWH test) DWH χ² (1) First level (2) Second level (3) Third level 1.367072479151179 25.31060874224525 80.15515017288125 p-value 0.2423 4.89e-07 3.52e-19 Result no significant endogeneity endogeneity exists endogeneity exists The results of the first-stage regression of instrumental variables are shown in Table 6 , which shows that the community participation rate at each level is highly significantly and positively correlated with the participation behavior of households at the corresponding level, and the F-statistic of the first-stage regression is much larger than the empirical critical value of 10, which suggests that there is no problem of weak instrumental variables, and that the selected instrumental variables are valid. Table 6 Results of the first stage regression of the instrumental variables approach rate_ins_1st (1) first order IV coefficient (2) second order IV coefficient (3) third order IV coefficient 0.890*** (0.010) rate_ins_2nd 0.858*** (0.027) rate_ins_3rd 0.899*** (0.023) Constant -1.094*** (0.022) -0.181*** (0.010) -0.117*** (0.011) N 96730 96730 96730 R 2 0.175 0.115 0.053 Adjusted R 2 0.175 0.115 0.053 F 1936.326 314.476 334.439 Table 7 reports the results of the two-stage least squares (2SLS) estimation based on the instrumental variables approach and compares them with the benchmark OLS results. The results show that the coefficient of the first level pension insurance becomes insignificant in the 2SLS estimation, which corroborates with the conclusion that it is not significantly endogenous in the DWH test, suggesting that the estimation of the effect of the benchmark fixed-effects model for this level is basically plausible. The impacts of the second and third levels of pension insurance are shown to be more significantly positive, which suggests that the OLS model seriously underestimates the real contribution of the second and third levels of pension insurance to the upgrading of household consumption structure. Further comparison of the strength of the impacts reveals that the promotion effect of the third level of commercial pension insurance is significantly larger than that of the second level of enterprise annuities, which implies that market-oriented and personalized supplementary protection may have a stronger marginal role in unleashing the potential of consumption and optimizing the consumption structure. Table 7 Comparison of instrumental variable method regression results with the baseline regression variables OLS 2SLS ins_1st 0.012*** (0.002) 0.009 (0.007) ins_2nd 0.050*** (0.004) 0.116*** (0.020) ins_3rd 0.056*** (0.004) 0.232*** (0.022) ln_finc 0.015*** (0.001) 0.012*** (0.001) hk 0.001 (0.002) -0.002 (0.002) gender -0.008*** (-0.002) -0.007*** (0.002) age -0.003*** (0.000) -0.004*** (0.000) age2 0.0000*** (0.000) 0.000*** (0.000) edu 0.061*** (0.003) 0.051*** (0.004) work 0.004* (0.002) 0.002 (0.002) size 0.020*** (0.001) 0.021*** (0.001) Constant 0.166*** (0.013) 0.195*** (0.014) N 96730 96730 R 2 0.089 0.067 4.4 Mechanism Test In order to test the channels of action proposed in the theoretical analysis, based on the relative income hypothesis in the social comparison perspective, this study introduces the mediator variable of household income disparity, and this paper uses the relative deprivation index of (Kakwani, 1984 ) as a measure of this variable, which is used to measure the degree of inequality of household incomes at the micro level, and which is able to reflect the relative income position of a single household within the community and its perceived relative deprivation. For a community Y containing n households, the Kakwani index is defined for the ith household by ranking household incomes in ascending order as (y1,y2,...,yn): $$\:\begin{array}{c}Household\:income\:inequality({y}_{i},{y}_{j})=\frac{1}{n{\mu\:}_{y}}\sum\:_{j=i+1}^{n}({y}_{j}-{y}_{i})={\gamma\:}_{yi}^{+}\left(\frac{{\mu\:}_{yi}^{+}-{y}_{i}}{{\mu\:}_{Y}}\right) (8)\end{array}$$ where µ y is the average income of all households in that community, µ + yi is the average income of households in community Y with incomes greater than y i , and γ + yi is the proportion of households in community Y with incomes greater than y i , expressed as a percentage of the total sample size. The theoretical logic of introducing this mechanism variable is that the participation in multi level pension insurance may alleviate the relative income deprivation felt by low-income households in the community by boosting their absolute and expected incomes. According to the Social Comparison Theory, when households' perceived income inequality decreases, their certainty about their future lives increases, which makes them more likely to allocate resources to development and enjoyment consumption, promoting the upgrading of consumption structure. The test results are shown in Table 8 . Column (1) of Table 8 shows the total effect of each level of pension insurance on the consumption structure of households when no mediating variable is included, and column (2) shows the effect of each level of pension insurance on the mediating variable. The results show that participation in all levels of pension insurance significantly reduces the relative deprivation index of households. This suggests that participation in pension insurance effectively improves the relative economic status of households within the community by boosting their actual and expected incomes, alleviating the sense of income deprivation resulting from social comparison. Column (3) incorporates both core explanatory variables and mediating variables. The results show that the relative deprivation index has a significant negative impact on the upgrading of consumption structure, i.e., the higher the household's perceived income inequality, the lower the share of its consumption that is developmental and enjoyment-oriented. More importantly, the coefficients of the first, second, and third levels of pension insurance all decrease compared to Column (1) after controlling for the relative deprivation index. Combined with the significant path relationships in Columns (2) and (3), this suggests that alleviating relative income deprivation is an effective mediating channel for pension insurance to promote the upgrading of household consumption structure. While providing economic security, old-age security also releases consumption potential for higher-level needs by alleviating households' social comparison pressure and sense of psychological inequality. In summary, the results of the mechanism test confirm the research hypothesis, and further calculation of the proportion of mediating effects shows that the effects realized by the first, second and third levels of pension insurance through the path of alleviating relative deprivation account for 7.7%, 9.6% and 5.4% of the total effect, respectively.The second level exhibits the highest proportion of mediation effect, demonstrating that enterprise or occupational pensions significantly enhance the relative income status of households and mitigate relative deprivation. In contrast, the third level shows the lowest proportion of mediation effect, primarily due to the fact that most participants are high-income households, which experience a diminished sense of relative deprivation. Consequently, the benefits derived from this level are more closely associated with absolute income increases. This finding validates that multi level pension insurance not only affects consumption through the absolute income path but also through the indirect path of alleviating relative income deprivation, which provides new mechanistic evidence for understanding the micro-socio-economic benefits of pension insurance from a psychological perspective. To further exclude the potential estimation bias of the stepwise regression method and verify the robustness of this mediation mechanism, we conduct an additional robustness test based on the Bootstrap method. To further verify the robustness of the above mediation effect results and avoid the potential estimation bias of the stepwise regression method, this paper adopts the bias-corrected non-parametric percentile Bootstrap method with 1000 repeated samples to re-test the significance of the mediation effect. The test results are shown in Table 9 . The results show that the 95% bias-corrected (BC) and percentile (P) confidence intervals of the mediation effects of the first, second and third levels of pension insurance do not contain 0, which means that the mediation effect of alleviating relative income deprivation is statistically significant at the 5% level for all three levels of pension insurance. This result is completely consistent with the conclusion of the stepwise regression test above, which further confirms the robustness of the mediation mechanism. It is fully verified that multi-level pension insurance can promote the upgrading of household consumption structure by alleviating households' relative income deprivation and reducing social comparison pressure, which provides more rigorous empirical evidence for the theoretical mechanism of this paper. Table 8 Mechanism test of the impact of pension insurance on the structure of consumption through the relative deprivation index ins_1st (1) total effect model (2) mediated variable model (3) complete model 0.013*** (0.002) -0.003*** (0.001) 0.012*** (0.002) ins_2nd 0.052*** (0.004) -0.037*** (0.001) 0.047*** (0.004) ins_3rd 0.056*** (0.004) -0.021*** (0.001) 0.053*** (0.004) ln_finc 0.012*** (0.001) -0.148*** (0.001) -0.006*** (0.001) hk 0.021*** (0.002) -0.048*** (0.001) 0.015*** (0.002) gender -0.014*** (0.002) -0.000 (0.001) -0.014*** (0.002) age -0.003*** (0.000) 0.002*** (0.000) -0.003*** (0.000) age2 0.000*** (0.000) -0.000*** (0.000) 0.000*** (0.000) edu 0.059*** (0.003) -0.050*** (0.001) 0.053*** (0.003) work 0.014*** (0.002) 0.005*** (0.001) 0.014*** (0.002) size 0.020*** (0.001) -0.010*** (0.000) 0.019*** (0.001) Household income inequality -0.121*** (0.008) Constant 0.248*** (0.013) 2.201*** (0.012) 0.513*** (0.022) N 96765 96765 96765 R 2 0.056 0.847 0.058 Adjusted R 2 0.056 0.847 0.058 Table 9 Robustness test of mediation effect based on Bootstrap method ind1 Observed coefficient Bias Std. err. 95% BC Confidence Interval 95% P Confidence Interval 0.01308344 -5.41e-06 0.00042739 0.0121497 0.0138153 0.0122199 0.0138772 ind2 0.03856864 -3.66e-06 0.00096563 0.0366919 0.0404712 0.0366697 0.0404289 ind3 0.02116122 − .0000491 0.00073643 0.0198347 0.0226631 0.0196525 0.0225509 Notes: BC represents Bias-corrected confidence interval; P represents Percentile confidence interval. The number of repeated sampling for Bootstrap is 1000. At the outset of the mechanism test design in this study, the classic economic mechanism—the precautionary savings mechanism—was incorporated. A mediating effect analysis was conducted using a standardized process aligned with the core mechanism test. However, the empirical results indicate that this traditional transmission mechanism did not achieve significance in the sample analyzed. Consequently, the proposed path of action is not supported. This finding highlights the applicability boundaries of classical consumption theory within the specific context of China, The underlying cause can be traced to the entrenched traditional savings concept of "preparing for the future" and the high savings habits that have developed among Chinese residents, which exhibit strong path dependence. Household savings motives encompass various life cycle risks, including retirement, medical care, children's education, and housing. Participation in endowment insurance alone addresses only certain uncertainties related to retirement and is insufficient to fundamentally alter residents' savings decisions (Zhang et al.,2025). 4.5 Heterogeneity Test With significant differences in development levels between urban and rural areas and geographic regions in China, and the differences in budget constraints and social security levels faced by households across regions or urban and rural areas (Gao et al.,2018), there may be group heterogeneity in the impact of pension insurance on consumption structure. In order to deeply reveal the boundary conditions of its effect, this paper further conducts group tests from the regional and urban-rural dimensions. In the regional dimension, the sample is divided into eastern, central and western regions based on the three major zones division method. The eastern region includes Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan; the central region includes Heilongjiang, Jilin, Shanxi, Anhui, Jiangxi, Henan, Hubei and Hunan; and the western region includes Sichuan, Chongqing, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, Guangxi and Inner Mongolia. In the urban-rural dimension, the sample is divided into urban and rural households, aiming to reveal possible systematic differences in the impact of pension insurance under the dual social security structure. 4.5.1 Analysis of regional heterogeneity Table 10 reports the regression results by East, Central and West regions. The analysis shows that the first level of basic pension insurance exhibits significant positive effects in all three regions, which reflects the general effectiveness of the basic system in China. It is very interesting that the significance of the second level and the third level pension insurance in the east, center and west regions is complementary. Specifically, the effect of the second level pension insurance is significantly positive only in the central region (0.0489, p < 0.05), while the third level of commercial life insurance plays a significant role in the east and west. The central provinces have developed industry and manufacturing, enterprises are relatively concentrated, the system of enterprise annuity covers a wide range of enterprises, and its consumption release effect is therefore prominent. In contrast, the economy of the eastern region is extremely developed, and the role of enterprise occupational pension in upgrading household consumption in the eastern region is not obvious. Table 10 Results of the analysis of regional heterogeneity ins_1st (1) East region (2) Central region (3) West region 0.0143** (2.2313) 0.0171** (2.2379) 0.0168** (2.2421) ins_2nd 0.0131 (0.9935) 0.0489** (2.3602) 0.0021 (0.1220) ins_3rd 0.0441*** (4.4299) 0.0202 (1.5401) 0.0502*** (3.4055) Constant 0.3226*** (55.2375) 0.3061*** (44.2544) 0.3005*** (43.7939) Time FE Yes Yes Yes Individual FE Yes Yes Yes Observations 44936 26514 25315 R 2 0.034 0.064 0.080 4.5.2 Analysis of urban-rural heterogeneity Table 11 shows the regression results by grouping urban and rural households. It is found that the first level pension insurance has a significantly stronger effect on the promotion of consumption upgrading for urban households than for rural households, directly reflecting the gap between urban and rural basic security levels. Whereas the second level pension insurance has a strong positive impact on the consumption structure of rural households, the impact on urban households is insignificant. The likely reason for this is that since the number of urban households participating in the second level pension insurance (2,445) is much larger than the number of rural households (150), the few rural domiciles with access to enterprise annuities have a large marginal improvement effect on consumption from the jump in long-term income security. The third level pension insurance shows a significant boost in both urban and rural areas. Table 11 Results of the analysis of urban-rural heterogeneity ins_1st (1) Rural households (2) Urban households 0.0137** (0.006) 0.0258*** (0.008) ins_2nd 0.1078*** (0.039) 0.0127 (0.010) ins_3rd 0.0335*** (2.5975) 0.0412*** (0.009) Constant 0.2945*** (0.005) 0.3313*** (0.007) Time FE Yes Yes Individual FE Yes Yes Observations 46033 50732 R 2 0.064 0.032 5. Conclusion, Policy Recommendations, and Research Limitations 5.1 Conclusions of the study Based on the life-cycle hypothesis and Maslow’s hierarchy of needs, this study constructs a theoretical framework to analyze the impact of multi-level pension insurance on household consumption structure and empirically tests it using panel data from the China Household Finance Survey (CHFS) for the years 2015, 2017, and 2019. The empirical results show that participation in any single level of pension insurance significantly improves household consumption structure, and that pension insurance not only expands the scale of consumption but also promotes consumption upgrading. A notable synergistic enhancement effect is observed when the first-level basic pension insurance and the third-level commercial life insurance are combined: households participating in both types experience a greater degree of consumption structure upgrading than the simple sum of the effects of participating in each separately. Moreover, the role of the third-level pension insurance in shaping consumption structure is contingent upon the presence of the first-level basic pension insurance. In contrast, the interaction between the first and second levels is not statistically significant; their effects operate independently. This is largely attributable to the functional similarity and overlapping objectives of the first- and second-level schemes, which diminish their complementary potential. Heterogeneity analysis reveals that the impact of pension insurance on consumption structure varies significantly across groups. Regionally, the second-level pension insurance plays a more prominent role in central China, while the third-level commercial insurance exhibits stronger effects in the eastern and western regions. In terms of urban–rural differences, the first-level insurance has a more pronounced effect on urban households, reflecting disparities in the level of basic protection between urban and rural areas. Notably, the second-level pension insurance shows an unusually strong marginal effect for rural households, suggesting that rural participants derive greater incremental security from such insurance compared to their urban counterparts, thereby significantly boosting their consumption confidence. Mechanism analysis further indicates that pension insurance enhances consumption confidence and facilitates consumption upgrading by alleviating households’ sense of relative income deprivation. These findings provide empirical support for optimizing the multi-level pension system and offer policy insights for strengthening social security as a lever to unlock consumption potential and promote high-quality economic development. 5.2 Policy recommendations Based on the findings above, this paper offers the following policy recommendations to strengthen the multi-level pension insurance system and promote the upgrading of household consumption: First, reinforce the foundational role of the first level to stabilize consumption expectations. It is essential to advance universal pension coverage and establish a dynamic mechanism for adjusting pension benefits. Although basic pension insurance has wide coverage, it has not yet achieved full universal coverage. Enhancing national pooling and linking benefit adjustments to economic growth and price changes can help secure residents’ basic living standards and underpin consumption confidence. Second, enhance the complementary role of the second and third levels to unlock consumption upgrading potential. For the second level, mechanisms such as automatic enrollment and tax deferrals should be introduced to expand coverage, particularly among small and medium-sized enterprises and employees in emerging industries. For the third level, policy efforts should promote the development of commercial pension insurance to meet diverse needs, while strengthening public awareness to encourage voluntary participation. These measures can reinforce the supplementary function of pensions and bolster household consumption capacity. Third, adopt targeted policy approaches tailored to regional and urban–rural disparities. In eastern regions, policy should focus on facilitating wealth planning and consumption among high-income groups through commercial insurance. In western and rural areas, a dual strategy combining improved pension coverage with employment support is recommended. Offering tax incentives to enterprises that provide supplementary pensions for rural workers could enhance their economic security and enable a shift toward consumption in health, education, and recreation. 5.3 Limitations This study has several limitations that also suggest directions for future research. First, in the context of rapid technological advancement and sustainable economic development, subsequent studies could explore the impact of pension insurance on emerging consumption areas such as digital consumption, green consumption, and health-related services, thereby providing theoretical insights for guiding the development of these sectors. Second, future research could employ longer-term panel data to examine the dynamic evolution and life-cycle patterns of how pension insurance influences household consumption structure. Third, further investigation into the linkages between multi-level pension insurance and household financial behaviors—such as asset allocation and risk preference—would contribute to a more comprehensive understanding of the micro-level mechanisms through which pension insurance promotes consumption upgrading. Finally, due to the absence of direct measures of commercial pension insurance in the China Household Finance Survey (CHFS) data, this study uses commercial life insurance as a proxy, which may introduce estimation bias. Commercial life insurance encompasses a range of products—including whole life, term life, and endowment insurance—only some of which serve a clear old-age security function. Data limitations prevent a full disentanglement of the heterogeneous effects of these insurance types. As a result, the estimated effect of third-level pension insurance may be biased: if life insurance products without pension functions have a weaker effect on consumption upgrading than dedicated commercial pension insurance, the coefficient may be underestimated; conversely, it may be overestimated. Accordingly, the estimated coefficients for third-level pension insurance in this study should be interpreted as the average effect of broadly defined market-oriented pension protection products. Declarations Funding statement This study was supported by the Scientific Research Interest Training Program of Sichuan Agricultural University. Declaration of conflicting interest The author(s) declared no potential conflicts of interest in this study, paper writing, and publication. Author Contribution Zhiqi Zhang :Conceptualization,Data curation,Methodology,Project administration,Writing – original draft,Writing – review & editing.Xingyu Xu:Data curation,Software,Validation,Writing – review & editing.Huidan Xu: Methodology,Resources,Supervision,Validation,Writing – review & editing.Qiye Meng:Data curation Data Availability The data used in this study were obtained from the China Household Finance Survey (CHFS) database, available at https://chfs.swufe.edu.cn/. References Agarwal, S., Qian, WL., & Zou, X. (2021). Thy neighbor's misfortune: Peer effect on consumption[J]. American Economic Journal: Economic Policy, 13(2): 1–25. https://doi.org/10.1257/pol.20170634 Angrist, J. D. (2014). The perils of peer effects[J]. Labour Economics, 30: 98–108. https://doi.org/10.1016/j.labeco.2014.05.008 Kakwani, N. (1984). The relative deprivation curve and its applications[J]. Journal of Business & Economic Statistics, 3(2): 171–173. Maslow, A. H. (1943). A theory of human motivation[J]. Psychological Review, 50: 370–396. https://doi.org/10.1037/h0054346 Cao, XB., Wang, F., Wang, Y., Wang, YX. (2022). An investigation of whether pensions increase consumption: Evidence from family portfolios[J]. Finance Research Letters, 47: 102591. https://doi.org/10.1016/j.frl.2021.102591 Sun, JY., Huang, JC., Jiang, XJ., Song, XL., Zhang, N. (2025). Spatiotemporal evolution and influencing factors of population aging in the Triangle of Central China at multiple scales[J]. Sustainability, 17(14), 6549. https://doi.org/10.3390/su17146549 Wei, YX., Chen, J. (2024). Sustainable design for the silver society: Developing the silver model for gerontechnology product innovation[J]. Sustainability, 17(1): 42. https://doi.org/10.3390/su17010042 Chen, DJ., Guo, XT. (2023). The impact of the digital economy on residents’ consumption upgrading—Evidence from Mainland China[J]. Sustainability, 15(10), 8041. https://doi.org/10.3390/su15108041 Hong MY., Wang J., Tian MJ. (2022). Rural social security, precautionary savings, and the upgrading of rural residents' consumption structure in China[J]. Sustainability 2022, 14(19), 12455; https://doi.org/10.3390/su141912455 Modigliani, F., & Ando, A.K. (1957). Tests of the life cycle hypothesis of savings: comments and suggestions[J]. Bulletin of the Oxford University Institute of Economics & Statistics, 19(2): 99–124. Stutzer, A.(2004).The role of income aspirations in individual happiness[J].JOURNAL OF ECONOMIC BEHAVIOR & ORGANIZATION.54: 89–109. https://doi.org/10.1016/j.jebo.2003.04.003 Zhao, MQ., Zhang, MX. (2025). Marketization and household consumption upgrading: Evidence from China[J]. Sustainability 2025, 17(21), 9373; https://doi.org/10.3390/su17219373 Huang, W., & Zhang, CC. (2021). The power of social pensions: Evidence from China's new rural pension scheme[J]. American Economic Journal: Applied Economics, 13(2),179–205. https://doi.org/10.1257/app.20170789 Larsen, L. S.,Nielsson, U.,Nutu, M.,&Rangvid, J.(2025). Mandatory pension contributions, household consumption, and savings[J]. Journal of Pension Economics & Finance, https://doi.org/10.1017/S147474722510005X Kang, JY., Park, S.,& Ahn, S Ahn, S(2022). The effect of social pension on consumption among older adults in Korea[J].Journal of the Economics of Ageing, https://doi.org/10.1016/j.jeoa.2021.100364 Cuong, NV. (2013). The Impact of Social Security on Household Welfare: Evidence from a Transition Country[J].European Journal of Development Research, 25(5), 737–757. https://doi.org/10.1016/j.asieco.2013.09.002 Wang, Y. Z., & He, X. Y. (2024). Does purchasing commercial insurance promote household consumption? Evidence from China[J]. International Review of Economics & Finance, 96: 103641. https://doi.org/10.1016/j.iref.2024.103641 Harenberg, D., & Ludwig, A. (2015). Social security in an analytically tractable overlapping generations model with aggregate and idiosyncratic risks[J]. International Tax and Public Finance, 22(4): 579–603. https://doi.org/10.1007/s10797-015-9368-x Yang, QN., Cai, YX., Zhang, PL., & Li, R. (2025). Set rainy days aside: How pension insurance affects household tourism consumption[J]. Journal of Hospitality and Tourism Management, 63, 3–14. https://doi.org/10.1016/j.jhtm.2025.03.001 Feldstein, M. (1974). Social Security, Induced Retirement, and Aggregate Capital Accumulation[J]. Journal of Political Economy, 82(5), 905–926. https://doi.org/10.1086/260246 Diamond, P. A. (1965). National Debt in a Neoclassical Growth Model[J]. American Economic Review, 55(5), 1126–1150. Zheng, HM., & Zhong, T. (2016). The impacts of social pension on rural household expenditure: evidence from China[J]. Journal of Economic Policy Reform, 19(3), 221–237. https://doi.org/10.1080/17487870.2015.1041524 Zheng, W., Lyu, Y., Jia, R.,& Hanewald, K. (2022). The impact of expected pensions on consumption: Evidence from China[J]. Journal of Pension Economics and Finance, 22(1), 69–87. https://doi.org/10.1017/S147474722100041X Bernard, T., Doss, C. R., Hidrobo, M., Hoel, J. B., & Kieran, C. (2020). Ask me why: Patterns of intrahousehold decision-making[J]. World Development, 125, 104671. https://doi.org/10.1016/j.worlddev.2019.104671 Gao, Q., Yang, S., Zhang, Y., & Li, S. (2018). The Divided Chinese Welfare System: Do Health and Education Change the Picture?[J]. Social Policy & Society, 17(2): 227–244. https://doi.org/10.1017/S1474746417000100 Cai Y. (2025).The pension insurance system, multi-level superimposed participation and the scale of household consumption[J].East China Economic Management, 39(9). https://doi.org/10.19629/j.cnki.34-1014/f.250401017 (in Chinese) Hogg M A. (2000). Social Identity and Social Comparison[A]//Suls J, Wheeler L. Handbook of Social Comparison[M]. Boston, MA: Springer https://doi.org/10.1007/978-1-4615-4237-7_19 Modigliani F,& Brumberg R. (1954). Utility analysis and the consumption function: An interpretation of cross-section data[A]//Kurihara K K. Post-Keynesian Economics[M]. New Brunswick, NJ: Rutgers University Press Zhang, Y., C Zhang, Q.X. (2023). Income Disparity, Consumption Patterns, and Trends of International Consumption Center City Construction, Based on a Test of China’s Consumer Market[J]. Sustainability, 15(4): 2862. https://doi.org/10.3390/su15042862 Zhang,LG.,Brooks,R.,Ding,D.,Ding,HY.,He,H.,Lu,J.,&Rui C.M.,(2025).China's high savings: Drivers, prospects, and policy implications[J]. 69.https://doi.org/10.1016/j.ememar.2025.101355 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 09 Apr, 2026 Reviewers invited by journal 08 Apr, 2026 Editor assigned by journal 28 Mar, 2026 Submission checks completed at journal 28 Mar, 2026 First submitted to journal 22 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9194470","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":620198727,"identity":"cc81eba4-6742-453a-ab2c-42e49e95a9fa","order_by":0,"name":"Zhiqi Zhang","email":"","orcid":"","institution":"Sichuan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Zhiqi","middleName":"","lastName":"Zhang","suffix":""},{"id":620198728,"identity":"b91abcb8-4e6c-4acb-9c7c-1190dfe080c8","order_by":1,"name":"Xingyu Xu","email":"","orcid":"","institution":"Sichuan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Xingyu","middleName":"","lastName":"Xu","suffix":""},{"id":620198729,"identity":"e38f841c-1d89-45e2-9e0f-febc657b2b48","order_by":2,"name":"Huidan Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIiWNgGAWjYBACNvb2gw8SKv7x2B9vbAQyaghr4eM5k2zw4MwBOYYzh5uBjGOEtchJJJhJPmw7YMxwI71N8mELMxEOk0hINkhsu5PYOCOxrSKxgY2Bv707Ab8WnodAv5x7ltjM87DtRuIOGQaJM2c34NfCDrQloYw5sY09EajlDBuDgUQuAS0MCWYSCWzMiT0MiW0FiW3MRGjhAGlpO2wswZHYxkCcFlAgJ5xJkzPgOdgskXDmGA9Bv8i3tx98+KPChseAvf3hxx8VNXL87b34tWAAHtKUj4JRMApGwSjACgApMFHm2zq2yQAAAABJRU5ErkJggg==","orcid":"","institution":"Sichuan Agricultural University","correspondingAuthor":true,"prefix":"","firstName":"Huidan","middleName":"","lastName":"Xu","suffix":""},{"id":620198730,"identity":"de216948-aa65-4e4a-9b26-51e33bfccb49","order_by":3,"name":"Qiye Meng","email":"","orcid":"","institution":"Sichuan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Qiye","middleName":"","lastName":"Meng","suffix":""}],"badges":[],"createdAt":"2026-03-23 01:53:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9194470/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9194470/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107480893,"identity":"094d8461-f0f4-4b43-be32-c3f256e37185","added_by":"auto","created_at":"2026-04-22 02:14:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1001012,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9194470/v1/1b306a16-878b-4a39-9fdb-c72fac0de107.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Impact of Multi-level Pension Insurance Participation on the Upgrading of Household Consumption Structure: Evidence from China","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eIn recent years, China has advanced consumption transformation alongside economic recovery, and the upgrading of household consumption has become the core focus of this study. At present, China\u0026rsquo;s per capita GDP exceeds 13,000 U.S. dollars, and household consumption has shifted from a survival-oriented model to a development- and enjoyment-oriented one. Consumption is gradually shifting from commodity consumption-led to service consumption-led (Chen \u0026amp; Guo, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), but future income and old age uncertainty still inhibit households\u0026rsquo; willingness to upgrade their consumption (Hong, Wang,\u0026amp; Tian. 2022). Pension insurance as social security can reduce residents\u0026rsquo; uncertainty about future income and other factors, increasing current consumption and thus supporting consumption upgrading (Wang \u0026amp; He, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAccording to the National Bureau of Statistics of China, as of 2025, 15.9% of the population was aged 65 and above, and China has entered a deeply aged society by international standards. Population aging poses a challenge to economic development, while the silver economy presents an opportunity for consumption growth (Wei \u0026amp; Chen, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Pension insurance can reduce pressure of old age and promote economic growth at the same time (Diamond,1965). China has established a \u0026ldquo;three-level\u0026rdquo; pension insurance system jointly developed by governments, employers, and individuals: the first level is government-led basic pension insurance, the second is employer- and market-led supplementary pension insurance, and the third is voluntary individual pension savings and commercial pension insurance.\u003c/p\u003e \u003cp\u003eExisting studies generally confirm that pension insurance exerts a significant positive effect on household consumption(Huang, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e;Cao et al.,2022). However, existing studies on the relationship between pension insurance and household consumption mainly focus on either its impact on a single type of consumption(Yang,Cai, Zhang,\u0026amp; Li,2025) or its effect on the overall scale of household consumption. There are studies on multi-level pension insurance and superimposed effect on household consumption(Cai, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e;Harenberg et al.,2015), but few studies focus on the relationship between pension insurance and consumption upgrading. Consumption transformation and consumption upgrading are prerequisites for China's high-quality development (Zhang, \u0026amp; Wang, 2025), focusing on multi-level pension insurance on household consumption structure can theoretically lead to reasonable policies to boost consumption. Multi-level pension insurance and superimposed effect on household consumption needs to be studied further.\u003c/p\u003e \u003cp\u003eAdditionally, relevant foreign studies offer cross-regional empirical support for the consumption-enhancing effects of pension insurance. At the level of developing countries, Cuong's (2013) research utilizing Vietnamese data demonstrates that both old-age insurance and social subsidies significantly increase household food and non-food consumption, with the consumption-boosting effect of old-age insurance being more pronounced, while also effectively reducing the poverty rate. At the level of developed countries in East Asia, Kang et al.2022) found that Korea's Basic Old Age Pension (BOAP) significantly influences the total consumption expenditure of impoverished and disadvantaged groups, underscoring the importance of consumption protection for vulnerable elderly individuals. In Europe and the United States, Larsen et al.'s (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) examination of the Danish mandatory pension system reveals a dual effect: 85% of workers can sustain 90% of their pre-retirement consumption levels through the system; however, mandatory contributions may reduce disposable consumption during the working years and limit non-retirement savings, illustrating a combination of \"work-period crowding out\" and \"retirement protection.\" This situation presents a policy trade-off between \"crowding out at work\" and \"protection at retirement.\" These cross-country studies affirm the universal impact of pension insurance on household consumption, providing critical references and a framework for this paper to investigate the role of pension insurance in enhancing household consumption and consumption upgrading through the mechanisms of alleviating future uncertainty and facilitating social comparison within the Chinese context.\u003c/p\u003e \u003cp\u003eAccording to the life-cycle hypothesis (Modigliani \u0026amp; Brumberg, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1954\u003c/span\u003e), pension insurance can alleviate residents\u0026rsquo; pressure around post-retirement life, provide stable pension income, and reduce uncertainty about future income, thereby boosting consumer confidence and current consumption (Feldstein, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1974\u003c/span\u003e;Zhang et al.,2025). Maslow\u0026rsquo;s hierarchy of needs demonstrates that once basic survival needs are secured by pension insurance, individuals will be more willing to upgrade their consumption and allocate more resources to higher-level needs for development and self-actualization. This is reflected in consumption as an increase in the proportion of expenditure on development and enjoyment(Maslow, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1943\u003c/span\u003e). Given the social nature of humans (Agarwal, Qian, \u0026amp; Zou,2021), household consumption behavior is often profoundly influenced by the communities to which households belong. Building on the operating mechanism of pension insurance, this paper introduces the social comparison theory (Hogg, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2000\u003c/span\u003e;Stutzer, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), which posits that pension insurance can alleviate residents\u0026rsquo;social comparison pressure by narrowing the income gap among residents within the same community. This, in turn, boosts their consumption confidence and enables them to spend more on meeting needs other than basic survival.\u003c/p\u003e \u003cp\u003ePanel data from 2015, 2017 and 2019 China Household Finance Surveys (CHFS) are used to analyze the effect of multi-level pension insurance and its superposition on household consumption structure by building the analysis of \u0026ldquo;protection portfolio-consumption structure\u0026rdquo; and further exploring mechanisms and heterogeneity of the effect. The main contributions of this study are as follows: First, based on data from the China Household Finance Survey, We integrated the life cycle hypothesis, social comparison theory and Maslow's hierarchy of needs theory to construct a theoretical framework of \"pension insurance portfolio - household income differences - consumption structure\". we explore the upgrading of Chinese households\u0026rsquo; consumption structure from the perspective of multi-level pension insurance coordination, providing theoretical support for the formulation and optimization of pension insurance policies. Second, we analyze the regional and urban-rural heterogeneity of the effects of pension insurance, and provide constructive suggestions for China\u0026rsquo;s balanced development and common prosperity. Finally, we integrate the social comparison theory and construct a mediation effect model, using the inter-household income gap as a mediator to explore the intrinsic mechanism of pension insurance\u0026rsquo;s impact on household consumption structure from a psychological perspective.\u003c/p\u003e"},{"header":"2. Theoretical hypotheses","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Impact of single-level pension insurance on household consumption structure\u003c/h2\u003e \u003cp\u003eAs a core part of the social security system, pension insurance can provide residents with stable long-term expectations of future income and reduce disincentives to consumption caused by old-age uncertainty. According to the life cycle hypothesis, rational consumers will smooth their consumption over their lifetime based on expected lifetime income. By providing a stable stream of future income, pension insurance alleviates residents\u0026rsquo; concerns about old age and increases their confidence to raise current consumption. Existing studies show that pension insurance has a boosting effect on household consumption. (Zheng, \u0026amp; Zhong,2016)found that the expansion of social security coverage significantly increased the consumption level of rural households. (Zheng et al.,2022) demonstrated that the individual pension system exerts a positive influence on the consumption structure optimization of urban households. These studies affirm the positive role of pension insurance in stimulating household consumption.\u003c/p\u003e \u003cp\u003eBased on Maslow's hierarchy of needs theory, when an individual's survival needs are basically guaranteed, the center of gravity of consumption will gradually shift to the needs for self-development and enjoyment. Pension insurance provides basic protection for old age and ensures the bottom line of residents' basic survival needs. As a result, families are able to allocate more resources in their consumption budgets from survival necessities such as food and clothing to education, health, culture and entertainment that can improve human capital and quality of life, thus promoting the upgrading of the consumption structure from survival to development and enjoyment.\u003c/p\u003e \u003cp\u003eAccordingly, this paper proposes research hypothesis H1: Participation in any single level of pension insurance can significantly promote the upgrading of consumption structure.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Impact of multi-level pension insurance on consumption structure\u003c/h2\u003e \u003cp\u003eChina's three-level pension insurance system, each complementing the other, may have interactions between different levels of pension insurance. In terms of their institutional positioning, the attributes and functions of each level of China's multi-level system are distinct. The first level basic pension insurance and the second level enterprise annuities and occupational pensions are institutionalized, supply-led forms of public protection. In contrast, the third level, commercial pension insurance is a market-oriented, demand-led form of protection. This type of protection is highly autonomous and citizens can select it according to their needs.\u003c/p\u003e \u003cp\u003eThe superposition of multi-level protection may have two effects: one is the enhancement effect between the first and the third level of the pension insurance. Since there exists an association between the two types of insurance where the first level provides basic protection and the third level offers supplementary coverage, they should complement each other in their roles, synergizing and enhancing each other to produce the effect \u0026ldquo;1\u0026thinsp;+\u0026thinsp;1\u0026gt;2\u0026rdquo;. The second effect is the substitution effect or role which is independent of each other, between the first and the second level of the pension insurance. Since these two types of pension insurance are similar in purpose and function, when they are combined, the diminishing marginal effect of their roles may lead to either a substitution effect or a situation where the two effects are independent of each other, such that their combined effect is merely a simple summation of the two.\u003c/p\u003e \u003cp\u003eAccordingly, this paper puts forward the research hypothesis H2: There are synergistic, substitution or independent effects of the impact of multi-level pension insurance on the upgrading of household consumption structure. Specifically, the superposition of the first level and the third level of pension insurance will produce a significant synergistic enhancement effect; the superposition of the first level and the second level of pension insurance will produce a substitution effect or the two roles are independent of each other.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Model setting and data","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n\u003ch2\u003e3.1 Data sources and sampling\u003c/h2\u003e\n\u003cp\u003eThis study uses panel data from the China Household Finance Survey (CHFS) 2015, 2017, and 2019. The CHFS is a large-scale national micro-survey organized and implemented by the SURVEY AND RESEARCH CENTER FOR CHINA HOUSEHOLD FINANCE, which adopts a scientific sampling design covering the whole country, except Tibet, Xinjiang, Hong Kong, Macao and Taiwan, and covers 29 provinces (autonomous regions and municipalities directly under the central government), with national and provincial representativeness. The survey systematically collected detailed information on household assets, liabilities, income, consumption, social security participation and other dimensions, providing a high-quality data base for studying the relationship between pension insurance and household consumption structure.\u003c/p\u003e\n\u003cp\u003eIn order to construct a data sample suitable for this study, this paper carried out the following processes. First, considering the primary role of the head of household in household decision making (Bernard et al.,2020), this paper excluded all non-head of household samples. Second, this paper vertically merged the household head data samples to construct an unbalanced panel dataset in a long format. For sample screening, this paper removed all samples with missing key variables to ensure that the observations that end up in the model are complete and consistent. After processing, the valid samples cover households in the eastern, central, and western regions of the country, resulting in a dataset containing 55,857 households with a total of 96,765 observations.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\n\u003ch2\u003e3.2 Definition of variables\u003c/h2\u003e\n\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\n\u003ch2\u003e3.2.1 Dependent Variable: Household Consumption Structure\u003c/h2\u003e\n\u003cp\u003eWith reference to Maslow's hierarchy of needs theory, this paper has categorized total household consumption expenditure into three types: survival, development and enjoyment. Survival consumption includes expenditures to satisfy basic physiological and safety needs such as food, clothing, housing, water, electricity, and fuel; developmental consumption includes expenditures to improve human capital and health such as education, training, and health care; and enjoyment consumption includes expenditures to improve the quality of life and to satisfy physical and mental pleasures such as culture and recreation, tourism, and beauty care. The dependent variable in this study is the household consumption structure, defined as the share of total consumption expenditures accounted for by the sum of developmental and enjoyment consumption expenditures. The higher the ratio, the more optimized the household consumption structure and the deeper the degree of consumption upgrading.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\n\u003ch2\u003e3.2.2 Core explanatory variables: multi-level pension insurance\u003c/h2\u003e\n\u003cp\u003eAccording to the framework for the construction of China's pension insurance system, China's pension insurance system includes the first level of government-led basic pension insurance, the second level of unit-supplemented enterprise annuities and occupational pensions, and the third level of individual voluntary commercial pension insurance. Since the CHFS dataset does not include direct measures of commercial pension insurance, this study uses commercial life insurance as a proxy for commercial pension insurance. Therefore, this paper constructs three dummy variables for whether residents participate in the first level of basic pension insurance, the second level of enterprise annuities and occupational pensions, and the third level of commercial life insurance. If participation in a certain level of pension insurance takes the value of 1 for that level, and 0 otherwise.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section3\"\u003e\n\u003ch2\u003e3.2.3 Control variables\u003c/h2\u003e\n\u003cp\u003eTo control for other factors that may affect the structure of household consumption, this paper includes total household income (taken in logarithmic form), household size (number of household members). age and its squared term, gender, education, work status, and type of hukou (rural or urban hukou) as control variables in the model.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\n\u003ch2\u003e3.2.4 Descriptive statistics\u003c/h2\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e shows the descriptive statistics of the main variables in this paper. The data show that the mean value of the dependent variable, the proportion of household development and enjoyment consumption, is 0.401, indicating that a considerable portion of the expenditure of the sample households has been used for higher-level needs such as education and recreation, and that there is a great deal of room for consumption upgrading, and its standard deviation is 0.253, indicating that there are significant differences in consumption upgrading among different households. The participation in multi-level pension insurance is in line with the reality that \"one level is dominant, while the development of the second and third levels is weak\" in China.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003ctable id=\"Tab1\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eDescriptive statistics of variables\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eVariable\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eSymbol\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMeasurement\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eObs\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMean\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eSD\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMin\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMax\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eExplained variable\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003econ_it\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eShare of development and enjoyment-oriented consumption\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.401\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.253\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"3\" align=\"left\"\u003e\n\u003cp\u003eCore explanatory variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eins_1st\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;participated; 0\u0026thinsp;=\u0026thinsp;did not participate\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.816\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.387\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eins_2nd\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;participated; 0\u0026thinsp;=\u0026thinsp;did not participate\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.027\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.162\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eins_3rd\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;participated; 0\u0026thinsp;=\u0026thinsp;did not participate\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.045\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.206\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"8\" align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eln_finc\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eln_total income\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e10.681\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1.430\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e16\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ehk\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;urban; 0\u0026thinsp;=\u0026thinsp;rural\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.524\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.499\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003egender\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;male; 0\u0026thinsp;=\u0026thinsp;female\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.787\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.409\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eage\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eAge (years)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e55.553\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e13.845\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e18\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e117\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eage2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eAge squared\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e3277.790\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1549.067\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e324\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e13689\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eedu\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;below bachelor\u0026rsquo;s degree; 2\u0026thinsp;=\u0026thinsp;bachelor\u0026rsquo;s degree or above\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1.076\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.264\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e2\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ework\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u0026thinsp;=\u0026thinsp;employed; 0\u0026thinsp;=\u0026thinsp;not employed\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.637\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e0.481\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003esize\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eHousehold size (persons)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e96765\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e3.177\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1.555\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e15\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n\u003ch2\u003e3.3 Empirical models\u003c/h2\u003e\n\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\n\u003ch2\u003e3.3.1 Benchmark regression model\u003c/h2\u003e\n\u003cp\u003eFirst, this paper uses a two-way fixed-effects model to estimate the independent effects of each level of pension insurance, with models (1)-(4) set up as follows:\u003c/p\u003e\n\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equa\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_1st+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(1\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equb\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_2nd+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(2\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equc\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_3rd+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(3\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equd\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_1st+{\\beta\\:}_{2}ins\\_2nd+{\\beta\\:}_{3}ins\\_3rd+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(4\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eIn the above model, Con\u003csub\u003eit\u003c/sub\u003e is a dependent Variable indicating the proportion of development and enjoyment consumption. ins_1st, ins_2nd, and ins_3rd indicate whether or not one participates in the first, second, and third level of pension insurance, respectively. X\u003csub\u003eit\u003c/sub\u003e is the set of control variables mentioned above. \u0026micro;\u003csub\u003ei\u003c/sub\u003e is an individual fixed effect, and \u0026theta;\u003csub\u003et\u003c/sub\u003e is a time fixed effect. \u0026epsilon;\u003csub\u003eit\u003c/sub\u003e is a random error term.\u003c/p\u003e\n\u003cp\u003eAfter determining the effect of a single level of pension insurance, in order to test whether there is a synergistic or substitutive superposition effect between different levels of pension insurance, this paper adds the interaction term of each level of insurance in the benchmark model. It should be noted that China's pension insurance system is based on basic pension insurance, and employees employed in enterprises and public institutions participating in the second level of pension insurance industry are also bound to participate in the first level, so there is no such thing as only participating in the second level and not the first level. This is also reflected in the data, the sample size of only participating in the second level without participating in the basic pension is zero. Models (5)-(7) are set up as follows:\u003c/p\u003e\n\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Eque\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_1st+{\\beta\\:}_{2}ins\\_2nd+{\\delta\\:}_{1}(ins\\_1st\\times\\:ins\\_2nd)+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(5\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equf\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equf\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_1st+{\\beta\\:}_{2}ins\\_3rd+{\\delta\\:}_{1}(ins\\_1st\\times\\:ins\\_3rd)+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(6\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equg\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equg\" class=\"mathdisplay\"\u003e$$\\:Co{n}_{it}=\\alpha\\:+{\\beta\\:}_{1}ins\\_1st+{\\beta\\:}_{2}ins\\_2nd+{\\beta\\:}_{3}ins\\_3rd+{\\delta\\:}_{1}(ins\\_1st\\times\\:ins\\_2nd)+{\\delta\\:}_{2}(ins\\_1st\\times\\:ins\\_3rd)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equh\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equh\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}+{\\delta\\:}_{3}(ins\\_1st\\times\\:ins\\_2nd\\times\\:ins\\_3rd)+\\lambda\\:{X}_{it}+{\\mu\\:}_{i}+{\\theta\\:}_{t}+{\\epsilon\\:}_{it} \\left(7\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eIn the above model, Ins_1st\u0026times;Ins_2nd denotes the interaction effect of the first level pension insurance with the second level pension insurance, Ins_1st\u0026times;Ins_3rd denotes the interaction effect of the first level pension insurance with the third level pension insurance, and Ins_1st\u0026times;Ins_2nd\u0026times;Ins_3rd denotes the interaction effect of the first level pension insurance, the second level pension insurance and the third level pension insurance, and other variables are consistent with the previous section. If '\u0026delta;\u0026thinsp;\u0026gt;\u0026thinsp;0' and significant, it indicates that there is a synergistic effect between the two types of protection, and the effect of joint participation is greater than the sum of the effects of separate participation; otherwise, it indicates that there is a substitution effect.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\n\u003ch2\u003e3.3.2 Endogenous treatment: instrumental variables approach\u003c/h2\u003e\n\u003cp\u003eTo address potential endogeneity issues caused by reverse causality and omitted variables between pension insurance participation and household consumption structure, this study uses the instrumental variable (IV) method combined with two-stage least squares (2SLS) for more rigorous causal identification.\u003c/p\u003e\n\u003cp\u003eDrawing on the research idea of community cohort effect(Angrist, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e), this paper chooses \"the participation rate of community (village/residence) in the same level of pension insurance\" as the instrumental variable. This instrumental variable belongs to the macro-level community characteristics, which only affects the consumption structure indirectly by influencing the participation decision of households, which meets the requirement of exogeneity; meanwhile, the community participation rate is highly correlated with the participation behavior of households, which meets the requirement of correlation(Zhang,\u0026amp; Zhang,2023). Since CHFS does not provide direct data on community participation rates, this paper uses the ratio of the sample size of a level of participation in a surveyed community to the sample size of the surveyed community as the estimation of the participation rate, and excludes communities with a sample size of less than 5 to avoid the interference of extreme values.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"4. Empirical results","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Benchmark regression results\u003c/h2\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e4.1.1 Impact of single-level pension insurance on household consumption structure\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the results of the benchmark regression on the impact of single level pension insurance on household consumption structure. Columns (1) through (3) examine the independent impact of the first, second, and third levels of pension insurance, respectively. The results show that participation in any level of pension insurance significantly increases the share of household development and enjoyment consumption, and the strength of the impact increases with the level of pension insurance participation. Column (4) confirms the independence of the promotion effect of each level of protection by including all three in the model at the same time, and the coefficients remain significant. Column (5) further incorporates all control variables and the core findings remain robust. Among the control variables, the positive effect of household size on consumption structure may be mainly reflected in the increase of developmental expenditures such as education. The above results provide support for research hypothesis H1.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSingle-level pension insurance and household consumption structure\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eModel 1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eModel 2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003eModel 3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003cp\u003eModel 4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(5)\u003c/p\u003e \u003cp\u003eModel 5\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0158***\u003c/p\u003e \u003cp\u003e(3.8434)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0157***\u003c/p\u003e \u003cp\u003e(3.8065)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0173***\u003c/p\u003e \u003cp\u003e(4.1852)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0189**\u003c/p\u003e \u003cp\u003e(2.0057)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0179*\u003c/p\u003e \u003cp\u003e(1.9036)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0176*\u003c/p\u003e \u003cp\u003e(1.8895)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0394***\u003c/p\u003e \u003cp\u003e(5.6110)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0393***\u003c/p\u003e \u003cp\u003e(5.6109)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0396***\u003c/p\u003e \u003cp\u003e(5.6528)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eln_finc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0067***\u003c/p\u003e \u003cp\u003e(4.6345)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehk\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0113**\u003c/p\u003e \u003cp\u003e(2.4828)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.0021\u003c/p\u003e \u003cp\u003e(-0.4102)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.0110***\u003c/p\u003e \u003cp\u003e(-8.8298)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0001***\u003c/p\u003e \u003cp\u003e(8.3916)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eedu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0108\u003c/p\u003e \u003cp\u003e(0.9553)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ework\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0024\u003c/p\u003e \u003cp\u003e(0.6211)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esize\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0242***\u003c/p\u003e \u003cp\u003e(13.2000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3144***\u003c/p\u003e \u003cp\u003e(84.3822)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3264***\u003c/p\u003e \u003cp\u003e(176.9837)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.3250***\u003c/p\u003e \u003cp\u003e(174.0005)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.3123***\u003c/p\u003e \u003cp\u003e(83.4194)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.4405***\u003c/p\u003e \u003cp\u003e(11.3037)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eNotes: t statistics in parentheses, \u003csup\u003e*\u003c/sup\u003e p\u0026thinsp;\u0026lt;\u0026thinsp;0.10, \u003csup\u003e**\u003c/sup\u003e p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u003csup\u003e***\u003c/sup\u003e p\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003e4.1.2 Overlapping effect of multi-level pension insurance on household consumption structure\u003c/h2\u003e \u003cp\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the test results of the superimposed effect of multi level pension insurance. The results show that the coefficient of the interaction term ins_1st\u0026times;_3rd is significantly positive (0.0368, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), indicating that there is a significant synergistic effect between the two, specifically, the households that participate in basic pension insurance and commercial life insurance at the same time have a consumption structure upgrade that is significantly larger than the sum of the independent effects of the two types of insurance. The consumption upgrading effect of commercial pension insurance is enhanced when basic pension insurance is owned. In contrast, after controlling for this interaction term, the independent effect of the third-level insurance becomes statistically insignificant. This suggests that the effect of the third level pension insurance in promoting consumption upgrading depends on the first level basic pension insurance.This also elucidates the fundamental reason for the sluggish development of China's third level: a segment of the population has not yet achieved comprehensive coverage under the first level and therefore lacks both the foundation and the incentive to invest in commercial pensions. For the interaction term ins_1st\u0026times;_2nd between the first and second levels, no valid regression coefficient is generated. Since there is no sample that only participates in the second level without the first level, the interaction term is completely collinear with the main effect and automatically omitted by the model, indicating that the effects of the first and second levels on consumption structure upgrading are independent of each other, with no synergistic or substitution effects. This arises from the significant overlap in their institutional attributes and the weak complementarity of their functions. Additionally, the second level serves as an extension of the first level, ensuring that the first level\u0026rsquo;s basic protection role remains intact. Consequently, both levels can be developed concurrently. In addition, the coefficient of the triple interaction is also insignificant, suggesting that having all three types of coverage at the same time did not generate additional nonlinear gains. These findings partially support research hypothesis H2 and highlight the complementarity that exists between basic and market supplemental protection, providing empirical evidence to optimize the structural design of the multilevel pension insurance system.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMulti-level pension insurance superposition and household consumption structure upgrading\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eModel 6\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eModel 7\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003eModel 8\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0174***\u003c/p\u003e \u003cp\u003e(4.1940)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0161***\u003c/p\u003e \u003cp\u003e(3.8361)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0159***\u003c/p\u003e \u003cp\u003e(3.7902)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0177*\u003c/p\u003e \u003cp\u003e(1.8975)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0183*\u003c/p\u003e \u003cp\u003e(1.8877)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0091\u003c/p\u003e \u003cp\u003e(0.5815)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0092\u003c/p\u003e \u003cp\u003e(0.5890)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_1st_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003cp\u003e(.)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003cp\u003e(.)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_1st_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0368**\u003c/p\u003e \u003cp\u003e(2.1905)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0371**\u003c/p\u003e \u003cp\u003e(2.1978)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_1st_2nd_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0074\u003c/p\u003e \u003cp\u003e(-0.2922)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eln_finc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0068***\u003c/p\u003e \u003cp\u003e(4.6916)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0067***\u003c/p\u003e \u003cp\u003e(4.6358)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0067***\u003c/p\u003e \u003cp\u003e(4.6399)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehk\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0111**\u003c/p\u003e \u003cp\u003e(2.4442)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0113**\u003c/p\u003e \u003cp\u003e(2.4793)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0113**\u003c/p\u003e \u003cp\u003e(2.4823)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0023\u003c/p\u003e \u003cp\u003e(-0.4668)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.0020\u003c/p\u003e \u003cp\u003e(-0.3961)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0021\u003c/p\u003e \u003cp\u003e(-0.4216)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0109***\u003c/p\u003e \u003cp\u003e(-8.7944)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.0109***\u003c/p\u003e \u003cp\u003e(-8.8023)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0110***\u003c/p\u003e \u003cp\u003e(-8.8234)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0001***\u003c/p\u003e \u003cp\u003e(8.3252)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0001***\u003c/p\u003e \u003cp\u003e(8.3628)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0001***\u003c/p\u003e \u003cp\u003e(8.3915)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eedu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0108\u003c/p\u003e \u003cp\u003e(0.9566)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0106\u003c/p\u003e \u003cp\u003e(0.9293)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0105\u003c/p\u003e \u003cp\u003e(0.9242)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ework\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0025\u003c/p\u003e \u003cp\u003e(0.6372)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0025\u003c/p\u003e \u003cp\u003e(0.6497)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0025\u003c/p\u003e \u003cp\u003e(0.6284)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esize\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0241***\u003c/p\u003e \u003cp\u003e(13.1451)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0242***\u003c/p\u003e \u003cp\u003e(13.2009)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0242***\u003c/p\u003e \u003cp\u003e(13.1998)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.4419***\u003c/p\u003e \u003cp\u003e(11.3408)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.4413***\u003c/p\u003e \u003cp\u003e(11.3150)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.4416***\u003c/p\u003e \u003cp\u003e(11.3266)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Robustness Tests\u003c/h2\u003e \u003cp\u003eIn order to verify the robustness of the benchmark regression results, this study examines the robustness of the benchmark regression results in turn by three methods: shrinking the tails, excluding the samples over 80 years of age, and replacing the core explanatory variables with the core explanatory variables lagged by one period. The results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. After shrinking the tails at the 1% level for continuous variables (Column 1) and excluding samples where the head of household is over 80 years old (Column 2), the promotional effect of pension insurance at all levels remains significant, consistent with the previous section. However, the results of the lagged one-period model (Column 3) show a change in the direction and significance of the effects of the core explanatory variables. This signal suggests that there may be a problem of reverse causality or endogeneity due to omitted variables between pension insurance participation and household consumption structure, which may not be fully overcome by the baseline fixed effects model. Therefore, a more rigorous causal identification using an instrumental variables approach is presented below.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRobustness test results\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eshrinking\u003c/p\u003e \u003cp\u003ethe tails\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eexcluding\u003c/p\u003e \u003cp\u003eelderly samples\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003eLagging\u003c/p\u003e \u003cp\u003eperiod\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0157***\u003c/p\u003e \u003cp\u003e(3.8294)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0170***\u003c/p\u003e \u003cp\u003e(4.0676)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0178*\u003c/p\u003e \u003cp\u003e(1.8968)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0160*\u003c/p\u003e \u003cp\u003e(1.6971)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0394***\u003c/p\u003e \u003cp\u003e(5.6279)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0389***\u003c/p\u003e \u003cp\u003e(5.5353)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_1st_lag\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0154**\u003c/p\u003e \u003cp\u003e(-2.5738)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd_lag\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0111\u003c/p\u003e \u003cp\u003e(0.8289)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd_lag\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0142\u003c/p\u003e \u003cp\u003e(-1.4295)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3114***\u003c/p\u003e \u003cp\u003e(83.4812)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3160***\u003c/p\u003e \u003cp\u003e(83.1211)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.4167***\u003c/p\u003e \u003cp\u003e(85.1091)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40899\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Endogeneity test\u003c/h2\u003e \u003cp\u003eThis paper first tests the endogeneity of the core explanatory variables through the Durbin-Wu-Hausman (DWH) test, and the results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. The results show that the p-value of the first level of pension insurance is 0.24, which cannot reject the original hypothesis of \"no endogeneity\" at the 5% significance level; The p-values of the second level and the third level of pension insurance are 4.88888916260e-07 and 3.52073428047e-19, respectively, which strongly reject the original hypothesis at the 1% level. This indicates that there is a significant endogeneity problem in the participation of the second and third level of pension insurance, therefore, this study used instrumental variable method to correct it.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of endogeneity test (DWH test)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDWH χ\u0026sup2;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eFirst level\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eSecond level\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003eThird level\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.367072479151179\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25.31060874224525\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.15515017288125\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.2423\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.89e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.52e-19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResult\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eno significant endogeneity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eendogeneity exists\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eendogeneity exists\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results of the first-stage regression of instrumental variables are shown in Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, which shows that the community participation rate at each level is highly significantly and positively correlated with the participation behavior of households at the corresponding level, and the F-statistic of the first-stage regression is much larger than the empirical critical value of 10, which suggests that there is no problem of weak instrumental variables, and that the selected instrumental variables are valid.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of the first stage regression of the instrumental variables approach\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003erate_ins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003efirst order IV coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003esecond order IV coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003ethird order IV coefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.890***\u003c/p\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003erate_ins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.858***\u003c/p\u003e \u003cp\u003e(0.027)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003erate_ins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.899***\u003c/p\u003e \u003cp\u003e(0.023)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.094***\u003c/p\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.181***\u003c/p\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.117***\u003c/p\u003e \u003cp\u003e(0.011)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e96730\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.175\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.175\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1936.326\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e314.476\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e334.439\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e reports the results of the two-stage least squares (2SLS) estimation based on the instrumental variables approach and compares them with the benchmark OLS results. The results show that the coefficient of the first level pension insurance becomes insignificant in the 2SLS estimation, which corroborates with the conclusion that it is not significantly endogenous in the DWH test, suggesting that the estimation of the effect of the benchmark fixed-effects model for this level is basically plausible. The impacts of the second and third levels of pension insurance are shown to be more significantly positive, which suggests that the OLS model seriously underestimates the real contribution of the second and third levels of pension insurance to the upgrading of household consumption structure. Further comparison of the strength of the impacts reveals that the promotion effect of the third level of commercial pension insurance is significantly larger than that of the second level of enterprise annuities, which implies that market-oriented and personalized supplementary protection may have a stronger marginal role in unleashing the potential of consumption and optimizing the consumption structure.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of instrumental variable method regression results with the baseline regression\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003evariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOLS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2SLS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.012***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.050***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.116***\u003c/p\u003e \u003cp\u003e(0.020)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.056***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.232***\u003c/p\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eln_finc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.015***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.012***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehk\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.002\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.008***\u003c/p\u003e \u003cp\u003e(-0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.007***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.004***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0000***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eedu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.061***\u003c/p\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.051***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ework\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.004*\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esize\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.020***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.021***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.166***\u003c/p\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.195***\u003c/p\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96730\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.067\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Mechanism Test\u003c/h2\u003e \u003cp\u003eIn order to test the channels of action proposed in the theoretical analysis, based on the relative income hypothesis in the social comparison perspective, this study introduces the mediator variable of household income disparity, and this paper uses the relative deprivation index of (Kakwani, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1984\u003c/span\u003e) as a measure of this variable, which is used to measure the degree of inequality of household incomes at the micro level, and which is able to reflect the relative income position of a single household within the community and its perceived relative deprivation.\u003c/p\u003e \u003cp\u003eFor a community Y containing n households, the Kakwani index is defined for the ith household by ranking household incomes in ascending order as (y1,y2,...,yn):\u003cdiv id=\"Equi\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equi\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}Household\\:income\\:inequality({y}_{i},{y}_{j})=\\frac{1}{n{\\mu\\:}_{y}}\\sum\\:_{j=i+1}^{n}({y}_{j}-{y}_{i})={\\gamma\\:}_{yi}^{+}\\left(\\frac{{\\mu\\:}_{yi}^{+}-{y}_{i}}{{\\mu\\:}_{Y}}\\right) (8)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u0026micro;\u003csub\u003ey\u003c/sub\u003e is the average income of all households in that community, \u0026micro;\u0026thinsp;+\u0026thinsp;yi is the average income of households in community Y with incomes greater than y\u003csub\u003ei\u003c/sub\u003e, and γ\u0026thinsp;+\u0026thinsp;yi is the proportion of households in community Y with incomes greater than y\u003csub\u003ei\u003c/sub\u003e, expressed as a percentage of the total sample size. The theoretical logic of introducing this mechanism variable is that the participation in multi level pension insurance may alleviate the relative income deprivation felt by low-income households in the community by boosting their absolute and expected incomes. According to the Social Comparison Theory, when households' perceived income inequality decreases, their certainty about their future lives increases, which makes them more likely to allocate resources to development and enjoyment consumption, promoting the upgrading of consumption structure. The test results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eColumn (1) of Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e shows the total effect of each level of pension insurance on the consumption structure of households when no mediating variable is included, and column (2) shows the effect of each level of pension insurance on the mediating variable. The results show that participation in all levels of pension insurance significantly reduces the relative deprivation index of households. This suggests that participation in pension insurance effectively improves the relative economic status of households within the community by boosting their actual and expected incomes, alleviating the sense of income deprivation resulting from social comparison. Column (3) incorporates both core explanatory variables and mediating variables. The results show that the relative deprivation index has a significant negative impact on the upgrading of consumption structure, i.e., the higher the household's perceived income inequality, the lower the share of its consumption that is developmental and enjoyment-oriented. More importantly, the coefficients of the first, second, and third levels of pension insurance all decrease compared to Column (1) after controlling for the relative deprivation index. Combined with the significant path relationships in Columns (2) and (3), this suggests that alleviating relative income deprivation is an effective mediating channel for pension insurance to promote the upgrading of household consumption structure. While providing economic security, old-age security also releases consumption potential for higher-level needs by alleviating households' social comparison pressure and sense of psychological inequality.\u003c/p\u003e \u003cp\u003eIn summary, the results of the mechanism test confirm the research hypothesis, and further calculation of the proportion of mediating effects shows that the effects realized by the first, second and third levels of pension insurance through the path of alleviating relative deprivation account for 7.7%, 9.6% and 5.4% of the total effect, respectively.The second level exhibits the highest proportion of mediation effect, demonstrating that enterprise or occupational pensions significantly enhance the relative income status of households and mitigate relative deprivation. In contrast, the third level shows the lowest proportion of mediation effect, primarily due to the fact that most participants are high-income households, which experience a diminished sense of relative deprivation. Consequently, the benefits derived from this level are more closely associated with absolute income increases. This finding validates that multi level pension insurance not only affects consumption through the absolute income path but also through the indirect path of alleviating relative income deprivation, which provides new mechanistic evidence for understanding the micro-socio-economic benefits of pension insurance from a psychological perspective. To further exclude the potential estimation bias of the stepwise regression method and verify the robustness of this mediation mechanism, we conduct an additional robustness test based on the Bootstrap method.\u003c/p\u003e \u003cp\u003eTo further verify the robustness of the above mediation effect results and avoid the potential estimation bias of the stepwise regression method, this paper adopts the bias-corrected non-parametric percentile Bootstrap method with 1000 repeated samples to re-test the significance of the mediation effect. The test results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eThe results show that the 95% bias-corrected (BC) and percentile (P) confidence intervals of the mediation effects of the first, second and third levels of pension insurance do not contain 0, which means that the mediation effect of alleviating relative income deprivation is statistically significant at the 5% level for all three levels of pension insurance. This result is completely consistent with the conclusion of the stepwise regression test above, which further confirms the robustness of the mediation mechanism. It is fully verified that multi-level pension insurance can promote the upgrading of household consumption structure by alleviating households' relative income deprivation and reducing social comparison pressure, which provides more rigorous empirical evidence for the theoretical mechanism of this paper.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMechanism test of the impact of pension insurance on the structure of consumption through the relative deprivation index\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003etotal effect model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003emediated variable model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003ecomplete model\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.013***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.012***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.052***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.037***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.047***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.056***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.021***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.053***\u003c/p\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eln_finc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.012***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.148***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.006***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehk\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.021***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.048***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.015***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.014***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.000\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.014***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.002***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eage2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.000***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eedu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.059***\u003c/p\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.050***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.053***\u003c/p\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ework\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.014***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.005***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.014***\u003c/p\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esize\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.020***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.010***\u003c/p\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.019***\u003c/p\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold income inequality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.121***\u003c/p\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.248***\u003c/p\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.201***\u003c/p\u003e \u003cp\u003e(0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.513***\u003c/p\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e96765\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRobustness test of mediation effect based on Bootstrap method\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eind1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eObserved coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBias\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStd. err.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e95% BC Confidence Interval\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e95% P Confidence Interval\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.01308344\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-5.41e-06\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.00042739\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0121497\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0138153\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.0122199\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.0138772\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eind2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.03856864\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-3.66e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00096563\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.0366919\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.0404712\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.0366697\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.0404289\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eind3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.02116122\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.0000491\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00073643\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.0198347\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.0226631\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.0196525\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.0225509\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003eNotes: BC represents Bias-corrected confidence interval; P represents Percentile confidence interval. The number of repeated sampling for Bootstrap is 1000.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAt the outset of the mechanism test design in this study, the classic economic mechanism\u0026mdash;the precautionary savings mechanism\u0026mdash;was incorporated. A mediating effect analysis was conducted using a standardized process aligned with the core mechanism test. However, the empirical results indicate that this traditional transmission mechanism did not achieve significance in the sample analyzed. Consequently, the proposed path of action is not supported. This finding highlights the applicability boundaries of classical consumption theory within the specific context of China, The underlying cause can be traced to the entrenched traditional savings concept of \"preparing for the future\" and the high savings habits that have developed among Chinese residents, which exhibit strong path dependence. Household savings motives encompass various life cycle risks, including retirement, medical care, children's education, and housing. Participation in endowment insurance alone addresses only certain uncertainties related to retirement and is insufficient to fundamentally alter residents' savings decisions (Zhang et al.,2025).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Heterogeneity Test\u003c/h2\u003e \u003cp\u003eWith significant differences in development levels between urban and rural areas and geographic regions in China, and the differences in budget constraints and social security levels faced by households across regions or urban and rural areas (Gao et al.,2018), there may be group heterogeneity in the impact of pension insurance on consumption structure. In order to deeply reveal the boundary conditions of its effect, this paper further conducts group tests from the regional and urban-rural dimensions. In the regional dimension, the sample is divided into eastern, central and western regions based on the three major zones division method. The eastern region includes Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan; the central region includes Heilongjiang, Jilin, Shanxi, Anhui, Jiangxi, Henan, Hubei and Hunan; and the western region includes Sichuan, Chongqing, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, Guangxi and Inner Mongolia. In the urban-rural dimension, the sample is divided into urban and rural households, aiming to reveal possible systematic differences in the impact of pension insurance under the dual social security structure.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e4.5.1 Analysis of regional heterogeneity\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e reports the regression results by East, Central and West regions. The analysis shows that the first level of basic pension insurance exhibits significant positive effects in all three regions, which reflects the general effectiveness of the basic system in China. It is very interesting that the significance of the second level and the third level pension insurance in the east, center and west regions is complementary. Specifically, the effect of the second level pension insurance is significantly positive only in the central region (0.0489, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), while the third level of commercial life insurance plays a significant role in the east and west. The central provinces have developed industry and manufacturing, enterprises are relatively concentrated, the system of enterprise annuity covers a wide range of enterprises, and its consumption release effect is therefore prominent. In contrast, the economy of the eastern region is extremely developed, and the role of enterprise occupational pension in upgrading household consumption in the eastern region is not obvious.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of the analysis of regional heterogeneity\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eEast\u003c/p\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eCentral region\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003cp\u003eWest\u003c/p\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0143**\u003c/p\u003e \u003cp\u003e(2.2313)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0171**\u003c/p\u003e \u003cp\u003e(2.2379)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0168**\u003c/p\u003e \u003cp\u003e(2.2421)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0131\u003c/p\u003e \u003cp\u003e(0.9935)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0489**\u003c/p\u003e \u003cp\u003e(2.3602)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0021\u003c/p\u003e \u003cp\u003e(0.1220)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0441***\u003c/p\u003e \u003cp\u003e(4.4299)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0202\u003c/p\u003e \u003cp\u003e(1.5401)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0502***\u003c/p\u003e \u003cp\u003e(3.4055)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3226***\u003c/p\u003e \u003cp\u003e(55.2375)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3061***\u003c/p\u003e \u003cp\u003e(44.2544)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.3005***\u003c/p\u003e \u003cp\u003e(43.7939)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44936\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26514\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25315\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.080\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e4.5.2 Analysis of urban-rural heterogeneity\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e shows the regression results by grouping urban and rural households. It is found that the first level pension insurance has a significantly stronger effect on the promotion of consumption upgrading for urban households than for rural households, directly reflecting the gap between urban and rural basic security levels. Whereas the second level pension insurance has a strong positive impact on the consumption structure of rural households, the impact on urban households is insignificant. The likely reason for this is that since the number of urban households participating in the second level pension insurance (2,445) is much larger than the number of rural households (150), the few rural domiciles with access to enterprise annuities have a large marginal improvement effect on consumption from the jump in long-term income security. The third level pension insurance shows a significant boost in both urban and rural areas.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of the analysis of urban-rural heterogeneity\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eins_1st\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003cp\u003eRural households\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003cp\u003eUrban households\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0137**\u003c/p\u003e \u003cp\u003e(0.006)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0258***\u003c/p\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_2nd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1078***\u003c/p\u003e \u003cp\u003e(0.039)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0127\u003c/p\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eins_3rd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0335***\u003c/p\u003e \u003cp\u003e(2.5975)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0412***\u003c/p\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.2945***\u003c/p\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3313***\u003c/p\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e50732\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"5. Conclusion, Policy Recommendations, and Research Limitations","content":"\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Conclusions of the study\u003c/h2\u003e \u003cp\u003eBased on the life-cycle hypothesis and Maslow\u0026rsquo;s hierarchy of needs, this study constructs a theoretical framework to analyze the impact of multi-level pension insurance on household consumption structure and empirically tests it using panel data from the China Household Finance Survey (CHFS) for the years 2015, 2017, and 2019. The empirical results show that participation in any single level of pension insurance significantly improves household consumption structure, and that pension insurance not only expands the scale of consumption but also promotes consumption upgrading. A notable synergistic enhancement effect is observed when the first-level basic pension insurance and the third-level commercial life insurance are combined: households participating in both types experience a greater degree of consumption structure upgrading than the simple sum of the effects of participating in each separately. Moreover, the role of the third-level pension insurance in shaping consumption structure is contingent upon the presence of the first-level basic pension insurance. In contrast, the interaction between the first and second levels is not statistically significant; their effects operate independently. This is largely attributable to the functional similarity and overlapping objectives of the first- and second-level schemes, which diminish their complementary potential. Heterogeneity analysis reveals that the impact of pension insurance on consumption structure varies significantly across groups. Regionally, the second-level pension insurance plays a more prominent role in central China, while the third-level commercial insurance exhibits stronger effects in the eastern and western regions. In terms of urban\u0026ndash;rural differences, the first-level insurance has a more pronounced effect on urban households, reflecting disparities in the level of basic protection between urban and rural areas. Notably, the second-level pension insurance shows an unusually strong marginal effect for rural households, suggesting that rural participants derive greater incremental security from such insurance compared to their urban counterparts, thereby significantly boosting their consumption confidence. Mechanism analysis further indicates that pension insurance enhances consumption confidence and facilitates consumption upgrading by alleviating households\u0026rsquo; sense of relative income deprivation. These findings provide empirical support for optimizing the multi-level pension system and offer policy insights for strengthening social security as a lever to unlock consumption potential and promote high-quality economic development.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Policy recommendations\u003c/h2\u003e \u003cp\u003eBased on the findings above, this paper offers the following policy recommendations to strengthen the multi-level pension insurance system and promote the upgrading of household consumption:\u003c/p\u003e \u003cp\u003eFirst, reinforce the foundational role of the first level to stabilize consumption expectations. It is essential to advance universal pension coverage and establish a dynamic mechanism for adjusting pension benefits. Although basic pension insurance has wide coverage, it has not yet achieved full universal coverage. Enhancing national pooling and linking benefit adjustments to economic growth and price changes can help secure residents\u0026rsquo; basic living standards and underpin consumption confidence.\u003c/p\u003e \u003cp\u003eSecond, enhance the complementary role of the second and third levels to unlock consumption upgrading potential. For the second level, mechanisms such as automatic enrollment and tax deferrals should be introduced to expand coverage, particularly among small and medium-sized enterprises and employees in emerging industries. For the third level, policy efforts should promote the development of commercial pension insurance to meet diverse needs, while strengthening public awareness to encourage voluntary participation. These measures can reinforce the supplementary function of pensions and bolster household consumption capacity.\u003c/p\u003e \u003cp\u003eThird, adopt targeted policy approaches tailored to regional and urban\u0026ndash;rural disparities. In eastern regions, policy should focus on facilitating wealth planning and consumption among high-income groups through commercial insurance. In western and rural areas, a dual strategy combining improved pension coverage with employment support is recommended. Offering tax incentives to enterprises that provide supplementary pensions for rural workers could enhance their economic security and enable a shift toward consumption in health, education, and recreation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Limitations\u003c/h2\u003e \u003cp\u003eThis study has several limitations that also suggest directions for future research. First, in the context of rapid technological advancement and sustainable economic development, subsequent studies could explore the impact of pension insurance on emerging consumption areas such as digital consumption, green consumption, and health-related services, thereby providing theoretical insights for guiding the development of these sectors. Second, future research could employ longer-term panel data to examine the dynamic evolution and life-cycle patterns of how pension insurance influences household consumption structure. Third, further investigation into the linkages between multi-level pension insurance and household financial behaviors\u0026mdash;such as asset allocation and risk preference\u0026mdash;would contribute to a more comprehensive understanding of the micro-level mechanisms through which pension insurance promotes consumption upgrading.\u003c/p\u003e \u003cp\u003eFinally, due to the absence of direct measures of commercial pension insurance in the China Household Finance Survey (CHFS) data, this study uses commercial life insurance as a proxy, which may introduce estimation bias. Commercial life insurance encompasses a range of products\u0026mdash;including whole life, term life, and endowment insurance\u0026mdash;only some of which serve a clear old-age security function. Data limitations prevent a full disentanglement of the heterogeneous effects of these insurance types. As a result, the estimated effect of third-level pension insurance may be biased: if life insurance products without pension functions have a weaker effect on consumption upgrading than dedicated commercial pension insurance, the coefficient may be underestimated; conversely, it may be overestimated. Accordingly, the estimated coefficients for third-level pension insurance in this study should be interpreted as the average effect of broadly defined market-oriented pension protection products.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding statement \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Scientific Research Interest Training Program of Sichuan Agricultural University.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of conflicting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author(s) declared no potential conflicts of interest in this study, paper writing, and publication.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eZhiqi Zhang :Conceptualization,Data curation,Methodology,Project administration,Writing \u0026ndash; original draft,Writing \u0026ndash; review \u0026amp; editing.Xingyu Xu:Data curation,Software,Validation,Writing \u0026ndash; review \u0026amp; editing.Huidan Xu: Methodology,Resources,Supervision,Validation,Writing \u0026ndash; review \u0026amp; editing.Qiye Meng:Data curation\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe data used in this study were obtained from the China Household Finance Survey (CHFS) database, available at https://chfs.swufe.edu.cn/.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAgarwal, S., Qian, WL., \u0026amp; Zou, X. (2021). Thy neighbor's misfortune: Peer effect on consumption[J]. American Economic Journal: Economic Policy, 13(2): 1\u0026ndash;25.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/pol.20170634\u003c/span\u003e\u003cspan address=\"10.1257/pol.20170634\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAngrist, J. D. (2014). The perils of peer effects[J]. Labour Economics, 30: 98\u0026ndash;108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.labeco.2014.05.008\u003c/span\u003e\u003cspan address=\"10.1016/j.labeco.2014.05.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKakwani, N. (1984). The relative deprivation curve and its applications[J]. Journal of Business \u0026amp; Economic Statistics, 3(2): 171\u0026ndash;173.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaslow, A. H. (1943). A theory of human motivation[J]. Psychological Review, 50: 370\u0026ndash;396. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/h0054346\u003c/span\u003e\u003cspan address=\"10.1037/h0054346\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao, XB., Wang, F., Wang, Y., Wang, YX. (2022). An investigation of whether pensions increase consumption: Evidence from family portfolios[J]. Finance Research Letters, 47: 102591. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.frl.2021.102591\u003c/span\u003e\u003cspan address=\"10.1016/j.frl.2021.102591\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun, JY., Huang, JC., Jiang, XJ., Song, XL., Zhang, N. (2025). Spatiotemporal evolution and influencing factors of population aging in the Triangle of Central China at multiple scales[J]. Sustainability, 17(14), 6549. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su17146549\u003c/span\u003e\u003cspan address=\"10.3390/su17146549\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei, YX., Chen, J. (2024). Sustainable design for the silver society: Developing the silver model for gerontechnology product innovation[J]. Sustainability, 17(1): 42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su17010042\u003c/span\u003e\u003cspan address=\"10.3390/su17010042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, DJ., Guo, XT. (2023). The impact of the digital economy on residents\u0026rsquo; consumption upgrading\u0026mdash;Evidence from Mainland China[J]. Sustainability, 15(10), 8041. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su15108041\u003c/span\u003e\u003cspan address=\"10.3390/su15108041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong MY., Wang J., Tian MJ. (2022). Rural social security, precautionary savings, and the upgrading of rural residents' consumption structure in China[J]. Sustainability 2022, 14(19), 12455; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su141912455\u003c/span\u003e\u003cspan address=\"10.3390/su141912455\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eModigliani, F., \u0026amp; Ando, A.K. (1957). Tests of the life cycle hypothesis of savings: comments and suggestions[J]. Bulletin of the Oxford University Institute of Economics \u0026amp; Statistics, 19(2): 99\u0026ndash;124.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStutzer, A.(2004).The role of income aspirations in individual happiness[J].JOURNAL OF ECONOMIC BEHAVIOR \u0026amp; ORGANIZATION.54: 89\u0026ndash;109.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jebo.2003.04.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jebo.2003.04.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao, MQ., Zhang, MX. (2025). Marketization and household consumption upgrading: Evidence from China[J]. Sustainability 2025, 17(21), 9373; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su17219373\u003c/span\u003e\u003cspan address=\"10.3390/su17219373\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, W., \u0026amp; Zhang, CC. (2021). The power of social pensions: Evidence from China's new rural pension scheme[J]. American Economic Journal: Applied Economics, 13(2),179\u0026ndash;205. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/app.20170789\u003c/span\u003e\u003cspan address=\"10.1257/app.20170789\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarsen, L. S.,Nielsson, U.,Nutu, M.,\u0026amp;Rangvid, J.(2025). Mandatory pension contributions, household consumption, and savings[J]. Journal of Pension Economics \u0026amp; Finance, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S147474722510005X\u003c/span\u003e\u003cspan address=\"10.1017/S147474722510005X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang, JY., Park, S.,\u0026amp; Ahn, S Ahn, S(2022). The effect of social pension on consumption among older adults in Korea[J].Journal of the Economics of Ageing, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jeoa.2021.100364\u003c/span\u003e\u003cspan address=\"10.1016/j.jeoa.2021.100364\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCuong, NV. (2013). The Impact of Social Security on Household Welfare: Evidence from a Transition Country[J].European Journal of Development Research, 25(5), 737\u0026ndash;757. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.asieco.2013.09.002\u003c/span\u003e\u003cspan address=\"10.1016/j.asieco.2013.09.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y. Z., \u0026amp; He, X. Y. (2024). Does purchasing commercial insurance promote household consumption? Evidence from China[J]. International Review of Economics \u0026amp; Finance, 96: 103641. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.iref.2024.103641\u003c/span\u003e\u003cspan address=\"10.1016/j.iref.2024.103641\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarenberg, D., \u0026amp; Ludwig, A. (2015). Social security in an analytically tractable overlapping generations model with aggregate and idiosyncratic risks[J]. International Tax and Public Finance, 22(4): 579\u0026ndash;603. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10797-015-9368-x\u003c/span\u003e\u003cspan address=\"10.1007/s10797-015-9368-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, QN., Cai, YX., Zhang, PL., \u0026amp; Li, R. (2025). Set rainy days aside: How pension insurance affects household tourism consumption[J]. Journal of Hospitality and Tourism Management, 63, 3\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhtm.2025.03.001\u003c/span\u003e\u003cspan address=\"10.1016/j.jhtm.2025.03.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeldstein, M. (1974). Social Security, Induced Retirement, and Aggregate Capital Accumulation[J]. Journal of Political Economy, 82(5), 905\u0026ndash;926. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/260246\u003c/span\u003e\u003cspan address=\"10.1086/260246\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiamond, P. A. (1965). National Debt in a Neoclassical Growth Model[J]. American Economic Review, 55(5), 1126\u0026ndash;1150.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng, HM., \u0026amp; Zhong, T. (2016). The impacts of social pension on rural household expenditure: evidence from China[J]. Journal of Economic Policy Reform, 19(3), 221\u0026ndash;237. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17487870.2015.1041524\u003c/span\u003e\u003cspan address=\"10.1080/17487870.2015.1041524\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng, W., Lyu, Y., Jia, R.,\u0026amp; Hanewald, K. (2022). The impact of expected pensions on consumption: Evidence from China[J]. Journal of Pension Economics and Finance, 22(1), 69\u0026ndash;87. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S147474722100041X\u003c/span\u003e\u003cspan address=\"10.1017/S147474722100041X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBernard, T., Doss, C. R., Hidrobo, M., Hoel, J. B., \u0026amp; Kieran, C. (2020). Ask me why: Patterns of intrahousehold decision-making[J]. World Development, 125, 104671. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.worlddev.2019.104671\u003c/span\u003e\u003cspan address=\"10.1016/j.worlddev.2019.104671\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, Q., Yang, S., Zhang, Y., \u0026amp; Li, S. (2018). The Divided Chinese Welfare System: Do Health and Education Change the Picture?[J]. Social Policy \u0026amp; Society, 17(2): 227\u0026ndash;244. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S1474746417000100\u003c/span\u003e\u003cspan address=\"10.1017/S1474746417000100\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai Y. (2025).The pension insurance system, multi-level superimposed participation and the scale of household consumption[J].East China Economic Management, 39(9). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.19629/j.cnki.34-1014/f.250401017\u003c/span\u003e\u003cspan address=\"10.19629/j.cnki.34-1014/f.250401017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (in Chinese)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHogg M A. (2000). Social Identity and Social Comparison[A]//Suls J, Wheeler L. Handbook of Social Comparison[M]. Boston, MA: Springer \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-1-4615-4237-7_19\u003c/span\u003e\u003cspan address=\"10.1007/978-1-4615-4237-7_19\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eModigliani F,\u0026amp; Brumberg R. (1954). Utility analysis and the consumption function: An interpretation of cross-section data[A]//Kurihara K K. Post-Keynesian Economics[M]. New Brunswick, NJ: Rutgers University Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, Y., C Zhang, Q.X. (2023). Income Disparity, Consumption Patterns, and Trends of International Consumption Center City Construction, Based on a Test of China\u0026rsquo;s Consumer Market[J]. Sustainability, 15(4): 2862. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su15042862\u003c/span\u003e\u003cspan address=\"10.3390/su15042862\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang,LG.,Brooks,R.,Ding,D.,Ding,HY.,He,H.,Lu,J.,\u0026amp;Rui C.M.,(2025).China's high savings: Drivers, prospects, and policy implications[J].\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e69.https://doi.org/10.1016/j.ememar.2025.101355\u003c/span\u003e\u003cspan address=\"69.10.1016/j.ememar.2025.101355\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"review-of-economics-of-the-household","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"reho","sideBox":"Learn more about [Review of Economics of the Household](http://link.springer.com/journal/11150)","snPcode":"11150","submissionUrl":"https://submission.nature.com/new-submission/11150/3","title":"Review of Economics of the Household","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Multi-level Pension Insurance, Household Consumption Structure Upgrading, Chinese Households","lastPublishedDoi":"10.21203/rs.3.rs-9194470/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9194470/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDuring China\u0026rsquo;s economic transition from high-speed growth to high-quality development, stimulating domestic demand and optimizing the consumption structure have emerged as critical economic priorities. Drawing on the life-cycle hypothesis, social comparison theory, and Maslow\u0026rsquo;s hierarchy of needs, this study employs panel data from the China Household Finance Survey (CHFS) to investigate the impact of multi-level pension insurance participation on the upgrading of household consumption structure. The findings reveal that participation in multi-level pension insurance significantly facilitates a shift in household consumption structure from subsistence-oriented to development- and enjoyment-oriented categories. Pension insurance enhances household consumption confidence both by increasing stable absolute expected future income and by improving households\u0026rsquo; relative economic standing within their communities. Mechanism analysis further indicates that pension insurance promotes consumption upgrading by alleviating residents\u0026rsquo; sense of relative income deprivation. This paper provides empirical evidence supporting the refinement of the multi-level pension insurance system in China, and offers policy insights for mitigating residents\u0026rsquo; financial insecurity to foster sustainable consumption upgrading.\u003c/p\u003e","manuscriptTitle":"The Impact of Multi-level Pension Insurance Participation on the Upgrading of Household Consumption Structure: Evidence from China","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-16 15:04:42","doi":"10.21203/rs.3.rs-9194470/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"127197951588489705228999738870604386899","date":"2026-04-09T09:25:04+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-08T17:49:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-28T14:44:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-28T14:44:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Review of Economics of the Household","date":"2026-03-23T01:44:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"review-of-economics-of-the-household","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"reho","sideBox":"Learn more about [Review of Economics of the Household](http://link.springer.com/journal/11150)","snPcode":"11150","submissionUrl":"https://submission.nature.com/new-submission/11150/3","title":"Review of Economics of the Household","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"311399d5-c5c1-4e92-b2df-59c0db46e447","owner":[],"postedDate":"April 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-16T15:04:42+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-16 15:04:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9194470","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9194470","identity":"rs-9194470","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}