Study on the Impact of China’s Pension Insurance System on the Savings Rate of Urban Residents Based on the Preventive Saving Perspective

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper investigates the relationship between China's pension insurance system and the urban residents' savings rate. It focuses on how the pension participation rate and pension replacement rate affect preventative savings. Using a two-period overlapping generations (OLG) model and empirical data from 31 provinces (2002-2021), the authors find that the pension participation rate positively correlates with savings, while the pension replacement rate negatively correlates with it. The study highlights regional disparities and the influence of aging populations on savings behaviors, providing a nuanced look at how different regions and demographic groups react to pension reforms.

While the paper provides a comprehensive quantitative analysis of China's pension system, addressing these limitations would significantly enhance its depth and applicability to policymaking.

1.      Assumptions of the OLG model: The model assumes homogeneous behavior among residents in terms of participation and savings, which may not accurately reflect the varied socio-economic factors influencing individual decisions. Assuming homogeneity in behavior may lead to inaccurate policy recommendations, as it overlooks income disparities and other factors that impact individual savings decisions.

2.      Limited exploration of alternative savings incentives: While the paper focuses on pension insurance, it doesn’t explore how other factors, such as healthcare or education costs, might influence savings. The exclusion of other significant costs (e.g., healthcare) reduces the depth of the analysis. These factors are integral to understanding the full scope of residents' saving behaviors, especially in an aging society where healthcare expenses rise.

3.      Overreliance on quantitative data: The analysis heavily leans on econometric models and panel data, potentially overlooking qualitative insights, such as residents' perceptions or satisfaction with the pension system. This limits the paper's ability to capture more nuanced human behaviors and perceptions, which are crucial for understanding the real-life impact of policy changes.

4.      Simplified categorization of regions: The division of China into eastern, central, and western regions overlooks the complexity of socio-economic development within each region, potentially obscuring more localized trends.

5.      “Variant” should be “Variable” in each table

6.      Methods and results sections are lengthy

7.      Heterogeneity tests should be conducted for high and low income provinces.

Comments on the Quality of English Language

See above

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The article is interesting. Appropriate research methods used. The conclusions, models and content of the article are appropriate. However, the subject of the article is addressed to a narrow group of readers, i.e. readers interested in China.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Clarity in Writing: The manuscript would benefit from clearer writing, particularly in the abstract and introduction sections. The current text is dense and occasionally difficult to follow, which might limit its accessibility to a broader audience. Simplifying the language and clarifying key concepts early on would enhance readability.

Introduction and Literature Review: The connection between the introduction and the literature review could be strengthened. The introduction should more explicitly set up the research questions and hypotheses, with the literature review then framing how the study fits into and contributes to the existing body of research.

Hypothesis Development: While the hypotheses are logically derived, the transition from the theoretical model to the empirical hypotheses could be smoother. It would be helpful if the authors more clearly linked the theoretical insights to specific testable predictions.

Discussion of Results: The discussion section should be expanded to provide a more in-depth interpretation of the empirical findings. Currently, the discussion is somewhat descriptive, focusing on the presentation of results without fully exploring their implications. A more nuanced discussion that connects the results back to the literature and policy implications would strengthen the manuscript.

Regional Heterogeneity: The manuscript does well to explore regional differences in the impact of the pension system, but the analysis could be deepened. It would be beneficial to explore why certain regions behave differently and what policy implications these differences might have. For instance, why does the central region show a positive correlation between pension participation and savings, while this is not significant in the eastern and western regions?

Comments on the Quality of English Language

• The quality of writing in the abstract is lower than the rest of the paper. Please rewrite it for clarity and to provide a concise summary of the paper.
• The introduction should conclude with a paragraph outlining how the rest of the paper is organized.
• The structure of the paper should be reconsidered. Please review related literature, as conclusions are typically a standalone section.
• The section "Indicator Selection and Model Construction" is too lengthy and should be divided into different sections. Consider consulting recent related literature for guidance on how to structure this part of the paper, especially in model calibration studies.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Summary

This study focuses on the Chinese savings rate and its relationship to the pension insurance system. This study presents a theoretical model of behavior in environments in which a pension insurance system exists and does not exist.  A series of empirical tests are presented to examine the relationship between pension insurance participation and the savings rate and the relationship between the replacement rate of old-age pensions and the savings rate.  The authors find that, contrary to their model’s prediction, pension insurance participation and the savings rate are positively related. Also, the authors find that the pension replacement rate is negatively related to the savings rate.  However, these associations vary across geographic regions and across regions with age differences.  The authors conclude with a series of policy recommendations to improve the pension system. 

General Comments

The theoretical  model represented by equation (19) posits that the savings rate is negatively related to the savings rate.  Given the empirical finding of a positive relationship and the conclusions presented in the study, are there parameter values that would reverse the sign of this relationship?   Can there be two representations of equation (19) based on parameter values; one where the relationship between participation and the savings rate is negative and another where the relationship is positive? If so, this would lead to a more coherent interpretation of the empirical findings. The unexpected findings need to be explained of course.  While readers may find that the explanations for why the findings are not aligned with the theoretical model to be plausible, they might ask why a theoretical model was presented if it does not capture actual behavior. It also seems to suggest that a key factor - the maturity of the pension system - is left out of the theoretical model.  A key recommendation from the study is that the pension insurance system should be strengthened.  However, this recommendation does not emerge from the theoretical model or directly from the empirical tests.  A reframing of the theoretical model is needed. 

The time period used for much of the analysis covers 2002 - 2021. It would be helpful to add a short section explaining the development of the Chinese pension insurance system. Was it a mature system at some point in the sample period? How much did the system change over this period of time? Lines 454-458 imply that it is still developing.  Knowing the broader patterns of participation in the pension system and savings rates over time would help the reader interpret the results.  Are there economy-wide factors that are driving the participation rate such that the income variable or the time fixed effects are capturing the pattern of rising participation?  And are there economy-wide factors that have led to large changes in the savings rate?  The minimum and maximum values of URSR, NPS, and RRI are far apart. Are the minimum values for cases early in the period and the maximum values for cases late in the period?

The analysis by geographic regions and age-based regions are useful sections in the study.  It would be helpful to include summary statistics for the key variables for the region subgroups.  This would help the reader appreciate the magnitude of the differences across the regions and to better interpret the economic significance of the coefficient estimates. 

Results based on OLS and FE static models are presented in Table 2.  However, shortcomings in these models are noted in the manuscript.  Since these models appear to be discredited by the authors, they should be removed from the manuscript.   

Additional Comments

46-47. The sentence: “In the process of promoting consumption, the pension insurance system has been highly expected” is confusing.  Is this the intention: “In the process of promoting consumption, the pension insurance system has been highly anticipated”?

100.   “IndiaAng” appears to be a typo.

137-138. The phrase: :”The proportion of labor force population is ..” is confusing.  Is the intention: “the share of the population in the labor force is ..”  

271-272. Is NPS based on the total population or the total adult population.  Presumably, children are not participating in the pension system. So NPS will be impacted by the share children comprise of the population in addition to the participation of adults in the pension system. 

500-503. Real interest rates are found to be negatively related to the savings rate. Slower expected growth is given as the explanation. However, why would one not expect that the savings rate would rise with higher real rates (e.g., greater compensation for savings)? 

505-510.Endogeneity may well exist. But it would be helpful to identify the variables for which endogeneity may be a concern.  

684-703. This text appears out of place.  It provides a summary of the study, which is provided elsewhere in the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have responded to my previous comments and make some changes as they said. But in the manuscript. I cannot find any changes as they said they made. I would like to see the revised files before I can make a decision. 

Comments on the Quality of English Language

Fine

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The authors have responded to most of the comments from my review of the first version of the study.  The responses to my comments were detailed and thoughtful. The new text and changes to the model were provided in the authors’ cover letter. 

The version of the study available on the portal was labeled “v2.”  However, it appears to be the original version of the paper.  While the cover letter provided the new text and model revisions, I was unable to review the updated version with these changes.  

Hypotheses 1 and 2 have been modified to align with the revised model.  Hypothesis 1 clearly flows from equation (19).  The paragraph leading into hypothesis 2 reads more like the alternative to the null statement of hypothesis 1 than a separate hypothesis.  The  authors might consider representing hypothesis 2 as simply the alternative hypothesis to hypothesis 1. It seems that the underlying cause for the rise in the participation rate and the savings rate is due to demographics and the maturing of the pension system.  For example, in explaining the gaps between the minimum and maximum values (in response to Comment 2), the authors note: “As for the large difference between the minimum and maximum values of the NPS (pension insurance participation rate), it reflects the process of continuous improvement and adjustment of the pension insurance system. For example, the policy of lowering the corporate contribution rate implemented in 2018 has not only reduced the burden on enterprises, but also contributed to the year-on-year increase in the pension insurance participation rate. The minimum value here represents the participation situation in the early years of the system, while the maximum value demonstrates the participation situation after a series of reforms.”  This argument explains why the participation rate is positively correlated with the savings rate. But it does not explain why an increase in the participation rate may lead to (cause) an increase in the savings rate as posited in hypothesis 2.  Section 6.1 seems framed to explain why hypothesis 1 is not supported and not why there is a positive causal relationship between the participation rate and savings rate. 

Additional discussion was added on the heterogeneity analysis.  However, I did not find summary statistics for the regional subgroups.  Providing mean values for RUDIGR, URSR, NPS for the three regions should be sufficient. 

Close to the end of section 6.1, the authors state: “Although this study focuses on China, the conclusions of this study can also provide valuable lessons and references for other countries that are facing high savings rates and are planning to solve this problem through pension insurance reforms.”  Do the authors mean: “Although this study focuses on China, the conclusions of this study can also provide valuable lessons and references for other countries where less-than-mature pension insurance systems are causing savings rates to be elevated.  Under these conditions, pension insurance reforms are desirable.”  The original statement suggests that high savings rates are necessarily undesirable.

Author Response

Please see the attachment.

Author Response File: Author Response.docx