Summary:
With the deepening of the global aging process, many countries are undertaking pension system reforms that tend to increase the uncertainty of future pension wealth. A primary trend in these reforms is the establishment of multi-pillar pension systems, a direction China has also emphasized in its national development strategy. This trend unfolds against the backdrop of China's persistently high savings rate, which has hovered around 45% since 2015. It is therefore crucial to investigate how pension uncertainty affects household savings and financial asset allocation. While existing literature has extensively studied the impact of changes in the level of pension wealth on household savings, research on how pension wealth uncertainty affects the structure of household financial assets remains scarce. This paper aims to fill this gap by leveraging the 2015 reform of the pension system for public sector employees in China as a policy experiment. This reform not only altered the benefit formula for public sector employees but also introduced an occupational annuity, thereby substantially increasing the uncertainty of their future pension wealth. This study conducts an empirical analysis using the 2015 public sector pension reform. Before the reform, the pension wealth for public sector employees was highly predictable, primarily determined by their final year's salary and years of service. The reform abolished this system, establishing a framework consistent with that for enterprise employees and introducing a second-pillar “occupational annuity” operating on a funded, individual account basis. A key feature of this annuity is that employer and employee are invested in the capital market, with returns based on actual investment performance. Compared to the relatively stable growth of wages, this market oriented reform significantly increased the uncertainty associated with future pension income. We employ a difference-in-differences (DID) methodology, analyzing data from the China Household Finance Survey (CHFS) from 2011 to 2019. Households headed by public sector employees constitute the treatment group, while those headed by enterprise employees serve as the control group. Our empirical results show that the reform caused public sector households to increase their savings rate by 6.1 percentage points. Simultaneously, the proportion of their financial assets held in risky instruments decreased by approximately 5.9 percentage points. To analyze the underlying mechanisms, we first rule out several potential alternative explanations. Regarding the wealth effect, our calculations show that while the reform did significantly reduce basic pension wealth, the newly established annuity compensated for this reduction, leaving the expected value of total lifetime pension wealth largely unchanged. Regarding the income effect, we find that although the pre-tax wages of public sector employees increased significantly post-reform, this increase was largely channeled into higher social security contributions, resulting in no significant change in their actual post-tax disposable income. Conversely, this paper provides direct evidence for the mechanism through which increased uncertainty in pension wealth altered household asset allocation decisions. First, in provinces that had explicitly implemented market-based investment management for their occupational annuity funds, the changes in household savings and financial asset structure were more pronounced, suggesting that the uncertainty stemming from the annuity's operation is a key driver. Second, the reform's impact is concentrated among younger employees who face a higher degree of uncertainty. For older employees within a transitional period, who were protected by a provision of guaranted benefits no lower than those under the old system, the effect was not statistically significant. The findings of this paper enrich the understanding of multi-pillar pension systems. The research indicates that under a market-based operation, increased uncertainty in pension wealth can crowd out household-level allocation to risky assets while simultaneously increasing household savings. These findings have important policy implications for improving China's multi-pillar pension system, reducing households' excessive savings motives, and informing the promotion of the third-pillar pension system. Policymakers need to leverage the long-term investment advantages of annuity funds to provide stable returns, and regulate and develop professional investment services to help households determine appropriate savings levels and investment structures.
Bottazzi, R., T. Jappelli and M. Padula, 2006, “Retirement Expectations, Pension Reforms, and Their Impact on Private Wealth Accumulation”, Journal of Public Economics, 90(12), pp.2187～2212.
[20]
Campbell, J. Y., 2006, “Household Finance”, The Journal of Finance, 61(4), pp.1553～1604.
[21]
Chetty, R., J. N. Friedman, S. Leth-Petersen, T. H. Nielsen and T. Olsen, 2014, “Active vs. Passive Decisions and Crowd-Out in Retirement Savings Accounts: Evidence from Denmark”, The Quarterly Journal of Economics, 129(3), pp.1141～1219.
[22]
Curcuru, S., J. Heaton, D. Lucas and D. Moore, 2010, “Heterogeneity and Portfolio Choice: Theory and Evidence”, Handbook of Financial Econometrics: Tools and Techniques: Elsevier, pp.337～382.
[23]
Etgeton, S., B. Fischer and H. Ye, 2023, “The Effect of Increasing Retirement Age on Households' Savings and Consumption Expenditure”, Journal of Public Economics, 221, 104845.
[24]
Feldstein, M., 1974, “Social Security, Induced Retirement, and Aggregate Capital Accumulation”, Journal of Political Economy, 82(5), pp.905～926.
[25]
Feng, J., L. He and H. Sato, 2011, “Public Pension and Household Saving: Evidence from Urban China”, Journal of Comparative Economics, 39(4), pp.470～485.
[26]
García-Miralles, E. and J. M. Leganza, 2024, “Public Pensions and Private Savings”, American Economic Journal: Economic Policy, 16(2), pp.366～405.
[27]
Heaton, J. and D. J. Lucas, 1996, “Evaluating the Effects of Incomplete Markets on Risk Sharing and Asset Pricing”, Journal of Political Economy, 104(3), pp.443～487.
[28]
Jappelli, T., I. Marino and M. Padula, 2021, “Social Security Uncertainty and Demand for Retirement Saving”, Review of Income and Wealth, 67(4), pp.810～834.
[29]
Lachowska, M. and M. Myck, 2018, “The Effect of Public Pension Wealth on Saving and Expenditure”, American Economic Journal: Economic Policy, 10(3), pp.284～308.
[30]
Li, J., X. Mi, C. Zhang and Y. Qin, 2024, “Social Pension Insurance and Household Risky Asset Investment: Evidence from China”, The Quarterly Review of Economics and Finance, 95, pp.219～233.
[31]
Van Santen, P., 2019, “Uncertain Pension Income and Household Saving”, Review of Income and Wealth, 65(4), pp.908～929.
[32]
Viceira, L. M., 2001, “Optimal Portfolio Choice for Long‐Horizon Investors with Nontradable Labor Income”, The Journal of Finance, 56(2), pp.433～470.
