An increase in the retirement age in China: the regional economic effects

ABSTRACT

China’s pension system is in need of comprehensive reform. One measure on which we focus is to increase the retirement age. It is likely that a change in retirement age will have significantly different effects across China’s regions. Interregional disparities are already very substantial in China and it will be important to know how changes in pension arrangements will affect disparities. We consider four policies to increase the retirement age from 60 to 61. They differ according the use made of the extra revenue generated by the policy. All four policies increase welfare and reduce the interregional welfare gap.

KEYWORDS:

• China
• pension system
• retirement age
• regional impacts

JEL CLASSIFICATION:

• R10
• R23
• R28
• H70
• H75

Ackownledgements

We are grateful for the comments received at the Fourth International Workshop on Regional, Urban and Spatial Economics at Tsinghua University, especially those of our discussant, Roberta Capello.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ In practice, most governments regulate the age at which pension benefits may be accessed rather than the retirement age as such. However, there is evidence that there is a close relationship between the two; see, e.g. Okumura and Usui (2014).

² For further exploration of regional disparities as well as a discussion of regional policy, see Chen and Groenewold (2013, 2014), Lin, Lin and Ho (2013), Rizov and Zhang (2014) and Herrerías and Monfort (2015).

³ While our structure drastically simplifies the structure of Chinese taxes, we would argue that it captures the salient features; see Jin, Qian and Weingast (2005), Zhang (2006), Shen, Jin and Zou (2012), and Shen and Zou (2015) for information on aspects of the Chinese public finances.

⁴ The complete specification of the model, including equations and the variable definitions, is given in Appendix.

⁵ We use ‘private’ to mean non-general government and include SOEs and other government-owned producers in these sectors.

⁶ Since there are no intermediate goods in the model, the VAT is equivalent to a tax on the value of final output, all of which is consumed and, since households are assumed to spend all their income, the VAT is also equivalent to an income tax. We therefore model households as ‘paying’ the VAT.

⁷ See Cheng and Selden (1994) for a general description and history of the hukou system and Bao et al. (2011) for a more recent discussion and modelling considerations.

⁸ See, e.g. Cai and Cheng (2014) and Song (2009).

⁹ We note that there have also been proposals to (eventually) raise the retirement age to 65 and we have re-run our simulations under this assumption. In this case, the shocks to r_I and r_C are a little over 4 times the shocks for the 60–61 case and the simulation results reflect this: by and large they are about 4–4.5 times those for the base case. We report the details in Appendix.

¹⁰ Details of the computation of shock sizes are provided in Appendix.

¹¹ Note, though, that the analysis ignores the likely disutility of working longer or the utility of retirement. That people value retirement as such is supported by the evidence that people self-identify as being happier on retirement as long as income does not fall, see Barrett and Kecmanovic (2013).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China; [71173092]; Ministry of Education of China; [NCET-12-0681].