Do State Pensions Crowd out Private Transfers? A Semiparametric Analysis in Urban China

Abstract:

This article investigates the relationship between state pension and intergenerational monetary transfers received by older parents in urban China. We used data from the Follow-up Sampling Survey of the Aged Population in Urban/Rural China (FUSSAPUR), a nationwide study of people age 60 and older from 20 provinces. Results indicate that the amount of pension was negatively correlated to the probability of receiving transfers, which are consistent with both altruism and exchange hypotheses. We apply a semiparametric method to identify the nonparametric responsiveness of transfers to recipients’ pension while still taking into account how the (linear) covariates affect such responsiveness. We found crowding out was not an important feature of transfer behavior, but crowding in was. There was initially a negligible negative relationship between the amount of transfer received and pension for the elderly at low pension income level. The correlation became significantly positive for older parents at median and high pension income level. Our findings suggest a coexistence of transfer motives, where the exchange motive and its implication, crowding-in effect of pension, dominate.

Acknowledgments

We would like to thank Albert Park, Maria Porter, Sarah Harper, Andreas Hoff, Kenneth Howse, and John Knight for their valuable advice. We would also like to thank the journal editor, Xiaogang Wu, and three anonymous reviewers for their helpful comments and suggestions during the revision process. All errors are ours.

Notes

Even in a one-to-one crowding-out case, it creates deadweight losses because government transfer programs usually involve administrative costs.

Data from the last two censuses show that the proportion of people 60 years of age and older grew from 10 percent in 2000 to 13.3 percent in 2010 (National Bureau of Statistics of China [NBSC] 2012).

“Normal goods” is an economics term referring to any goods for which demand increases when income increases and falls when income decreases but price remains constant (e.g., holidays, cars, and hi-tech products). A counterpart of normal goods is inferior goods, the demand for which decreases when income increases (e.g., cheap cars and potatoes).

The opportunity cost here refers to forgone leisure. For example, wealthier parents are able to pay for their holiday trips, and are willing to do so rather than babysitting their grandchildren, while poor parents may have to babysit their grandchildren in order to make children pay for their holiday trip. Alternatively, we can understand if workers becomes wealthier, they will have higher expectations in terms of remuneration. Because they are the only workers in the labor market (remember, the theory assumes parents are the only providers of services), their wages will eventually increase.

For more detailed descriptions of the data, please see Guo and Chen (2009).

This makes the survey more formal and is likely to make interviewees feel more responsible and less likely to give false responses. We believe that, in this sense, the data here is reliable, and should not be subjected to more measurement errors than any other survey data in the same field.

“Urban areas” in the survey only include cities and suburbs of the city, rather than towns or suburbs of towns, as in other statistics (e.g., China Statistics Yearbooks).

Given any variable L, ΔL indicates the difference between L_r and L_r−1, where r indicates a sample, and r − 1 indicates the one next to it (ranked by Ir).

Specifically, we use a bandwidth of 0.45 with observations weighted using a tri-cube weighting function as calculated by the Locally Weighted Smoothed Scatterplot (LOWESS) command in Stata. The LOWESS estimator was developed by Cleveland (1979), and is used as opposed to the kernel estimator because it does not suffer from bias at the endpoints.

Additional information

Notes on contributors

Taichang Chen

Taichang Chen is an assistant professor at the School of Public Administration and Policy, Renmin University of China, Beijing. Taichang earned his Bsc in Economics from Queen Mary, University of London and received his master (MPhil in Economics) and doctorate (DPhil in Sociology) degrees from the University of Oxford. His research focuses on the socioeconomic aspects of population aging in China. In particular, he considers the impact at societal and individual levels of the age-structural shift from predominantly young to predominantly older societies, on the well-being of older people in China.

George W. Leeson

George W. Leeson is co-director of the Oxford Institute of Population Ageing, University of Oxford, United Kingdom. Dr. Leeson's main research interests include sociodemographic aspects of ageing populations, covering both demographic modeling of population development, and the analysis of national and international data sets.

Jun Han

Jun Han is a postdoctoral research fellow at Georgetown University. His research interests include sociology, nonprofit studies, social innovation, public policy, and China studies.

Shuai You

Shuai You is a final-year undergraduate student at the Maxwell School of Citizenship and Public Affairs, Syracuse University, New York.