Skip to main content

Family support or social support? The role of clan culture


This paper examines the impact of cultural norms on arrangements made for old-age support. Using data from a recent national household survey in China, I show that clan culture is significantly associated with a set of individual values about the purpose of childbearing. I then find that, among older people in rural China, clan culture is positively related to birth rates and the likelihood of having a son, coresidence with adult or married sons, and receiving financial transfers from non-coresident children. Finally, I find a significantly negative relationship between clan culture and enrollment in social pension programs. The overall results indicate that cultural norms have a significant influence on arrangements for old-age support.

This is a preview of subscription content, access via your institution.


  1. Calculations are based on data from the China Health and Retirement Longitudinal Study (CHARLS), which is a biennial survey aimed at collecting a nationally representative sample of Chinese residents aged 45 years and older and is designed to be complementary to the Health and Retirement Survey (HRS) in the USA. More details about the survey can be found at its official website:

  2. See Table 1 in Ebenstein (2014).

  3. Using a life cycle analysis, Liu et al. (2005) show that on-the-job wage-rising potential and consumption smoothing may be the key factors in explaining why many young eligible individuals choose not to participate in means-tested welfare programs.

  4. For instance, Almond et al. (2013) and Ebenstein (2014) examine the effect of culture on sex selection, and Borck (2014), Fernandez (2007), and Fernandez and Fogli (2009) examine the effect of culture on fertility.

  5. Studies on the determinants of living arrangement usually focus on factors like family structure (Aquilino 1990), income (Kahn et al. 2013; Manacorda and Moretti 2006), and assistance needs (Choi 2003; Matsudaira 2016; Ward et al. 1992).

  6. In the patrilineal family system, the timing of property transfer and the transfer of power from one generation to the next varies depending upon the marriage dates of the younger generation, at the earliest, to the death of the older generation, at the latest. In China, the ideal time is at the death of the older generation (Harrell 2002)

  7. According to an interview with NRPS officials, this binding policy was removed shortly after the first round of the pilot in order to promote the scheme’s take up among China’s elderly population.

  8. Please see Lei et al. (2013) for more institutional details about the NRPS.

  9. I also use a sample of elderly people, defined as persons 60 years of age and older, as a robustness check and obtain quite similar results. These results are available upon request.

  10. In formal sectors, such as governments and state-owned firms, 50 is also the mandatory retirement age for female workers.

  11. For the fertility behavior regressions, I also control for the family planning policy regime, which is an important determinant of fertility.

  12. It is worth noting that the number of children is probably endogenous, as parents who prefer a son may want to give birth to an additional child if they do not yet have a son.

  13. For example, Chen et al. (2013), Ebenstein (2010), Ebenstein and Leung (2010), Das Gupta (2006), Li et al. (2011), and Qian (2008)

  14. Ulker (2008) shows that coresiding with others effectively supplements social pensions, and private saving, and helps the elderly to smooth consumption.

  15. One possible concern is that married sons are still not necessarily economically independent. As a further robustness, I examine coresidence with adult sons 30 years of age and older. In this age group, adult sons are arguably economically independent. I obtain quite similar results, which are available upon request.

  16. I did not control for household income in columns 1 and 2, because this variable is highly endogenous and may have a “bad control” problem. More importantly, we can also tell whether the results are robust through comparing the results with and without controlling for household income.

  17. To explicitly test the substitutes between children and social pensions, I estimate the impact of the number of children on pension participation using the instrumental variable approach, where I instrument the number of children using the presence of an ancestral temple. The two-stage least squares (2SLS) estimation results, as reported in Appendix Table 10, suggest a significantly negative relationship between the number of children and pension participation. However, one should keep in mind that the presence of a temple is not an ideal instrument, as it may directly affect the take-up of social pensions and therefore violates the exclusion restriction assumption.

  18. It is worth noting that some costs are inevitable as the elderly need to apply to local agents and complete the necessary forms.


Download references


I would like to thank Wei Huang as well as seminar participations at Fudan University, Shanghai University of International Business and Economics, and the 2017 Chinese Economists Society (CES) annual conference for their helpful comments. I am also grateful to three anonymous referees and the editor Junsen Zhang for helpful comments and suggestions.


This study was funded by the National Natural Science Foundation of China (71503282), the Young Elite Teacher Project of Central University of Finance and Economics (QYP1609), and the Fundamental Research Funds for the Central Universities.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Chuanchuan Zhang.

Ethics declarations

Conflict of interest

The author declares that he has no conflict of interest.

Additional information

Responsible editor: Junsen Zhang



Fig. 1
figure 1

The percentage of communities having temples in each county in the CFPS. The calculations are weighted using sampling weights

Table 8 Clan culture and sex ratios at the county level
Table 9 Clan culture and fertility patterns by the sex of existing children
Table 10 Children and social pension enrollment: 2SLS estimations

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhang, C. Family support or social support? The role of clan culture. J Popul Econ 32, 529–549 (2019).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Culture
  • Clan
  • Living arrangement
  • Intergenerational transfer
  • Social pension
  • China

JEL classification

  • H55
  • J13
  • J14
  • Z13