![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
ABSTRACT
This study examines the relationship between China’s contraception policy and married women’s (20–49 years) choice between long-acting and short-acting contraceptive methods during two periods–1980–1994 and 1995–2012. The aim is to examine the link between strictness of the contraception policy and married women’s contraceptive choice. Using data from the 1988 and the 2006 National Family Planning and Reproductive Health Surveys, we estimated the effect of contraception policy in the tightened and the loosened policy periods using a permutation test. The results show that while contraception policy promoted long-acting contraceptive methods, many married women themselves preferred short-acting contraceptive methods. During the tightened policy period, we found married women on average were 2.7 times more likely to use long-acting methods. The effect was only 1.7 times during the loosened policy period. The effect was also parity dependent. The more stringent the contraception policy was, the more likely married women used a long-acting contraceptive method. This study provides the first ever proper estimation of China’s contraception policy effects during the two periods.
Acknowledgement
The analysis was done as part of Wang's postdoctoral research at the University of Illinois, and was supported by the Program for New Century Excellent Talents in University, China; and the Program of the National Social Science Foundation of China (14BRK025). The authors contributed to the paper equally, and the order of the authorship was determined by a random draw.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. We use the term ‘contraceptive choice’ to mean the use of a particular contraceptive method even though women’s choices may be constrained, in the sense of ‘constrained contraceptive choice’ as described by Barrett and Buckley (Citation2007), especially during the tightened policy period.
2. The long-acting contraceptive methods include female and male sterilisation, the intra-uterine device (IUD) and subdermal implants. We grouped sterilisation and the IUD together because the government did not allow women to remove the IUD on their own at least in the tightened period. The short-acting contraceptive methods include oral contraception, condom usage, contraceptive jelly, contraceptive suppositories, withdrawal and other methods (Ren & Zheng, Citation2006; Wang, Citation2012a, Citation2014a).
3. How ‘tight’ or ‘loose’ a policy period can only be relative. We use the term of ‘tightened’ or ‘loosened’ to refer to the policy period under consideration as compared to its preceding period.
4. The division into of the tight and loosened policy periods was based on the following considerations: Official documents indicating a sharp policy change: The categorisation of the tight and loosened policy periods was firstly based on the two official documents: an open letter from the Central Committee of the Communist Party of China (CPC) to advocate the one-child policy in 1980, and a client-centered contraceptive approach of informed choice in 1994. Long-acting contraception has been one of the prioritised methods of preventing unauthorised births under the Chinese family planning policy since the early 1980s (Greenhalgh, Citation1994; Wang, Citation1988, Citation2012a, Citation2014a, Citation2014b). However, its implementation changed from a mandatory approach during the 1980s and early 1990s to a non-mandatory approach from the mid-1990s onward (Attane, Citation2002; Wang, Citation2014a, Citation2014b). This transition was made possible by two contraception policies with stringency levels looser than the one in the 1980s (Wang, Citation2012b, Citation2014a, Citation2014b). Based on the conventions of official documents and previous studies, this study divides the years from 1980 to 2006 into two periods: the tightened policy period of 1980–1994 and the loosened policy period of 1995–2012 (see e.g. Peng, Citation1997; Zhang, Citation2007; Wang, Citation2012a, Citation2014a).
5. Strictly speaking, induced abortion is not a method of contraception but a post-conceptive means to remove the pregnancy. Because its effect is rather immediate, it was considered to be in the same category with the other three contraceptive operations by local birth control officials.
6. Based on the method for calculating contraception rates used by the United Nations. When a woman reports her or her partner’s using a contraceptive male or female oriented method, it is counted as the utilisation of a female contraceptive measure (that is, a male contraceptive measure is counted as a female contraceptive measure), and the denominator is the total population of eligible childbearing-aged females (aged 15–49 years). The sampling unit was woman for the surveys though a man could answer the questionnaire as a proxy when his spouse was not available.
7. The Program of Action (‘Cairo Program’) of the ICPD offered solutions to population-related social problems and set limits on state population policies through its recognition of informed choice. As a member state in the ICPD, China pledged to initiate the informed choice policy, which meant that its population policies would empower women’s reproductive health rights by offering women access to voluntary decision making on contraception and child-bearing and fertility regulation.
8. The IC policy included the provision of updated, scientifically accurate and easy-to-understand information on contraceptive methods and standard technical services. First officially recognised in 1995 by the State Family Planning Commission, the reorientation of family planning switched policy implementation from a target-driven to a client-centered approach.
9. According to the Cairo Conference on Population and Development (International Conference on Population and Development held in Cairo in September 1994), informed choice is twofold. First, couples or individuals should be able to determine the number of children that they wish to have, space the births of those children freely, and are responsible for their childbearing decisions. Second, couples are entitled to informed choice concerning contraception. That is, men and women have the right to relevant information on and choice of safe and effective contraceptive methods. Since the ICPD in 1994, most countries have initiated informed-choice family planning, with the exception of China, which launched a client-centered policy of informed-choice contraception only.
10. The data used in this research are from the published secondary statistics. There is no information in the data sources that can be used to identify respondents, and human subject protection is not an issue here. Furthermore, as a state organisation, the NPFPC has an IRB to approve data collection related to human subjects. The above disclaimers also apply to the data from the National Family Planning and Reproductive Health Surveys in China (NFPRHS) used in this study.
11. Contraceptive prevalence is defined as the percentage of women who are currently using or whose sexual partner is currently using at least one method of contraception, regardless of the method used. It is usually reported for married or in-union women aged 15–49 (United Nations, Citation2012). There is no reporting error in married couples on contraceptive methods use in 1988 and the 2006 National Family Planning and Reproductive Health Survey (Department of Planning and Finance of National Population and Family Planning Commission, Citation2003; Wang, Citation2012b).
12. While the 2006 survey may underrepresent unmarried women, our analysis is done on married women. This dataset has been widely used by many studies (see e.g. Wang, Citation2014a, Citation2014b).
13. The percentages of the non-contraceptors are 2.6 per cent in 1988 data and 3.4 per cent in 2006 data, respectively.
14. Because using time period as the instrument is problematic, we rely on a surrogate of the treatment represented by the interaction between ethnicity and parity. The F-test in a 2SLS analysis was 14.63, which was greater than the critical value of 10. Therefore, the null hypothesis of a poor instrumental variable is rejected.
15. Comparisons of logit coefficients across models can be affected by differences in residual variances (Allison, Citation1999). To make certain our comparisons here and later in the section are valid, we conducted a straightforward test of equal residual variances from the counterpart linear probability models and found no significant difference in residual variance by Levene’s test between the pairs of models involved in our comparisons in this section.
16. Note, however, the estimate for non-Han married women with three or more children is insignificant and smaller than that for non-Han married women with two children (and is significant). This can be explained by the observation that those with three or more children were also older and would not feel the need for a long-acting contraceptive method especially since they were not official required to use such a method. They could simply choose an easier, short-acting method.