Religion and Fertility in East Asia: Evidence from the East Asian Social Survey

AuthorDonata Bessey
Published date01 August 2018
Date01 August 2018
DOIhttp://doi.org/10.1111/1468-0106.12209
Pacic Economic Review,••:•• (2016) pp. 1-29
doi: 10.1111/1468-0106.12135
bs_bs_banner
RELIGION AND FERTILITY IN EAST ASIA: EVIDENCE
FROM THE EAST ASIAN SOCIAL SURVEY
DONATA BESSEY*Yonsei University, EastAsia International College
Abstract. This article analyses the effect of religious afliation on fertility in Japan, Korea and the
Republic of China (Taiwan). It adds to the sparse empirical evidence on the effects of religious afl-
iation on fertility in East Asia, for both Christian and other religions. It uses an identity-economic
model and analyses prescriptions among the different religions to derive testable hypotheses. In the
empirical section, the East Asian Social Survey and a generalized Poisson model are used to estimate
the effects of different religions on fertility. In line with theoretical predictions, the empirical results
suggest that a positive effect of Catholicism on fertility is still present in East Asia. In line with the
sparse previous empirical evidence on Buddhism, it seems to have no effect on fertility.
1. INTRODUCTION
There is a substantial amount of research on the fertility effects of different reli-
gions, analysing the effects of Judaism and different Christian denominations,
and using data from the United States and Europe. Usually, these studies nd
that there used to be a pronounced difference between Catholics and Protestants
in the past, with the former having had much higher fertility rates, but that this
difference has all but disappeared. See, for example, Lehrer (1996) for an over-
view. Different norms among religions are often cited as the explanation for
these differences in behaviour (e.g. the anti-contraception stance of the Vatican
as the reason for higher fertility among Catholics in the past).
Much less is known about the possible fertility effects of other religions, and
of the fertility effects of the Christian religions in other regions. The aim of this
article is to close part of this research gap by adding to the sparse previous liter-
ature. First, it analyses the possible fertility effects of Christian religions in East
Asian countries and compares their effects to those reported in the previous
literature on fertility effects of Christian religions in Western countries. Second,
it analyses the possible effects of other religions, especially Buddhism, on fertility
behaviour in East Asian countries, and compares the results to those reported in
the previous literature. Finally, it also analyses the effect of having a traditional
attitude on fertility.
For the empirical analysis, I use the 2006 wave of the East Asian Social Study
(EASS) and data for Japan, Korea and the Republic of China (Taiwan),
*Address for Correspondence: Yonsei University, East Asia International College, Yonseidae-gil 1,
220-710 Wonju, Republic of Korea. E-mail: dbessey@yonsei.ac.kr. I would like to thank Hyun-ok
Lee, Kimiko Osawa, James Paradise, Sungho Park, and session participants at the 14th International
Convention of the East Asian Economic Association in Bangkok for helpful comments. The East
Asian Social Survey (EASS) is based on the Chinese General Social Survey (CGSS), Japanese Gen-
eral Social Surveys (JGSS), the Korean General Social Survey (KGSS) and the Taiwan Social
Change Survey (TSCS), and is distributed by the EASSDA (East Asian Social Survey Data Archive).
I gratefully acknowledge nancial support from Yonsei Universitys New Faculty Research Grant.
Pacic Economic Review,••:•• (2017)
doi: 10.1111/1468-0106.12209
© 2017 John Wiley & Sons Australia, Ltd
© 2017 John Wiley & Sons Australia, Ltd
Pacific Economic Review
, 23: 3 (2018) pp. 517–532
doi:10.1111/1468-0106.12209
countries well known for their ultra-low fertility rates (Jones et al., 2009). Fertil-
ity is measured as the number of children born to married women in their rst
marriage, with two additional robustness checks. The rst one limits this sample
to women above age 45 who should have completed their fertility decisions, and
the second one uses the respondents ideal number of children as a dependent
variable to take into account possible differences between intended and actual
fertility. Estimation results suggest that being Catholic is associated with higher
fertility, while there are no effects for being Buddhist or Protestant, and no
effects of having a traditional attitude either. Being a follower of other Christian
religions is associated with signicantly lower fertility.
The remainder of this paper is organized as follows. Section 2 provides a brief
review of the relevant literature. In Section 3, I give a brief discussion of theoret-
ical considerations, and in Section 4, I describe the data set used for the empir-
ical analysis. Section 5 presents and discusses the estimation results, while
Section 6 concludes.
2. LITERATURE REVIEW
It is a well-known empirical fact that East Asia has ultra-low fertility rates, and
they have been the focus of much research (see e.g. Jones et al. (2009) or Suzuki
(2013)). Some of the reasons discussed in the literature include the trend to
marry later or to remain single (Jones and Gubhaju, 2009), the high costs of
education per child which prevent couples from having more than one or two
children (Anderson and Kohler (2013) for the case of Korea), or institutional
features of married life which become less and less attractive compared to living
as a single person with ones parents, especially for women (Bumpass et al.
(2009) for the case of Japan). Those well-documented low fertility rates and their
underlying causes should be seen as the background for the following discussion
of previous relevant literature on the effect of religion on fertility. I restrict the
following literature review to the few studies that also focus on non-Christian
religions and/or on non-Western countries.1For an overview of results for
the United States, see Lehrer (2004) and Zhang (2008), and for a comparison
of results from Europe and the United States, see Frejka and Westoff (2008).
Using case studies, McQuillan (2004) provides a useful starting point for
understanding the conditions under which religion affects fertility. He shows
that religion matters for fertility when the religion actually has norms that guide
fertility behaviour, when the religion is able to enforce these norms and to com-
municate them to members, and when membersreligious afliation matters
greatly in their daily lives. When these three conditions are not met, it is unlikely
that religious afliation matters for fertility. McQuillans work can help to
explain why results for the effects of religion on fertility vary widely for different
countries and different religions.
1In addition to religious afliation, other studies also analyse the effect of religiosity, measured as
the frequency of religious attendance or praying, on fertility. As there is no information on religiosity
in the EASS, I will not discuss previous research on religiosity here.
D. BESSEY2
© 2017 John Wiley & Sons Australia, Ltd
In the following paragraph, I summarize results from previous studies that fo-
cus on the effect of Buddhism before turning to the effects of Christian religions
in non-Western countries. For the effects of Buddhism on fertility, Knodel et al.
(1999) analyse differences between Buddhist and Muslim fertility in Thailand
and nd that Buddhists have experienced a more signicant fertility transition
than Muslims (i.e. they have lower fertility rates). De la Croix and Delavallade
(2014) use micro-level data from South East Asia (Cambodia, Indonesia,
Malaysia, the Philippines, Thailand and Vietnam) and nd that being Buddhist
increases fertility compared to being religiously unafliated. Kojima (2014),
however, uses data from the Survey on Comparative Study of Family Policies
in East Asia: South Korea, Singapore and Japanand nds only moderate pos-
itive effects of being Buddhist on fertility for Korean men. Skirbekk et al. (2015)
analyse the effect of Buddhism on fertility in Cambodia, India, Japan,
Mongolia, Nepal and Thailand. Controlling for a host of background variables,
they nd that Buddhism is not related to nor signicantly decreases fertility,
except in Cambodia. Interestingly, the percentages of Buddhists among the pop-
ulation vary widely in the countries that they analyse, suggesting that the effect
does not depend on the minority or majority status of the religion. The scarce
empirical evidence for the possible effects of Buddhism on fertility shows widely
different results for different countries and samples.
For Christian religions in non-Western countries, de la Croix and Delavallade
(2014) nd that being Catholic or Protestant increases fertility compared to being
religiously unafliated. Kojima (2014) nds that for Catholicism and Protestant-
ism, there are no effects on fertility in Korea and Singapore. Heaton (2011) uses
data from 30 developing countries and compares Catholic and Protestant fertility
behaviour. He nds only small differences between Catholics and Protestants and
concludes that those differences vary widely between the countries included in his
sample, leaving large parts of the variation unexplained by his regression models.
As for Buddhism, the effects of Christian religions on fertility in the previous
studies summarized here differ widely between different countries.
Again, the theoretical model by McQuillan (2004) might explain the mixed
previous evidence for the effects of religion on fertility in the existing literature.
While all of the analysed religions have prescriptions on fertility, their capability
of communicating them to members and enforcing them, as well as individual
importance, might differ. Varying degrees of enforcement (capability) from
religious institutions and individual differences in the importance of religion in
general or their fertility-related prescriptions in the studies cited above might
be able to explain the mixed evidence for the effects of religions on fertility.
While this article cannot control for all of these factors either, it has at least
two advantages over previous research. First, the present study uses a sample
of three countries that, despite important differences, share the cultural inuence
of Confucianism, and for which its principles are still present in everyday life
(Chang, 1997; Tu, 1996), so the possible confounding effects of other cultural
inuences should be smaller. Second, it uses additional information on an
individuals traditional attitudes that might matter for fertility decisions as well,
but have not been analysed in the previous literature.
RELIGION AND FERTILITY IN EAST ASIA 3
© 2017 John Wiley & Sons Australia, Ltd
© 2017 John Wiley & Sons Australia, Ltd
D. BESSEY
518

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT