The unexpected effect of social status on reproduction: a case study in Joseon Korea from the nineteenth to the twentieth centuries

Abstract

We examine how parents have made decisions about the number of children they have, given their social status in accordance with residential location (either urban or rural areas) and time (either the pre-modern or modern periods). We use two sets of microdata – Jokbo and Jejeokbu – spanning the early nineteenth to mid-twentieth centuries in Korea. Combining the two data-sets, we use multiple imputation to fill the missing entries of some observations and apply a Poisson regression model on the augmented data. Our empirical results reveal statistically significant evidence that higher socioeconomic status is related to having more children. Additionally, our findings indicate that: (1) all else being constant, among high-status people, rural residents had more children than urban families; (2) for people born between 1800 and 1945, those born closer to the 1940s tended to have fewer children; and (3) during modernization, there was still a significant trend for high-status families to have more children.

Keywords:

• Socioeconomic status
• reproduction
• Jokbo
• Jejeokbu
• multiple imputation
• Poisson regression

Acknowledgement

Earlier versions of this article were presented to 41st Annual Meeting of the Social Science History Association at Chicago, Illinois, United States. We would like to thank anonymous referees for their valuable suggestions, Gyudong Kim for his research assistance, and Byung-giu Son who allowed authors to use the raw data for this article.

Notes

1. Stys (1957) showed that the fertility choice of peasant women is strongly correlated with economic conditions and the expected productivity of a household with one more child.

2. Hajnal (1965), Schofield and Wrigley (1981), and Wrigley and Schofield (1989) reported conventional outcomes that economic development has encouraged marriage at older ages and decreased the cohort-level marriage rate.

3. Some researchers may agree that the relationship between fertility and socioeconomic status remains positive (Fieder et al., 2005; Stys, 1957).

4. Much literature on the demographic transition in Korea remarks on a drastic decline in the fertility rate since the early 1960s alongside rapid industrialization and urbanization. A major socioeconomic cause of continuous low fertility is related to an increase in the cost of raising a child (e.g. Kim, 2009; Kye & Park, 2016).

5. According to Dribe et al. (2014), despite the difficulty in finding a definite pattern, a higher social status could have been related to a later transition from leaving home to reproduction in Europe and to an earlier transition to reproduction in Asia. While persisting with their attitudes and values toward the traditional sense of a child, Asian families had considered a son as a critical factor in promoting a family’s values in a patriarchal society, resulting in a slow fertility transition relative to Europe. In fact, patrilocal residential arrangements were found in Chinese family formations in Lee and Campbell (1997). We also find consistent evidence from Wolf (1984) and Tsuya, Wang, George, and Lee (2010).

6. The annual urban growth rate from 1935 to 1940 was 129.4% according to Kim (1994).

7. There were two land systems in both societies: a stipend land system and a rank land system (Lee, 2004a).

8. According to Noh et al. (1485), there was a limited number of officials (approximately 2300) in the Great Code of Administration of Joseon (Gyeongguk–Daejeon). Among these positions, only a few were considered highly prestigious. While the status of yangban, the two strata of the literary and military elites, was inheritable, de facto social positions were determined by rankings within government offices. This means that most elite families had to compete with each other fiercely to obtain government offices. This situation was very similar to that of the powerful elite families in Tang China (Bol, 2003).

9. For example, Han (2013) showed that 24% of the total population were officials in early Joseon Korea in contrast to 53% in the late periods of the dynasty. This empirical evidence is consistent with the research on pre-modern China in the study by Ho (1962).

10. In Lee’s (2004b) study, only 30% of elite families could have at least one of their number in a government office in Joseon Korea.

11. Some studies have treated the weak points of the genealogy as the data source of historical demography and quantitative analysis (Hanley & Wolf, 1985; Harrell, 1987, 1996; Park & Lee, 2008; Telford, 1990). In this article, we try to compensate for the weakness of the genealogy through statistical methodologies.

12. Park and Lee (2008) presented the pros and cons of Korean genealogy as a source of historical demography. Most of the description of Korean genealogy given here originates from this study.

13. The description of Jejeokbu data is based on Son and Lee (2013).

14. Other countries, colonial Taiwan for example, created a civil register with details of temporary migration data and lifetime events from birth to death.

15. 60% of people in Jejeokbu were removed from Hojeok at the termination of their lives. Migration (15%), a family’s branching out or separation (10%), and marriage (10%) followed death.

16. Myrskyla, Kohler, and Billari (2009) argued that industrialization with higher incomes correlated to a lower population. Whether industrialization leads to a lower population, or a lower population leads to industrialization, is an ongoing debate.

17. The searching mechanism programmed by R is available from the authors on request.

18. Our crude birth rates (CBRs) from 1809 to 1889 are consistent with those from 1910 to 1960 suggested by Kim (1994). We do not believe that our estimates of CBRs from 1929 to 1969 represent the correct values because we have such a limited number of observations of a particular region in recent years. The total fertility rate (TFR) based on our data possibly overestimates the rates compared with the suggestions after the 1920s in Kim (1994). The overestimation is due to the lack of data for individuals who were younger than 15 years old and female. Despite the limitation of our data-set, the CBR and TFR trends explain the smooth demographic transition in the late Joseon dynasty in accordance with a decrease in CBR and TFR. See Lee and Son (2012) for overall mortality rates in Joseon Korea.

19. The urban city graveyard theory describes unsustainable urbanization before the mid-seventeenth century in which population loss due to deaths in cities exceeded population growth due to births. Urban populations grew largely because of rural to urban migration until the advent of public health measures. Wrigley (1967) argued that high (infant) mortality rates in an overly crowded city (for example in the seventeenth and eighteenth centuries in London) were due to over-urbanization.