Changes and Trends in Family Structure in Taiwan,1990 to 2010

Abstract

Taiwan has experienced significant demographic changes since the late 20th century, including decreasing fertility rate, increasing life expectancy, postponement of marriage and childbearing, reduced marriage rate, and rise in divorce. These demographic trends are likely to change the family structure. The objective of this article is to describe and summarize the changes in family structure between 1990 and 2010 in Taiwan using the Population and Housing Census data. During this period, the average household size decreased from 4.1 to 3.0 persons per household. Also, the share of couple households with unmarried children decreased by 15.1%, and the share of single-person households increased by 8.6%. Nuclear-family households, though they became less common, remained the dominant household type. Single-person households surpassed stem-family households as the second most common household type in 2000 and 2010. Households that are at greater risk of economic hardship, such as single-parent households and skipped-generation households, also increased their respective share during this period.

Keywords

family structure households Taiwan

Taiwan has experienced significant demographic changes since the late 20th century. Taiwan’s total fertility rate (TFR) has declined rapidly since the late 1950s (Figure 1), when the TFR was at the level of 6 to 7 (Department of Household Registration Affairs, 2016d). In 1984, the TFR further dropped below the replacement level (2.1), marking the completion of the first demographic transition (Y.-H. Chen, 2012; Department of Household Registration Affairs, 2016d; Selya, 2004). The TFR remained below the replacement level ever since, and even dipped to 0.90 in 2010. In addition to the subreplacement TFR, since the late 1980s, Taiwan has also shown some other characteristics of the second demographic transition (Atoh, Kandiah, & Ivanov, 2004; Lesthaeghe, 2010; Raymo, Park, Xie, & Yeung, 2015; Yang & Dong, 2007), such as postponement of marriage and childbearing, decreased marriage rate, and rise in divorce (Y.-C. C. Chen & Li, 2014; Department of Household Registration Affairs, 2016d) (Table 1). Moreover, the increasing life expectancy led to the growth of the elderly population, which exceeded 7% by 1993. Given the low fertility rate and the increased proportion of people aged 65 years or older, Taiwan has become an “aging” society (Department of Household Registration Affairs, 2016d).

Figure 1.

The total fertility rate in Taiwan, 1951 to 2015.

Table 1.

The Average Age at First Marriage, the Average Age of Mothers, the Crude Marriage Rate, and the Divorce Rate in Taiwan, 1990 to 2010.

Year	Average age at first marriage (years)		Average age of mothers (years)		Crude marriage rate	Crude divorce rate
Year	Men	Women	First birth	All births	Crude marriage rate	Men	Women
1990	29.0	25.8	25.4	27.0	7.1	6.3	6.3
2000	30.3	26.1	26.7	28.2	8.2	10.6	10.8
2010	31.8	29.2	29.6	30.6	6.0	11.3	11.6

Note. Data for average age at first marriage are from the Department of Household Registration Affairs (2016a). Data for average age of mothers are from the Department of Household Registration Affairs (2016b). Data for crude marriage rate are from the Department of Household Registration Affairs (2017b). Data for divorce rate are from the Department of Household Registration Affairs (2017a).

All these demographic changes are likely to have changed the family structure and household composition as well. In a report released in 2011, the Organisation for Economic Co-operation and Development (OECD) found that, because of the decrease in TFR over the past few decades, the OECD countries experienced a general decline in household size between the mid-1980s and mid-2000s and that a greater portion of households was without children (OECD, 2011). Taiwan is at the forefront of this trend. In addition, the postponement of marriage and the rise in divorce have led to an increase in single-person households and single-parent households (Yang & Dong, 2007).

The objective of this article is to document the changes in family structure in Taiwan during the late 20th century and the early 21st century. Using data from the 1990, 2000, and 2010 Population and Housing Census in Taiwan (hereafter Census), we describe and summarize the changes in household composition, as a proxy of family structure, during these 20 years. Although, unlike survey data, Census data only have basic information of the respondents, the strength of Census data is that they cover almost everyone in Taiwan (C.-N. Chen & Liu, 2002) and can provide a complete picture of families and households. In the following sections, we first document the changes in the number of households, household sizes, and distribution of household types between 1990 and 2010. Then, we describe the characteristics of household heads for each household type for the year 2000 and 2010. We do not include information on household heads for the year 1990 because such detailed information was not available in the 1990 Census.

Changes in Number of Households, Household Size, and Distribution of Household Types

The population in all households and the number of households steadily grew from 1900 to 2010 (Table 2). The number of households increased by 50.3%, from 4.9 million to 7.4 million. However, the population within these households only grew by 12.5%; from 19.7 million to 22.2 million. The reason for the discrepancy is that the average number of persons in a household decreased by 26.8% during this period: from 4.1 persons per household in 1990 to 3.3 in 2000 and to 3.0 in 2010.

Table 2.

Population in all Households, Number of Households, Average Household Size, and Distribution of Household Types in Taiwan in 1990, 2000, and 2010.

	1990	2000	2010	Differences (2010 − 1990)
Population in all households	19,682,565	21,383,503	22,151,900	12.5%
Number of households	4,932,763	6,470,225	7,414,371	50.3%
Average household size (persons)	4.1	3.3	3.0	−26.8%
Household type (%)
Nuclear family	63.6	55.1	54.3	−9.3
Couple with unmarried children	50.9	41.5	35.8	−15.1
Couple	6.9	7.8	11.0	4.1
Single parent	5.8	5.8	7.5	1.7
Stem family	16.2	15.7	16.5	0.3
Three generations	12.2	10.5	11.0	−1.2
Couple with married children	3.1	4.0	4.1	1.0
Skipped generation	0.9	1.2	1.4	0.5
Single person	13.4	21.5	22.0	8.6
Other	6.8	7.8	7.3	0.5

Note. Data retrieved from the 1990, 2000, and 2010 Population and Housing Census.

During 1990 and 2010, the nuclear-family household was the dominant household type in Taiwan, although its share decreased by 9.3%, from 63.6% in 1990 to 55.1% in 2000 and 54.3% in 2010. The decreasing share of nuclear-family households was mostly due to the decreasing share of couple households with unmarried children, which account for the majority of nuclear-family households and were also the most common household type. A couple with unmarried children accounted for half (50.9%) of households in 1990, but its share of total households fell to 41.5% in 2000 and 35.8% in 2010, a 15.1% decrease between 1990 and 2010. This trend was in line with the decreasing TFR in this period. The TFR was 1.81 in 1990, and it fell to 1.68 in 2000 and further to 0.90 in 2010 (Department of Household Registration Affairs, 2016c).

Not all types of nuclear-family households decreased in the share of all households during this period: couple households and single-parent households saw an increase in their shares between 1990 and 2010. The share of couple households increased by 4.1%, from 6.9% in 1990 to 7.8% in 2000, and further to 11.0% in 2010. Couple households consist of either empty nesters or couples who do not have children either voluntarily or involuntarily. Given the decreasing TFR, the increasing share of couple households is not surprising. The increase in empty nesters reflects changes in adults’ attitude and practice toward coresidence with parents (Chu & Yu, 2010; Yasuda, Iwai, Yi, & Xie, 2011). Traditionally, adult son, usually the oldest one, supports and lives with parents after marriage. However, recently fewer adults live with their parents, and more parents choose not to live with their children (Chu & Yu, 2010)—especially parents with more economic resources or higher socioeconomic status (Y.-J. Chen & Chen, 2012; Chu, Xie, & Yu, 2011).

The share of single-parent households remained constant at 5.8% between 1990 and 2000, but it increased to 7.5% in 2010. This pattern reflects the increasing divorce rate and decreasing remarriage rate between 2000 and 2010. Between 2000 and 2010, the divorce rate increased from 10.6% to 11.3% for men and from 10.8% to 11.6% for women (Table 1). In the same period, the remarriage rate decreased from 48.1% to 28.0% for men and from 16.1% to 11.9% for women (Department of Household Registration Affairs, 2017b).

The share of all stem-family households did not change much between 1990 and 2010, remaining at roughly 16%. The majority of stem-family households were three-generation households (10.5% to 12.2%), and along with couple households with married children, these two types of households accounted for about 15% of all households between 1990 and 2010. These two household types are relatively common in Taiwan compared with the Western societies because filial piety is highly valued in Taiwanese and Chinese culture, and adult children, especially sons, living with parents is regarded as a preferable way to fulfill the filial norm (Lin & Yi, 2013). Between 1990 and 2010, the share of three-generation households decreased by 1.2%, and the share of couple households with married children increased by 1.0%. This pattern might be attributed to the low TFR; fewer children were born in couple households with married children, and hence it was mathematically harder for these households to become three-generation households and many remained two-generation households.

Skipped-generation households refer to households composed of one or two grandparents and unmarried grandchildren only. In Taiwan, most skipped-generation households were formed because the sole parent or both parents were too ill to work, incarcerated, working in another place, or divorced (Hsu, 2016). The composition of only young children and the grandparents makes skipped-generation households more likely to be economically disadvantaged. Though skipped-generation households only accounted for a small portion of total households, the share increased from 0.9% to 1.4% between 1990 and 2010. Because of the vulnerabilities of skipped-generation households, the future trend of these households warrants more attention.

Last, the share of single-person households grew from 13.4% in 1990 to 21.5% in 2000, and further to 22.0% in 2010. With this increase, single-person households surpassed stem-family households as the second most common type of household in 2000 and 2010. This pattern possibly reflects the postponement of marriage, rise in divorce rates, and decrease in re-marriage rates during this period; many adults were not married and did not live with a spouse, and thus those adults were more likely to live alone.

To better understand the characteristics of each household type, in the following sections, we describe the household head’s gender, age, marital status, and employment status for each household type in 2000 and 2010. We first describe the characteristics of nuclear-family households followed by descriptions of the characteristics of stem-family and single-person households.

Nuclear-Family Households

Nuclear-family households include couple households with unmarried children, couple households (with no children), and single-parent households. Table 3 shows the characteristics of these nuclear-family household heads in 2000 and 2010. Regarding the gender of household heads, the majority of couple households with or without unmarried children were headed by men, but roughly two-thirds of the single-parent households were headed by women. Age-wise, the heads of couple households were the oldest of all nuclear-family households in 2000 and 2010: About 40% of the male household heads of couple households were aged 65 years or older, and close to half of the female household heads of couple households were aged 55 or older. In contrast, more than half of the heads of couple households with unmarried children and single-parent households were 35 to 54 years old.

Table 3.

Characteristics of Nuclear-Family Household Heads by Family Structure, 2000 and 2010.

	Couple household with unmarried children				Couple household				Single-parent household
	2000		2010		2000		2010		2000		2010
	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female
N	1,982,617	701,365	1,829,571	825,877	426,074	78,054	650,252	167,719	128,829	245,466	172,883	382,203
Row % of N by gender	73.9	26.1	68.9	31.1	84.5	15.5	79.5	20.5	34.4	65.6	31.1	68.9
Age in years (column %)
<25	0.5	1.2	0.3	0.5	0.5	1.2	0.1	0.3	2.0	1.0	1.2	0.5
25-34	14.0	19.6	8.4	10.9	14.3	16.3	7.7	10.6	13.8	9.1	12.2	7.3
35-44	38.0	45.3	28.3	35.7	13.0	14.7	9.6	12.6	29.1	31.4	25.8	24.3
45-54	31.6	26.6	36.6	36.6	13.4	21.1	18.2	26.5	27.0	32.7	31.5	34.3
55-64	10.6	5.9	19.8	13.5	18.5	23.3	24.3	27.9	12.5	14.4	17.3	20.7
≥65	5.4	1.5	6.6	2.8	40.4	23.4	40.0	22.1	15.6	11.5	12.0	12.9
Marital status (column %)
Unmarried (never married)	1.4	1.9	2.3	2.8	0.2	0.6	0.2	0.3	22.2	8.0	22.2	9.3
Married or cohabiting	98.6	98.1	97.7	97.2	99.8	99.4	99.9	99.7	0.0	0.0	0.0	0.0
Divorced or separated	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	50.3	38.0	58.5	50.3
Widowed	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	27.5	54.0	19.3	40.4
Employment status (column %)
Employed	86.6	62.3	82.1	62.1	60.4	45.4	53.8	45.3	68.9	58.9	72.0	56.3
Not employed	13.4	37.7	17.9	38.0	39.7	54.6	46.2	54.7	31.1	41.1	28.0	43.7

Note. Data retrieved from 2000 and 2010 Population and Housing Census.

As for marital status, not surprisingly, almost all the heads of couple households with or without children were married or cohabiting. The marital status of single parents, though, varied by their gender. Most single fathers were divorced or separated in 2000 and 2010. On the other hand, most (54.0%) single mothers were widowed in 2000, but in 2010 only 40.4% were widowed, and half of them were divorced or separated. The share of single fathers who were never married was more than twice that of single mothers (22.2% vs. 8.0% or 9.3%).

Regarding employment status, heads of couple households with unmarried children had the highest employment rate, followed by heads of single-parent households and couple households without children. The lower employment rate of single parents than the heads of couple households with unmarried children indicates that single-parent households and the children in them were more economically disadvantaged. Also, the employment rate and the age pattern indicate that by 2010 the majority of couple households without children in Taiwan were old empty nesters or old childless couples, who might be retired and thus did not work. Young couples without children were not the majority of couple households.

Between 2000 and 2010, some characteristics of nuclear-family household heads changed. First, more nuclear-family households were headed by women. Second, household heads were older: The share of household heads aged 44 years or less decreased for all types of nuclear-family households. Third, as mentioned earlier, based on the marital status of single mothers, more single mothers became a single parent because of divorce or separation instead of widowhood in 2010. Last, in couple households with or without unmarried children, fewer male household heads were employed in 2010, but the percentage of employed female household heads remained about the same. On the other hand, in 2010, more single fathers were employed, but fewer single mothers were employed.

Stem-Family Households

Stem-family households include three-generation households, couple households with married children, and skipped-generation households. Table 4 shows the characteristics of stem-family household heads in 2000 and 2010. Most of the stem-family households were headed by men, but a good share (43.2% to 45.2%) of skipped-generation households were headed by women. Most of the stem-family household heads were 55 years or older, and more than half of the skipped-generation household heads were 65 years or older. In terms of marital status, the majority of male heads of stem-family households were married or cohabiting. However, a good portion (37.9% to 51.7%) of female heads in these types of households were widowed. This pattern indicates that many women only became household heads in stem families because their spouses were deceased.

Table 4.

Characteristics of Stem-Family Household Heads by Family Structure, 2000 and 2010.

	Three-generation household				Couple household with married children				Skipped-generation household
	2000		2010		2000		2010		2000		2010
	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female
N	487,149	193,955	530,604	283,911	173,519	83,939	187,561	112,989	42,752	32,497	55,692	45,934
Row % of N by gender	71.5	28.5	65.1	34.9	67.4	32.6	62.4	37.6	56.8	43.2	54.8	45.2
Age in years (column %)
<25	0.2	0.3	0.1	0.1	0.3	0.4	0.1	0.2	1.2	1.3	1.0	0.9
25-34	5.4	3.2	3.8	2.8	5.4	2.6	4.2	3.0	4.7	3.1	2.9	1.8
35-44	15.7	8.6	14.1	7.8	8.8	6.4	8.2	4.4	4.2	3.6	1.0	0.7
45-54	19.1	20.5	18.9	16.4	18.6	23.6	18.7	20.3	5.4	9.0	4.9	7.9
55-64	27.0	30.6	25.5	31.7	29.6	27.9	32.9	32.5	23.5	26.8	27.8	30.4
≥65	32.6	37.0	37.6	41.2	37.3	39.0	35.9	39.7	61.2	56.2	62.5	58.4
Marital status (column %)
Unmarried (never married)	1.0	1.5	1.1	2.1	1.5	1.4	1.2	2.2	4.9	4.0	4.7	4.5
Married or cohabiting	89.0	44.8	87.7	46.2	89.3	48.9	90.0	53.6	82.6	39.9	85.3	40.4
Divorced or separated	2.3	4.4	4.6	8.2	1.4	3.7	2.1	6.4	2.0	4.4	2.9	9.3
Widowed	7.7	49.3	6.6	43.5	7.8	46.0	6.7	37.9	10.6	51.7	7.1	45.8
Employment status (column %)
Employed	61.9	28.8	54.6	26.9	59.7	30.7	52.5	27.2	41.2	21.2	33.2	16.2
Not employed	38.2	71.2	45.4	73.1	40.3	69.3	47.5	72.8	58.8	78.8	66.8	83.8

Note. Data retrieved from 2000 and 2010 Population and Housing Census.

Among stem-family households, the heads of three-generation households had the highest employment rate, followed by the heads of couple households with married children and the heads of skipped-generation households. More than half of the male heads of three-generation households and couple households with married children were employed. On the other hand, most male heads of the skipped-generation households and most female heads of all stem-family households were not employed. For instance, only 33.2% of grandfathers and 16.2% of grandmothers in skipped-generation households were employed in 2010. In fact, the percentage of household heads who were employed was lowest in skipped-generation households, compared with other types of households with unmarried children, including couple households with unmarried children, single-parent households, and three-generation households. This employment pattern indicates that members of skipped-generation households, grandparents and children, were at high risk of being economically disadvantaged.

Similar to the changes of the nuclear-family household heads between 2000 and 2010, the heads of stem-family households were older and less likely to be employed in 2010 than they were in 2000. Also, in 2010 more stem-family households were headed by women than in 2000.

Single-Person Households

Table 5 shows the characteristics of single-person households (i.e., people who live alone) in 2000 and 2010. In 2000, men accounted for more than half (53.7%) of the population living alone. However, the pattern reversed in 2010, when more women than men lived alone. With regard to age distribution, in 2000, most people who lived alone were aged 35 to 44 years (23% to 24%), regardless of gender. However, in 2010, most men who lived alone were aged 45 to 54 years (21.8%), and most women who lived alone were 65 years or older (25.5%). Overall, compared with 2000, the percentage of the population living alone got older in 2010. In terms of marital status, the majority of people who lived alone were unmarried, married, or cohabiting. A good share (23.3% to 23.7%) of women who lived alone were widowed in 2000 and 2010. Also, the percentage of people who lived alone who were divorced or separated increased in 2010: Only 9.0% of men and 10.0% of women who lived alone were divorced or separated in 2000, but the percentages increased to 16.7% and 16.9%, respectively, in 2010. This trend indicates that divorce and separation became more important causes for single-person households, regardless of gender. More than half of men who lived alone were employed in 2000 and 2010, although the percentage fell by 3.0%, from 65.9% to 62.9%, in 2010. However, for women who lived alone, a little more than half (51.0%) of them were employed in 2000, but the percentage fell by 6.8% (to 44.2%) in 2010.

Table 5.

Characteristics of Single-Person Household Heads, 2000 and 2010.

	Single-person household
	2000		2010
	Male	Female	Male	Female
N	747,073	645,220	772,207	857,763
Row % of N by gender	53.7	46.3	47.4	52.6
Age in years (column %)
<25	5.9	6.8	4.6	4.3
25-34	20.7	17.8	18.3	13.9
35-44	24.1	23.0	19.3	16.3
45-54	18.3	18.2	21.8	21.6
55-64	10.2	11.8	16.9	18.4
≥65	20.7	22.4	19.2	25.5
Marital status (column %)
Unmarried (never married)	41.9	28.7	38.7	28.2
Married or cohabiting	41.2	38.0	37.2	31.3
Divorced or separated	9.0	10.0	16.7	16.9
Widowed	8.0	23.3	7.4	23.7
Employment status (column %)
Employed	65.9	51.0	62.9	44.2
Not employed	34.1	49.0	37.1	55.8

Note. Data retrieved from 2000 and 2010 Population and Housing Census.

Discussion and Conclusion

Overall, between 1990 and 2010, due to the decreasing TFR in Taiwan, the household size got smaller and the share of childless households increased. The most significant change was the decreasing percentage of typical nuclear families, where the family consists of one couple and their unmarried children; the share was 50.9% in 1990 but fell to 35.8% in 2010. The above pattern is similar to the pattern observed in the OECD countries (OECD, 2011). Nevertheless, the relatively high share of multigenerational households in Taiwan was different from most OECD countries. The stem-family households remained about 16% of all households in Taiwan. In addition, Taiwan’s average household size in 2010 (3.0 persons per household) was higher than the average household size in 31 OECD countries in the mid-2000s (2.6 persons per household). Another significant change during this time was the increasing share of single-person households. Overall, although most people still lived with their spouses, children, or parents, less traditional households, such as single-parent households, skipped-generation households, and single-person households, became more common. In other words, the landscape of family structures has become more diverse in Taiwan. While the traditional two-parent nuclear family with children has declined dramatically, and while several types of less traditional family types are becoming more common, the proportion of families that maintain the traditional multigenerational stem-family structure has remained remarkably stable. This latter trend appears to be unique to Taiwan.

As for the characteristics of households and household heads, the heads of couple households with unmarried children were relatively young, and they had the highest employment rate. Couples in couple households without children were most likely to be childless older adults and empty nesters instead of young couples who did not have children. Single-parent households were characterized by the highest percentage of female household heads among all types of households. Also, compared with 2000, divorce and separation replaced widowhood as the main reason for single motherhood in 2010. The heads of stem-family households were older than those in nuclear-family households. Also, many women became the heads of stem-family households because of widowhood. The skipped-generation households were characterized by the old age and a low employment rate of the household heads. Last, in 2010, more women lived alone than men, and about 24% of women who lived alone were widowed. For men and women, divorce and separation became more important reasons for single-person households. The population living alone also got older.

Moving forward, what will households in Taiwan look like in the following decades? Will the trend we observed between 1990 and 2010 continue? Yang and Dong (2007) used a multidimensional household projection model to estimate the number and composition of households in Taiwan from 1990 to 2050. Similar to the trend we observed from the 1990, 2000, and 2010 Census, the share of one-generation households (single-person households and couple households) is projected to keep rising. Also, due to low fertility, half of all one-generation households are projected to consist of people aged 65 years or older. They also found that the share of couple households with children, including married and unmarried children, will decrease over time. Single-parent households, especially single-mother households, will also see a significant increase: The share of single-mother households will grow from 4.4% in 1990 to 20.1% in 2050. Although the study by Yang and Dong (2007) was conducted almost a decade ago and did not include skipped-generation households, their estimates are still good references for future changes in households because key demographic factors that determine household composition have not had any drastic changes from earlier estimates. Their estimation is similar to the OECD forecast. As many OECD countries are facing low fertility and an aging population, the OECD report forecasts that the number of single-person households and couples without children will increase from now to 2025-2030 (OECD, 2011). The increase in divorce and separation will contribute to the increase in the number of single-parent households (OECD, 2011).

It is important to monitor family and household structures. They enable us to anticipate the socioeconomic conditions of children, support for the elderly, and housing accommodations for the population (OECD, 2011). Although there are fewer children given the low fertility, there is an increased share of children living in more economically disadvantaged households, such as single-parent households and skipped-generation households. This trend warrants future research and attention from the government. Moreover, although elder care is still mainly considered a responsibility or even an obligation of families in Taiwan, adult children’s and parent’s willingness to live with each other has decreased over time; thus, how to provide proper care to elderly couples and old people living alone has become and will remain a pressing social issue as the population ages. Last, as the number of single-person households increases, there will be a higher demand for smaller and more affordable housing options.

In the end, we would like to address the limitations of using Census data. First, some households are not easy to categorize, and thus the Census uses looser criteria to categorize them. For instance, a couple household may include relatives other than the couple, such as a brother-in-law or sister-in-law. In some cases, the other adults are the head of the couple households. Therefore, we observed that a small portion of heads of couple households, of couple household with unmarried children, and of skipped-generation households were unmarried/never married/non-cohabiting. Nevertheless, roughly only 1% to 4% households had other adults as the head of household. Second, without access to the raw data of the Census, we could only describe the heads of households based on the tables and reports made by the Directorate General of Budget, Accounting, and Statistics, and no further analysis could be conducted.

Third, about 7% of the households were categorized as “other” households, but the Census did not provide further information about “other” households, such as their composition and characteristics of household heads. Fourth, increasing cohabitation is one of the key components of second demographic transitions (Lesthaeghe, 2014), but we could not estimate the changes in cohabitation households because the design of the 2000 and 2010 Census did not separate cohabiting couples from married couples. Last, the 2010 Census was a sampling survey that sampled 16% of the original census areas, and thus the 2010 Census may not be as accurate as a full census. Nevertheless, the 2010 Census was nationally representative by the Directorate General of Budget, Accounting, and Statistics’ design and postcensus estimation.

Although these limitations should be noted, the trends in family structure change in Taiwan are robust, and the limitations are not likely to affect either the decade-to-decade portrait of family structures or the observed changes over time.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Atoh

Kandiah

Ivanov

(2004). The second demographic transition in Asia? Comparative analysis of the low fertility situation in East and South-East Asian countries. Japanese Journal of Population, 2, 42-75.

Chen

C.-N.

Liu

P. K.-C.

(2002). Taiwan 2000 nian hukou pucha jieguo de pingjia: Changzhu renkou yu huji dengji renkou de bijiao fenxi [An evaluation of Taiwan census 2000: A comparison of census and registered populations]. Renkou Xuekan, 25, 1-56.

Chen

Y.-C. C.

J.-C. A.

(2014). Family change in East Asia. In Treas

Scott

Richards

(Eds.), The Wiley Blackwell companion to the sociology of families (pp. 61-82). Hoboken, NJ: Wiley.

Chen

Y.-H.

(2012). Trends in low fertility and policy responses in Taiwan. Japanese Journal of Population, 10, 78-88.

Chen

Y.-J.

Chen

C.-Y.

(2012). Living arrangement preferences of elderly people in Taiwan as affected by family resources and social participation. Journal of Family History, 37, 381-394. doi:10.1177/0363199012440948

Chu

C.-Y. C.

Xie

R.-R.

(2011). Coresidence with elderly parents: A comparative study of southeast China and Taiwan. Journal of Marriage and Family, 73, 120-135. doi:10.1111/j.1741-3737.2010.00793.x

Chu

C.-Y. C.

R.-R.

(2010). Co-residence and family size. In Understanding Chinese families: A comparative study of Taiwan and Southeast China (pp. 40-61). New York, NY: Oxford University Press.

Department of Household Registration Affairs. (2016a). Average age at marriage. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC. Retrieved from http://www.ris.gov.tw/346?p_p_id=populationdata_FastDwnld&p_p_lifecycle=2&p_p_state=normal&p_p_mode=view&p_p_resource_id=getFile&p_p_cacheability=cacheLevelPage&p_p_col_id=column-1&p_p_col_count=1&m4=y4s5&e3=xls&d5=

Department of Household Registration Affairs. (2016b). Average age of the mother at first birth. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC. Retrieved from http://www.ris.gov.tw/346?p_p_id=populationdata_FastDwnld&p_p_lifecycle=2&p_p_state=normal&p_p_mode=view&p_p_resource_id=getFile&p_p_cacheability=cacheLevelPage&p_p_col_id=column-1&p_p_col_count=1&m4=y2s4&e3=xls&d5=

10.

Department of Household Registration Affairs. (2016c). Fertility rates of childbearing age women. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC. Retrieved from https://www.moi.gov.tw/files/site_stuff/321/2/year/y02-04.ods

11.

Department of Household Registration Affairs. (2016d). Population policy data collection. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC.

12.

Department of Household Registration Affairs. (2017a). Divorce rates. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC. Retrieved from http://www.ris.gov.tw/346

13.

Department of Household Registration Affairs (2017b). Number and rates of birth, death, marriage and divorce. Taipei, Taiwan: Department of Household Registration, Ministry of the Interior, ROC. Retrieved from https://www.moi.gov.tw/files/site_stuff/321/1/month/m1-02.ods

14.

Hsu

Y.-H.

(2016) Gedai jiaoyang jiating shenghuo xuqiu ji fuli fuwu yanjiu diaocha chengguo baogao [A study on skipped-generation family’s need and welfare service]. Taipei, Taiwan: Social and Family Affairs Administration, Ministry of Health and Welfare, ROC.

15.

Lesthaeghe

(2010). The unfolding story of the second demographic transition. Population and Development Review, 36, 211-251. doi:10.1111/j.1728-4457.2010.00328.x

16.

Lesthaeghe

(2014). The second demographic transition: A concise overview of its development. Proceedings of the National Academy of Sciences of the United States of America, 111, 18112-18115. doi:10.1073/pnas.1420441111

17.

Lin

J.-P.

C.-C.

(2013). A comparative analysis of intergenerational relations in East Asia. International Sociology, 28, 297-315. doi:10.1177/0268580913485261

18.

Organisation for EconomicCo-operation and Development. (2011). Doing better for families. Retrieved from http://www.oecd.org/els/soc/doingbetterforfamilies.htm

19.

Raymo

J. M.

Park

Xie

Yeung

W.-J. J.

(2015). Marriage and family in East Asia: Continuity and change. Annual Review of Sociology, 41, 471-492. doi:10.1146/annurev-soc-073014-112428

20.

Selya

R. M.

(2004). Development and demographic change in Taiwan (1945-1995). Singapore, Singapore: World Scientific.

21.

Yang

C.-L.

Dong

Y.-J.

(2007). Taiwan de jiahu zucheng bianqian: 1990-2050 [Taiwanese household projection scenarios, 1990-2050]. Taiwan Shehui Xuekan, 38, 135-173.

22.

Yasuda

Iwai

C.-C.

Xie

(2011). Intergenerational coresidence in China, Japan, South Korea and Taiwan: Comparative analyses based on the East Asian social survey 2006. Journal of Comparative Family Studies, 42, 703-722.