1 For a Swedish example, see Rogers, J. and Norman, H., eds., The Nordic family: perspectives on family research, Meddelanden från familjehistoriska projektet, no. 4, Department of History, University of Uppsala, 1985.Google Scholar

2 The following papers presented at the International Union for Scientific Study of Population seminar on ‘The Demography of the later phases of the family life-cle’, held in Berlin in 1984, were very relevant to my topic: Kathleen Kiernan, ‘The departure of children’; Karl Ulrich Mayer, ‘The process of leaving home: a three-cohort comparison for West Germany’; Karl Schwarz, ‘When do children leave the home of parents: an analysis of cohort data in the FRG for the years 1972–1982’; Christabel Young, ‘The effect of children returning home on the precision of the timing of the leaving-home stage of the family life-cycle’. Other recent studies of interest are Sandra, Hofferth, ‘Children's life course: family structure and living arrangements in cohort perspective’, The Urban Institute (Washington, D.C., 1981)Google Scholar, unpublished mimeograph; Frances, Kobrin and Julie, DaVanzo, ‘Leaving home and the transition to adulthood’, Demography (in press)Google Scholar; Stefan, Lundberg and Arne, Modig, ‘Ungdomars etablering på bostadsmarknaden’, Report R 182, National Council of Building Research (Stockholm, 1984)Google Scholar. For the historical perspective, see Richard, Wall, ‘The age at leaving home’, Journal of Family History 3 (1978), 181–202Google Scholar. A study of Swedish women born between 1936 and 1960 is provided by Öystein, Kravdal, ‘Flytting fra foreldrehemmet: regionale og sosiale forskjeller blant svenske kvinner fodt 1936–1960’, Stockholm Research Reports in Demography 26, Section of Demography, University of Stockholm (1985).Google Scholar

3 Ethel, Shanas et al. , Old people in three industrial societies (London 1968)Google Scholar; Dorrian, Apple Sweetser, ‘Asymmetry in intergenerational family relationships’, Social forces (1963), 346–52; idem., ‘Love and work: Intergenerational household composition in the United States in 1900’, Journal of Marriage and the Family 46 (1984), 289–93.Google Scholar

4 See, however, Gerdt Sundström, ‘Gamla som bor med sina barn: ett alternativ till sammhällsomsorg?’ [Aged who coreside with their children: an alternative to formal care?], Report to The Commission for Social Research, Ministry of Social Affairs, Stockholm (1982), unpublished mimeographGoogle Scholar; Daniel, Scott Smith, ‘Modernization and the family structure of the elderly in the United States’, Zeitschrift für Cerontologie 17 (1984), 13–17.Google Scholar

5 An attempt to solve the situation of such daughters in the 1960s in Sweden is described in Gerdt, Sundström, ‘De gamla, deras anhöriga och hemtjänsten’ [The aged, their family, and the social services], Report 22, The School of Social Work, University of Stockholm (Stockholm, 1984).Google Scholar

6 Thomas, Burch, ‘The size and structure of families: a comparative analysis of census data’, American Sociological Review 32 (1967), 347–63.Google Scholar

7 See Table 4 above.

8 Already a classic is Eugene Litwak, Geographical mobility and extended family cohesion', American Sociological Review 25 (1960), 385–94CrossRefGoogle Scholar. This has also been a focus of my own gerontological research; for references, see Gerdt, Sundström, ‘Caring for the aged in welfare society’, Stockholm Studies in Social Work, 1, Liber (Stockholm, 1983); idem, ‘Intergenerational mobility and the relationship between adults and their aging parents in Sweden’, The Gerontologist 26 (1986), 367–72.Google Scholar

9 Francis, Lang-Kwang Hsu, ‘The myth of Chinese family size’, American Journal of Sociology (1943), 555–62Google Scholar; Sundström, , ‘Caring for the aged’; Ma Shia, ‘An analysis on the size of family household and family structure in China’, unpublished paper presented at the International Seminar on China's 1982 Population Census, Beijing, 26–31 March 1984.Google Scholar

10 By ‘independence’, what is meant here is merely the physical separation of the generations.

11 Robert, Moroney, ‘Families, care of the handicapped, and public policy’, Home Health Care Services Quarterly 3 (1982), 188–212.Google Scholar

12 James, Morgan, ‘The redistribution of incomes by families and institutions and emergency help patterns’, in Duncan, G. and Morgan, J., eds., Five thousand American families: patterns of economic progress, X, Analyses of the first thirteen years of the panel study of income dynamics, Institute for Social Research, University of Michigan (Ann Arbor, 1983).Google Scholar

13 Leslie, Kennedy and Dennis, Stokes, ‘Extended family support and the high cost of housing’, Journal of Marriage and the Family 44 (1982), 311–18Google Scholar. See also the discussion of Chudacoff, Howard P., ‘Newlyweds and family extension: the first stage of the family cycle in Providence, Rhode Island, 1864–5 and 1879–80’, in Hareven, Tamara K. and Vinovskis, Maris A., eds., Family and population in nineteenth-century America (Princeton, 1978)Google Scholar. For the United States in 1960, there are interesting data in James, Morgan et al. , eds., Income and welfare in the United States (New York, 1962)Google Scholar: chapter 14 describes ‘The economics of living with relatives’. A more recent study on the same theme is Ronald, Angel and Marta, Tienda, ‘Determinants of extended household structure: cultural pattern or economic need?’, American Journal of Sociology 87 (1982), 1360–83.Google Scholar

14 Gerdt, Sundström, ‘100 years of coresidence between the generations: the case of the younger generation in Sweden’, report to the National Council of Building Research (Stockholm, 1983)Google Scholar; Gunilla, Kjellman, ‘Kultur och åldrande: en etnologisk studie av boendemönster och livsfomer bland äldre på svensk landsbygd’, Projektet Äldre i samhället – förr, nu och i framtiden, Rapport 17 (Uppsala, 1984). The rates of coresidence between married persons (they were very high everywhere for the unmarried) appear to have varied with such factors as the availability of land and farming units: Sundström, ‘100 years of coresidence’.Google Scholar

15 This was purchased through the Inter-university Consortium for Political and Social Research (ICPSR), No. 7825.

16 Sweetser, ‘Love and work’.

17 Chudacoff, ‘Newlyweds and family extension’. At least two other studies are relevant here. For the United States, Susan, Bloomberg et al. , ‘A census probe into nineteenth-century family history: Southern Michigan, 1850–1880’, Journal of Social History (1971), 26–45Google Scholar; Rudy, Seward, The American family: a demographic history (Beverly Hills, 1978Google Scholar). Both studies demonstrate that coresidence and extension were on the increase in the nineteenth century. An interesting study of shifts in ‘nest-leaving’ in the twentieth century is Frances, Kobrin Goldscheider and Celine, LeBourdais, ‘The decline in age of leaving home, 1920–1979’, Sociology and Social Research 70 (1986), 143–45.Google Scholar

18 In the nineteenth-century Swedish source, I have manually inspected the census lists (compiled from the catechetical examination records by the priests) for 1880. The difference from the computerised tabulations (based on another source) lies between one and three per cent.

19 See above, note 2.

20 But see David, Gaunt, ‘Household typology: problems, methods, results’, in Akerman, S. et al. , eds., Chance and change (Odense, 1978Google Scholar; Scott, Beck and Rubye, Beck, ‘The formation of extended households during middle age’, Journal of Marriage and the Family 46 (1984), 277–87.Google Scholar

21 An exception is one of the reports on the United States 1970 Census: PC(2)–4B, which has been very useful for me. Regrettably, the 1980 Census will not produce a corresponding volume, due to financial restrictions (personal communication from Steve Rawlings, Population Division, Bureau of the Census).

22 Sundström,’ 100 years of coresidence’.

23 The source may differ due (conceivably) to under-representation of the very ill in surveys. As will be shown later, persons incapacitated by illness make up a large proportion of coresidents.

24 Wall, ‘The age of leaving home.’

25 Halliman, Winsborough, ‘Statistical histories of the life cycle of birth cohorts: the transition from school-boy to adult male’, in Taeuber, K. et al. , eds., Social demography (New York, 1978).Google Scholar

26 E.g. Peter, Uhlenberg, ‘Death and the family’, Journal of Family History (1980), 313–20Google Scholar; Louise, Gaunt and David, Gaunt, ‘The family in Scandinavia’, in Klapisch, C. et al. , eds., Histoire de lafamille (Paris, 1985).Google Scholar

27 This is a survey of 1,064 old (67 and older) people in Sweden in 1954. The study is representative and of very high quality, with excellent information on all household members, including their occupations and employment status. The data are now available on magnetic tape at Statistics Sweden (transferred by the author with a grant from the Scientific Council of Statistics, Sweden).

28 Sundstrom, ‘Gamla som bor med sina barn’.

29 Ibid.

30 But see Ingrid, Sjöberg, ‘Ensamhet eller gemenskap’, Rapport 18, Undersökningen av levnadsförhållanden, Statistics Sweden, S.O.S. (Stockholm, 1980).Google Scholar

31 Sundström, ‘Intergenerational mobility’.

32 Ibid.

33 Gerdt, Sundström, ‘Hur nära? Avstånd och närhet i svenska familjer [How close? Distance and propinquity in Swedish families]’, HSB:s Riksförbund and the School of Social Work (Stockholm, 1984), unpublished mimeograph.Google Scholar

34 Lundberg and Modig, ‘Ungdomars etablering på bostadsmarknaden’.

35 Personal communication from Ingrid Sjöberg, IU/I, Statistics Sweden (based on ULF-data).

36 Personal communication from Ingrid Sjöberg (note 35). Schwarz, ‘When do children leave’, suggests that currently rising coresidence rates in West Germany are due to falling nuptiality rates; single persons stay with their parents ‘instead’.

37 Marcia, Guttentag and Paul, Secord, Too many women? The sex-ratio question (Beverly Hills, 1983).Google Scholar

38 See Sweetser, ‘Love and work’.

39 See especially Kobrin and DaVanzo, ‘Leaving home’; Young, ‘The effect of children returning home’.

40 Sundström, ‘100 years of coresidence’, used a longitudinal survey, but the information on household structure in the level-of-living survey prior to 1981 was unsatisfactory. I thus had to ‘reverse’ through the datasets, which necessarily gives an underrepresentation of those who left their parents. These data should not be confused with the cross-sectional survey (ULF) which is the basis of this study.

41 Personal communication from Martha Hill at the Institute for Social Research, University of Michigan, which administers the Panel Study of Income Dynamics.

42 Sundstrom, ‘Caring for the aged’.

43 ‘Ensamstående mödrars sociala och ekonomiska förhallanden’, S.O.S. Socialvård (Stockholm, 1957).Google Scholar

44 Computations from the 1954 survey (see Note 27).

45 James, Sweet, ‘The living arrangements of separated, widowed, and divorced mothers’, Demography 9 (1972), 143–59Google Scholar; Mary, Ann Scheirer, ‘Household structure among welfare families: correlates and consequences’, Journal of Marriage and the Family 45 (1983), 761–71.Google Scholar

46 Lundberg and Modig, ‘Ungdomars etablering på bostadsmarknaden’.

47 To discover the situation in the United States in 1900 would require considerable additional research.

48 Smith, ‘Modernisation and family structure’.

49 Sundström, ‘Gamla som bor med sina barn’.

50 Computations from the 1954 survey (see note 27).

51 There is a good discussion of this from the elderly parent's perspective, in Daniel, Scott Smith, ‘Life course, norms, and the family system of older Americans in 1900’, Journal of Family History 4 (1979), 285–98.Google Scholar

52 This raises the issue, not only of parents and children-in-law, but also of women (in this case mothers) rather than men being care-givers. In this case it may be the child who receives the care: see below.

53 Sundström, ‘Gamla som bor med sina barn’.

54 Ibid.

55 Sundström, ‘100 years of coresidence’.

56 Michael, Anderson, ‘Family, household and the industrial revolution’, in Anderson, , ed., Sociology of the family (Harmondsworth, 1971).Google Scholar

57 Sundstrom, ‘100 years of coresidence’.

58 Smith, ‘Life course, norms, and family structures’.

59 This has been highlighted in a few studies of the older generation in these and other ‘extended’ households. The parent usually remained head of the household if he (or even she) was married and/or employed. Headship, then, was more a matter of status than of age in itself. See Smith, ‘Life course, norms, and family structures’; Michael, Dahlin, ‘Perspectives on the family life of the elderly in 1900’, The Gerontologist 20 (1980), 99–107. I have largely foregone the issue of ethnicity. However, in 1900 elderly non-whites (aged 55 and over) had coresidence rates above average: Smith, ‘Life courses, norms, and family structures’. Yet my data show the younger generation of non-whites to have lived less often than average with their parents in 1900, at least up to the ages of 40–45, although older non-whites had slightly higher rates.Google Scholar

60 Dorrian Apple Sweetser, in a number of works, has suggested that the changing division of labour, modernisation, and the affinity between mother and daughter raises the proportion of married women as against married men among coresident offspring (see note 3).

61 Folk- och Bostadsräkningen 1935/36, 6, 51–2. When fertility data were standardised for duration of marriage, it was shown that persons of this description had roughly as many children themselves as had those starting out directly as established farmers.Google Scholar

62 Sundström, ‘Gamla som bor med sina barn’.

63 See Table 9. These were too few to be subject to meaningful statistical analysis.

64 These percentages are based upon groups numbering, respectively, 662, 567, 86, 41, 97 and 63.

65 See Folk- och Bostadsrükningen 1935/36, 5, ch. 1.

66 Sundström, ‘Intergenerational mobility’.

67 The one American survey source I know of which provides this, the National Longitudinal Survey (scattered years), does not on the other hand include persons of all ages in their sample.

68 Computations on the elderly in the 1980–1981 ULF show that one-tenth of those who live together with their children report that the latter needed some care and tending within the previous three months! In the survey of elderly in Stockholm in 1976–1977, it was found that about one-quarter of those who lived with children (as did about three per cent of all elderly) reported their children to suffer from some illness: Sundström, ‘Gamla som bor med sina barn’. It appears that the majority of these were sons. Daughters more frequently had the traditional role of care-giver. Likewise, a British study found high mortality rates (higher for men than for women) among a group of adults coresiding with parents: Fox, A. J. and Goldblatt, P. O., Longitudinal study: Socio-demographic mortality differentials, 1971–1975, HMSO (London, 1982), Table 4.5, ‘other child’.Google Scholar

69 Figures are obtainable from the author. A good source is also Table 10 in the United States Census 1970: PC(2)–4B. This shows that about 10 per cent of persons aged 25 and older who are labelled as ‘child of head’ had no income at all, as against just about one per cent in the entire age-group. Table 5 in the same publication also shows the head of household in these families to have very low median incomes (especially when female).

70 The most comprehensive empirical study of kin transfers in the United States is Morgan, ‘The redistribution of income’.

71 Alvin, Schorr, Explorations in social policy (New York, 1968)Google Scholar; ibid. , ‘“…thy father and thy mother…”: a second look at filial responsibility and family policy’, United States Department of Health and Human Services (Washington, 1980).

72 The same has been observed in Australia. Young, ‘The effect of children returning home’.

73 Scheirer, ‘Household structure among welfare families’.

74 There is one American study dating from the 1960s which shows the expected pattern in some ingenious tabulations: the higher the income of the elderly, the less often relatives were present, and the more of these relatives were non-married children only. At least, this held for couples and non-married women; non-married elderly men had no uniform pattern in this respect. Leonore, Epstein and Janet, Murray, ‘The aged population of the United States’, The 1963 Social Security survey of the aged, United States Department of Health, Education and Welfare (Washington D.C., 1967).Google Scholar

75 See also Moroney, ‘Families, care and public policy’.

76 Linda, Brunn, ‘Elderly parent and dependent adult child’, Social Casework 66 (1985), 131–8.Google Scholar

77 Daniel, Scott Smith, ‘Historical change in the household structure of the elderly in economically developed societies’, in Fogel, R. et al. , eds., Ageing, stability and change in the family (New York, 1981); my own calculations from Table 5.A1.Google Scholar

78 For a discussion and references, see Sundström, ‘Caring for the aged’.

79 Elina, Haavio-Mannila, ‘Caregiving in the welfare state’, Acta Sociologica 36 (1983), 61–82.Google Scholar

80 See Sundström, ‘100 years of coresidence’.

81 Gerdt Sundström, Kari Waerness, and Dagfinn Ås, ‘Familjen, staten och de äldre’ [The family, the state, and the elderly], unpublished manuscript. Few scholars have suggested that coresidence may sometimes or often be to the advantage of the younger generation: Schorr, Explorations in social policy; Clare, Wenger, ‘Surtout des meres et des filles: Des vieux parents vivant avec des enfants adultes’, Penelope/Vieillesses des femmes 13 (1985), 111–15.Google Scholar