Sammendrag
The unifying theme of this thesis is the past, present and future household and family structure in various Nordic countries. It provides a forecast of household structure in Denmark and Finland. An application of household forecasts is to study the implications for energy use – a topical issue in an environmental context. Here the effect of divorce is considered. Another strand of the research on household structure is to examine potential links between selected aspects of family structure and mortality. This thesis studies two such aspects.
A finding from the detailed probabilistic household forecast is that the proportion of households that are one-person households is likely to increase during the coming decades as it has done in the past decades, leading most probably to a further decline in the average household size. On the other hand, the share of households consisting of married couples is likely to fall. Part of the explanation is high divorce and union dissolution rates. Splitting one household into two leads to higher domestic energy use as a result of loss of economies of scale. However, taking into account the depressing effect of divorce on fertility, higher divorce rates may not entail mounting total domestic energy use in the long term; indeed, it may even lead to lower household energy consumption. The results presented here show that in the short run higher dissolution rates lead to higher domestic energy consumption, but in the long run it might drive down energy use due to low fertility.
The thesis also addresses the possible link between mortality and two aspects of the family structure, namely the gender composition of one’s children and whether or not one is a grandparent. Neither has so far received much attention. I find only limited support for the hypothesis that having children of only one sex is associated with a mortality disadvantage - for mothers with two or more children. The link between grandparenthood and mortality is stronger, with grandfathers in general experiencing higher mortality than fathers who are not grandfathers. The association varies with characteristics of the grandparents, especially age, with young grandmothers also displaying a mortality disadvantage, and those who became grandmothers after age 50 having a mortality advantage.