AbstractBACKGROUNDThere is still much to learn about the explanation for the often-found negative association between sibship size and different child outcomes. A plausible explanation is resource competition between siblings in larger families, as suggested by the resource dilution hypothesis.OBJECTIVEThis study contributes to our understanding of these mechanisms by investigating the association between sibship size and height before, during, and after the fertility decline to test predictions based on the resource dilution hypothesis.METHODSThe investigation is conducted using information from universal conscript inspections linked to a longitudinal demographic database. Regression analyses estimate a model derived from the resource dilution explanation and analyze the association between sibship size and height among men born in 1821-1950 in southern Sweden.RESULTSThe results show that the association between sibship size and height was negative from the mid-nineteenth century until the mid-twentieth century. There is no association in the early nineteenth century. The strength of the association is gradually weakened over time for men born from the 1840s until the 1940s. It is most consistent among men born from 1881-1921, corresponding closely to the time for the fertility decline in the area. The association is not a result of confounding by observable demographic or socioeconomic differences between families.CONCLUSIONSThe results are in line with resource dilution being an important explanation for the negative association between sibship size and height. Resource dilution in larger families still seems to be dependent on the societal and historical context.1. IntroductionA negative association between the number of children in a family, the sibship size, and the heights of the children has been shown many times for 20th century populations in high-income (Douglas and Simpson 1964; Olivier and Devigne 1983; Mednick et al. 1984; Lawson and Mace 2008; Suliga 2009) and low-income countries (Desai 1995; Jordan et al. 2012; Manley, Fernald, and Gertler 2012), even though no association has been found in some populations (Desai 1995). A negative association between sibship size and the children's outcomes has not only been observed with regard to height, but also, and in an even larger number of studies, with regard to the children's educational outcomes and social mobility (Blake 1981, 1985; Downey 1995, 2001; Sacerdote 2007; see also the review by Steelman et al. 2002).The association between sibship size and child outcomes is thus firmly established, at least in 20th century high-income populations. Much less is known about the mechanisms that lead to these negative associations. The most-used explanations are resource dilution in larger families and confounding. The mechanisms behind the negative associations have implications for our understanding of family dynamics and behavior, and also for our understanding of the demographic transition (e.g., Becker 1993). The presence of the associations in historical populations also has implications for, for example, unified growth theory (Galor 2012) or the theory of the technophysio evolution (Floud et al. 2011). Both theories assume that resource dilution affects child outcomes and this is the most important (Galor 2012), or one of several (Floud et al. 2011), mechanism(s) generating dynamic effects in the models.The fertility decline has also been proposed as an explanation for the secular increase in height and improving health of children from the late 19th century onwards (Reves 1985; Weir 1993; Schneider 1996; Hatton and Martin 2010a). Hatton and Martin (2010a), for example, retrospectively extrapolate based on cross-sectional data from 1930s Britain and suggest that about 25% of the increase in height should be attributed to the fertility decline. These suggestions also make it interesting to test how the association between the sibship size and the living conditions of children has developed longitudinally. …