Abstract
Helle et al. (2002) used data from Finnish parish records to study the cost of bearing sons vis-à-vis daughters in terms of postmenopausal longevity and found a large and significant cost associated with sons. In this paper, we replicate and extend their analysis on a larger dataset of pre-modern Swedish women and find no evidence of a negative relative impact of sons. Neither do we find any evidence for the resource competition hypothesis put forth by Van de Putte et al. (2004), despite the relative poverty of our study population. This suggests that the effects found in Helle et al. (2002) were not a general feature of life in pre-modern populations. Finally, we raise some concerns regarding the methodology used and inferences made in previous studies on the topic.