Raising Children: Private Expenditure and Foregone Earnings *

. From the point of view of an individual household there are two major types of private cost of children, viz. direct household expenditure for the offspring plus (usually) mothers' foregone earnings. Both aspects have been studied theoretically and empirically, but thus far only separately. The joint empirical analysis of these main costs of children are the subject of this paper. Employing the Austrian Consumer Survey of 1984/85 and 1983's Microcensus, household expenditure on one, two and three children and mothers' foregone earnings are estimated for a comparable sample of households. In order to estimate household expenditure on children a modified Prais‐Houthakker model has been employed. Labour force participation functions as well as human‐capital wage equations were estimated using a multinomial logit model and allowing for both the impact of career interruptions and part‐time work. Through simulation on the basis of these estimates we arrive at comparable life‐time earnings of women with no, one, two or three children. We then add the relevant differences to the expenditure on children to obtain the “total” private cost of children. Being aware of the limitations of our analysis–not all monetary private outlays and no public expenditure are included, no consideration of children's benefits to parents, etc. ‐ we can nevertheless draw some tentative conclusions: there are decreasing marginal costs for the second child and rather constant costs for the third child, except for the case of “low attachment“ to the labour force, where marginal cost also declines for the third child. Assuming utility maximizing households, one could arrive at a policy prescription for raising birth rates: find a means to reduce foregone earnings which appear to be a major deterrent to having more than one child.