How lone mothers combine being mothers with employment has become a central policy issue in many western countries. But policy making is dominated by the operational assumption that lone mothers are ‘rational actors’ who make individual utility calculations about the costs and benefits of taking up paid work. This is what we have called the ‘rationality mistake’, for evidence shows that decisions are still made rationally, but with a different sort of rationality. In this article we use a case-study from Norway to explore this issue further. Norway is unusual in having a ‘designer benefit’ – the transitional allowance – exclusively for lone mothers. Before 1998, the transitional allowance positioned lone mothers as mothers at home, whereas after 1998 policy places lone mothers – after the first three years of motherhood – as workers in employment. The policy lever of benefit change was supposed to change how lone mothers – as rational economic actors – behave. Using an intensive research design, we find that lone mothers employ a moral and relational rationality which usually means that both the pre and post-1998 policy would act more as a constraint than an enabler on their behaviour.

This paper evaluates the model of ideal type care regimes, as introduced by Jane Lewis and Barbara Hobson, as a means of understanding the comparative position of lone mothers in different welfare regimes. It does this in the light of an application of the model to the case of Austria. We find that the model offers insights and advances in understanding the comparative situation of lone mothers in Austria, but the analysis also points to certain shortcomings in the model. Partly these are problem of range and, to adequately represent outcomes for lone mothers, a 'negative' dimension of poor outcomes needs to be developed for both the ideal types postulated -the 'Caregiver Social Wage' model and the 'Parent/Worker' model. Secondly, there is a problem of mis-specification of process, as revealed by the case of Austria where lone mothers seemed to be positioned in the 'Parent/Worker' model (if somewhat negatively), while married mothers were positioned within the 'Caregiver Social Wage' model. An answer may be to modify the specifications of the second ideal type as a 'Caregiver/ Breadwinner Social Wage' model. Résumé Cet article évalue le modèle de régimes idéal typiques de soins (care), tel qu'introduit par Jane Lewis et Barbara Hobson, comme un moyen de comprendre de manière comparative la position des mères célibataires dans les différents régimes de protection sociale, en testant ces hypothèses au cas autrichien. Nous aboutissons à la conclusion que si ce modèle nous permet de faire des progrès dans la compréhension du phénomène en Autriche, il a également des faiblesses. Ces problèmes proviennent en partie du dimensionnement. En effet, pour représenter de manière adéquate les résultats pour les mères célibataires, une dimension négative doit être développée tant pour le modèle «Caregiver Social wage» que pour celui «Parent/Travailleur. Ensuite, il existe un problème de mauvaise spécification, comme le relève le cas autrichien qui fait que les mères célibataires semblent devoir être positionnées dans la catégorie «Parent/ Travailleurs (même si cela est de manière négative) tandis que les mères mariés se trouvent placées dans le modèle «Caregiver Social Wage». Une réponse peut être de modifier les spécifications du second modèle idéal type comme une modèle «Caregiver/Breadwinner Social Wage».