The purpose of this study is to analyse a certain line of work on the interpretation of unemployment in the theoretical and empirical analysis of female labour supply in Italy.
This is achieved by estimating two stochastic specification models for labour force participation which are valid under more general theoretical assumption than previous approaches: the Double Hurdle and the Sequential Probit models.
The proposed methodology is applied to a sample of married women drawn from the 1995 Bank of Italy's Survey of Household Income and Wealth.
The empirical results show that the persistence of relative regional unemployment rates has been particularly striking among overqualified married women in the South.
They support the hypothesis that in absence of any information concerning demand-side constraints, labour force estimates can be misspecified. Finally, the estimation based on the sample-separation information provides some
evidence of a widespread discouragement among married women "voluntarily" out of labour market (i.e. housewives), implying that a large proportion of individuals in the South of Italy would like to choose to participate if a job were available. In this respect, the notion of voluntary unemployment is unconvincing