MICSIM: A behavioural microsimulation model for the analysis of tax-benefit reform in the Netherlands

Read also the used presentations ((Jongen, Blundell, Figari and Decoster).

The core of the model is a discrete choice model for labour supply. We estimate preferences for a large number of subgroups, using an exceptionally large and rich administrative household-panel dataset. We find that men in couples have much smaller labour-supply elasticities than women in couples, in particular when young children are present. Furthermore, cross-elasticities of women in couples are non-negligible. The labour-supply elasticity is relatively high for single parents with young children, much lower for single parents with older children, and also relatively low for singles without children. The decision whether or not to participate is much more responsive to financial incentives than the hours per week decision, though for women in couples with young children, the hours worked per week response is also relevant.

We illustrate the working of the model with a number of policy simulations. Because intensive margin responses are small, and cross-effects of the income of the husband on the labour supply of the wife are non-negligible, the impact of marginal tax rates on total hours worked is limited. Because people are more responsive along the extensive margin, changes in the participation tax rate, via e.g. changes in social assistance and in-work tax credits, have larger effects on total hours worked. An in-work tax credit targeted more at lower incomes is more effective in increasing labour supply than an across-the-board in-work tax credit. Fiscal incentives to work for mothers with young children are particularly effective. Childcare subsidies are also an implicit subsidy for work for mothers with young children, but they are expensive from the perspective of the government because they encourage substitution of informal for formal care.