ABSTRACT
I am more optimistic on the behaviour side, and I think we have already made some progress there. It is possible, but not with the standard traditional macro type of model, but with different types of models to study human behaviour, consumers’ behaviour, household behaviour. But you also need different data for that. There I go along with Professor Lynch very much: we need a lot more data than is available right now (although a lot more is available now than 20 years ago). But still, if it is possible and feasible to get a lot more data with information on those issues, and with modern computing technology to manage the data, then we can also expand the range of models that we have now. We already have a much broader range than the focus on macro-type models discussed in this conference. We have models for special issues, for labour market issues, for market structure issues, etc. To give two examples: we made special purpose models for the issue of shop-opening hours and for the issue of the regulation of the broadcasting industry in the Netherlands. These are very small models, especially made for a specific policy question. But we have also developed the larger ones like the MIMIC model, which tries to model household behaviour in the context of the social security and tax system. This is a very good example of how we try to get a better grip on behaviour, but to get more data, especially micro data, would help to improve that further.
