Mr. Brendan O'Connor:

I hope I am not horribly wrong about this, but to the best of my knowledge, the SWITCH model is not strictly used. Let me get this point right. There are some household types in there, but primarily it is used for looking at income groups. I refer to households split into income groups. There is also some analysis within the SWITCH model of households, such as impacts in cash terms on different household structures. Regarding how confident we are in this regard, it is a difficult question to answer. As I understand it, this approach uses data from things like CSO household budget surveys and information on incomes of household groups, etc., to try to model the impact of tax changes. Therefore, it is a function of the data going into the model. What the model, or the modeller, then tries to do is to adjust these things for changes in wages or inflation. A starting point is modelled, then adjustments are made over time, based on forecasts-----