Housing service are enormously subsidized in the Netherlands (e.g. Conijn, 2008). Home-owners are subsidized mainly through mortgage interest deductibility and tax free capital gains over home equity. Renters receive two kinds of subsidies in the Netherlands: first, there is an income dependent housing allowance to support low income households. Second, there is an implicit subsidy that results from i) rent regulation and ii) rent setting behavior of dominant market players. Koning et al. (2006) and Romijn and Besseling (2008) report that in each sector housing was subsidized for over 14 billion euro in 2006. The extensive subsidization of housing services has strongly disruptive effects on the housing market (e.g. Conijn and Schilder, 2009). Recently the debate on housing market reforms has increased (partly) due to the strongly increasing government expenditures on mortgage interest deductions (Rouwendal, 2007). Spurred by this vivid debate on housing market reforms a number of studies have applied welfare analysis in the Dutch housing market. In general all studies report overconsumption of housing services, across all income groups and in both sectors of the market. Two sets of publications have recently been published concerning subsidization and housing demand. There is a series of papers by Ras et al. (2005, 2006) and Eggink et al. (2007) that model implications of several policy changes on housing demand. In these papers a regional constant quality price index is used to define housing demand. The other series of papers, including Koning et al. (2006) and Romijn and Besseling (2008), do not differentiate regionally. In our analysis we estimate demand curves using and comparing both approaches. Furthermore, we follow Ras et al. (2005, 2006) in applying a Heckman procedure for correcting for a non-random distribution of households over both sectors. We add to current literature by modeling explicitly the impact of home equity. There is a tax incentive to roll over (tax exempt) capital gains on owner-occupied housing into new dwellings. Hoyt and Rosenthal (1992)report increasing excess burdens with increasing incentives to roll over capital gains, even when controlling for other subsidizations of user cost of owning. By explicitly taking up home equity in our models we attempt to control for the impact that home equity has on the demand for housing services and thus on our welfare cost estimates.