Housing affordability and the effect of housing subsidies

摘要

This dissertation looks at the affordability of housing and the distributional impact of housing subsidies, in Flanders and the Netherlands. We analyze housing affordability according to income groups and tenure type. Furthermore, we explore how housing subsidies are distributed over these groups, and to what extent they affect income inequality and poverty. Additionally, the aim is to demonstrate the value of a comprehensive conceptual approach in a comparative analysis. In both the affordability and subsidy analysis two methodological approaches are used: a cash-flow (short-term) and a user cost (long-term) approach. Flanders and the Netherlands are interesting cases for comparison since the tenure structure and housing subsidies largely differ, as a result of different policy choices in the past. The analyses are based on survey data for the period 2005/2006 and administrative data of housing subsidies in 2008. The study reveals that in both Flanders and the Netherlands, about 14% of households are confronted with unaffordable housing, according to the budget method (cash-flow approach). In both cases, the group with an affordability problem is relatively larger for tenants than for owner-occupiers and the largest in the lowest income quintile. The user cost analysis points out that owner-occupiers – in a period of rising house prices – also have a significantly lower user cost of housing than tenants. For both Flanders and the Netherlands in 2005/2006 the expected value increase of owner-occupied dwellings exceeded all the costs involved. Furthermore, for Flanders, the results point at a clear duality between private and social housing with regard to income composition, actual rent and subsidy impact. The Flemish social housing is rather small and, compared to the private rented sector, rents are relatively low and strongly subsidized. Due to a weak income profile of social tenants in Flanders, affordability based on the budget method turns out worse for this group than for private tenants. In the Netherlands, the differences between private and social renting in terms of actual rent, affordability and subsidy impact are far more limited. Rent in social housing is subsidized to the same extent as in Flanders (half of the market rent) whereas subsidization in the private rental market is considerable. While tenants in the Flemish private rented sector are hardly subsidized, in the Netherlands the discussed housing subsidies are received by both private and social tenants. In Flanders social housing is the measure that targets the poor, while housing allowances fulfill this role in the Netherlands. With regard to the owner-occupied sector, the heavily debated mortgage tax relief is found to increase income inequality, more so in the Netherlands than in Flanders. In the Netherlands, overall, housing subsidies are relatively more directed at tenants and lower income groups. In contrast, in Flanders overall housing subsidization is disproportionally received by higher incomes groups and owner-occupiers. This means that in Flanders housing subsidies are generally not directed at the groups with the most affordability problems.