mixed logit estimation in table 11 shows a negative sign, meaning a dislike for bedrooms.
However, all heterogeneity aspects are positive, leading to a positive bedroom valuation in
general (see table 8). Table 11 indicates that commuters with a higher education level, value
an extra bedroom relatively higher than respondents with a lower education level. As
expected commuters with a working partner or with children next to that have a positive
valuation for an extra bedroom. Furthermore, respondents living in a municipality with 50.000
or more inhabitants value an extra bedroom more positively than respondents living in smaller
municipalities. Finally, table 11 shows that people living in an apartment value an extra
bedroom higher than people living in another type of house. This might be due to a sort of
selection effect of having relatively (too) little space in an apartment.
Monthly housing costs are valued negatively. However, people having a higher level of
education dislike monthly housing cost relatively less than people with a lower education. The
same goes up for respondents living in a house they own. Respondents having children on the
other hand dislike housing costs even more than people without children. This might be
explained by the on average higher expenditures households with children have to make.
Looking at the location variable, table 11 shows a strongly negative coefficient for living in a
big city. This dislike is even stronger for respondents having a partner, living in a (semi-)
detached house and/or having a lower gross household income than 56000 euro/year. First of
all, respondents with a partner might often need more space. This space can in general better
be found outside the big cities. Secondly, the extra dislike of living in a big city for people
living in a larger house might partly be seen as a sort of self-selection effect: larger houses
occur more often in smaller towns/cities (at least in this dataset). And people living outside a
big city might quite likely have a reason for living outside a big city. Thirdly, respondents
with a higher income, in general quite often live in nicer neighbourhoods in a big city,
whereas relatively lower income households are forced (by housing prices) to live in less
preferred neighbourhoods. This last group then might prefer to live (for the same price) in a
smaller city or more rural region.
Furthermore, the dislike of living in a big city is relatively lower for people having children,
living in a municipality with more than 50.000 inhabitants, owning a house, living in an
apartment and/or for people not having to pay fuel costs. The fact that households with
children have a relatively lower dislike of living in a big city is somewhat opposite to the
22