The surveys do not record support provided recently and in the more distant past equally well. The distribution of the time of support provision suggests that the overwhelming majority, more than 60% of respondents, provided support to their children in the 15 years preceding the survey and around 90% of support is recorded in the preceding 25 years. For this reason, the 2003 dataset records support provided in the past two decades of state socialism, the first post-state socialist decade and the beginning of the housing boom around the millennium. The 2015 survey provides data mostly about the post-state socialist periods and the share of support providers during state socialism is very low. Hence, due to the surveys’ relatively short-term perspective of transactions, the overlap between their time span is not too large.

 

 

Figure 7 displaying periods of housing system formation together with the development of housing construction testifies to the significant fluctuation of construction by periods of housing system formation. Figure 8 (in the bottom scale of the graph) displays the distribution of the share of support-providing households by years passed since the provision of parental support. On the top scale of the graph, years of support provision from the 2015 survey are displayed first, years of provision from the 2003 survey are displayed second. Years from the 2003 sample are always twelve years behind as the survey was recorded twelve years earlier. Periods where the blue and the orange line diverge the most are characterised by significant differences in the frequency of support-provision. Fortunately, the two surveys were recorded in two very different periods of housing system formation: the 2003 one was recorded at the time of the expansion of mortgage lending while the 2015 one was recorded at a time when the housing market just started to leave behind the depression following the crisis and mortgage lending was subject to much stricter regulations. For this reason, the two curves record very different periods and if periods of the development of the housing system impact parental support, the two curves should diverge during different cycles. If the orange curve displays higher values than the blue one following the millennium characterised by the expansion of the mortgage market, H1 based on the transition perspective proves wrong. If it displays higher values than the blue line, the assumption that easily accessible mortgages replace parental support, that is, commodification reduces familialism, may prove correct.

 

The figure shows that the occurrence of parental support increases with boom cycles of the housing market. A much bigger share of parental support was provided in the four years preceding the 2003 survey conducted during the housing boom, than the 2015 one which was carried out at the end of the “crisis management” period and the start of the current housing market uptake. The effect of the 1999-2008 housing market prosperity is clear from the 2015 data which display higher values than the 2003 line depicting the years following the regime change. The other period marked by a higher share of parental support is the 1980s when subsidies and preferential loans for single-family housing construction were introduced.

 

According to the transition approach, a housing system with easily available mortgages poses an alternative to the post-state socialist “housing-welfare regime by default” (Norris and Domański, 2009; Stephens, Lux and Sunega, 2015; Csizmady, Hegedüs and Vonnák, 2019). In this sense, reliance on the family in housing access, and parental support in particular, should have decreased in the period of the millennial mortgage boom. The fact that the opposite occurred and a significant rise could not be observed in periods characterised by limited access to mortgages suggests that housing price appreciation during mortgage booms increases parental support while limited availability of market finance actually relieves families from financing the housing of young adults.