Housing wealth inequality, intergenerational transfers and young households in the super-homeownership system

Abstract

This article focuses on the problem of housing wealth inequality as an increasingly important dimension of social inequality in post-socialist super-homeownership countries and uses the Czech Republic as a case study. The article shows that housing wealth inequality is higher than income inequality and that it tends to grow in time, especially due to the spatially uneven appreciation of house prices. However, intergenerational housing-related within-family resource transfers have a mitigating effect on growing wealth inequality. Using a microsimulation model, the article shows that any discontinuation of transfers could significantly increase housing wealth inequality in the future, especially among young families. Unlike in some Western societies, intergenerational resource transfers thus work to equalise rather than catalyse housing wealth inequality. These findings have significant implications for measuring social inequalities, demographic trends, and the stability of housing systems in post-socialist countries.

KEYWORDS: :

• Household inequality
• housing wealth
• homeownership
• post-socialist

Acknowledgement

We are grateful to Jiří Šatava (CERGE-EI) and Ladislav Kažmér (NIMH) for valuable research assistance during the preparation of the microsimulation model. The research for this paper was sponsored by the Czech Science Foundation (Grantová Agentura České Republiky) with grant number 19-07402S.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The amount was calculated as a weighted average using tenure structure data from Pittini et al. (2017) and the number of inhabitants in 2017 from Eurostat. The average includes also Norway as a non-EU state.

2 Co-op housing is a kind of hybrid tenure between homeownership and renting. Specifically, the flats are owned by a co-op, and their occupants, formally tenants, have a share in the co-op. There are different forms how such co-op housing work in reality. Sometimes the disposal rights of co-op members (such as the right to sell occupancy rights and the co-op share on free market) are restricted and co-op housing is closer to tenancy; sometimes the disposal rights of co-op members are very strong and co-op housing is then closer to homeownership. In the post-socialist region, the latter case is a more common option.

3 The Czech mortgage market is relatively advanced compared to other post-socialist societies (Lunde & Whitehead, 2016). Outstanding residential debt increased to 22% of GDP in 2019 from almost zero in 1997, i.e. in 22 years (Hypostat 2020). Mortgages have been playing an increasing role over time; while in 2001 only 5% of respondents in the survey Housing Attitudes 2001 stated that they had used a mortgage loan to obtain housing, in 2013 the share was 17% according to a similar survey. Despite this, an estimated one-quarter of home sales are made without a mortgage and the typical loan-to-value ratio was only 56% at the end of 2015 (Hypostat, 2016). Intergenerational transfers are still the main source of finance, and they are acquiring even more importance among the young generation. According to Housing Attitudes 2013, 30% of homeowners had inherited their housing or received it as a gift, and of those who bought their home at market price or built it themselves, 42% had received financial assistance (usually in the form of a gift they did not have to pay back) from their parents. Moreover, the biggest intergenerational assistance is observed among the youngest cohorts. Among homeowners aged 18-44, 33% of them had inherited their housing or received it as a gift and 53% of those who had bought their housing at market price or built it had received financial assistance from their parents and 35% of them had also received financial assistance from the parents of their partner. Ultimately, almost two-thirds of young homeowners acquired their housing with financial or in-kind help from their family.

4 This estimate is made by the household itself; the exact wording of the question in the survey questionnaire is: ‘The following question is intended to help estimate the current prices of housing in the Czech Republic. Please try to estimate the market price of the flat/house you live in (price in CZK)’. Home price estimates may not be accurate, especially during the market correction phase or when made by households with low income and equity who probably overstate the amount of their residential wealth. However, due to the limitations of the data, the self-reported value of assets in large household surveys is commonly used in comparisons of household income and wealth inequality trends in the literature. Our tested period contains both price correction and boom periods, thus partially controlling for the cyclical effect on self-reporting. And, most notably, our paper also includes the real housing price index to measure the effect of price appreciation on the past residential wealth inequality changes or to estimate the residential wealth inequality trend in the future.

5 A value of 1 was assigned to the head of the household, 0.3 to each child aged 0 to 13 years, and 0.5 to the other children or other persons in the household.

6 The data on the provision of intergenerational resource transfers from grandparents to parents and from parents to their descendants, differentiated by regions, were obtained from the 2015 wave of the Czech Household Panel Survey.

7 The subjective perception of income of the parental household (i.e. whether the income situation of the household is perceived as good or difficult) has also been a statistically significant factor in determining transfer-giving (Lux et al., 2021a), but it would be hard to simulate this subjective measure into the future. ‘Objective’ household income itself has not been a significant predictor of transfer giving – because of the high number of missing values and the fact that many parents provide their transfers to their adult children around the time they are finishing their economic activity (or in retirement). Unfortunately, the data did not include information about the previous income of parents. Finally, the parental educational status that could replace parental income in determining transfer giving also proved not to be a significant factor (as demonstrated in Lux et al., 2021a). For all these technical, factual, and formal reasons, we could not include parental income among the predictors of transfer giving in the microsimulations. This may lead to an overemphasis of the equalising effect of resource transfers in our results. However, we found that both regional differentiation and past transfer giving within the family (which determined transfer giving in the model) likely also reflect the impact of parental income on transfer giving. Specifically, the transfers are less frequent in less wealthy regions with a higher number of lower-income households, and households with a tradition of transfer giving have higher income than households that have no such tradition.

8 It is very probable that if transfers were interrupted, price appreciation trends would also change because price trends and resource transfers are endogenous. However, the impact of resource transfers on housing price appreciation has not, to the best of our knowledge, been tested in any country to date, and it cannot be simulated because of the lack of continuous data on transfer-giving. Our model intends to show only the major trends in housing wealth inequality; and these would not significantly change if only appreciation levels were smaller in the future than they were in the past but the inequality in gains was preserved.

9 Wealth is a stock (accumulated flow) and income is a flow; and there are more people in society without any residential wealth than without any income. It is therefore somewhat natural that inequality in the residential wealth is going to be higher than inequality in incomes, especially when residential value is increasing.

10 For the final sensitivity test, we used net income (before pensions and transfers), and the net value of households’ residential wealth, i.e. residential wealth net of the outstanding principal of the (mortgage) loan used to acquire the property. Unfortunately, the EU-SILC dataset does not contain precise information on this, and the Czech Statistical Office gave us (at our special request) only the annuity amount, the loan maturity, and the year in which a household started to repay a mortgage. Using these data, we estimated the outstanding balance, and we assumed that all households use the (most common) annuity repayment type (details of the analysis are available upon request). The results of testing again confirmed that in all the years analysed housing wealth inequalities were higher than income inequalities: the average ratio between the coefficients of variation of two inequalities was 1.2 for all households and 2.6 for young households in this case.