Home-ownership and asset-based welfare: the case of Belgium

Abstract

In this article, using policy documents and both qualitative and quantitative data sources, we evaluate the extent to which the Belgian welfare system conforms to trends towards asset-based welfare involving the promotion of home-ownership as an alternative to social security provision. We conclude that, following the explicit and ongoing sponsorship of home-ownership since the end of the 19th century, in Belgium, an asset-based approach to welfare has actually been in place for some time. Most Belgian elderly people are income-poor (mainly due to low public pensions) but asset-rich. While the risk of poverty for home-owners in old age is somewhat higher than that for the general population, it is much higher for elderly renters. As far as the preconditions for a possible restructuring of the Belgian welfare state in the direction of greater reliance on asset-based welfare are concerned, we find that most of them are fulfilled. Public debt is high with increasing costs of population ageing looming large on the economic horizon. However, although some politicians have raised the issue, so far, virtually no initiatives have been taken to tap into existing housing wealth. Our qualitative evidence shows that this can be partly explained by the fact that Belgians have a rather conservative attitude towards the welfare state, which is expected to provide adequately for ‘traditional’ life-course risks such as unemployment and old age. Housing is considered a private issue, separated from the social security sphere.

1 Introduction

In this article, we evaluate the role and development of so-called ‘asset-based’ approaches to welfare in Belgian housing and social security policy, within the context of a welfare state plagued by high public debt and with the retirement of the baby boom generation looming large on the horizon. Taking cross-national variations in housing regimes as a starting point, we can actually identify two main ‘strategies’ for governments moving (or planning to move) in the direction of a higher reliance on asset-based welfare. Firstly, in those countries where, during the course of the post-war decades, home-ownership has become the norm, such as Britain and also Belgium, governments under pressure might look in the direction of elderly home-owners to extract at least some of their housing wealth in order to supplement income from (public) retirement pensions. Secondly, in a number of countries where until the 1980s owner-occupation was less widespread, such as the Netherlands, policy-makers have since been looking to increase home-ownership, as a means to promote wealth accumulation and better housing quality for the lower income classes, and to reduce social inequalities. A number of authors (e.g., Gurney 1999; Ronald 2008) have shown how this shift in the direction of more owner-occupation was accompanied by an underlying ‘ideology of home-ownership’, resulting in the ‘normalisation’ of one form of housing consumption at the ‘expense’ of other tenure forms. Furthermore, this process has also led to a more intimate relationship between housing (policy) and other parts of the welfare state, in such a way that housing has evolved from a ‘wobbly pillar’ to a ‘cornerstone’ of modern welfare states (Malpass 2008).

However, as Malpass (2008: 17) has argued, and as has become clear from recent economic developments, ‘housing is unlikely to become a robust and long-term cornerstone of a modernised welfare state, for a number of reasons to do with the uneven distribution of housing wealth, especially in relation to need, and uncertainties concerning macro-economic performance, consumer behaviour and the capacity of the market to develop appropriate and attractively priced financial products’. Put differently, and as the recent turn of economic events has shown, in both types of countries the promotion of owner-occupation has been accompanied by a shift of risks from the welfare state onto private households.

In this article, we assess the role and development of housing-related assets in the Belgian welfare state and the implications thereof for housing policy, welfare state development and individual economic well-being, with a focus on the elderly population. In the first section, we provide a brief overview of Belgian housing policy, which has historically been directed at the promotion of home-ownership. We use large-scale quantitative evidence from the EU-SILC (Statistics on Income and Living Conditions) to evaluate whether, as is generally assumed, the elderly have significant financial/housing wealth which could be called upon, and how this wealth influences their financial situation. Secondly, we situate the Belgian welfare state within the context of the general debate on welfare state retrenchment. We take a closer look at the development over time of public finances, contrasting the ‘poverty’ of the federal government (which is in the current Belgian constellation still responsible for all matters related to social security) with the wealth, both financial and in terms of real estate, concentrated in private households. Next, we assess whether and how this paradox between a poor welfare state and rich households has been translated into recent policy developments aiming to incorporate housing in the debate on the financing of the social security system. We conclude our article with some qualitative evidence on this issue, and more in particular, how home-owners themselves feel about their position, and how they conceive the role of the state in acquiring a property and protecting home-ownership. In the last section, we weigh up the evidence and put forward some issues for debate and further research.

2 Setting the scene: Belgian housing policy and its outcomes in old age

2.1 State promotion of home-ownership in Belgium

As outlined in the introduction, we start this article with a brief review of Belgian housing policy. Belgium is, to paraphrase Saunders (1990), a nation of home-owners. But unlike many countries that recently became nations of home-owners (Doling and Ford 2007; Ronald 2008), Belgium has been one for a long time. The 1889 Housing Law marked the start of an ongoing promotion of home-ownership and the discouragement of other forms of tenure (Smets 1977; Goossens 1982; Mougenot 1988; De Decker 2008). This policy was underpinned by three motives, although their importance varied over the course of time: (1) promoting family life and more in particular raising children, (2) disciplining the work force and (3) after the Second World War, supporting the construction industry by stimulating individual households to construct their own new houses (Goossens 1982).

This pro-ownership policy was supported by both the (conservative) Liberal Party (from an obvious ideological point of view) and the powerful Catholic/Christian Party (combining the ideological strand with anti-socialism). In the long run, it was only the Socialist Party that aimed to promote alternative forms of tenure like social rental housing and opted for more regulation of the private rental market, be it with little success. Social housing today has a market share that, depending on the region, does not exceed 6–8%, and the private rental market remains quasi unregulated. Although they defended ‘pro-rental’ housing policies, including an increase of social rental dwellings and a regulation of private renting, the socialists never contested (the promotion of) home-ownership.

2.2 The financial situation of Belgian elderly: income-poor, asset-rich

We use quantitative data sources to assess whether this policy track has resulted in, as is generally assumed, the majority of the elderly having significant financial wealth which could be called upon in times of financial austerity. The idea that housing policies can reduce poverty in later life by providing a ‘hidden’ source of income resulting from outright ownership has been addressed by several authors (e.g., Fahey 2003; Ritakallio 2003; Fahey et al. 2004). Inspired by Kemeny (1981), Castles (1998, 13) pointed at the existence of a possible trade-off between the extent of home-ownership and the generosity of old-age pensions. As the author puts it:

by the time of retirement, for a large percentage of owners, the process of home purchase is likely to be complete, leaving them with a net benefit equivalent to the rent they would otherwise have to pay on the property minus outgoings for maintenance and property taxes. In other words, when individuals own their own homes, they can get by on smaller pensions.

Furthermore, the author found that in countries with high ownership rates, the lower income groups are more successful in accumulating housing assets.

In a recent study comparing the situation in Flanders to the other ‘leading’ egalitarian welfare states in Northern and Western Europe¹, Cantillon et al. (2009) found that whilst the Flemish elderly had the highest income poverty risk of all, they had one of the lowest scores on a cumulative deprivation index. This paradox was confirmed by Capéau and Pacolet (2009), who found that Belgian elderly have a high level of consumption, higher than can be expected on the basis of their public pension income—with an empirical replacement rate of 55.7% (Dewilde and Raeymaeckers 2008), from a comparative perspective the Belgian pension system is not particularly generous. Belgian elderly are thus income-poor but asset-rich, as both studies situate the explanation for this paradox in the level of assets, both financial and in terms of real estate, built up over the life course.

Earlier comparative research (Ritakallio 2003; Dewilde and Raeymaeckers 2008) has shown that, when taking account of housing costs by deducting expenditure on rent and mortgages from total disposable household income, in most countries income poverty figures² for the elderly are lower than before taking account of housing costs. Based on data from the 2001 European Community Household Panel for Belgium, the number of elderly (≥65 years of age) ‘at risk of income poverty’ drops from 25.6 to 15.8% (Dewilde and Raeymaeckers 2008). This result can be linked to the fact that many elderly (74.2%) are outright home-owners, and thus have no housing costs in terms of rent or mortgage payments. In Table 1 we present data for 2006 based on the EU Statistics on Income and Living Conditions (population of 60 years and older³), currently the main source of information on living conditions in the European Union.

Table 1 Tenure status of the elderly (SILC Belgium 2006, weighted %)

In Table 2 we present a range of financial indicators according to tenure status. As can be expected, we find that owners have the highest disposable income, while social housing tenants have the lowest income. The household income of private sector renters is not so far below that of owner-occupiers, but their housing costs are obviously much higher. Nevertheless, as housing costs in the SILC data also include insurances, services and charges, taxes, regular maintenance and repairs and utility costs (water, electricity, gas and heating), we see that elderly home-owners spend quite a large sum (about € 200 a month, equivalised) on housing, a sum that is actually larger than their income from other assets (€ 135). This has as a consequence that, when taking housing costs into account in the calculation of the income poverty risk, the elderly poverty risk no longer declines for home-owners: the poverty risk of owners-occupiers now increases from 19.4 to 24.7%. As the housing stock in Belgium is relatively old, many elderly live in (too) large houses which are badly equipped and badly isolated (Goossens et al. 2006, Vanneste et al. 2007), and are consequently faced with large utility bills and maintenance costs. Nevertheless, it is the group of renters who face the largest poverty risk: taking housing costs into account, for renters in the private sector the income poverty risk increases from 42.9 to 58.8%, and for renters in social housing from 33.1 to 64.6%. Put differently, roughly two-thirds (!) of elderly renters, whether they live in social or private rental housing, are at risk of income poverty. As is clear from their score on a composite index of life-style deprivation (see “Appendix”), this also affects their standard of living (to compare, the deprivation score in the total population is 0.09). Elderly renters are furthermore almost completely dependent on (public) pension income, as their income derived from assets (income from renting out a property or land/interests, dividends, profit from capital investments) is negligible.

Table 2 Financial situation of elderly owners and tenants (SILC, Belgium-2006, weighted results)

To summarise, the figures show that in so-called home-ownership countries, being an outright owner-occupier in old age is perhaps the most important dimension of social stratification, having a profound influence on economic well-being and living conditions. Those elderly who did not succeed on the housing market are markedly less well-off in terms of other indicators, such as disposable income and the composition thereof, and the extent to which they suffer from life-style deprivation. In the next section, we describe how the accumulation of wealth in private household’s contrasts with the precarious financial situation of the Belgian welfare state and to what extent this relates to policy initiatives promoting a higher reliance on asset-based welfare.

3 Poor state, rich households: fertile ground for asset-based welfare strategies?

3.1 Welfare state retrenchment and population ageing

‘Rolling back’, ‘withdrawal’, ‘retrenchment’, ‘from providing to enabling’: at the end of the 1980s, these were concepts widely used by scholars and commentators trying to grasp the changes in the development of advanced welfare states. For the first time since the post-war expansion of the welfare state, governments were looking to cut back rather than expand. This process was fuelled by a situation of ‘permanent austerity’, partly caused by ‘transnational’ developments such as economic globalisation and the economic and monetary integration of the European Union (e.g., Duffy 1999; Rhodes and Mény 1998; Scharpf 1998), with the associated stability pact and the acceptance of high levels of unemployment. It was furthermore intensified by a range of internal pressures caused by the transition from an industrial to a post-industrial society, such as growing service sector employment and low economic growth, changing family patterns and population ageing (Korpi 2003; Pierson 1996).

However, looking back, the general picture is that welfare states are fairly resilient: ‘the welfare state is here to stay, at least in the mid-term’ (Starke 2006, 206). It was Pierson (1996) who first pointed out that welfare state retrenchment is a development which is qualitatively different from welfare state expansion, and should thus be explained in a different way. According to this author, the welfare state has produced a large number of interest groups, who can essentially vote against any initiative that is too clearly aimed at cutting back existing rights. Therefore, governments have developed different strategies—obfuscation, division and compensation—in order to avoid the blame which results from unpopular policy decisions. Furthermore, the possibilities for reform seem strongly limited by the boundaries of existing institutions, policy traditions and conflicts of interest, a phenomenon which is generally known as ‘path dependency’ (also see Esping-Andersen 1999; van Kersbergen 1999).

More recently, using better data (i.e., based on social rights and benefit generosity rather than on social expenditure) for a larger number of countries, several studies have found that the 1980s did in fact ‘mark something of a watershed’, characterised by a relatively great deal of retrenchment in unemployment and sickness insurance (Allan and Scruggs 2004, 500). For instance, Korpi and Palme (2003) found that in Britain, following the Conservative government since 1979, by 1995 replacement rates in sickness insurance were reduced to their 1930 level, while unemployment and work-accident insurance benefits plunged to half (!) of the 1930 level. In Belgium, unemployment replacement rates dropped from 74 to 62%. In times of rising numbers of dependants, social expenditure was kept under control by small, incremental changes such as new forms of selectivity (during the 1980s) and later by the gradual erosion of welfare transfers (during the 1990s) (Cantillon et al. 2004, 2007; De Lathouwer 1996).

Although the above-mentioned studies mainly focus on social transfers for the ‘active’ population, most advanced welfare states have also changed their retirement-income systems during the last 25 years (Whiteford and Whitehouse 2006). However, as many authors point out (e.g., Green-Pedersen and Haverland 2002), welfare arrangements in policy sectors such as pensions and health care (in which a large part of the budget is consumed by the elderly) are very difficult to trim, as they enjoy population-wide support. Furthermore, especially in countries where pensions are funded on the basis of a pay-as-you-go system, such as in Belgium, privatisation tendencies are strongly curtailed by the so-called ‘double-payment’ problem, as the working generation cannot be asked to pay for both current retirees’ pensions as well their own future benefits. Basically, people expect to receive the same ‘quality’ of pensions as the ones they are currently paying for. Nevertheless, in most countries governments have improved the affordability of pension systems by exploiting their complexity in ‘less-than-transparent ways’—i.e., the obfuscation strategy identified by Pierson: by changes in the number of years used in benefit calculation; in the valorisation of past earnings; in indexation mechanisms; in the pension eligibility age… (for an overview, see Whiteford and Whitehouse 2006). In Belgium however, changes since 1990 have been limited to the alignment of the pension age for women with that for men and to the tightening of contribution conditions for early retirement.

Still, population ageing has fostered ongoing public debate on the affordability of pension systems and health care. While in 2002 the number of elderly (+60) amounted to 2.2 million people (in a population of 10.3 million Belgians), this is expected to increase to 3.3 million by 2030 and 3.6 million by 2050 (Festjens 2005). Although the increase in life expectancy, which is in combination with the ageing of the post-war baby-boom generation, the driving force behind this pension ‘bomb’, has for a large part been an increase in healthy life expectancy (Deboosere et al. 2005), there is no doubt that population ageing will have a significant impact on both pension and health expenditure.

Rather than cutting back on welfare, the Belgian government has planned for the retirement pensions of the baby-boom generation from 2010 to 2030 by financing the temporary high expenses on pensions through the so-called ‘Silverfund’ (established in 2001), a separate savings fund in which the government deposits any budgetary leeway at the end of the fiscal year (Festjens 2005). However, as there was a budget deficit in 2007, for the first time since its establishment in 2001, no funds have been transferred to the Silverfund. In all likelihood, this will also be true for 2008 and 2009 (ZVF 2008).

This ‘collectivist’ reaction to the budgetary implications of population ageing does not exclude the promotion of individual safeguards. Already in 1987, the federal government introduced a tax deduction for private pension saving (either at the individual or company level). And at the regional Flemish level, the government established a compulsory long-term care insurance in 2001.

Above these practicalities, the ideological dominance of the so-called ‘Washington consensus’ (Hertz 2001; Klein 2007) has encouraged tendencies towards deregulation, privatisation and the limitation of public debt for governments suffering from financial austerity. As a consequence, it is no surprise that governments, when confronted with the possible future inability of paying for collective services and with the unpopularity of downsizing them, are increasingly looking for new ways of providing social security (e.g., through active labour market policies instead of direct support through benefits) and/or for additional and/or alternative sources of wealth which can possibly complement traditional social security arrangements. Given the fact that most welfare states are under financial pressure and that through the post-war decades households and individuals have increasingly accumulated wealth, it is no wonder that in Western societies the idea of an ‘asset-based’ and/or ‘property-based’ welfare system has become central to debates on the restructuring of the welfare state (Regan and Paxton 2001; Sherraden 2003; Malpass 2008). The underlying argument for an asset-based approach to welfare is to rely more on the investment of individual households in financial products and property, which over the long term tend to increase in value over time, and thus can potentially provide a base from which to procure welfare services from the market.

At the end of September 2008, Belgian public debt amounted to more than € 295.36 billion, equalling 90.9% of the 2007 GNP. At the end of October 2008, debt rose further to € 305.95 billion. If we take state assets into account, the net debt is then € 295.07 billion, which implies that the assets of the state equal € 10.88 billion. As a consequence, the repayment of this debt consumes 8% of the yearly public expenditure. Compared to its neighbouring countries (Fig. 1), during the last decades Belgian central government debt has been significantly higher—it is however, decreasing thanks to the so-called reversed ‘interest snowball effect’⁴ caused by an accelerated repayment, which was made possible by the relatively favourable economic conditions during the 1990s. Given the current economic climate, the government might however, decide to resort to deficit spending, in order to prevent a recession.

Fig. 1

Central government debt as % of GDP (1980–2007). Source: OECD

The ‘poverty’ of the federal government contrasts strongly with the financial wealth⁵ accumulated in private households (Fig. 2). Mid-2008, Belgian households possessed € 864 billion of financial assets, of which € 265 billion in cash and deposits—all this amounts to approximately € 86,400 per person (Table 3).

Fig. 2

Net financial wealth in million € (1992–2007). Source: National bank of Belgium

Table 3 Belgium, households’ financial assets, mid-2008

On top of these financial assets, Belgian households possess a huge amount of real estate. In 2001, 70% of households were owner-occupiers (Vanneste et al. 2007). Around 4 million owner-occupied houses and 600,000 second homes represent an estimated value of approximately € 825 billion. And particularly in Flanders, private households possess 49,118 hectares of building land with an estimated value of € 6.3 billion (all 2006 prices).

3.2 Policy responses with respect to housing

Although there are, given the huge public debt, worries about the future of the welfare state with its increasing costs, there is in Belgium so far no ongoing public debate on the (further) ‘incorporation’ of the huge private wealth of households into the social security system. The most notable change has been that the Flemish socialists have recently become the firmest advocates of home-ownership^{6 , 7}. The (Flemish) Socialist Party today echoes the Christian Democrats⁸ (e.g., CD&V 2007) and the former liberal Flemish housing minister Keulen (2009) and strives for as much home-ownership as possible since ‘this is the best guarantee for good dwellings and the best manner of pension saving’ (Party manifesto—www.s-p-a.be; SP.a 2008). It was the former socialist party chairman Stevaert (2003a, b, 2004, 2005) who persuaded the party to start promoting home-ownership, and who was the first to make the link between housing policy and social security.

Most Flemish dream of owning their house…An owned house is not a luxury that is only for the rich, it is for everybody. An owned house is after all a form of social security. It is a money-box for the future and offers protection and security, warmth and cosiness. As a consequence, renewing housing policy goes beyond social housing. Of course, we need enough modern and well-equipped social dwellings, and of course the tenant needs to be protected. But those who buy a house ‘rent’, so to speak, for a period of twenty years. They pay rent to the bank. (…) But “renters from the bank” have one benefit: in the end, they own the house (Stevaert 2003a).

If you repay a loan, you live for free. If you keep on renting, you never live for free. So, the idea is that ordinary people can buy a house and they can borrow at normal interest rates. (…) My ideal is of a society with as many owners as possible and as few tenants as possible. As a socialist, I am convinced that such a system makes people feel responsible. (…) I want more money for social housing, but less people living in social housing (Stevaert 2003b, translation by the authors).

Despite these bold statements, specific policies so far rarely link home-ownership to issues concerning social security in an explicit way. Housing policies are still predominantly aimed at increasing owner-occupation through a range of subsidy schemes and at keeping low-income owners in the system. The latter occurs through an insurance⁹ which covers part of the mortgage in case of unemployment. Furthermore, ‘conservative’ mortgage legislation—that also forbids reversed mortgages and limits interest increases/decreases in the case of variable interest rates—prohibits the introduction of ‘exotic’ mortgage products, with the consequence that there is much less speculation on house prices in Belgium compared to some other European countries and that mortgages are not used to back up consumption like elsewhere (see e.g., Elsinga et al. 2007).

This prudent attitude does not imply an unawareness on the part of the different governmental levels (of which the federal and regional levels are the most important) of the huge wealth stocked in housing. For instance, former Flemish and current federal Prime Minister Yves Leterme, a Christian-democrat with ties to the Christian trade union, raised the issue when stating that the paradox ‘poor government—rich citizen’ cannot last forever and that households will one day have to take responsibility for themselves (Leterme 2009). However, when policy-makers have sought to stimulate the use of this wealth, this has been limited to encouraging home-ownership among the children of home-owners. Some years ago, the federal government drastically lowered taxes on the donation to descendants of dwellings or plots of land for housing construction. More recently, the Flemish government agreed to introduce a so-called ‘mom-and-dad’ loan, allowing the donors an extra tax exemption for financing the purchase and renovation of a house by children or other relatives. It has to be stressed that this is basically nothing more than rewarding a practice that has been firmly in place for a long time (see for instance, Doms et al. 2001; Gedas et al. 2001; De Decker 2005). The achievement of owner-occupation by younger generations is partly financed by older generations, with around 18.2% of new ‘projects’ (purchase or construction) involving gifts or loans from family (mainly partial finance; Heylen et al. 2007).

To summarise, in this section we have established that poor governments/political parties are indeed thinking of home-ownership as a form of financial security which could be called upon in old age. However, so far there is no evidence that policy-makers are actively seeking to free the capital stocked in housing. Policy changes in the domain of housing have been mainly limited to making it easier—at least in theory—for the younger generations to purchase their own house. In part, this lack of policy developments towards more asset-based welfare might be linked to the fact that there is no societal consensus on this issue, and that any movement in this direction might be punished by voters.

4 Housing as a source of social security? Qualitative evidence from the OSIS research project

‘For me, my home is first of all a place of rest. A place where you can hide after work. It is like “coming home”. A place where you are at ease. A safe place. A place to retreat with the people you like’ (home-owner, woman, 50 years).

‘We are not investors’ (home-owner, woman, 37 years).

The main conclusion from our analysis so far is that, although there was and is no public or political debate on the issue, the asset-based welfare state in Belgium has been firmly in place for decades. Given the low level of public pensions, Belgian elderly provide for old age by accumulating assets, in the form of both owner-occupation and financial wealth. Due to the continuous promotion of home-ownership, housing thus represents a large share of households’ assets. But what do people think about the (possible) relationship between their housing wealth and issues of social security? Do they follow the reasoning that, given their wealth, they should pay more for things like health and elderly care or provide for their own (additional) pension?

To explore public attitudes on the potential of use of housing wealth in order to complement more traditional forms of social security, we use data from the OSIS Research Project¹⁰ (Origins of Security and Insecurity). This internationally comparative research project had two main objectives. The first was to analyse the factors that have impacted upon individual households and have consequences for their position as home-owners. The second was to establish: (a) how households perceive patterns of security and insecurity, advantage and disadvantage associated with different housing positions; (b) how these perceptions have moulded their personal strategies with respect to housing and jobs, family size, education and pensions; and (a) how these housing positions have provided them with material security or insecurity.

These research questions were framed by two international trends. The first is the ongoing increase of home-ownership rates in Europe, making it a continent of home-owners (Doling and Ford 2007). This increase is as much linked to increasing economic affluence as it is to government policies such as the sale of social rental housing (see Jones and Murie 2006) or the introduction of tax incentives (e.g., the Netherlands—see Elsinga 1995). The second is that in some countries (e.g., the United States, the United Kingdom, Finland), house prices have been volatile, manifesting booms and busts, with the consequence that owners were caught (at least temporarily) with negative equity or even experienced repossession of their home (Ford et al. 2001). Since house prices are linked to the economy, volatility is a basic feature, which raises the question of whether collective provisions like social security can or should be (partly) linked to something that is intrinsically unstable.

Given the fact that, in the long run, housing wealth tends to increase, given the forced austerity of the Belgian welfare state and given that ownership is increasingly associated with issues of social security (see Stevaert earlier), using a qualitative research design we probed our respondents on their feelings towards home-ownership, their strategies to secure home-ownership and their attitudes towards using the assets of their home for social security.

It is important to stress that the Belgian OSIS sample was dominated by urban residents. ‘Urban’ not only by location but also by attitude, since the large majority of respondents enjoy city life and, for the majority, moving to the countryside or to the suburbs is not a consideration. Thirty in-depth interviews with 20 owners and 10 tenants of an ‘active’ age were carried out in Ghent in June and July 2005. Of our 30 respondents, 27 lived in inner-city neighbourhoods, two in the suburbs and one further out. Six respondents had lost their home before and seven could be considered ‘poor’ owners. Only two respondents lived in detached houses and only one in a semi-detached house. All the others lived in either a terraced house or a flat. With respect to the interpretation of our findings, it is important to emphasise that this sample is atypical, as Belgian housing policy has a strong anti-urban bias and has channelled people towards buying detached and semi-detached houses in non-urban areas (Mougenot 1988; De Decker et al. 2005; De Decker 2008). A second cautionary note should also be made as, although this kind of qualitative research does not seek to be representative, in our sample, highly educated respondents are overrepresented.

The main research instrument was a topic guide which structured the in-depth interviews. This guide was designed to cover issues relevant to households’ appraisal of their housing security and insecurity. In each country, the same topic guide and analytical coding framework was used to structure the research and analysis (Toussaint et al. 2007).

The OSIS research findings shed considerable light on the conclusions reached in the previous sections. Several issues were highlighted concerning home-ownership among individually and publicly funded means of social security. Also significant were the opinions of respondents on the quality of the Belgian social security system, including their attitudes towards using housing wealth to supplement pension incomes. Before reporting our results concerning these two issues, we have to stress that nearly all respondents used a diverse range of strategies to become an owner-occupier in an affordable way in the first place (De Decker 2007).

4.1 Private safety nets are important

Although rational calculation and confidence in future earnings are prominent in becoming a home-owner (De Decker 2007), we also observed some ambiguity—or even anxiety—since most of our respondents used a wide array of private safety nets. Some even combined a range of more or less private, additional arrangements (see Table 4): different private insurances, building up savings, calculations of the increased value of the dwelling in case of a forced sale, social security benefits and other state aid (e.g., insurance against income loss—see earlier), private pension funds, insurances against illness and hospitalisation, and finally, help from family and friends¹¹. What was important here was that the higher the professional status, household earnings and household stability, the more safety measures these households considered or had—this was particularly true for hospital insurance, which is typically seen as a luxury¹². This means that the wealthier people are, the more they are protected, and the poorer they are, the less they are protected. The latter group mainly consists of lone mothers with children, which brings relational instability to the forefront as a major source of housing and financial insecurity (also see De Decker and Geurts 2003).

Table 4 Ghent sample, resources of home-owners

Private safety nets are not only of a monetary nature, as they also include the possibility of financial support from family and, to some extent, friends. Although all respondents stressed that they would only call on family and friends when all other roads are closed, a large majority did consider it a viable option. Note that this is not a virtual safety net but a real one, already used by different respondents. For example, following a divorce, one of our respondents stayed in the marital home. During the period between breaking up with his first wife and meeting his second partner, he faced hardship, including a threat of being placed on the ‘black list’¹³. During the hardest times, he received a lump-sum from his parents to see him through. Another self-declared ‘poor’ homeowner, who had faced a divorce, unemployment and a house fire, was helped out on different occasions by friends. She got informal loans that she could repay without interest, at a suitable time. A further example was provided by a 32-year-old woman with two young children who faced difficulties in paying the mortgage when her husband moved out. She received some financial support by her brother, whom she had to repay, and from her mother, who was caring for her children when she was at work. A final example was a disabled person facing divorce. She was also pretty sure that she could count on her parents to help out in case of financial problems.

4.2 Is social security securing?

From the previous sections, we have learned that two major strategies overlap. On the one hand, there is the ‘risk-avoidance strategy’ when entering home-ownership, and on the other hand, we have seen that an overwhelming majority of respondents use a wide spectrum of private safety net arrangements (although most of these are not directly linked to home-ownership). This shows confidence and prudence at the same time. If we probe deeper into the attitudes and strategies of households with respect to risk, different factors come to the forefront. The first is that most respondents still strongly believed in the social security system. This was often because they had already enjoyed the benefits: some respondents had already benefited from unemployment transfers, disability/illness payments or an adaptation of the interest rate of a social loan following an income drop. This belief in the social security system becomes even clearer when analysing the response to questions on a possible introduction of reversed mortgages¹⁴ as a supplement to (public) pension income. Most respondents not only considered it a bad idea to use housing wealth to complement their pensions, but also expressed the hope that the social security system would never reach a state when this would be necessary. A majority of respondents agreed that the Belgian social security system offers adequate protection against major set-backs like unemployment and illness, and provides sufficiently high pensions (although most are saving for old age). And if not, they believed that the state should offer decent protection. The following quotations illustrate respondents’ aversions concerning reversed mortgages and their opinions on the role of the social security system.

‘they [the banks] are a gang of thieves. I think that what is left should go to the children. (…) It is a scheme for couples without children’ (homeowner, woman, 63 years).

‘Basically you give your house to the bank. It is beneficial to the bank. I’ll try to save myself for my pension. It is not a good system’ (tenant, woman, 24 years).

‘It is not a good scheme if you have children. Above that, it is a dangerous scheme since you become dependent on a bank’ (homeowner, man, 38 years).

‘The banks profit from the fact that pensions are low. I hope that it will not happen over here. I count on our social security system. We pay for it… and what’s more, I do not want the power over my pension to go to the bank’ (tenant, woman, 26 years).

‘I do not think it exists over here. But I think it will come. If you listen to the statements of the likes of Vande Lanotte ¹⁵ on ownership and you look at the pensions. I interpret this as a clear hint that there is not enough money for paying our pensions. (…) It is not a good system. The owner-occupied house is the classic method for the worker to acquire some property. They have no access to other forms of capital formation. They don’t do that. It is not fair. If they need money, they should look at other forms of capital formation’ (homeowner, man, 29 years).

The importance attached to the social security system becomes even more obvious when people were asked whether the state has any responsibility for helping people secure owner-occupation. However, answers indicated that people distinguish between housing issues (including home-ownership) and social security (a system that protects against set-backs like unemployment, illness and old age). So, when asked whether the state has any role or responsibility if an owner fails to pay for his/her mortgage loan, most respondents were very cautious. To a large extent, they felt that owners are responsible for their own decisions. For some, it was a matter of principle that the state should not help owners (one 37-year-old civil servant had, as he put it, ‘a Thatcherite view on it’). The general line of reasoning was that if you decide to become a home-owner you are aware of the consequences. And ‘if something goes wrong, you should limit your dinner to some slices of bread with jam for a while’.

Respondents saw no need for safeguarding mortgage repayments of ‘would-be owners’ since there exists an elaborate social security system. According to a 39-year-old man who owned a large house where he lived with his wife and two children, Belgium has a ‘pretty good social security system’ and that should be sufficient. One respondent feared misuse and suggested that no more than minimal support should be provided: ‘a temporary roof, not more’. For a 35-year-old woman the situation was such that, ‘in our country, when you lose your income, there is always something. You will not end up on the streets’.

In elaborating on this question, only one respondent, a single mother now working for the housing administration of the city of Ghent, linked her opinion to housing policy more generally. Since the government has stimulated and keeps stimulating ownership, according to her, the government has to be consistent and provide a safety net. This implies that she is in favour of the already existing free insurance against income loss for low-income households (see De Decker 2005). According to her, however, it should be more general and generous.

5 Conclusion and discussion

In this article, based upon the consideration of policy documents and both qualitative and quantitative data sources, we have provided a sketch of Belgian housing asset and welfare systems and assessed the extent to which the state has joined the trend towards the promotion of home-ownership as an alternative form of social security provision.

Although it is often assumed that in countries with high home-ownership rates, pensions can be less generous because of low housing costs in old age (Castles 1998; Kemeny 1981), our quantitative evidence shows that, at least for Belgium, this claim might turn out not to be completely true. As opposed to earlier research on this topic, our quantitative information includes all housing costs (not just rent or mortgage payments). Using this information to recalculate income poverty risks, we actually found that the poverty risk increases for all tenures, although much more for both private and social renters. In fact, the housing costs of elderly home-owners turn out to exceed their income from other assets. This somewhat puzzling finding can be linked to the fact that the Belgian housing stock is rather old, with many elderly households concentrated in badly-equipped and badly-isolated houses, which are often too large for their needs. We could thus say that for these elderly, freeing up housing wealth by, for instance, moving to a smaller, more modern house, could have a positive impact on both the quality of housing and housing costs. From a policy perspective, this avenue might be worth looking at. On the other hand, based on the fact that compared to elderly home-owners, elderly renters face extremely high poverty risks, we could say that in Belgium, an asset-based approach to welfare has actually been in place for decades. While the elderly poverty risk for home-owners is somewhat higher than that for the general population, it is particularly high for elderly renters. As this group receives virtually no help with their housing costs in Belgium, economic well-being in old age is already to a fairly large extent based on assets. Or, as found in our study, Belgian elderly are income-poor but asset-rich.

As far as the preconditions for a possible restructuring of the Belgian welfare state in the direction of more reliance on asset-based welfare are concerned, we can conclude that most of them are fulfilled. Although decreasing during the last 15 years, from a comparative perspective public debt is high. Furthermore, the costs of population ageing (both in terms of pensions and in terms of health care) are looming large on the horizon, while the so-called Silverfund—which was supposed to bridge the most difficult years through very high social spending—is threatening to turn into an empty box. On the other hand, both financial and housing wealth accumulated in private households has been increasing for decades, and currently about three-quarters of the elderly are outright home-owners.

Nevertheless, recently the only major political party defending the interests of (social) renters has also turned to the promotion of home-ownership as the ‘ideal’ tenure. Belgium is in many ways an extremely conservative country. While some housing policies have made it easier to pass on housing wealth to the future generations (again promoting home-ownership), virtually no initiatives have been taken to convert housing into other forms of wealth, neither for the younger generations (as there are for instance no ways to convert housing wealth into consumption through re-mortgaging), nor for older generations (no equity release schemes). This lack of policy interest in the possibilities of using the housing wealth concentrated in private households is also reflected in the views of the respondents in our qualitative research. Most of them use a whole range of risk-avoiding strategies when they enter home-ownership: by making sure they can afford repayments and by protecting themselves through a range of (private) insurances. Furthermore, they see no role for the state when owners do not manage to pay their mortgage, and they are extremely resistant to the suggestion of using housing wealth in order to supplement (public) pension income. To summarise, we can say that in Belgium, housing is considered as a private issue, separated from the social security sphere. Hence, it seems that there exists a real trade-off between housing and pensions, as the class of owners-occupiers in Belgium is very reluctant to trade housing wealth for a higher pension.

This brings us to our last point of discussion, which might be circumscribed as ‘home-ownership, limits to growth?’ During the last decades, home-ownership has increased from 65.4% in 1991 to 69.5% in 2001, but this has only been possible through the combined effects of low interest rates, the increase in dual-earner households and because of intergenerational transfers which have made it possible to keep up with rising house prices. There are however, a number of trends which might limit a further increase in home-ownership, such as the increase in single-person and single-parent households (for instance because of rising divorce rates), and increasing job insecurity. At the same time, older generations are living longer and might have to consume more of their wealth in order to sustain their living standards or to pay for their health care needs. Thus, there is a distinct possibility that those people who do not manage to achieve owner-occupation—which currently amounts to about one-quarter of the population—will no longer diminish, while their financial position might actually be worse than in the past. For these people, there are no assets to base any welfare on.

Appendix

Our life-style deprivation index is based on 18 so-called non-monetary indicators referring to financial stress and life-style deprivation arising from a lack of economic resources (Dewilde and Vranken 2005; Whelan et al. 2001). Items are coded as 1 (deprived) or 0 (not deprived); the deprivation score is calculated as the weighed sum of all items, with the weights corresponding to the proportion of non-deprived. This way, situations of deprivation which are less common in the population, and thus lead to stronger feelings of relative deprivation, are weighted more heavily. We divide the total sum by the sum of the weights, so that our index ranges from 0 (no deprivation) to 1 (completely deprived). Cronbach’s Alpha (a measure of ‘internal consistency’) for this index is 0.76 (see Table 5).

Table 5 Non-monetary indicators of poverty/deprivation (SILC Belgium 2006)