‘New’ but ‘Squeezed’: Middle Class and Mortgaged Homeownership in Croatia

Abstract

Some recent anthropological accounts of middle classes centred on their indebted home-ownership. They stressed its two contrastive logics fitting a wider binary – exposing ‘squeezed’ middle classes in the global North to increasing risks, and supporting the ascent of their ‘new’ counterparts in the South. The genealogy of middle-class housing debt in Croatia presented in this article reveals another, post-socialist trajectory where mundane and opaque institutional practices regulating access to housing finance, such as bank credit scoring and the allocation of state housing benefits, were key in steering a middle class inherited from socialism towards mortgaged home-ownership. The latter was articulated as a middle-class experience only after the 2000s credit boom had come to an end and the consequences of rampant predatory lending became visible and subject to contestation. The resulting middle-class subjectivities are ambiguous and, as comparisons with other Eastern European cases suggest, accessible for a range of political projects.

The past two decades have seen a wave of anthropological accounts of populations (self-)identified as ‘middle class’ (Donner, 2017; Weiss, 2019: 17–18). Broadly in line with public discourses on the matter, this work has tended to depict two faces of contemporary middle classes: the ‘squeezed’, struggling and declining middle classes in rich countries of the global North, and their ‘new’, booming and aspirational counterparts in the global South. Some anthropological accounts highlighted the key role of housing finance in this double dynamic. To illustrate, Noelle Stout (2019: 38–44, 107–15) argued that middle-class Americans, due to legacies of post-war Keynesian policies, used to see secure and affordable state-backed mortgages as their entitlement. They held on to this notion as they accumulated unprecedented levels of debt to offset the neoliberal stagnation of incomes. But their struggles with over-indebtedness and dispossession after the global financial crisis (GFC) made them realize that the erstwhile ‘moral economy’ was gone, and with it their ability to attain middle-class life projects such as holding on to one’s home and bequeathing it to children (Stout, 2019: 14–18, 133–8; see also Zaloom, 2019: 10–11, 49, and passim). In contrast, Deborah James (2015) traced the aspirations of South Africa’s ‘new black middle class’ that fuelled the debt boom after apartheid. Among other purposes, its members took on debt to buy homes in higher status suburbs, build houses in their villages of origin, or make profit-oriented investments in property (James, 2015: 14, 51, 159, 173–200). While debt was a fragile basis for their upward social mobility, some mobility undeniably occurred. As Hadas Weiss (2019: 46–7) notes, these seemingly contradictory effects reflect different logics of the same global process of financialization in different sites of the unequal world economy – its mature forms in cores expose established home-owning middle classes to ever-higher levels of debt and volatile asset prices while its expansive stages in peripheries widen access to home-ownership and middle-class lifestyles. This helps to make sense of the contrastive impacts of financialization on the capitalist ideology of the middle class as representing a possibility for workers to escape their dependence on wage labour and achieve security through private ownership of assets such as housing (Weiss, 2019) – its erosion in the North and its boom in the South.

This article examines the relationship between mortgaged home-ownership and the middle class in post-socialist Croatia, part of the Eastern European region that has been less represented in this debate. The few relevant anthropological studies on Poland and Czechia focused on particular groups of mortgagors, especially young couples with children inhabiting new suburbs and (self-) identified as middle class in terms of education, profession, income and social habitus (Halawa, 2015; Olcoń-Kubicka and Halawa, 2018; Samec, 2016). These people were described as pursuing the norm of owner-occupied housing as the appropriate kind of housing, and simultaneously an investment object, for the middle class. Research on housing practices of Slovak and Hungarian middle classes stressed their preference for suburban or rural family houses and ‘organicist’ aesthetics (Buzalka, 2021; Fehérváry, 2013). The emphasis on the aspirational motivations of Eastern European mortgagors/home-owners and on public discourses portraying them as the quintessential subjects of the post-socialist market democracy (Halawa, 2015: 720) puts them in the same category as the new middle classes of the global South. Their shared recent experiences of economic liberalization, assumed to benefit middle classes, also justify this treatment.

In what follows ¹ , I offer an alternative account that (1) stresses the role of institutions regulating access to housing and housing finance in shaping the association between the middle class and mortgaged home-ownership in Croatia and (2) fleshes out the complex interplay of continuities and discontinuities that characterize this relationship and prevent it from being slotted neatly into the new/squeezed dichotomy. Although both housing finance and middle classes existed already in socialist Yugoslavia, I will argue their coupling took place only in the post-socialist period and initially mainly as the structural effect of state housing policies and bank credit scoring models that steered established middle-class groups into mortgaged home-ownership rather than as intentional, distinction-seeking strategies of a self-identified middle class. In the key period of the mortgage boom in the 2000s, hegemonic public narratives and common sense naturalized mortgaged home-ownership and obscured its linkages with class. It was only after the boom, as the consequences of predatory lending became apparent and subject to contestation, that mortgage indebtedness has been publicly articulated as a middle-class experience. This dovetailed with a wider narrative about the Croatian middle class being in decline and victimized by the recession and austerity after the GFC. In this context, mortgagors identified themselves as middle class while simultaneously questioning this identification. The Croatian trajectory of housing financialization has thus contributed to a formation of a middle-class subjectivity that is ‘new’ in terms of its ideological construction and economic experiences but ‘squeezed’ already at birth – bearing a sense of lack, disappointment and insecurity quite similar to the ‘old’ middle classes in the West. The analysis reveals a distinct post-socialist trajectory where financialization neither provokes the decline of an old middle class nor fuels the rise of a new one but rather reshapes the existing middle class by wedding it to mortgaged home-ownership with distinct implications for middle-class subjectivities and politics. It also demonstrates the role of overlapping classification practices of financial and state institutions in class formation.

Mortgaged home-ownership and transformations of Croatia’s housing regime

From the Second World War up to independence in 1991, Croatia was a sovereign republic within the Socialist Federal Republic of Yugoslavia. Yugoslav housing policy privileged the provision of public rental housing. Tenants used ‘social flats’ (društveni stanovi) on the basis of a right of occupancy that worked similarly to private ownership in practice while conferring the benefits of highly subsidized costs (Archer, 2016: 62). Despite massive construction of public housing, owner-occupation remained the dominant housing tenure up to the end of Yugoslav socialism. According to a 1990 survey, 69.6% of the housing stock in Croatia was owner-occupied and only 25.5% was in public ownership (Hegedüs et al., 1996: 115). Due to intense rural–urban migration, large cities experienced chronic housing scarcity (Archer, 2016: 60; Duda, 2009: 118). Search for adequate and stable housing was thus a major preoccupation for many Yugoslavs, as pop culture also reflected (Patterson, 2014: 277–8).

Despite frequent criticism, the allocation of convenient and prestigious public housing openly favoured white-collar workers (managers, administrators and experts) and, to some extent, highly qualified manual workers (Archer, 2016; Duda, 2009: 34, 116; Marčetić, 2020: 30). This reflected competition between companies for scarce ‘cadres’ (Archer, 2016: 70). Those unable to access public housing, such as manual workers and those outside the public sector (peasants, small business owners), sought to acquire housing into private ownership through self-construction or purchase. Various forms of credit were commonly used for both purposes. In Zagreb, credit-funded self-construction was common in peripheral areas with poor infrastructure and abundant cheap land (Marčetić, 2020: 34–7). Mirela ² , my second case study below, recalled how her grandmother used to complain about the injustice of not receiving a ‘state flat’ (unlike her parents) and having to take out a loan. Again, those higher up in socialist hierarchies enjoyed better treatment: housing loans issued to insiders by companies and self-management organizations were cheaper than those from commercial banks, and complaints about their unfair distribution and abuses were rampant (Duda, 2009: 118; Kujović, 1985: 323–5). Those unable to offer a down payment likely could not take a housing loan at all and had to resort to consumer credit (Archer, 2016: 67). Overall, credit seems to have been somewhere in the middle of the hierarchy of means of accessing housing, with public housing being the most desirable solution and the unregulated, low-quality and exploitative subtenancy being at the opposite end of the spectrum (Archer, 2016: 67; Bežovan, 2018: 151–2).

After socialism, in line with the general Eastern European trend (Stephens et al., 2015), Croatia developed a housing regime based on the dominant role of the family and an increasing role of the market in housing provision (Rodik et al., 2019: 342). Rodik et al. (2019: 326–9), identify three stages of this housing regime transformation: ‘transition’ in 1991–99, ‘financialization’ (more precisely the mortgage boom) in 2000–8, and ‘crisis’ (or bust) in 2009–15. To be added is a stage of ‘recovery’ since 2016, now likely coming to a close. The transition overlaps with the right-wing authoritarian regime of Franjo Tuđman and its first half was marked by the Croatian War of Independence. Nearly all public housing stock was privatized, mostly at discounted prices to the sitting tenants. The state cut its role in housing provision to social housing for welfare recipients and housing benefits for war veterans and returnees (Bežovan, 1998: 25–7). The wholesale privatization of housing resulted in very high rates of home-ownership on par with other ‘super-homeownership’ societies in Eastern Europe (Stephens et al., 2015). Nearly 83% of households were owner-occupiers already in 2001 (DZS, 2001). The rate of home-ownership reached nearly 89% as of 2011, around which level it has remained since (DZS, 2011; Rodik et al., 2019: 320).

Housing construction and the market stagnated during the 1990s owing to (post-) war conditions, including the inaccessibility and very high costs of credit. The latter changed quickly after the regime change in 1999, as swift financial liberalization and foreign privatization of the banking sector set the scene for the peripheral financialization of Croatia’s economy. This process was largely driven by Western European banking groups that acquired nearly the entire banking sector and imported large volumes of interest-bearing capital (Mikuš, 2019b). Lending to households and real estate developers was one of the key investment outlets for this capital. In a setting of pent-up demand, an unregulated and speculative housing market, and insufficient supply in high-demand areas, a major mortgage and housing boom ensued. Housing debt grew eightfold and total household debt sevenfold in the period from 2000 to 2008 (Rodik and Žitko, 2015: 60). House prices increased by about 63% during the same period, reaching highly unaffordable levels (Vizek, 2009).

The state also expanded its role in housing provision. Subsidies for housing savings had been in place already since the late 1990s (Marčetić, 2020: 63). After the regime change, income tax deductions for mortgage interest payments were introduced in 2000 (Bežovan, 2018: 155) and the State-Subsidized Housing Construction Programme ensued in 2001. Better known under its abbreviation POS, the programme supports recipients in buying newly built apartments by subsidizing both construction and mortgage borrowing from private banks (Bobovec et al., 2016: 220–1). The programme’s scope has remained limited: only 2.8% apartments built between 2001 and 2017 were POS apartments (Rodik et al., 2019: 326). Still, its concept indicates the state’s support for private home-ownership and commercial mortgage finance. The share of ‘housing transitions’ (changes in individuals’ housing status) resulting in indebted home-ownership increased from 8% during the transition to 18% during the boom (Rodik et al., 2019: 333).

Croatia’s debt boom, as in other Eastern European countries at the time, was characterized by a ‘mainstreaming of predatory lending’ – targeting of both subprime and prime borrowers with practices such as foreign currency loans and variable interest rates that banks adjusted at their discretion (Mikuš, 2019a: 297). Swiss franc loans (mostly housing loans), with principals indexed to the franc but repaid in the local currency, proved particularly destructive as the appreciation of the franc after the GFC drastically inflated principals and repayment instalments (Rodik, 2019; Rodik and Žitko, 2015). Mortgage lending, housing transactions and prices plummeted already in 2009 and remained at low levels during the severe recession in 2009–15. Non-performing household loans, defaults, debt enforcement proceedings and foreclosures shot up, and became significant public and political issues (Mikuš, 2019a, 2020). While private renting became somewhat more common, most people still practice it only as a temporary solution on a path to home-ownership (Rodik et al., 2019: 333, 334–5, 343). The state’s orientation to mortgage finance has become even more apparent as several new housing programmes expanded their subsidies to mortgage loans for the purchase of any housing (Bobovec et al., 2016: 221–2; Marčetić, 2020: 126–7).

This trend in housing policy continued in the post-recession period after 2016, which saw a strong but geographically uneven growth of house prices (Tica, 2020). While beneficiaries of the earlier housing programmes would typically take out two separate mortgage loans – one from a commercial bank and the other, at a fixed and subsidized rate, from the state Agency for Legal Transactions and Mediation in Real Estate (APN) – the most recent subsidy programmes introduced in 2017 marketized the process even further by requiring applicants to sign a bank mortgage contract before applying for the subsidy. The projected cost of the four cycles in 2017–20 is close to the cost of the much longer POS programme without any construction being subsidized. What is more, the main aggregate effect was to boost the growth of house prices (Kunovac and Žilić, 2020; Marčetić, 2020: 127–33).

Case study 1: Katarina and Zoran, POS beneficiaries

Katarina and Zoran were the only POS beneficiaries among the approximately 35 mortgage debtors whom I interviewed during my fieldwork in 2016–17. Zoran was an aviation technician in his early sixties and Katarina was a court clerk in her fifties. They lived together with their two adult children, university students, in an apartment in Špansko-Oranice, one of three largest POS estates in Zagreb. Špansko-Oranice is located in a semi-peripheral location west of the city centre, outside of the tram zone and next to Zagreb’s main arterial road. It was built in 2004–9 and includes some 1500 apartments in several dozen densely built multi-storey blocs. Katarina and Zoran purchased their three-room, 65-square-metre apartment in early 2004, before it was even completed. They paid €930 per square metre – about two thirds of the Zagreb average of €1369 in that year (Vizek, 2009: 283), and the lowest price that any of the mortgagors I interviewed paid during the boom.

Katarina and Zoran used what is still the typical POS model of mortgage funding (APN, n.d.): they took out two mortgages, one from a private bank and the other from the APN. The bank loan was denominated in euros and had a variable interest rate starting at 6.66%, about one percentage point above the market average (Rohatinski, 2008). Katarina and Zoran repaid it in 2016. The APN loan was kuna-denominated and had a fixed 3% interest rate, on top of which came a 1% interest rate for the period of 12 years when the bank loan was being repaid and the APN loan was waiting in queue. The couple did not present their repayment conditions as privileged; on the contrary, they mentioned they had to tighten their belts due to higher interest rates during the recession. I estimated that in 2016 the household spent about 13% of its net income for debt repayment, which was one of the lowest debt-to-income ratios among the debtors I interviewed. When I video-called with Katarina in 2020, she told me that Zoran passed away in 2018 but she was able to manage the repayment with her salary and the family pension to which she became entitled after his death. Overall, then, despite being exposed to market conditions through the bank arm of the loan, Katarina and Zoran did benefit from the POS programme and avoided over-indebtedness, unlike many other Croatian mortgagors. Their experience illustrates the higher degree of decommodification in the POS programme through the double subsidy – for the house price and the interest rate – compared to the post-GFC mortgage subsidy programmes, in which only the second component remained.

The mortgage boom in hegemonic narratives and common sense

I found little evidence of the link between mortgaged home-ownership and middle-classness being articulated during the 2000s debt boom. As I discuss in more detail elsewhere (Mikuš, 2022), elite actors such as the financial community, experts and the media employed two hegemonic public narratives to make sense of the expansion of household debt. First, a naturalizing narrative explained this process as an inevitable part of the transition to liberal democracy, free-market democracy and Western standards of living. A naturalization of the mortgage and housing boom specifically can be discerned in business press articles that, despite noting poor regulation and rampant speculation, predicted that the growth of house prices would happily continue until a market equilibrium is reached (Brkanić Kulenović, 2006; Dokonal, 2005). Commentators also pointed to the growing housing wealth of Croatian households as evidence that they were not over-indebted and used credit to make sound investments (Martinović, 2007; Stojić, 2006). Some of my research participants echoed this narrative when they remembered the ‘atmosphere’ of the boom and the expectations it generated. Ivan, an IT worker in his forties, talked about growing salaries and how these fed aspirations to reach ‘European values’ also in one’s standard of living. His friends kept telling him ‘Why haven’t you bought a flat yet? What are you waiting for? The price for a square metre goes up a 100 [euros] a month.’ Several others mentioned how the vertiginous growth of house prices in Zagreb pushed them to buy before it would become entirely unaffordable.

The second hegemonic narrative was an individualizing one with strong culturalizing and moralistic overtones. It attributed the debt boom to an alleged ‘consumer mentality’ of Croatians – a desire for status goods even at the cost of excessive indebtedness (Burić, 2010: 11, 15–16, 23–5). However, such critiques of ostentatious consumption usually focus on consumer goods like cars, clothing and mobile phones or on vacations. If housing is mentioned within this frame, it is usually as something that, in contrast, would have made a sensible investment (Burić, 2010: 24, 189). This moral distinction between rational and prudent borrowing for assets such as housing and education, and irrational and imprudent borrowing for consumer goods, is far from unique for Croatia, although the exact boundary between the two varies depending on context (e.g. James, 2015: 14; Samec, 2016: 43). It is found in expert and lay discourses alike and it reflects precisely the kind of middle-class rentier strategies based on assets expected to conserve – and hopefully increase in – value (Weiss, 2019).

Some of my interlocutors hinted at these ideas when they told me that they believed, or used to believe, that real estate is always a good ‘investment’. However, they understood this rather as saving – accumulating by acquiring an asset instead of ‘throwing money away’ and ‘having nothing in the end’ as with renting. Mortgages motivated by clear expectations of financial profit were rare; I interviewed only two households who took out two mortgages each to buy ‘investment flats’, that is, flats intended for commercial renting. Most mortgagors bought homes for themselves and their families, and cited ideas about affective and use values of owner-occupied housing as their motivations. They stressed the full ownership and control over such housing, and therefore its higher stability and quality compared to unregulated and unstable private rentals as the only real alternative. They often related the need for owner-occupied housing to a stage in their life when they wished or needed to ‘resolve their housing question’ because they wanted to set up an independent household with a partner, wanted or already had children, or were simply exhausted after years of renting (see also Rodik, 2019: 85–92). Most decisions in favour of mortgaged home-ownership were thus motivated by a mix of considerations (reproductive needs, economic common sense and social values) and taken under conditions of a housing regime that offers young people limited, if any, alternatives to owner-occupation – and consequently mortgage borrowing, if they can afford it, and unless they are the lucky few to have received substantial wealth from parents.

In relation to the mortgage and housing boom specifically, a third narrative has emerged as hegemonic: one that explains these processes and the dominance of home-ownership as the result of Croatians’ ‘traditional’, cultural preference for owning their home, described with phrases such as ‘to be oneself in one’s own place’ (biti svoj na svome), ‘fetish for real estate’, ‘something in our genes’ or ‘our mentality’ (see also Marčetić, 2020: 10–12; Rodik, 2019: 85, 89). A business journalist thus argued that one of the causes of the boom was ‘Croats’ traditional relationship towards real estate. Having one’s own flat or house is the milestone of success in life’ (Dokonal, 2008: 23). Obviously, this narrative renders the ‘super-homeownership’ regime a matter of irrational, atavistic preferences while obscuring its structural and political causes. The narrative is frequently evoked in the media and I have heard it time and again from experts, finance professionals and debtors alike. Nevertheless, debtors rarely used it to account for their own mortgaged home-ownership. In sum, then, the mortgage boom was naturalized in part through hegemonic public narratives about the household debt expansion in general, but more importantly through narratives and common sense that evoke the unique cultural, affective, use and financial values of owner-occupied housing.

Credit scoring, subsidy eligibility and middle-classness

Although the relationship between mortgaged home-ownership and the middle class was rarely explicitly articulated in the 2000s, this section shows that middle-class selectivity was built into the way in which mortgaged home-ownership was institutionalized. It is important to note that local folk concepts of middle class have been shaped by its local scholarly understandings. Since the 1970s, Yugoslav sociology increasingly recognized the existence of a socialist middle class of white-collar managers, administrators, technical experts and so forth, typically denoting it with the less contentious term ‘middle stratum’ (srednji sloj). Its chief attributes were the Weberian criteria of occupation (labour-market position), status and lifestyle, while the criterion of ownership was not so relevant. In post-Yugoslav sociology, another stratification approach to social class, a neo-Bourdieusian one, with an emphasis on cultural distinctions, has emerged as dominant (Tomić-Koludrović and Petrić, 2014a, 2014b). Broadly in line with these approaches, contemporary folk models of social class centre on educational, occupational and income strata, and lifestyle groupings. ‘Middle class’/’middle layer’ is understood as positioned high in educational and occupational hierarchies but more in the middle, between the ‘upper’ and ‘lower’ classes/layers, in income and consumption hierarchies. Partly overlapping with this, it is also associated with sophisticated urban lifestyles and two broad occupational categories – white-collar professionals with a strong foothold in the public sector (doctors, teachers, academics, etc.) and private entrepreneurs.

The selectivity of mortgaged home-ownership for a middle class in this sense was institutionalized most of all through bank credit scoring practices that determined access to mortgage credit. Economic sociologists Fourcade and Healy (2013) note that the risk-based classification of individuals in credit scoring results in their differential treatment through granting or refusing access to credit and through pricing. Crucially, these practices reflect pre-existing social inequalities while generating new ‘artificial’ classifications with potential class-like effects. My analysis of credit scoring in Croatia, more often called ‘creditworthiness assessment’ or ‘credit risk assessment’, relies mainly on interviews with banking professionals. The general consensus among them was that credit scoring during the 2000s lending boom was lax. The Croatian Credit Information Registry (HROK) only became operational towards the end of the boom, in May 2007 (Banka, 2007). Prior to that, only limited data on individuals’ irregular repayment had been available via the Information Exchange System (informally the ‘blacklist’) run by the Croatian Banking Association (HUB) since 2003 (Bratoš, 2005). Apart from these technical considerations, some interlocutors admitted that banks operated loose credit policies. To mention one major example, credit risk assessments for Swiss franc loans systematically underrated the actual risk of repayment (Mikuš, 2019a: 308; Rodik, 2019: 100–2).

Đurđa, a woman working in banking since 2002, offered the deepest insight into bank credit scoring procedures. From 2009 onwards, she worked as a retail risk manager in a small bank and designed its credit scoring model around 2010. Although this was already after the debt boom, the model was based on other banks’ older models. According to Đurđa, these models generally shared the same main criteria. The most important criteria of her model were: the level of the applicant’s income; their HROK credit report; the duration of current employment; and the creditworthiness of the employer. I found the latter two components less obvious. Đurđa explained that a longer duration of current employment attracted a higher score. The assessment of the employer’s creditworthiness focused on the company’s history, size and business results. Public-sector organizations and large joint-stock companies were automatically seen as more creditworthy than small private companies. As in other contexts (e.g. James, 2015: 2–3 and passim), and putting aside the small demographic of those with high but irregular incomes (successful entrepreneurs and freelance professionals), the scoring model preferred those who earned regular and relatively high salaries based on a long-term employment in the public sector or large and stable private companies. Although these characteristics cannot be simply equated to the middle class, the overlap with its local folk concepts is clear.

State housing subsidy programmes likewise incorporated some criteria favouring the middle class in the framework of welfarist and pro-natalist policies. From the start, eligibility for the POS programme was limited to citizens who met the creditworthiness criteria of the APN and the partner bank. In other words, those unable to satisfy the credit scoring criteria of private banks were excluded by default. The prioritization of individual applicants had to reflect their housing status but beyond this, it was up to local governments to determine exact criteria (Croatian Assembly, 2001). The original POS rulebook of the City of Zagreb lists 11 criteria, with most potential points being attached to the applicant’s housing status, age, degree of disability, and household size. However, the not insignificant ‘qualifications’ criterion shows a curious continuity with the socialist-period privileging of those with higher education. While those with primary education get no points for this item, every additional level of educational achievement, including every additional postgraduate qualification, attracts extra points, and there are even extra points for honours-grade degrees (City of Zagreb, 2002). In addition, a pro-natalist logic can be discerned implicitly in the criteria of age (those aged 23–40 get most points) and household size, and explicitly in another criterion awarding extra points for each child. Already since the mid-1990s, and increasingly under right-wing governments since the mid-2010s, Croatian policies stress a need for subsidized mortgages and other housing benefits for ‘young families’ (child-bearing heterosexual families) in order to counter ‘demographic problems’ of population ageing and decline (Bežovan, 1998: 27; Bobovec and Mlinar, 2013: 155; Marčetić, 2020: 125–6, 127; SIFMYP, 2002: 27, 43). This natalist discourse intersects with a discourse about squeezed middle classes when it identifies a ‘brain drain’ as one ‘demographic problem’ that such policies should mitigate (Marčetić, 2020: 126). The qualifications criterion in the Zagreb rulebook can be seen as putting this connection into practice.

The same focus on the highly qualified was central in several cycles of mortgage subsidies for employees of public research and higher education institutions, which were implemented from 2003 to 2008. While administered by universities, this was a central government programme whose repeatedly stated purpose was to prevent the emigration of ‘young scientists’ (Hina, 2007; NZS, 2003a). Despite demands of the research and higher education labour union (NZS, 2003a), the programme always distinguished two benefit tiers: subsidized mortgage loans for those in research and lecturing positions, and non-subsidized loans based on partnership agreements between universities and banks for all other personnel. The difference between the benefits was considerable: in 2003, academic staff in the Zagreb region could get loans with fixed 2.1% rate (a true bargain at that time), while non-academics could only obtain 6.29% (NZS, 2003b). More than 2600 subsidized loans have been approved (Hina, 2007).

Statistics on the Croatian mortgagor population are consistent with the presumed effects of these components of bank credit scoring procedures and the eligibility and prioritization criteria for mortgage subsidies. A 2011 survey showed that housing debt was highly concentrated in households in the tenth decile of the income scale (Rodik, 2019: 55). A 2014 survey revealed that those with graduate and postgraduate degrees were particularly likely to have a housing loan, as well as those with above-average incomes but not those in the highest income rubric (Rodik, 2019: 53). While not referring to housing loans specifically, two surveys in 2005 and 2008 showed that public-sector employees were the employment category with the highest probability of repaying bank loans, which the authors explained by their high creditworthiness (Herceg and Šošić, 2011: 211).

Case study 2: Mirela and Nenad, beneficiaries of the academic mortgage subsidy

Mirela, a researcher in a public research institute in her forties, lived together with her husband Nenad, also an academic, and their high school-aged daughter on one of the many socialist-period estates in the southern, chiefly residential district of Novi Zagreb. Mirela and Nenad bought their two-room apartment in 2001, using a combination of a commercial mortgage loan and savings. The starting interest rate was 6.8% – still reasonable at the time – but it soon fell to about 4%. Some two years later, the opportunity came up for Mirela to refinance it with what she described as a special loan of the Ministry of Science for research novices. This was the very first wave of the subsidized loans for academics with fixed 2.1% rate. In fact, Mirela and Nenad originally considered buying a larger apartment with the new loan, but decided to merely refinance since the deadline was so short. Indeed, in Zagreb in 2003, the official application documentation was published in mid-October and applications had to be submitted by November – complete with a signed purchase contract for the apartment (NZS, 2003b). Mirela remembered the breakneck hunt for flats in the academic community that fall and believed that it contributed to the price boom in Zagreb, a claim repeated by several other interlocutors. A Croatian National Bank (CNB) study confirmed this kind of causality for the mortgage subsidies introduced in 2017 – the short deadline concentrates demand for housing in one part of the year and distorts price negotiations by pushing buyers to accept whatever prices they can get (Kunovac and Žilić, 2020).

Mirela and Nenad accumulated considerable savings for the apartment that they bought in their late twenties; Mirela had lived at her parents’ to save most of her salary. The flat was ‘within the limits of [their] possibilities’, to the point that they later regretted not buying a larger, more comfortable apartment. As the main debtor, Mirela made sure that the instalment did not exceed a third of her salary. In addition, they opted for a loan with the straight-line amortization method. Most Croatian borrowers choose mortgage-style amortization for the benefit of paying (approximately) the same instalment each month, but the price for that is that they pay more in total interest and amortization is longer. Straight-line amortization is less convenient since one starts off by paying larger instalments but repayment is shorter and less total interest is paid. Nenad carefully calculated their amortization schedule before they signed the contract. The modest principal (about €43,000) and the straight-line amortization reduced the maturity to 12 years.

Mirela and Nenad’s approach to mortgage borrowing was part of their wider prudent attitude to finance. They never took any other loan and did not even use credit cards. Instead, they had savings in multiple forms – time deposits, private pension savings, and shares. Mirela believed that many Croatians got over-indebted due to their compulsive need for status items, which she firmly rejected. Her opinions on home-ownership were also in line with the aforementioned common sense. Nenad and she apparently never contemplated renting – she considered it a ‘pure waste of money’ while Nenad spent his student years as a tenant and had enough of the precarity. Mirela and Nenad’s ‘financial subjectivities’ thus point to an overlap between the common sense about mortgaged home-ownership and the kind of middle-class subject that Weiss (2019) analysed as an ideological fiction – one seeking to accumulate and maintain value in assets while keeping in check current consumption.

Post-boom discourses on middle-class indebtedness

Apart from the narratives naturalizing the debt boom, the weak articulation of class in Croatian politics has likely also contributed to the limited public recognition of mortgaged home-ownership as a middle-class experience at the time of the boom. Croatian post-socialist politics has been characterized by a primacy of so-called ‘cultural’ and ‘ideological’ cleavages over class relations. The paradoxical outcome of this state of affairs is that the Croatian Democratic Union (HDZ), the leading right-wing party, has been supported predominantly by working-class people motivated by Croatian nationalism and traditionalism, while the Social Democratic Party (SDP) attracted middle-class voters motivated by secular and liberal values (Bagić, 2006; Dolenec, 2012). It is indicative that the link between middle-classness and mortgage debt started to be explicitly articulated in public discourse in parallel with a Croatian version of the narrative about a squeezed middle class (Škokić and Potkonjak, 2016). The financial difficulties and dispossession of many debtors, and wider economic hardship during the long period of recession and austerity (2009–15), prompted the politicization of both indebtedness and the middle-class predicament. The high-profile case of Swiss franc loans (chiefly mortgages) was particularly instrumental. From the early 2010s onwards, activists have been mobilizing the idea that Swiss franc debtors were essentially middle-class people – highly educated, hardworking, productive and meritocratic – and as such vital for the Croatian economy and nation, and morally deserving of public assistance (Mikuš, 2022). The franc debtors’ discourse of middle-classness matched their orientation to institutional means of contestation (consumer rights litigation, party politics) and their cooperative relationship with the SDP, which passed several ‘unorthodox’ measures in their favour while in government (Mikuš, 2019a: 310–11).

In this period, media and commentators increasingly presented mortgage debt as an attribute of the Croatian middle class. For example, the leading daily Jutarnji ran a piece entitled ‘How do Croatian families live: “We are the middle class. To live as before, we have to work freelance each weekend”’ in its 2013 Christmas special. This was a human-interest story of a ‘typical Croatian family of the middle stratum’. The parents are ‘both intellectuals’ and live with their two children and a dog in a spacious, stylish apartment in a ‘fine Zagreb neighbourhood’ that they bought with a subsidized loan for ‘young scientists’. It looks idyllic on the surface but the reality is harsh – repayment of their two loans takes 40% of their income, and since their salaries (modest in international comparison) stagnated while prices went up, the husband needs to freelance on weekends and the family has to implement all manner of everyday frugalities to maintain a decent standard of living (Novak, 2013). Such images were part of a wider public discourse about the Croatian middle class as ‘disappearing’, ‘lost’, or much smaller than in, supposedly, Scandinavia and the United States (e.g. Grenac, 2013). As the story of the ‘typical’ middle-class family already implied, debt was seen as one of the causes of this dynamic. An article thus cited expert opinion that hardest-hit by the recession and austerity were ‘families that make up the middle class’ – ‘households that have average incomes and that most often opted for taking out housing loans and higher personal consumption in the good years’ (Jurasić, 2016). In this period, the two leading political parties, the SDP and the centre-right HDZ, both started to present themselves as the champion of the Croatian middle class/stratum while accusing its opponent as hurting that same class with its policies (HDZ, 2014; Špoljar, 2020; tportal.hr, 2013).

When I raised the issue of their class self-identification, a sizable minority of my mortgagor interviewees said that they did not identify as members of any class. A tiny majority identified as middle class/layer, which they understood in line with the local folk concepts of middle class – in reference to income, level of education and lifestyle. Marija, a marketing agency manager, told me that she was generally content with her ‘kind of normal life of the middle layer’, which she illustrated by being able to afford petty optional purchases such as pricy Pokémon cards for her son without too much worry. Revealingly, several interlocutors related this self-identification explicitly to their mortgaged home-ownership. Domagoj, a social worker, told me: ‘Objectively, I’d place myself in the middle class. Owning an apartment and repaying the loan every month, I definitely cannot place myself in a class below that.’

At the same time, many interlocutors who identified themselves as middle class/layer questioned or qualified this in various ways. Most commonly, they felt solidly middle class in terms of education and profession, but hardly so in terms of income and therefore consumption standard and lifestyle. They often compared this to a loose idea of middle-class standards in, say, Germany. Apart from the low incomes in Croatia, they related their own ambiguous middle-classness to the financial burden of repayment, which necessitated cutting down on optional (and in some cases and periods even essential) consumption. This made it impossible to achieve a middle-class lifestyle defined in terms of benchmarks such as being able to afford a foreign vacation once a year, being carefree about everyday consumption in the way described by Marija, or being able to go out for drinks with friends. As a result, they described themselves in ambiguous terms such as ‘perhaps some [sort of] lower-middle class’ or ‘an edge’ of the middle class.

Conclusions

The analysis presented in this article highlights that localized middle classes – both their notions and the actual social groups they describe – have complex local genealogies that do not necessarily fit into the squeezed/new binary and that may exert a long-term influence. It also points to the importance of mundane, often opaque institutional practices in middle-class formation, alongside the more visible identitarian discourses and practices stressed by the existing anthropological studies of Eastern European middle classes. The Croatian case shows that socialist societies had generated their own middle-class concepts and categories that were partly carried over to post-socialist class relations instead of being replaced wholesale. The classificatory practices of state and financial institutions continued the tendency to privilege middle classes (defined in similar terms as under socialism) in accessing the most desirable form of housing under the existing housing regime, now owner-occupied housing instead of public rental housing. However, mediating this access by mortgage finance turned out to be a mixed blessing for the middle class due to the resulting high levels of debt, the effects of extensive predatory practices, and volatile housing prices, giving rise to a public recognition of mortgage debt as a middle-class experience and to a formation of middle-class politics that have more in common with the disappointed and defensive middle classes in the West than the ascending and economically liberal new middle classes in the global South. While mainstream mortgage debtors in Croatia as well as Serbia gravitated to institutional and consumer rights-based politics (Dolenec et al., 2021; Mikuš, 2019a), their Hungarian counterparts exposed to the same predatory lending practices did not shy away from far-right registers and alliances as part of their more eclectic and insurgent politics (Molnár, 2016; Szabó, 2018). The Croatian case is thus likely relevant for other Eastern European countries with comparable material conditions shaped by similar transformations of housing systems and mortgage finance and ownership-oriented housing policies skewed in favour of the middle class (Florea et al., 2022; Makszin and Bohle, 2020; Stephens et al., 2015). At the same time, it is important not to overgeneralize and overstate the similarities without a systematic comparative analysis of middle classes and their relationships with mortgaged home-ownership in each setting. Some countries have been spared the proliferation of predatory mortgages that was clearly central to the distinctive politicization of the indebted middle class in Croatia, while the different outcomes in otherwise comparable Hungary point to the role of collective actors, such as social movements, and the properties of the political arenas in which they operate, in shaping this politicization.