Inside Eurobank's Pillar Portfolio:
Are auctions disproportionately affecting Athens' poorer?

An investigation into the auctions' system through RU's published database, alongside an analysis of Athens' historical class distribution questions who are those most affected by the bank speculation scheme.

By Dafni Karavola

Cover image by Ioannis Dedes

In June 2024, Reporters United published an investigation into real estate auctions in Greece. Using Eurobank's Pillar portfolio of non-performing loans as a case study, the investigation uncovered a financial mechanism for appropriating real estate from over-indebted households by a complex network of companies. Additionally, the investigation's second part revealed that most of the auctioned properties were purchased by Eurobank itself, through its subsidiary Pillar Estate. This process allows the bank to acquire extensive real estate at low prices through its own auctions, while the indebted households continue to owe money to the bank.

Support RU's independent investigative journalism here

While the investigation illuminated the ways this network of companies generates, conceals, distributes, and transfers large capital gains to tax havens, subsidized by the Greek state, a critical question remains: Who are those most affected by this bank speculation scheme?

A closer examination of the auctions' published dataset, alongside an analysis of Athens' historical class distribution across its urban landscape, reveals a troubling reality: Banking speculation through auctions disproportionately impacts residents of low-income neighbourhoods, while the richer Athenian suburbs remain almost untouched.

Homes held hostages

Eight out of ten auctioned properties across Greece are residential homes, with 45% located in Athens or its broader region, Attica. In contrast, commercial properties account for only 14% of all auctions, and half of these are storage spaces of less than 10 sq. m., typically accompanying residential properties. In Attica, auctioned land is most frequently parking spaces, again usually tied to residential properties, while in the countryside, many plots of agricultural land are being auctioned. Large tourist complexes or industrial facilities are rare, highlighting that Eurobank's auctions almost exclusively target houses and their accompanying infrastructure, such as small storage and parking spaces.

These are just the numbers representing the published auctions of only one bank’s non-performing loan portfolio. If multiplied by the number of banks and portfolios we understand that thousands of households saw or soon will see their house being sold-off.

The auctions’ horizontal class distributions across Attica

After the urbanization of the 1960s, Attica's demographic landscape underwent a complete transformation, evolving into the structure we observe today. The east/west axis became the primary social divide in Athens and its suburbs. Thomas Maloutas professor of social geography at Harokopio University of Athens, explains that in this new model of horizontal class distribution, "the lower social classes, apart from their increased presence in the center, remained dominant in most of the western suburbs and the broader periphery of Attica, while the higher-income population retreated to the eastern suburbs."

Does this urban characteristic apply to the real estate data in Pillar’s portfolio? If so, what does this reveal about the portfolio’s demographics?

This choeroplyth map shows the number of auctioned properties in each of Attica's municipalities.

As expected, most auctioned properties are concentrated in the densely populated cities of Athens and Piraeus.

From the suburban areas, the traditionally low-income neighborhoods of Nikaia - Rentis, Keratsini - Drapetsona, and Acharnes have the highest number of auctions. Other working-class residential areas such as Peristeri, Aegaleo, and Kallithea follow closely. Notably, all top-ten high concentration areas are located in the western suburbs.

In contrast, the northeastern suburbs with higher-income populations, such as Filothei - Psychiko, Vrilissia, and Dionysos and the southern municipalities across the costal line have significantly fewer auctioned properties.

Overall, the poorer the neighborhood, the more vulnerable the houses are to the auction system. But what about the prices?

Do all social classes receive an equal share when it comes to paying off their loans?

This choreoplyth map show the price ranges of auctioned properties across Attica's municipalities.

The two major cities score around 86-89.000 euros per property. Considering that the average market price for a 75 sq.m. appartment in Athens city center is 150.000 euros and the equivalent for Piraeus is 185.000 euros, these prices are significantly low.

Similarly, the average market prices for medium sized apartments in the western suburbs fluctuate around 145.000 euros. However, in most low-income neighborhoods, auction prices barely reach half of the market value.

For northeastern and southeastern suburbs market prices differ and the houses' surface is larger. A 100 sq.m. house in the northeastern of Attica can be sold for around 300.000 euros and auctioned for the average of 250.000, a much smaller difference than the western low-income equivalent. Municipalities of the costal front present larger gaps, as similar house could be sold for around 375.000 euros but auctioned for only 2/3 of this price.

Overall, auction prices remain significantly below market prices but tend to follow market trends, favoring upper-class regions. In some cases, the more affluent northern suburbs can secure much better deals than the western suburbs. This disparity can be catastrophic for poorer households, as in many instances, if the auction price is not high enough to fully repay the mortgage, the family will lose their home and still owe money to the bank.

Average market prices from spitogatos.gr

The auctions’ vertical class distributions across Attica

According to Maloutas , vertical division is another strong characteristic of the class distribution across Athens, and is related to both the apartment building’s construction year and surface. The professor explains: “Only buildings constructed until 1980 have a structure and internal layout that favour vertical social separation. In these buildings, we find small and degraded apartments in the basement and ground floor, in contrast to the privileged and large apartments on the upper floors. In apartment buildings from this period, a significant percentage of residents (40%) are concentrated in these "extreme" positions.” This means that the poorer the household the closer to the street level the house.

It is important to note that those most affected by vertical class separation are migrant communities. However, the rule also applies to lower-income homeowners, as evidenced by Pillar's portfolio. Eight out of ten auctioned houses are located between the ground floor and the second floor. Basement and mezzanine apartments are rarely auctioned, likely because the migrant communities occupying them are renters, not homeowners. This additional observation highlights that the affected low-income homeowners are somewhat better off compared to the non-local residents of the lower floors.

Moreover, the data analysis reveals that most auctioned houses are 75 sq.m. apartments built between the 1970s and 1990s, typical of the Athens city center. This is followed by larger apartments up to 150 sq.m. in the same older buildings, as well as newer, smaller houses built after 2000.

In summary, the analysis of Pillar's portfolio reveals a stark social divide, with lower-income homeowners bearing the brunt of banking speculation through the auction system, while more affluent areas remain less affected, underscoring the systemic inequalities in Athens' housing market.