Vacant dwellings in the Municipality of Athens: spatial dynamics, methodological challenges and reactivation policies to address the housing crisis

This article examines the geography, characteristics and temporalities of vacant dwellings in the Municipality of Athens, drawing on two primary data sources: (a) the Hellenic Statistical Authority (ELSTAT) Population-Housing Census (2011 and 2021), and (b) the HEDNO (Hellenic Electricity Distribution Network Operator) registry of non-electrified (N.E.) properties (September 2024 and 2025). The aim is, on the one hand, to develop a more nuanced understanding of the scale and forms of vacancy across different neighbourhoods of the city and, on the other, to contribute to the policy debate on mobilising the vacant housing stock within the framework of social housing policies.

The article argues that the activation of vacant dwellings cannot be achieved through horizontal, centrally administered measures. Instead, it should be embedded in locally grounded housing strategies, developed in synergy with local urban planning frameworks. The quantitative and spatial analysis of available datasets constitutes a starting point for a broader methodological approach, which requires complementary targeted qualitative and field-based research, as well as the creation of a Housing Observatory for the systematic monitoring of property conditions, housing market dynamics and residents’ housing needs.

Reframing vacancy in contemporary housing policy

Vacant dwellings are a structural feature of real estate markets across Europe. Increasingly, they are associated with speculative practices and investment strategies of capital, but also with the specificities and historical trajectories of built environment production, as well as with public policies in each local context. A range of studies propose different typologies of vacancy: according to duration out of use (FEANTSA 2016), property condition and use (Caramaschi & Chiodelli 2022), or the processes through which vacant properties are produced, maintained and removed in relation to owners’ strategies and the role of the state (O’Callaghan 2023). These typologies provide an analytical framework for understanding a complex phenomenon and can inform more nuanced policy responses, moving beyond simplified interpretations that treat vacant properties merely as a supply-demand imbalance, detached from social relations, property regimes and dynamics of urban restructuring.

At the European level, the mobilisation of underused and inactive building stock has become a central policy direction in the recommendations of organisations such as FEANTSA, Housing Europe and the Urban Agenda, and more recently within the objectives of the European Affordable Housing Plan in addressing the housing crisis [1]. This concerns not only residential properties but also commercial or former industrial buildings that could be repurposed into housing. The reuse of vacant properties is promoted as a more sustainable alternative to new construction, alongside methodologies for mapping, classifying and assessing vacant dwellings in line with goals of energy retrofitting, sustainable development and equitable access to housing.

Recently, the issue of vacancy has moved to the forefront of public debate and housing policy also in Greece, linked to the intensification of the housing crisis, rising rents, the expansion of short-term rentals and post-crisis urban transformation dynamics. Policy responses have largely focused on horizontal incentive schemes -grants and tax exemptions for owners of vacant dwellings aimed at “returning” them to the market- without substantial results. Beyond the need to strengthen the social conditionality of public support, particularly regarding rent levels and long-term use, there is a pressing need for systematic analysis of the geography, characteristics and ownership structure of vacant dwellings, as well as of the differentiated reasons why a significant share of the housing stock remains outside the market despite high demand.

The “Mediterranean paradox” and the spatial dynamics of vacant dwellings in Greece

The large number of vacant dwellings in Southern European metropolises such as Athens is not a recent phenomenon. The “Mediterranean paradox” -the coexistence of a substantial stock of unused properties alongside high demand and rising housing prices- has been identified for decades as a distinctive feature of Southern European housing systems (see, indicatively, Hoekstra & Vakili Zad 2006). High vacancy rates are associated with the touristic and second-home character of Mediterranean countries, the central role of construction in their economies, the prevalence of small-scale ownership, the importance of family property, and the historically pivotal role of real estate in processes of social reproduction.

In Athens, numerous studies have examined vacant properties; both residential (Maloutas 2000; Maloutas & Spyrellis, 2015; Αράπογλου κ.ά., 2019; Αράπογλου & Σιατίτσα, 2019; Dimitrakou, 2021) and non-residential (Triantafyllopoulos 2015; Belavilas & Prentou 2015). Similar research in Thessaloniki has identified comparable patterns and dynamics producing vacancy (Chatziprokopiou et al. 2024; 2025; Vrantsis 2025). These studies highlight the relationship between the spatial concentration and timing of vacant dwellings and broader processes of investment and disinvestment, real estate cycles, public policies, demographic and urban change, as well as the social relations and household practices structured around property ownership. Understanding vacancy requires combining various datasets and research methods, an increasingly necessary approach in Greece, where fragmented small ownership, accumulation by investment funds and REITs, inactive public properties and inheritance renunciations together shape a highly complex field of property regimes and uses. An effort that is further hindered by the lack of transparency in the real estate market and the absence of systematic public monitoring mechanisms.

The significant increase in vacant dwellings in central areas of Athens during 2001-2011 (reaching up to 200% in some neighbourhoods; see Table 3) has been attributed to wider metropolitan restructuring processes: suburbanisation, long-term population decline (since the 1980s; see Table 2), and reduced demand linked to declining purchasing power in these areas (Maloutas & Spyrellis 2015). It has also been associated with structural characteristics of the Greek property regime, such as fragmented and multiple ownership, while disinvestment during the economic crisis contributed to further abandonment (Triantafyllopoulos 2015).

Photos 1-7: Vacant Dwellings in Athens

This article revisits the phenomenon using data from the 2021 Census and complements them with data on non-electrified residential properties, employed as a proxy indicator of vacancy, following a similar approach to that adopted by Chatziprokopiou et al. (2024) for Thessaloniki. Since each data source constructs the object of analysis differently, the article highlights both the methodological limitations and the distinct interpretative potentials of the two approaches [2]. It focuses on the quantitative and spatial mapping of vacant dwellings in the Municipality of Athens, as well as on their temporalities and characteristics, treating vacancy not as a static category but as a process shaped over time through intertwined social, economic and institutional dynamics.

Vacant Dwellings in the Municipality of Athens Based on the Census (2011-2021)

Housing censuses conducted every ten years constitute the most common reference source for vacant dwellings. Vacant dwellings are recorded as conventional dwellings that do not constitute the principal usual residence of any person at the time of the census, for various reasons [3].

In 2021, 437.188 conventional dwellings were recorded (only 9.363 more than a decade earlier), of which 117.137 were vacant (26,8% of the housing stock), located predominantly in apartment buildings (94,6%). Of these, 49.837 (42,5%) were for rent [4] and 7.176 (6,1%) for sale, while 34.842 (29,7%) were recorded as secondary residences, 869 (0,7%) as holiday homes, and 24.215 (20,8%) were vacant for other reasons. As in other Southern European Mediterranean metropolises, a high proportion of both seasonal and occasional non-use of residential property is recorded, indicating systematic underutilisation of the available housing stock [5].

Table 1. Vacant dwellings in the Municipality of Athens by category of non-use, 2011–2021

Source: ELSTAT censuses 2011 & 2021

Overall, during the period 1991-2021, Athens experienced significant demographic decline (particularly in the decade 2001-2011), combined with an increase in the number of households and in the housing stock (especially during 2001-2011) (Table 2). Vacant dwellings peaked in 2011 and decreased in the decade 2011-2021, while remaining at high levels [6].

Table 2: Population, households and conventional dwellings in the Municipality of Athens, 1991–2021

Source: ELSTAT, Censuses 2021, Panorama ELSTAT/ΕΚΚΕ on census 2001 &1991

With regard to spatial distribution across the seven Municipal Communities (Table 3, Figure 1), the 6th Community (Patissia, Kypseli) recorded in 2021 the highest number of vacant dwellings (in both absolute and relative terms), linked to its large housing stock and the significant increase in vacancy observed during the previous decade in the area. High concentrations are also recorded in the 7th (Ampelokipoi, Polygono) and 2nd (Pangrati, Neos Kosmos) Communities, while in the 1st (City Centre, Kolonaki, Exarchia, Koukaki) the high proportion of vacant dwellings is also associated with the strong presence of commercial and professional uses.

Table 3: Vacant and conventional dwellings by Municipal Community, 2021, and change 2011–2021

Source: ELSTAT census 2021, Panorama ELSTAT/ΕΚΚΕ on census 2001 &1991

Figure 1. Number of vacant dwellings by Municipal Community per decade (2001–2021)

With the exception of the 7th Community, where vacant dwellings increased by 15,5%, and the 4th, which recorded a smaller increase of approximately 5%, it appears that during the period 2011-2021 a significant share of vacant dwellings returned to use, possibly covering -together with newly built units- part of the demand generated by the increase in households in the Municipality. However, this assessment must also take into account dwellings that shifted from conventional residential use to tourist or short-/medium-term rental use, as well as vacant dwellings converted into tourist accommodation, a process that is difficult to capture due to statistical ambiguities and definitional and recording limitations in the census categories.

The significant change in the 3rd Municipal Community (Petralona, Thiseio, Votanikos, Gazi) may be explained by the comparatively small size of its housing stock, but also by patterns of abandonment and underinvestment in the previous decade, in anticipation of more profitable upgrading dynamics, as well as by the extensive development of leisure and tourism-related uses. The 2nd and 6th Communities lie at the centre of gentrification and touristification trends in central neighbourhoods, while the reduction of vacant dwellings in the 5th Community (Ano Patissia, Rizoupoli) may be associated with the displacement of residents from more central areas. In the 1st Community, despite the high concentration of tourist accommodation and investment activity, the limited change may reflect the shift of dwellings from vacancy to short-term rental use. More robust conclusions, however, require further field-based and in-depth research.

Map 1. Change 2011–2021 and number of vacant dwellings by Municipal Community

The age of vacant dwellings constitutes a critical factor, as it is associated with the high cost of rehabilitation and energy upgrading that may be required for their reuse. 82% of vacant apartments are located in buildings constructed during the period 1946-1980, particularly during the phase of mass apartment-block production under the antiparochi system (1960-1980) (Table 4). The high proportion of vacancy in more recently constructed buildings also requires further investigation. Although new housing production -as reflected in the census- is relatively limited compared to demand, a substantial share of the newly built stock does not enter into use. This may be related either to the very high sale and rental prices of newly built dwellings or to the profile of their purchasers, who may have investment-oriented motives (e.g. golden visa schemes).

Table 4. Construction period of vacant dwellings in relation to the total housing stock

Source: Census 2021

With regard to the size of vacant dwellings, half range between 50 and 80 sq.m., in proportion to the overall housing stock, while a significant share is also found in dwellings below 50 sq.m. This category may correspond to a comparatively more devalued segment of the housing stock, given that in the typical antiparochi apartment-building model that dominates the Municipality of Athens, smaller dwellings are generally located on the lower floors, which have historically been associated with lower value and lower housing quality (see Figure 1; Maloutas & Spyrellis, 2015).

Table 5. Size of vacant dwellings in relation to the total housing stock

Source: Census 2021

Vacant Dwellings in the Municipality of Athens Based on HEDNO Data (2024–2025)

Non-electrified properties constitute a critical complementary source, as the absence of electricity connection is directly associated with actual non-use, in contrast to ELSTAT categories, a large share of which may involve occasional or seasonal use. They therefore provide a more reliable indication of vacant properties, given that almost all non-electrified properties are vacant, although not all vacant properties are non-electrified. This indicator underestimates the overall scale of the phenomenon but can capture its minimum extent with relative accuracy.

The HEDNO database provides information on properties that remain disconnected from the electricity network at the date of data extraction [7]. The following analysis examines the concentrations and spatial distribution of non-electrified properties in the Municipality of Athens by use (former residential or commercial), the duration of disconnection, as well as the size and age of these properties. According to data from September 2025 (Table 6), a total of 622.849 electricity connections were recorded in the Municipality of Athens, of which 81,7% were residential and 18,3% commercial [8]. Of these, 11,7% were disconnected. The share of non-electrified properties is considerably higher among commercial uses, with nearly one in three properties disconnected, while the corresponding rate for residential connections is 7,7%, significantly lower than the 26,8% of vacant dwellings recorded by ELSTAT in 2021.

Table 6. Electrified and non-electrified properties by use

Source: Municipality of Athens (HEDNO data), September 2025

The total number of non-electrified properties in September 2025 amounted to 75.289, representing an increase of 11,9% compared to September 2024 (Table 7). This increase is significantly influenced by the large number of disconnections recorded during the first nine months of 2025 (13.837 disconnections) [9] of which 81% concerned residential properties and correspond to 27,5% of all non-electrified residential properties. If the new disconnections carried out in 2025 are excluded, the properties that were already disconnected in 2024 show an overall decrease of 8,7% between 2024 and 2025. Reconnection appears to be more immediate for residential properties (12,9%) compared to commercial ones (4,3%).