Decisions to Renovate: Identifying Key Determinants in Central and Eastern European Post-socialist Countries

Abstract

Housing renovation is a topical issue in CEE countries facing the need to tackle their troubled post-war housing estates and improve their energy efficiency. In this paper the renovation decisions of households living in multidwelling buildings are modelled to identify the key determinants of such decisions and to gain a better insight into the reasons for the insufficient extent of renovation in CEE. Considering general factors as well as CEE specific factors, and specifically adding variables of social capital, renovation decision-making is modelled by applying a discrete choice framework of analysis. The results clearly show that, next to the physical characteristics of the stock, such as its age, an important role in the renovation process is played by residents and particularly their relations in terms of social capital. The results thus identify some of the key missing pre-conditions for renovating multidwelling buildings across CEE.

1. Introduction

Housing renovation is a highly topical issue in transitional central and eastern European (CEE) countries, where the need to repair the housing stock is very pronounced. Several analysts of post-socialist housing developments have pointed out the dilapidated and troubled large housing estates, with the large post-war housing estates being particularly problematic in that regard (Struyk, 1996; Hegedüs and Tosics, 1996; Černič-Mali et al., 2003; Lowe and Tsenkova, 2003; Soaita, 2012). According to Murie et al. (2003, p. 12), this situation involves as much as 20–35 per cent of the total population in CEE. Also highlighted are the generally poor housing conditions and high incidence of unfit housing in this region, indicating a large and continuous gap between the new and pre-enlargement states in this respect (Domanski et al., 2006; Norris and Shiels, 2007; Ball, 2006; Mandič, 2010). Housing renovation in CEE and other post-socialist countries has also appeared on the European policy agenda; namely, Structural Funds have been made available to help improve the poor housing conditions in new member-states after the European Commission (EC, 2005) recognised this was needed. Moreover, according to the UNECE (2007), the renewal, management and maintenance of multidwelling buildings represent some of the most critical problems in the region.

The recent need to renovate and repair the housing stock in CEE is therefore huge. Considering that improved housing management and maintenance were proclaimed a strong motive for the privatisation of state-owned housing (Clapham et al., 1996; Struyk, 1996; Hegedüs et al., 1996), the current state of affairs following mass privatisation is surprising and a challenge to explain. Much of the discussion on the renovation process and its obstacles has been marked by particular issues of the transition processes and the related outcomes. Most notably, the privatisation of state-owned housing that occurred during the early 1990s across CEE countries brought a drastic and very fast change in the ownership structure of multifamily housing, introducing individual ownership of a dwelling and collective ownership of the common parts (such as roofs, stairways etc.), together with all responsibilities for their maintenance. The widespread privatisation has led most CEE countries to become a ‘super homeownership system’ (Stephens, 2003; Hegedüs and Teller, 2006), with homeownership rates there of around 90 per cent. As a result, recently over 75 per cent of the urban population in the region can be found in owner-occupied apartments in multidwelling buildings (Schweinichen, 2006). While the management and renovation of such multi-owned housing has been found to be very demanding, even in such long-standing institutional settings of Western democracies as the UK (Blandy et al., 2006), this is an even larger problem in the newly established CEE institutional frameworks and has been frequently reported (see Murie et al., 2003; Rabenhorst and Ignatova, 2009; Bouzarovski et al., 2011). Compared with multifamily housing, single-family housing in CEE has not drawn much attention with regard to its maintenance and renovation.

Housing renovation as an important economic activity has also been a topical issue outside CEE countries and significant efforts have been made to analyse it in more detailed and methodologically sophisticated ways. There is a growing body of literature on renovation decisions and their determinants, while some go even further and propose comprehensive explanatory econometric models (see Baker and Kaul, 2002; Plaut and Plaut, 2010). In contrast, new concepts developed in the social sciences are also being introduced to the discussion of renovation activity, most notably that of social capital (see, for instance, Taylor, 2000; Middleton et al., 2005; Yau, 2010).

In addition, housing renovation is becoming increasingly important for yet another reason. Within the framework of the fight against climate change, buildings are the first target of the EU’s energy efficiency policy. In 2007, residential buildings represented 24 per cent of total final energy consumption in the EU (ADEME, 2009). About 80–90 per cent of the total energy used during the whole life of a building is consumed during the use phase, mostly for space and water heating (EEA, 2007). Technologies enabling greater energy efficiency in the housing sector are available and also largely turn out to be economically viable as well as yielding strong environmental benefits (IEA, 2006; Jakob, 2007). Although there is a great need for renovation, the housing sector is facing a range of maintenance and renovation problems that have been hindering interventions. The reasons for the inadequate actual investment levels or so-called energy efficiency gap have to do with the institutional framework within which buildings are financed, designed, constructed and operated (UNECE, 2009; Jakob, 2007; Golove and Eto, 1996; Jaffe and Stavins, 1994). Many problems and benefits of greater energy efficiency are common to EU countries; however, there are also important differences. Particularly in transitional countries, policy-makers face a specific challenge that relates not only to overcoming the efficiency gap but the more serious challenge of the energy inefficiency trap. According to the UNECE, the energy inefficiency trap refers to

a situation in which countries with lower energy efficiency are unable to change their respective status due to lack of funds, experience, technology, motivation and initiative (UNECE, 2009, p. 7).

In this paper, we turn to the question of why the renovation of owner-occupied housing in CEE is insufficient and observe the renovation decisions of owner-occupier households living in multidwelling buildings. We focus on the relationship between household characteristics and the likelihood of renovation in the sample population. The analysis is restricted to the renovation of roofs and facades, which hold the largest potential for energy savings. We seek to identify the key determinants of renovation decisions considering the whole array of relevant variables noted in the general literature as well as for CEE in particular. Moreover, we wish to combine these factors and model renovation decision-making by applying a discrete choice framework of analysis. In addition to factors such as location, dwelling and household socioeconomic characteristics already considered in previous renovation econometric modelling, we also specifically include social capital to estimate its relative strength as a renovation predictor.

Building on previous particular findings on housing renovation in CEE, we take Slovenia as an indicative case and use the Slovenian Housing Survey 2005, a unique source of micro data in CEE. We believe that findings from these data can reveal the situation in CEE as they are about household renovation decisions, which have already been identified as one of the major barriers to renovation in CEE—namely, and as further documented in the literature review, the unwillingness or inability of households to bear common costs and to accept responsibility for the common parts of multidwelling buildings are widely recognised phenomena in privatised multidwelling buildings across CEE, along with the inability to enforce the rules and obligations of individual apartment owners.

We start with an overview of the literature on renovation and the factors that affect it in section 2. In section 3 the methodology, model and data description are presented. Estimation results and a discussion of the main findings are provided in section 4. Finally, section 5 concludes the paper.

2. Literature Review

In CEE countries, the discussion of the renovation of owner-occupied housing has been marked by the specific situation arising from the common recent experience with transitional housing reforms. Accordingly, the current renovation problems have mostly been analysed by focusing on the specific generators and causes of the decay of the privatised housing stock, ranging from historical to contemporary. The historical cause of the recent deteriorated multidwelling housing stock and the acute need for its renovation has been identified as the legacy of socialism. Namely, during socialism, state-owned housing was often criticised for its low construction standards and the use of prefabricated panel technology and was also subject to the phenomenon of maintenance being deferred for years (Clapham et al., 1996; Hegedüs and Tosics, 1996; Struyk, 1996; Lowe and Tsenkova, 2003). However, problems with management and maintenance also continued to mount during the transition period when the housing stock served as a ‘shock absorber’ (Struyk, 1996) and housing production and investment declined (Hegedüs and Tosics, 1996; Buckley and Tsenkova, 2003); moreover, as a result of unclear responsibilities concerning the management and financing of common repairs, even greater delays in maintenance were reported (Ball, 2006).

Besides the historical reason, a number of recent generators of problems have also been recognised by housing analysts. These reasons can be broadly grouped into three different, but not mutually exclusive, categories. The first is the individual attitudes of the new owners to the common property. Here, we note that the aims of privatisation also included improving residents’ attitudes to maintenance (Struyk, 1996). After privatisation, new individual owners were expected to assume responsibility for their buildings and common property, but the results were greatly disappointing. While accepting responsibilities for their individual units, the common areas and parts are frequently reported to be in a poor condition, often due to individual owners neglecting or avoiding their responsibilities for the common parts and thus making co-ordinated action difficult (Černič-Mali et al., 2003; Lowe and Tsenkova, 2003; Bouzarovski et al., 2011; Milstead and Miles, 2011).

The second issue commonly identified as a barrier to renovation is financial constraints due to a households’ unwillingness or inability to bear the common costs for maintenance or renovation. This is often related to the ‘poor homeowner’ phenomenon (Hegedüs and Teller, 2006). Among homeowners, a significantly higher incidence of economic hardship and an inability to sustain their assets have been found in CEE countries than in western Europe (Mandič, 2010). In addition, housing-related costs, particularly utility costs, have increased after the transition much faster than real incomes, so a significant proportion of households is in arrears even when it comes to paying for basic services such as water and heating (Hegedüs and Teller, 2007).

The third common issue related to the problematic renovation and maintenance of housing in multidwelling buildings can be broadly termed ‘institutional restraints’, which are related to the social context. According to Rabenhorst and Ignatova (2009), owners in CEE often lack the financial, legal and technical tools to fulfil their obligations as owners and continue to encounter problems such as an unclear delineation of their rights and responsibilities, and a lack of meaningful choices in contracting management and maintenance services. Financial constraints mean that it is difficult to provide loans for development and to secure them, mostly due to inadequate laws and procedures related to title and interest registration in condominium housing. In organisational terms, they find the greatest constraints in the inability to enforce the rules and obligations of individual apartment owners, thus leading to the poor management of common areas and limiting the renovation of common parts.

Another feature identified in CEE countries is that, in carrying out repairs and improvements to owner-occupied housing, self-help or do-it-yourself practices have also been frequently employed in multifamily buildings, not only in single-family housing (Mandič, 2001; Bouzarovski et al., 2011; Milstead and Miles, 2011; Soaita, 2012). More specifically, in their study of the determinants of do-it-yourself activity in maintenance and home improvements in privatised housing in Lithuania, Milstead and Miles (2011) found that the perception of ‘the neighbourhood as a good place to live’ is a key determinant, while the impacts of household disposable income, education level and the physical location of the dwelling were not confirmed.

In addition to the specific discussion in CEE countries, a number of studies elsewhere have examined the renovation and factors affecting it in a more complex way and by relying on new concepts. Social capital is one of the most prominent new concepts in the discussion of factors, impacting housing renovation and urban renewal. It has been increasingly employed to denote the ability of the actors involved to participate, co-operate and mobilise to attain a common goal through co-ordinated action. Following the notion of social capital as the

features of social organization, such as networks, norms and trust, that facilitate coordination and cooperation for mutual benefit (Putnam 1993, p. 36)

several analyses have highlighted its significance in neighbourhood change and regeneration (Forrest and Kearns, 2001; Purdue, 2001; Bridge, 2006; Kleinhans et al., 2007), as well as urban regeneration (Cento Bull and Jones, 2006), slum clearance (Taylor, 2000) and the managing of multi-unit residential buildings (Yau, 2010). However, because social capital has often remained insufficiently defined, Middleton et al. (2005) raised important reservations about its use as an analytical concept.

One of the complex approaches to an operational definition of social capital identifies its components within a particular framework, also relevant for multi-owner residential buildings. Starting from the context in which civic capacity—i.e. collective action—is required to replace the state’s provision of services, Saegert et al. (2002) identify three components of social capital: participation in resident associations, pro-social norms and a building’s formal organisation. A similar operational definition of social capital and of its components was used by Temkin and Rohe (1998). In their study of fostering community development, social capital is defined as consisting of two main components. The first is the socio-cultural milieu, measuring residents’ commitment and attachment to the neighbourhood. The second is the institutional infrastructure, measuring the organisational ability of the neighbourhood to act in its common interest and to turn commitment into effective collective action. A combination of both components was found to be the key determinant, making neighbourhoods with larger social capital less likely to decline. In addition, in his study of two multidwelling residential buildings in Hong Kong, Yau (2010) found the level of maintenance in the building was positively correlated with homeowners’ participation in the management and with the sense of community, measured as social connections, mutual concerns and community values. Such an approach to an operational definition of social capital also seems quite applicable to the issue of renovation in multidwelling buildings in CEE, where the previously state-dominated management structure needs to be substituted by the collective action of individual homeowners and where problems are reported in both residents’ commitment and attitudes (i.e. in the socio-cultural milieu) as well as in their formal organisation, facing financial, legal and technical constraints (i.e. institutional infrastructure).

Among many other approaches to renovation, particularly relevant to this study are those aiming at more comprehensive explanations of renovation activity where significant efforts have been made statistically to model the factors impacting it. Mayer (1981) developed a theoretical capital-stock adjustment model to study rental housing rehabilitation. This model was further extended by Helms (2003) by applying profit maximisation not only to landlords but to utility-maximising households. He shows that the optimal investment in renovation depends on households’ characteristics (their utility functions and budget constraints) as well as on a building’s structural attributes and neighbourhood characteristics. In the model of Gyourko and Saiz (2004) the supply side of renovation is also emphasised. Their renovation decision model is based on comparing the current value of a property with the equivalent of its construction materials and costs. Housing renovation models have also been subjected to extensive empirical testing, particularly in the US (see, for example, Mendelsohn, 1977; Baker and Kaul, 2002; Helms, 2003; Plaut and Plaut, 2010; Jakob, 2007; and Banfi et al., 2008, for Switzerland).

3. Methodology, Model and Data

3.1 Model Specification and Methodology

To model the renovation decision-making of households living in multidwelling buildings, a discrete choice framework is applied. The relationship between a polytomous response variable representing a set of unordered alternatives and a set of regressor variables that are individual-specific is modelled in order to analyse the determinants of renovation decisions. A random utility approach is employed (Louviere et al., 2000). We consider household i choosing among J alternatives in a choice set of renovation types. The utility of household i choosing renovation type j can be represented by the following expression

U_{i j} = β_{j}^{’} X_{i} + e_{i j}

(1)

where x _i represents a vector of explanatory variables that relate to household i (i = 1, …, N) and are constant across alternatives; β _j is a vector of respective coefficients that are specific to each choice of renovation type j (j = 1, … J); and e_ij stands for a random error.

Equation (1) represents a specific utility equation for each choice j. The probability that alternative j is chosen by household i is

Pr (U_{ij} > U_{ik}) \forall k \neq j

(2)

where, Pr indicates probability; and U_ij stands for the utility of household i choosing renovation type j.

We are particularly interested in examining household decisions on renovations of multidwelling buildings that hold the greatest potential for energy savings—namely, renovations of the roof and facade. The roof and facade are important elements of the building envelope that not only have a very large energy efficiency potential, but also determine the level of useful energy requirements, facilitate efficient heating systems and generate benefits such as greater housing comfort, an improvement of the building’s condition and increased building value (Jakob, 2007; Martinaitis et al., 2004). Hence, the dependent variable RENOV_i is a random variable indicating the choice made and is defined as follows

R E N O V_{i} = {\begin{matrix} 1 & if the roof and / or facade have been renovated, \\ 0 & otherwise . \end{matrix}

(3)

The dependent variable defined thus is in fact dichotomous and reflects the presence or absence of roof or facade renovations of multidwelling buildings performed by households. Other renovations possibly made by households are not considered in the study. Therefore, a binomial logit model is applied.¹ The probability that a renovation decision is made by household i is

\Pr (R E N O V_{i} = 1 | X_{i}) = \frac{e^{X_{i}^{’} β}}{1 + e^{X_{i}^{’} β}}

(4)

The maximum likelihood (ML) estimation method is used to estimate the model. Since the right-hand side of equation (4) is a non-linear function of explanatory variables, the parameters of the logit model do not represent marginal effects. Therefore, marginal effects are obtained by taking the derivative of equation (4) with respect to x_i.

In line with the existing literature on housing renovations using personal and household characteristics, dwelling characteristics and location characteristics and the noted importance of social capital, vector x_i consists of four groups of explanatory variables relating to those dimensions. The first group refers to dwelling location characteristics, such as living in a big city (CITYL), small city (CITYS) or rural area (RURAL). The three dummy variables take a value of 1 if a respondent resides in the indicated locality, and a 0 value otherwise. Further, the dummy variable for the least-developed Slovenian region is considered (REGLOW), as is the dummy variable for the capital city Ljubljana (CAPITAL) which represents the most developed part of the country. Macro location dummy variables serve three important functions. They capture potential differences in characteristics of households living in different places and also the differences in local housing markets. Moreover, access to information on renovation as well as the supply of renovation support services may differ with location. Housing markets also define the value of homes which can, in relation to the construction costs, influence the household’s willingness to renovate (Gyourko and Saiz, 2004). Unfortunately, no reliable data on the value of homes and the liquidity of local housing markets are available for the period studied; therefore, this effect may be also captured by the aforementioned location dummy variables. Besides macro location characteristics, two micro location characteristics are introduced to measure noise (NOISE) and the feeling of safety (SAFE) in the neighbourhood. Noise in the neighbourhood is measured on a scale from 1 (very low) to 5 (very high), while safety is measured on a scale from 1 (very safe) to 4 (very dangerous).

The second group of variables encompasses dwelling and building characteristics. The first one is the age of the building in years (AGEBUILD). The building-age effect consistently emerges from the empirical literature and is found to have a positive influence on renovations. The second is the perceived quality of the residence (QUALHOME); combining various amenities and space, three categories were defined: poor, good and very good. In addition, to distinguish between different building types two dummy variables were introduced: a multidwelling building with up to four floors (BUILDS) and a multidwelling building with more than four floors (BUILDB). On the one hand, economies of scale are present in bigger multidwelling buildings due to the higher intensity of use measured by the housing surface to inhabitant ratio and positively affect the financial burden of a renovation. Moreover, in terms of energy efficient renovations, larger multidwelling buildings also have a more favourable ratio between the heated volume and the envelope surface. On the other hand, it appears to be more likely to renovate in the case of smaller buildings as it is usually easier to reach an agreement among a smaller number of decision-makers.

The third group of variables accounts for socio-demographic and economic characteristics of households, such as the age of the respondent (AGE), a dummy for respondents older than 64 years (OLD), the education of the respondent (EDU), household size (HSIZE), the number of household members aged less than 18 years (MINOR), household income (INCOME) and whether a given household needs to repay a housing loan (LOAN). Since income is particularly sensitive information and many missing values were reported, the perception of households on how well they manage to cope with their monthly income was used as an approximation. We also added a dummy variable indicating households that have recently moved into the current residence in the period in which we observe renovation activity (MOVEIN).

While the socio-demographic and economic characteristics of households are generally recognised as crucially affecting their ability and decisions to renovate, in multidwelling buildings still another condition needs to be met. Unlike households in single-family houses, the agreement of the majority of other households is also needed and household characteristics may prove to be less significant relative to variables reflecting social capital. Therefore, the fourth group of variables stands for social capital and its two components. The first component is the socio-cultural milieu, denoting a resident’s commitment and attitudes to his/her neighbours. Accordingly, respondents were asked to indicate whether they know the majority of their neighbours (NEIGHBOUR) and whether they are attached to their neighbourhood (ATTACH). Respondents were further asked to estimate the fluctuation rate of residents (FLUCT) and to estimate differences in the residents’ economic situation (ECONDIF), indicating the homogeneity and stability of residents in the building. The second component of social capital—the institutional infrastructure, measuring the organisational ability of residents to act—was covered by the following variables. The variable (AGREE) represents an estimation of how easy it is to reach an agreement on renovation among the residents. Quality of the building manager is captured by a dummy variable reflecting dissatisfaction with the work of the building manager (POORBM). Further, a dummy was introduced to denote regular payments to renovation funds (RFUND), indicating the availability of common financial sources for maintenance.

3.2 Data Description

The data were obtained from the Slovenian Housing Survey carried out in 2005, consisting of a representative sample of 4009 Slovenian households living in family houses and multidwelling buildings. Hence, a revealed preference method is used, based on an examination of the actual choice decisions made by households.² Due to missing data, 366 observations were excluded from the sample, resulting in a sample size of 3643 households. As the paper focuses on renovation decisions of a sub-sample of households living in multidwelling buildings, the final sample consists of 1248 households which represents 34.3 per cent of the households in the full sample. For households in multidwelling buildings, the decision to renovate the roof or facade is not autonomous since the majority of households must first agree on such renovations in order for them to be carried out. Namely, according to the Slovenian Housing Act, for the renovation of a roof or a facade, 50 per cent of residents’ interest consent (based on the share of their interest in the building) is needed and, if the renovation also includes improvements in terms of an energy efficiency upgrading, consent at the level of 75 per cent is needed.

Further, by restricting our sample to households residing in multidwelling buildings, we avoid the potential problem of a different understanding and classification related to renovation as opposed to maintenance activities performed by respondents. While maintenance of the common parts of a building is a regular activity carried out by a building manager in line with the annual plan approved by the owners’ supervisory board, renovation requires the consent of (at least) the majority of owners. There is also a notable difference in funding; in the former case, regular monthly fees charged by the building manager are used, while renovations are typically associated with one-off and considerably higher expenses funded from an earmarked renovation fund and often require additional payments by owners. Therefore, it is believed that a satisfactory understanding of ‘renovation’ has been reached among respondents by reliance on formal provisions.

Table 1 relates to our sample data and sets out renovations carried out in the period of five years before the survey was performed. This limited period of observation³ may be argued to be insufficient since many buildings may have been renovated before or after the observed period. However, statistical data on the existing housing stock in Slovenia show that over 70 per cent of residential buildings are more than 30 years old (with a large share built in the 1970s) and 71 per cent of those buildings have never been refurbished (Sitar et al., 2009). Therefore, the observed period after 2000 is regarded as capturing the main renovation cycle in multidwelling buildings in Slovenia reasonably well and limits the potential bias stemming from the limited observation period.

Table 1.

Household renovations living in multidwelling buildings in the 2000–05 period

Renovation type	Number of renovations	Percentage share of renovations
Facade and roof	87	6.97
Facade only	52	4.17
Roof only	377	30.21
Facade and/or roof (RENOV)	516	41.35
Full sample	1248	100.00

Source: Slovenian Housing Survey (2005), the authors’ calculations.

The data in Table 1 show that 41.4 per cent of households living in multidwelling buildings undertook a facade and/or roof renovation in the observed period.⁴ A more detailed breakdown of renovations reveals that the most common renovation type was a roof-only renovation, while roof and facade and facade-only renovations were performed less frequently. Table 2 reports descriptive statistics of the four groups of explanatory variables included in the model for the households residing in multidwelling buildings. It can also be noted that the estimated bivariate correlation coefficients do not indicate the presence of a multicollinearity problem in our dataset.

Table 2.

Descriptive statistics of the explanatory variables

Variable description	Variable	Mean	S.D.	Minimum	Maximum
1. Locational characteristics
Big city dummy	CITYL	0.5128	0.5000	0	1
Small city dummy	CITYS	0.3702	0.4830	0	1
Rural area dummy	RURAL	0.1170	0.3215	0	1
Dummy for the least developed region	REGLOW	0.0337	0.1804	0	1
Capital city dummy	CAPITAL	0.2756	0.4470	0	1
Noise in the neighbourhood (scale 1–5)	NOISE	2.4159	1.3620	1	5
Safety of the neighbourhood (scale 1–4)	SAFE	1.9487	0.6952	1	4
2. Dwelling characteristics
Smaller multidwelling building	BUILDS	0.6579	0.4746	0	1
Bigger multidwelling building	BUILDB	0.3421	0.4746	0	1
Age of building (years)	AGEBUILD	37.84	26.56	0	205
Quality of residence (0 – poor; 1 – solid; 2 – very good)	QUALHOME	0.8998	0.5271	0	2
3. Socio-demographic and economic characteristics of households
Recent move-in (2000–05)	MOVEIN	0.1402	0.3474	0	1
Respondent’s age (years)	AGE	47.83	14.83	14	90
Age 65 years or more (dummy)	OLD	0.1458	0.3531	0	1
Education of respondent (scale 1–7)	EDU	4.2292	1.2729	1	7
Number of household members	HSIZE	2.7724	1.2019	1	7
Number of minors	MINOR	0.8862	0.9426	0	4
Household income (scale 1–5)	INCOME	3.1707	1.0461	1	5
Apartment loan (dummy)	LOAN	0.4159	0.4931	0	1
4. Social capital
Knowing most of the neighbours (dummy)	NEIGHBOUR	0.7372	0.4403	0	1
Attachment to the neighbourhood (scale 1–5)	ATTACH	3.4119	1.3511	1	5
High fluctuation of residents (dummy)	FLUCT	0.3710	0.4833	0	1
Different economic situation of residents (dummy)	ECONDIF	0.6883	0.4634	0	1
Easy to reach an agreement with neighbours (dummy)	AGREE	0.4671	0.4991	0	1
Renovation fund (dummy)	RFUND	0.7428	0.4373	0	1
Dissatisfaction with building manager (dummy)	POORBM	0.0577	0.2333	0	1

Source: Slovenian Housing Survey (2005), the authors’ calculations.

4. Results

The estimated coefficients of the logit model indicating a positive or negative influence on the probability to renovate the roof and/or facade of multidwelling buildings are provided in Table 3. Table 4 reports the marginal effects of explanatory variables which represent the marginal effect on the probability to renovate by increasing a given explanatory variable by one unit while holding the other variables constant. In the case of dummy variables, marginal effects denote a discrete change from the base level.

Table 3.

Estimation results of the logit model (N = 1248)

Dependent variable (RENOV)	Parameter estimate
	Coefficient	S.E.
Constant	−2.3667***	0.6350
1. Locational characteristics
CITYS	−0.3714**	0.1549
RURAL	−0.2785	0.2209
REGLOW	0.3762	0.3385
CAPITAL	0.4095**	0.1604
NOISE	−0.0267	0.0464
SAFE	0.0466	0.0933
2. Dwelling characteristics
BUILDB	0.2007	0.1388
AGEBUILD	0.0066***	0.0025
QUALHOME	−0.1115	0.1195
3. Socio-demographic and economic characteristics of households
MOVEIN	−0.5039**	0.2037
AGE	0.0143**	0.0066
OLD	−0.3444	0.2440
EDU	0.0265	0.0533
HSIZE	0.0136	0.0929
MINOR	−0.0644	0.1130
INCOME	0.0012	0.0660
LOAN	−0.0615	0.1262
4. Social capital
NEIGHBOUR	0.2534*	0.1494
ATTACH	0.1271***	0.0480
FLUCT	0.0291	0.1314
ECONDIF	−0.0047	0.1330
AGREE	0.5374***	0.1237
RFUND	0.3902**	0.1583
POORBM	−0.4228	0.2693
logL	−787.4
Pseudo R²	0.0696
LR test (χ²)	117.74***
D.f.	24

Notes: ^* significant at 10 per cent; ^** significant at 5 per cent; ^*** significant at 1 per cent.

Source: the authors’ calculations.

Table 4.

Estimated marginal effects

Variable	Marginal effect
1. Locational characteristics
CITYS	−0.0821**
RURAL	−0.0604
REGLOW	0.0844
CAPITAL	0.0925**
NOISE	−0.0059
SAFE	0.0103
2. Dwelling characteristics
BUILDB	0.0445
AGEBUILD	0.0014***
QUALHOME	−0.0246
3.Socio-demographicandeconomic characteristics of households
MOVEIN	−0.1082***
AGE	0.0032**
OLD	−0.0737
EDU	0.0058
HSIZE	0.0030
MINOR	−0.0142
INCOME	0.0003
LOAN	−0.0135
4. Social capital
NEIGHBOUR	0.0554*
ATTACH	0.0280***
FLUCT	0.0064
ECONDIF	−0.0010
AGREE	0.1202***
RFUND	0.0852**
POORBM	−0.0901

Notes: ^* significant at 10 per cent; ^** significant at 5 per cent; ^*** significant at 1 per cent.

Source: the authors’ calculations.

With respect to the locational characteristics, the results in Table 3 suggest that living in a smaller (CITYS) as opposed to a bigger city (CITYL) significantly decreases the likelihood of renovating the roof and/or facade, while in the case of a rural area (RURAL) the negative effect is not shown to be significant. Marginal effects in Table 4 reveal that living in a small city as opposed to a big city decreases the probability of renovating by 0.08. It can also be found that renovations of multidwelling buildings are significantly more likely for households living in the capital (CAPITAL). In regional centres and in the capital, there is better infrastructure to support renovation, such as the presence of energy efficiency consulting offices, a more competitive market among housing maintenance managers and the presence of a local subsidy scheme (in place in Ljubljana). Living in the least-developed region located in the north-east of Slovenia does not have a significant effect on renovations (REGLOW).

Moreover, the two micro location characteristics, noise (NOISE) and safety of the neighbourhood (SAFE), are not found significantly to influence renovations of the roof and/or facade. In general noise levels are relatively low, with the exception of living in the city centres of the biggest Slovenian cities. In addition, Slovenia is considered to be a relatively safe place to live and therefore household decisions do not seem to be seriously affected by safety issues.

In line with theoretical expectations, the older a building (AGEBUILD), the more likely it is that the roof and facade will be renovated as the coefficient proved to be highly statistically significant. Living in a bigger (BUILDB) as opposed to a smaller multidwelling building increases the likelihood of renovation. Nevertheless, this relationship is not found to be significant. The same holds for the third building characteristic considered, the perceived quality of the home (QUALHOME).

As expected, the influence of households’ socio-demographic and economic characteristics on renovation decisions in the case of multidwelling buildings is relatively weak. Renovations are significantly less likely to be carried out in the case of recent move-ins (MOVEIN). The result can be ascribed either to the explanation that households move into dwellings with less need for substantial renovation or move-ins are typically associated with the purchase of a dwelling and therefore require substantial funding, which results in a lack of funding for renovations. It can also be concluded that the likelihood to renovate multidwelling buildings is significantly positively affected by the age (AGE) of the respondent, while other socio-demographic and economic variables carry the right sign but do not prove to be significant.

Our results confirm the important role of social capital in the renovation decisions of households living in multidwelling buildings.⁵ The variables standing for the socio-cultural milieu—namely, attachment to the neighbourhood (ATTACH) and knowing the neighbours (NEIGHBOUR)—significantly increase the likelihood of renovating. In addition, two variables denoting the organisational infrastructure and the ability to act collectively were also found to be significant—namely, the ability easily to reach an agreement with one’s neighbours (AGREE) and regular payments to the renovation fund (RFUND). However, satisfaction with the building manager (POORBM) was not confirmed to impact significantly on the likelihood of renovation. The fluctuation of residents (FLUCT) and economic differences among residents (ECONDIF) also do not seem to be significantly associated with the probability of renovation.

Based on the highly statistically significant chi-squared statistic, the null hypothesis that all coefficients in the model are equal to zero is rejected. The estimation results confirm that locational and building characteristics importantly influence renovation decisions. While household characteristics are much less important, social capital is found to play the key role. Nevertheless, the pseudo-R ² value of 0.070 indicates that the predictive power of the model is not particularly high. Likewise, Helms (2003) reports that none of the pseudo-R ² values in his study were found to be higher than 0.022, which was argued as having been expected since, similar to our study, only improvements made within a five-year period were investigated. In the future, the research could be improved by extending the investigated period and obtaining more detailed data on renovations, in particular distinguishing between energy-efficient and regular renovations.

5. Conclusion

Central and eastern European countries face a pronounced need for the renovation of multidwelling buildings. The low construction standards and delayed maintenance in the past as well as contemporary issues—namely, problematic individual attitudes of owners to common property, financial constraints and institutional limits—have resulted in a large amount of deteriorated housing stock. Since energy efficiency has been gaining ever more importance in the European Union in recent years, the countries are also increasingly challenged by the need to improve the energy efficiency of their housing stock.

In our paper, we model the renovation decisions of households living in multidwelling buildings to identify the key determinants of such decisions and gain a better insight into the reasons for the insufficient extent of renovation in CEE. The results clearly show that, next to the physical characteristics of the stock such as its age, an important role in the renovation process is played by residents and particularly their relations in terms of social capital. The renovation probability was found to be positively affected by both components of social capital. The first component—defined and measured as the socio-cultural milieu, encompassing a positive attitude to one’s neighbours and an ability to reach an agreement—revealed a positive impact on renovation. The second component—the organisational ability of owners to act collectively—also turned out to be very important. This organisational feature, embodied in the existence of a common fund for renovation, was proved to be a necessary prerequisite for a collective owner to act responsibly. Considering that income and poverty indicators were not found to be significant, what looms large is the ability to enforce common decisions when individual contributions for a renovation fund are collected.

What clearly emerges as the key result of our analysis is the significance of the mezzo level—i.e. the organisational scheme that facilitates not only co-operation for mutual benefit, but also the collection of common resources to fund the action. It seems that privatisation policies in CEE have underestimated the role of this mezzo organisational level compared with both the individual and macro structural levels when shifting many responsibilities from the state to individuals. We argue that, when transferring ownership into private hands, the difficulties in the operation of collective ownership and challenges in the development of efficient organisational schemes were underestimated, representing a common weakness of efforts of past housing reforms in CEE. Our results thus identify some of the key missing pre-conditions for renovating multidwelling buildings across CEE. They clearly indicate that policies aimed at encouraging renovation in the region should address both the physical and organisational bases of renovation, together with their legal and financial pre-conditions.

There is not enough evidence about the particular organisational schemes applied across CEE to conclude whether only minor organisational fine tuning is needed. However, considering the important role of politics and policies in creating norms (Rothstein, 1998), it seems that the proper subsidy targeting energy efficiency improvements may also serve as a powerful tool to encourage the co-operation of individual owners and improve the organisational performance of the management of privatised multidwelling buildings across CEE.

Footnotes

Acknowledgements

The authors are grateful to three anonymous referees for their helpful comments.

Funding

The paper is based on research supported by the Agency for Research and Development of the Republic of Slovenia, PS P5-0200.

Notes

References

ADEME (Agence de l’environnement et de la maîtrise de l’énergie) (2009) Energy efficiency in the European Union: overview of policies and good practices. French Environment and Energy Management Agency, Paris.

Baker

Kaul

(2002) Using multiperiod variables in the analysis of home improvement decisions by homeowners, Real Estate Economics, 30(4), pp. 551–566.

Ball

(2006) European housing review. RICS, London (www.rics.org).

Banfi

Farsi

Filippini

Jakob

(2008) Willingness to pay for energy-saving measures in residential buildings, Energy Economics, 30, pp. 503–516.

Blandy

Dixon

Dupuis

(2006) Theorising power relationships in multi-owned residential developments: unpacking the bundle of rights, Urban Studies, 43, pp. 2365–2383.

Bouzarovski

Salukvadze

Gentile

(2011) A socially resilient urban transition? The contested landscapes of apartment building extensions in two post-communist cities, Urban Studies, 48, pp. 2689–2714.

Bridge

(2006) Perspectives on cultural capital and the neighbourhood, Urban Studies, 43, pp. 719–730.

Buckley

R. M.

Tsenkova

(2003) Housing reforms and market performance, in: Lowe

Tsenkova

(Eds) Housing Change in East and Central Europe: Integration or Fragmentation?, pp. 3–20. Aldershot: Ashgate.

Cento Bull

Jones

(2006) Governance and social capital in urban regeneration: a comparison between Bristol and Naples, Urban Studies, 43, pp. 767–786.

10.

Černič-Mali

Sendi

Filipovič Hrast

Boškić

. (2003) Large housing estates in Slovenia: overview of developments and problems in Ljubljana and Koper. Report No. 2g, RESTATE, Faculty of Geosciences, Utrecht University.

11.

Clapham

Hegenüs

Kintrea

Tosics

Kay

H. I.

(1996) Housing Privatization in Eastern Europe. Westport, CT: Greenwood Press.

12.

Domanski

Ostrowska

Przybysz

Romaniuk

Krieger

(2006) First European quality of life survey: social dimensions of housing. European Foundation for the Improvement of Living and Working Conditions, Dublin.

13.

EC (European Commission) (2005) Joint report on social inclusion in the 10 new member states. Office for Official Publications of the European Communities, Luxembourg.

14.

EEA (European Environment Agency) (2007) Sustainable consumption and production in the south east Europe and eastern Europe, Caucasus and central Asia. EEA, Copenhagen.

15.

Forrest

Kearns

(2001) Social cohesion, social capital and the neighbourhood, Urban Studies, 38, pp. 2125–2143.

16.

Golove

W. H.

Eto

J. H.

(1996) Market barriers to energy efficiency: a critical reappraisal of the rationale for public policies to promote energy. Energy & Environment Division, University of California at Berkeley, CA.

17.

Government of the Republic of Slovenia (2008) National Energy Efficiency Action Plan 2008–2016. Ljubljana, 31 January.

18.

Greene

W. H.

(2000) Econometric Analysis, 4th edn. Upper Saddle River, NJ: Prentice Hall.

19.

Gyourko

Saiz

(2004) Reinvestment in the housing stock: the role of construction costs and the supply side, Journal of Urban Economics, 55, pp. 238–256.

20.

Hegedüs

Teller

(2006) Managing risks in the new housing regimes of the transition countries: the case of Hungary, in: Doling

Elsinga

(Eds) Home Ownership: Getting In, Getting From, Getting Out, Part II, pp. 175–200. Amsterdam: IOS Press.

21.

Hegedüs

Teller

(2007) Hungary: escape into home ownership, in: Elsinga

M. G.

de Decker

Teller

Toussaint

(Eds) Home Ownership beyond Asset and Security, pp. 133–173. Amsterdam: IOS Press.

22.

Hegedüs

Tosics

(1996) The disintegration of the east European housing model, in: Clapham

Hegedüs

Kintrea

Tosics

, with Kay

(Eds) Housing Privatization in Eastern Europe, pp. 15–40. Westport, CT: Greenwood Press.

23.

Hegedüs

Mayo

Tosics

(1996) Transition of the housing sector in the east-central European countries. USAID, Metropolitan Research Institute, Budapest.

24.

Helms

A. C.

(2003) Understanding gentrification: an empirical analysis of the determinants of urban housing renovation, Journal of Urban Economics, 54, pp. 474–498.

25.

IEA (International Energy Agency) (2006) World energy outlook. IEA, Paris.

26.

Jaffe

A. B.

Stavins

R. T.

(1994) The energy-efficiency gap: what does it mean?, Energy Policy, 22, pp. 804–810.

27.

Jakob

(2007) The drivers of and barriers to energy efficiency in renovation decisions of single-family home-owners. Working Paper No. 56, Centre for Energy Policy and Economics, Zurich.

28.

Kleinhans

Priemus

Engbersen

(2007) Understanding social capital in recently restructured urban neighbourhoods: two case studies in Rotterdam, Urban Studies, 44, pp. 1069–1091.

29.

Louviere

J. J.

Hensher

D. A.

Swait

J. D.

(2000) Stated Choice Methods: Analysis and Application. Cambridge: Cambridge University Press.

30.

Lowe

Tsenkova

(2003) Housing Change in Central and Eastern Europe. Aldershot: Ashgate.

31.

Maddala

G. S.

(1983) Limited-dependent and Qualitative Variables in Econometrics. Cambridge: Cambridge University Press.

32.

Mandič

(2001) Residential mobility versus ‘in-place’ adjustments in Slovenia: viewpoint from a society in transition, Housing Studies, 16, pp. 53–73.

33.

Mandič

(2010) The changing role of housing assets in post-socialist countries, Journal of Housing and the Built Environment, 25, pp. 213–226.

34.

Martinaitis

Rogoža

Bikmaniene

(2004) Criterion to evaluate the ‘twofold benefit’ of the renovation of buildings and their elements, Energy and Buildings, 36, pp. 3–8.

35.

Mayer

(1981) Rehabilitation decisions in rental housing: an empirical analysis, Journal of Urban Economics, 10, pp. 76–94.

36.

Mendelsohn

(1977) Empirical evidence on home improvements, Journal of Urban Economics, 4, pp. 459–468.

37.

Middleton

Murie

Groves

(2005) Social capital and neighbourhoods that work, Urban Studies, 42, pp. 1711–1738.

38.

Milstead

Miles

(2011) DIY home improvements in a post-socialist housing market: a socio-spatial analysis of Vilnius, Lithuania, Housing Studies, 26, pp. 403–421.

39.

Murie

Knorr-Siedow

Kempen

R. van

(2003) Large-scale housing estates in Europe. Report No. 1, RESTATE, Faculty of Geosciences, Utrecht University.

40.

Norris

Shiels

(2007) Housing inequalities in an enlarged European Union: patterns, drivers, implications, Journal of European Social Policy, 17, pp. 65–76.

41.

Plaut

P. O.

Plaut

S. E.

(2010) Decisions to renovate and to move, Journal of Real Estate Research, 32, pp. 461–484.

42.

Purdue

(2001) Neighbourhood governance: leadership, trust and social capital, Urban Studies, 38, pp. 2211–2224.

43.

Putnam

(1993) Making Democracy Work. Princeton, NJ: Princeton University Press.

44.

Rabenhorst

Ignatova

(2009) Condominium housing and mortgage lending in emerging markets: constraints and opportunities. Working Paper No. 2009-04, Urban Institute Center on International Development and Governance, Washington, DC.

45.

Rothstein

(1998) Just Institutions Matter: The Moral and Political Logic of the Universal Welfare State. Cambridge: Cambridge University Press.

46.

Saegert

Winkel

Swartz

(2002) Social capital and crime in New York City’s low-income housing, Housing Policy Debate, 13, pp. 189–226.

47.

Schweinchen

(2006) Challenges in the housing sector: experiences from western and eastern countries. Paper presented at the European Network for Housing Research Conference, Housing in an Expanding Europe: Theory, Policy, Implementation and Participation, Ljubljana, July.

48.

Sitar

Žegarac Leskovar

Krmelj

(2009) Stimulating integral refurbishment principles in social housing as a challenge for sustainable development: case study: energy saving strategies of the Municipality of Maribor, Slovenia. Paper presented at the European Network for Housing Research Conference, Changing Housing Markets: Integration and Segmentation, Prague, June/July.

49.

Soaita

A. M.

(2012) Strategies for in situ home improvement in Romanian large housing estates, Housing Studies, 27, pp. 1008–1030.

50.

Stephens

(2003) Globalisation and housing finance systems in advanced and transition economies, Urban Studies, 40, pp. 1011–1026.

51.

Struyk

(1996) Economic Restructuring of the Former Soviet Bloc: The Case of Housing. Washington, DC: The Urban Institute Press.

52.

Tax Administration of the Republic of Slovenia (2006) Slovenian tax authority data. Ljubljana.

53.

Taylor

(2000) Communities in the lead: power, organisational capacity and social capital, Urban Studies, 37, pp. 1019–1035.

54.

Temkin

Rohe

(1998) Social capital and neighborhood stability, Housing Policy Debate, 9, pp. 61–88.

55.

UNECE (United Nations Economic Commission for Europe) (2007) UNECE Report 2006–2007: Achievements and Challenges. Geneva: UNECE.

56.

UNECE (2009) Toward energy-efficient housing: prospects for UNECE member states. Committee on Housing and land Management, United Nations Economic Commission for Europe, Geneva.

57.

Yau

(2010) Sense of community and homeowner participation in housing management: a study of Hong Kong, Urbani izziv, 21, pp. 126–135.