Review of Constructions and Materials Used in Swedish Residential Buildings During the Post-War Peak of Production

buildings

Article Review of Constructions and Materials Used in Swedish Residential Buildings during the Post-War Peak of Production

Björn Berggren * and Maria Wall Department of Architecture and Built Environment, Division of Energy and Building Design, Lund University, Box 118, 221 00 Lund, Sweden; [email protected] * Correspondence: [email protected]

Received: 13 March 2019; Accepted: 19 April 2019; Published: 23 April 2019

Abstract: One of the greatest challenges for the world today is the reduction of greenhouse gas emissions. As buildings contribute to almost a quarter of the greenhouse gas emissions worldwide, reducing the energy use of the existing building stock is an important measure for climate change mitigation. In order to increase the renovation pace, there is a need for a comprehensive technical documentation that describes different types of buildings in the existing building stock. The purpose of this study is to analyse and describe existing residential buildings in Sweden. The data are based on published reports from 1967 to 1994 that have not been publicly available in a database for other researchers to study until now. Data from the reports have been transferred to a database and analysed to create a reference for buildings and/or a description of building typology in Sweden. This study found that there is a rather large homogeneity in the existing residential building stock. However, it is not possible to use a single reference building or building technique to cover the majority of the existing buildings. In Sweden, common constructions for exterior walls in multi-dwelling buildings which should be used for further studies are insulated wood infill walls with clay brick façades, lightweight concrete walls with rendered façades and concrete sandwich walls. The most common constructions for one- and two-dwelling buildings are insulated wooden walls with clay brick façades or wooden façades. Furthermore, roof constructions with insulated tie beam and roof constructions where the tie beam is a part of the interior floor slab are frequently used and should be included in further studies.

Keywords: renovation; residential buildings; reference building; building stock data base

1. Introduction One of the greatest challenges for the world today is the reduction of greenhouse gas emissions. Worldwide, energy use in buildings accounts for over 40% of the world’s primary energy usage and 24% of the greenhouse gas emissions [1]. Within the European Union (EU), Switzerland and Norway, the largest portion of the housing stock is residential houses, with a current growth rate of around 1% [2]. Thus, even if policy-makers set strict energy requirements for the new construction of residential houses, the eﬀect may be low. Hence, reducing the energy use in the existing building stock is an important measure for climate change mitigation. Speciﬁcally, in Sweden, there are 4.5 million homes and the Swedish National Board of Housing, Building and Planning (Boverket), estimated that 75 percent of these dwellings must undergo major renovations before 2050 [3]. Boverket further concluded that the pace of renovation of existing buildings must increase as an important climate change mitigation measure. Several projects and studies have demonstrated that there is a large potential to reduce the energy demand in existing buildings by improving the building envelopes and technical installations [4–23]. A

Buildings 2019, 9, 99; doi:10.3390/buildings9040099 www.mdpi.com/journal/buildings Buildings 2019, 9, 99 2 of 21 recent study from Italy also described different approaches for U-value assessment, including infrared thermography, which may be used to assess the U-value of constructions in existing buildings [24]. Detailed analyses for energy performance can be found in different studies [9,12–16,20–23] where energy performance was investigated for cities, boroughs and buildings. However, the description/input data collected from the buildings are generic in most of the mentioned studies. Descriptions of building envelopes are presented as U-/R-values, rather than describing the construction. Only four of the mentioned studies present the building envelopes and/or renovation measures in more detail [20–23]. Studies have also shown that many of the existing buildings in Europe were built between 1940 and 1980, and opportunities to use prefabricated building systems for the energy renovation of these building envelopes have been identified [17–19,25]. These opportunities are partly based on the finding that there is a large homogeneity in this building segment (buildings constructed between 1940 and 1980) [4,18,19,26,27]. In Sweden, this homogeneity is partly because during the most intense construction period, the so-called “Miljonprogrammet” (the million homes programme), the undertaking of projects with 1000 or more apartments in similar buildings was encouraged through rules on mortgages [26]. The million homes programme represents the era that was driven by the 1966 Swedish Parliament’s decision that one million new homes would be built in a decade (1965–1974) [28]. However, there is a risk that the assumption of homogeneity is too general, since it is largely based on the architectural design of the buildings, i.e., not looking into differences among different regions within Sweden and differences regarding load bearing structure. For example, it may be possible to renovate a lightweight concrete exterior wall with a rendered façade by applying additional insulation on the exterior side of the wall followed by a new layer of rendered façade. The same measure cannot be applied on a lightweight infill wall with a ventilated façade. A more regional analysis is important in the Swedish context, since many contractors focused on renovation may be active in a limited region. Another issue concerning building envelope improvement is that even if an increased thermal resistance of building envelopes can improve the energy performance of existing buildings, it is seldom profitable [4,9], especially concerning the exterior walls. Thus, there is a need for the development of more cost-effective methods or prefabricated building elements that can substantially increase the thermal resistance of the building envelopes of existing buildings. To enable this development, there is a need for a comprehensive technical document that describes different types of buildings in the existing building stock in Sweden as well as the building systems and materials used. In Sweden, such a technical description exists and includes the method of production and materials used, among other aspects. Between 1967 and 1994, Statistics Sweden (SCB) [29] published a compilation of the data every year, covering from 1962 up until 1992 [30]. However, the yearly-published data are not available in a database, which makes the information difficult to analyse in a generic way for renovation purposes. The data come from applications for state loans, where technical descriptions of the buildings were made based on a predefined template. The loans, granted by the state, ended in 1992 [28]. It should be noted that the idea of describing a building stock or reference buildings is not new. The recast of the energy Performance of Buildings Directive (EPBD) [31] required that member states set minimum requirements for energy performance based on optimal cost levels, highlighting the need for reference buildings to allow optimal cost levels to be defined. An extensive research project, TABULA [32], was conducted in 2009–2012. It included the creation of residential building typologies for 13 European countries in order to make the energy refurbishment processes in the European housing sector transparent and effective. Within this project, a report describing the Swedish situation was published. The report defined three different geometries for residential buildings. Based on these three building geometries, the effects of different energy-saving measures were investigated [33]. In many of these cases, the construction of the existing exterior walls was not defined. Two general Swedish reference buildings were defined by The Swedish National Board of Housing, Building and Planning (Boverket) in 2010 [34]. Later, based on the EPBD recast, Boverket defined ten reference buildings in order to calculate optimal cost levels for energy performance in 2013 [35]. Buildings 2019, 9, 99 3 of 21

However, only five different geometries and three different exterior wall constructions were defined, as other parameters varied, such as ventilation and heating systems. A Special Issue of the journal Energy and Buildings was recently published with a focus on monitoring, mapping and modelling the existing building stock in Europe [36–46]. However, of the previous studies that compiled and described a specific building typology, only one Spanish case [46] where the constructions of different buildings were described more in detail was found. There are several Swedish-based studies that describe the building typology and renovation measures more in detail [20,21,47–49]. However, some concerns still exist. Three of the studies [20,47,48] based their topologies on previous work carried out by Boverket [34], who produced an inventory of 1400 residential buildings. However, it was pointed out that it is unclear whether these buildings are statistically representative or not [20]. Furthermore, none of these three studies present differences in building typology based on geographic location. The fourth study [21] uses reference houses without any specification regarding underlying data for these. The last of the mentioned studies [49] bases its building typology on architectural books [50,51], which describe architectural trends rather than statistical data. As mentioned at the beginning of the introduction, Boverket estimated that 75 percent of the existing homes in Sweden must undergo major renovation before 2050 [3]. There are no specific data related to how existing homes are renovated today. However, the inventory carried out by Boverket in existing buildings [34] shows that the amount of insulation in existing buildings is low compared to Swedish regulations today. Buildings before 1990 can be expected to have less than 100 mm of insulation in ground constructions, less than 200 mm of insulation in exterior walls and less than 300 mm of insulation in roof constructions. No Swedish-based study has conducted a bottom-up analysis to create reference buildings and/or a description of building typologies, including type of buildings, load bearing constructions, materials used, etc., for different regions in Sweden. A recent Swedish-based study presents a method for creating a synthetic building stock [52]. The study underlines the lack of data available for building stock characterization, which necessitates the use of a synthetic model. Therefore, in this article, data from the SCB reports [30] were compiled to enable generic bottom-up analyses, which describe the data in order to create references for further studies. This may enable strategic development of more cost-effective and robust methods or prefabricated building elements that can substantially increase the thermal resistance of building envelopes in existing buildings. All of the compiled data is available for other researchers for further studies. The first part of this article introduces the research problem and the purpose of the article. The second part gives an overview of definitions and nomenclature used in this study. The third part describes the available data. The fourth part describes the method used to analyse data followed by the fifth part, which presents the results. The sixth part discusses the results and compares them with previous research. In the final part of this article, conclusions and recommendations for further research are given.

2. Deﬁnitions and Nomenclature Swedish residential buildings have some characteristic aspects, which were registered by SCB. These characteristics are explained below and are presented in Figures1–3. Multi-dwelling building A building containing three or more dwellings. The building may be • a balcony access building, point block building, slab block building or terraced building, as explained below and in Figure1. One- or two-dwelling building A building containing one or two dwellings. The building may be a • one-dwelling building, two-dwelling building, linked building or terraced building, as explained below and in Figure2. Balcony access building A multi-dwelling building with one (or more) common staircase. The • dwellings are accessed through a common balcony on each storey. Buildings 2019, 9, x FOR PEER REVIEW 4 of 23

• Balcony access building A multi-dwelling building with one (or more) common staircase. The dwellings are accessed through a common balcony on each storey. • Point block building A multi-dwelling building with one central core/common staircase in the centre of the building. • Slab block building A multi-dwelling building with two or more common staircases. • Terraced building BuildingsA 2019multi-dwelling, 9, 99 building or two-dwelling building, usually with almost identical dwellings,4 of 21 which shares one or two walls with a neighbouring dwelling. • One-dwelling building Point block building A multi-dwelling building with one central core/common staircase in the • A building containing one dwelling. centre of the building. • Two-dwelling building Slab block building A multi-dwelling building with two or more common staircases. • A building containing two dwellings, usually with almost identical dwellings stacked on top of Terraced building A multi-dwelling building or two-dwelling building, usually with almost • each other. identical dwellings, which shares one or two walls with a neighbouring dwelling. • Linked building One-dwelling building A building containing one dwelling. • A number of buildings (may be more than two) which are connected via a complementary Two-dwelling building A building containing two dwellings, usually with almost identical • building (not used as a dwelling), such as a garage or storage area. • dwellingsTransverse stacked load-bearing on top of each other. Linked building A number of buildings (may be more than two) which are connected via a • A superstructure of a building (usually slab block buildings or balcony access building) based complementaryon a system where building the gable (not usedwalls asand a dwelling),interior walls such are as aload-bearing. garage or storage The load-bearing area. walls Transverseare oriented load-bearing transversely in A relation superstructure to the building’s of a building dominant (usually longitudinal slab block direction. buildings or • • balconyLongitudinal access load-bearing building) based on a system where the gable walls and interior walls are load-bearing.A superstructure The of load-bearinga building where walls the are load-bea orientedring transversely walls are oriented in relation in the to same the building’s direction dominantas the building’s longitudinal dominant direction. longitudinal direction. The gable walls may also be load-bearing. • LongitudinalColumn construction load-bearing A superstructure of a building where the load-bearing walls are • orientedA superstructure in the same of a direction building as where the building’s the dominant dominant load-bearing longitudinal wall direction.is based on The columns. gable walls • mayMalmö also region be load-bearing. ColumnIncludes constructionthe municipalities A superstructure of Bara, ofBurlöv, a building Dalby, where Genarp, the dominant Kävlinge, load-bearing Lomma, wallLund, is • basedLöddeköpinge, on columns. Malmö, Månstorp, Räng, Skannör, Staffanstorp, Svedal, Södra Sandby, MalmöTrelleborg, region Veberöd Includes and the Vellinge. municipalities of Bara, Burlöv, Dalby, Genarp, Kävlinge, Lomma, Lund, • • Löddeköpinge,Göteborg region Malmö, Månstorp, Räng, Skannör, Staﬀanstorp, Svedal, Södra Sandby, Trelleborg, VeberödIncludesand the municipalities Vellinge. of Askim, Fjärås, Göteborg, Härryda, Kungsbacka, Kungälv, Lerum, GöteborgLöftadalen, region Mölndal, Includes Nödinge, the municipalities Onsala, Partille of Askim,, Skepplanda, Fjärås, Göteborg, Starrkärr,Härryda, Stenungsund, Kungsbacka, Styrsö, • Kungälv,Tjörn and Lerum,Öckerö. Löftadalen, Mölndal, Nödinge, Onsala, Partille, Skepplanda, Starrkärr, • Stenungsund,Stockholm region Styrsö, Tjörn and Öckerö. StockholmIncludes the region municipalities Includes the of municipalities Botkyrka, Dand of Botkyrka,eryd, Djurö, Danderyd, Ekerö, Djurö, Gustavsberg, Ekerö, Gustavsberg, Haninge, • Haninge,Huddinge, Huddinge, Järfälla, Lidingö, Järfälla, Nacka, Lidingö, Salem, Nacka, Sigtuna, Salem, Sollentuna, Sigtuna, Solna, Sollentuna, Stockholm, Solna, Sundbyberg, Stockholm, Sundbyberg,Tyresö, Täby, Tyresö, Upplands-Bro, Täby, Upplands-Bro, Upplands-Väsby, Upplands-Väsby, Vallentuna, Vaxholm, Vallentuna, Värmdö Vaxholm, and Värmdö Österåker. and • Österåker.Non-metropolitan regions Non-metropolitanIncludes all municipalities regionsIncludes except for all the municipalities ones listed above. except for the ones listed above. •

Buildings 2019, 9, x FOR PEER REVIEW 5 of 23

Point block Balcony access Slab block building Terraced building Figure 1. Different types of multi-dwellingbuilding buildings. The orange/dark sections represent the common building areas/staircases/entrances where the residents enter their dwellings (presented in light blue). Figure 1. Diﬀerent types of multi-dwelling buildings. The orange/dark sections represent the common areas/staircases/entrances where the residents enter their dwellings (presented in light blue).

One- Two- dwelling dwelling Linked building Terraced building building building

Figure 2. Diﬀerent types of one- and two-dwelling buildings. The orange/dark areas represent entrances Figurewhere the2. Different residents entertypes their of one- dwellings and two-dwelling (presented in lightbuildings. blue). The orange/dark areas represent entrances where the residents enter their dwellings (presented in light blue).

Transverse load bearing Longitudinal load bearing Column construction

Figure 3. Different load bearing systems. The green/dark areas represent the load-bearing constructions.

3. Available Data The available data from state loans cover the years 1962–1992 and are described in Appendices A and B. This research focuses on residential buildings produced from 1960 up until 1990. During this period, roughly 1,250,000 dwellings were produced in multi-dwelling buildings [53]. The available data from the Statistical Reports (SR) from the state loans cover almost 1,151,000 dwellings, i.e., 92% of the produced dwellings. During the same period, roughly 900,000 dwellings were produced in one- and two-dwelling buildings. The available data cover almost 620,000 dwellings (69%). There are two main reasons for the lower coverage. The first reason is that SCB did not publish statistics from state loans for one- or two dwelling buildings in 1960–1965. The second reason is that in 1988–1990, they only published data for dwellings where the applicant of state loan was not the same as the final resident. For 1966–1987, the published data cover 83% of the dwellings. A comparison of the number of newly constructed dwellings (NC) and the data from Statistical Reports (SR) is presented in Figure 4.

90 000 80 000 70 000 NC one- and two-dwelling buildings 60 000 50 000 NC Multi-dwelling buildings 40 000 SR one- and two-dwelling buildings 30 000 SR Multi-dwelling buildings 20 000 Number dwellings of Number 10 000 0 1960 1965 1970 1975 1980 1985 1990 Year Buildings 2019, 9, x FOR PEER REVIEW 5 of 23

Figure 1. Different types of multi-dwelling buildings. The orange/dark sections represent the common areas/staircases/entrances where the residents enter their dwellings (presented in light blue).

Buildings 2019, 9, x FOR PEER REVIEW 5 of 23

Figure 1. Different types of multi-dwelling buildings. The orange/dark sections represent the common areas/staircases/entrances where the residents enter their dwellings (presented in light blue).

One- Two- dwelling dwelling Linked building Terraced building building building

Figure 2. Different types of one- and two-dwelling buildings. The orange/dark areas represent One- Two- entrancesdwelling where the residentsdwelling enter their dwellingsLinked (presented building in light blue). Terraced building building building

Figure 2. Different types of one- and two-dwelling buildings. The orange/dark areas represent entrances where the residents enter their dwellings (presented in light blue). Buildings 2019, 9, 99 5 of 21

Transverse load bearing Longitudinal load bearing Column construction

Figure 3. Different load bearing systems. The green/dark areas represent the load-bearing constructions. Transverse load bearing Longitudinal load bearing Column construction

3. AvailableFigure 3. DiDataﬀerent load bearing systems. The green/dark areas represent the load-bearing constructions.

3. AvailableFigureThe available 3. Data Different data fromload bearingstate loans systems. cover Thethe yearsgreen 1962–1992/dark areas and represent are described the load-bearing in Appendices A andconstructions. B. TheThis available research data focuses from stateon residential loans cover buildings the years 1962–1992 produced and from are described1960 up until in Appendices 1990. DuringA and Bthis. 3.period, AvailableThis roughly research Data 1,250,000 focuses dwellings on residential were buildings produced produced in multi-dwelling from 1960 buildings up until 1990.[53]. The During available this period, roughly 1,250,000 dwellings were produced in multi-dwelling buildings [53]. The available dataThe from available the Statistical data from Reports state (SR) loans from cover the the state years loans 1962–1992 cover almost and are1,151,000 described dwellings, in Appendices i.e., 92% data from the Statistical Reports (SR) from the state loans cover almost 1,151,000 dwellings, i.e., 92% Aof and the B. produced dwellings. During the same period, roughly 900,000 dwellings were produced in of the produced dwellings. During the same period, roughly 900,000 dwellings were produced in one-This and researchtwo-dwelling focuses buildings. on residential The available buildings data produced cover almost from 620,000 1960 up dwellings until 1990. (69%). During There this are one- and two-dwelling buildings. The available data cover almost 620,000 dwellings (69%). There are period,two main roughly reasons 1,250,000 for the dwellingslower coverage. were producedThe first re inason multi-dwelling is that SCB didbuildings not publish [53]. Thestatistics available from two main reasons for the lower coverage. The ﬁrst reason is that SCB did not publish statistics from datastate from loans the for Statistical one- or two Reports dwelling (SR) buildingsfrom the state in 1960–1965. loans cover The almost second 1,151,000 reason isdwellings, that in 1988–1990, i.e., 92% state loans for one- or two dwelling buildings in 1960–1965. The second reason is that in 1988–1990, ofthey the only produced published dwellings. data for During dwellings the wheresame period,the appl roughlyicant of state900,000 loan dwellings was not the were same produced as the final in they only published data for dwellings where the applicant of state loan was not the same as the ﬁnal one-resident. and two-dwelling For 1966–1987, buildings. the published The available data cover data 83 cover% of the almost dwellings. 620,000 A dwellings comparison (69%). of the There number are resident. For 1966–1987, the published data cover 83% of the dwellings. A comparison of the number twoof newly main constructedreasons for thedwellings lower coverage. (NC) and Thethe datafirst fromreason Statistical is that SCB Reports did not (SR) publish is presented statistics in Figure from of newly constructed dwellings (NC) and the data from Statistical Reports (SR) is presented in Figure4. state4. loans for one- or two dwelling buildings in 1960–1965. The second reason is that in 1988–1990, they90 only 000 published data for dwellings where the applicant of state loan was not the same as the final resident.80 000 For 1966–1987, the published data cover 83% of the dwellings. A comparison of the number of newly70 000 constructed dwellings (NC) and the data from Statistical Reports (SR) is presented in Figure NC one- and two-dwelling buildings 4. 60 000 50 000 NC Multi-dwelling buildings 9040 000 000 SR one- and two-dwelling buildings 8030 000 000 SR Multi-dwelling buildings 7020 000 000 Number dwellings of Number NC one- and two-dwelling buildings 6010 000 000 50 0000 NC Multi-dwelling buildings Buildings40 000 20191960, 9, x FOR 1965 PEER REVIEW 1970 1975 1980 1985 1990 SR one- and two-dwelling buildings6 of 23 30 000 Year SR Multi-dwelling buildings 20 000

Number dwellings of Number Figure 4. Newly constructed dwellings in Sweden. Comparison of data from newly constructed (NC) 10 000 dwelling statistics [[53]53] andand StatisticalStatistical ReportsReports (SR)(SR) [[27].27]. 0 1960 1965 1970 1975 1980 1985 1990 Figure4 4 shows shows thatthat thethe numbernumber ofof dwellingsdwellings reportedreported inin statistical statistical reportsreports (SR) (SR) was was sometimes sometimes Year higher than reported number of newly constructed dwellingsdwellings (NC). This is probably due to the offset oﬀset between the grant ofof aa statestate loanloan forfor aa buildingbuilding andand thethe completioncompletion ofof thatthat building.building.

4. Method The studystudy waswas carriedcarried outout in in four four steps, steps, as as described described in in Figure Figure5. Each5. Each step step is furtheris further described described in thisin this section. section.

1. 3. 4. 2. Comparison and Transfer of data Analysis of data Defining technical discussion of from reports to and creation of aspects of interest results in relation to database figures and graphs previous studies

Figure 5. Diagram of the method used for this study. Figure 5. Diagram of the method used for this study.

4.1. Step 1—Transfer of Data Specific data were gathered from the annual reports from SCB and transferred into a database (Excel [54]) for analysis. As data from 56 reports were transferred manually, imposing a risk of error, a quality check was carried out after the data transfer by randomly choosing ten reports and comparing the data contained in them with the data in the database.

4.2. Step 2—Defining Technical Aspects of Interest Previous literature was examined to identify the most interesting technical aspects of multi- dwelling buildings and one- or two-dwelling buildings. For multi-dwelling buildings, the type of building, number of storeys, type of superstructure and materials used for exterior walls were determined to be the most interesting technical aspects. As the roof constitutes a relatively small share of the building envelope in multi-dwelling buildings, roofs were not included in the analysis. For one- and two-dwelling buildings, the type of building, number of storeys (including the presence of a cellar) and materials used were determined to be the most interesting technical aspects.

4.3. Step 3—Analysis of Data Based on the technical aspects defined in the previous step, the existing data were analysed for different regions to create a basis for reference buildings and/or a description of building typology in Sweden. The results from step 3 are presented in the results section.

4.4. Step 4—Comparison of Results in Relation to Previous Studies The results from the analysis were compared with previous research related to building typology to enable a discussion about differences between the results from this study and previous research. The comparison is presented in the discussion section.

4.1. Step 1—Transfer of Data Speciﬁc data were gathered from the annual reports from SCB and transferred into a database (Excel [54]) for analysis. As data from 56 reports were transferred manually, imposing a risk of error, a quality check was carried out after the data transfer by randomly choosing ten reports and comparing the data contained in them with the data in the database.

4.2. Step 2—Deﬁning Technical Aspects of Interest Previous literature was examined to identify the most interesting technical aspects of multi-dwelling buildings and one- or two-dwelling buildings. For multi-dwelling buildings, the type of building, number of storeys, type of superstructure and materials used for exterior walls were determined to be the most interesting technical aspects. As the roof constitutes a relatively small share of the building envelope in multi-dwelling buildings, roofs were not included in the analysis. For one- and two-dwelling buildings, the type of building, number of storeys (including the presence of a cellar) and materials used were determined to be the most interesting technical aspects.

4.3. Step 3—Analysis of Data Based on the technical aspects deﬁned in the previous step, the existing data were analysed for diﬀerent regions to create a basis for reference buildings and/or a description of building typology in Sweden. The results from step 3 are presented in the results section.

4.4. Step 4—Comparison of Results in Relation to Previous Studies The results from the analysis were compared with previous research related to building typology to enable a discussion about diﬀerences between the results from this study and previous research. The comparison is presented in the discussion section.

5. Results As previously mentioned, many of the existing buildings in Europe were built between 1940 and 1980. In Sweden, many of the existing dwellings were built during the so called “Miljonprogrammet” (the million homes programme). Figure4 shows that roughly 70% of the dwellings in Sweden were built as multi-dwelling buildings during the 1960s and early 1970s. At the beginning of the 1970s, the production of dwellings in multi-dwelling buildings dropped signiﬁcantly, while the production of one- and two-dwelling buildings increased, and in 1974, the production of dwellings in one- or two-dwelling buildings became higher compared to multi-dwelling buildings.

5.1. Multi-Dwelling Buildings In Figure6, the distribution of multi-dwelling buildings is presented by region and year. In 1960–1965, SCB did not present the distribution of dwellings by region. However, the total number of dwellings in Sweden was presented. To indicate the possible distribution, the data for the total amount of dwellings were based on available data for the overall distribution of dwellings in the speciﬁc decade [55]. Hence, data for that period are presented as hatched, as there is uncertainty regarding the distribution. In 1966–1975, 59% of the dwellings in multi-dwelling buildings were produced in non-metropolitan regions. This means that they were not produced in the regions of Malmö, Göteborg or Stockholm. The production of multi-dwelling buildings dropped harshly in the mid-1970s. After a long period of low production, from the mid-1970s to the mid-1980s, an increase in the non-metropolitan regions occurred in the end of the 1980s. For the same period, no signiﬁcant increase occurred in the metropolitan regions. During the million homes programme, more than 80% of the dwellings were slab block buildings (see Figure7). After the mid-1970s, the number of dwellings in slab block buildings remained constant Buildings 2019, 9, x FOR PEER REVIEW 7 of 23 production of dwellings in multi-dwelling buildings dropped significantly, while the production of one- and two-dwelling buildings increased, and in 1974, the production of dwellings in one- or two- dwelling buildings became higher compared to multi-dwelling buildings.

5.1. Multi-Dwelling Buildings In Figure 6, the distribution of multi-dwelling buildings is presented by region and year. In 1960– 1965, SCB did not present the distribution of dwellings by region. However, the total number of dwellings in Sweden was presented. To indicate the possible distribution, the data for the total amount of dwellings were based on available data for the overall distribution of dwellings in the specific decade [55]. Hence, data for that period are presented as hatched, as there is uncertainty regarding the distribution. BuildingsIn 20191966–1975,, 9, 99 59% of the dwellings in multi-dwelling buildings were produced in 7non- of 21 metropolitan regions. This means that they were not produced in the regions of Malmö, Göteborg or Stockholm. The production of multi-dwelling buildings dropped harshly in the mid-1970s. After a longat a lowperiod level of for low a long production, period of from time, the with mid-1 a small970s increase to the atmid-1980s, the end of an the increase 1980s. Instead, in the othernon- metropolitanbuilding types regions became occurred more common. in the end The of share the of198 point0s. For block the buildings same period, and balcony no significant access buildingsincrease occurredincreased, in but the there metropolitan were also regions. increases in other types of buildings.

80 000 70 000 60 000 50 000 40 000 30 000 20 000

Number of dwellings10 000 0 1960 1965 1970 1975 1980 1985 1990 Year A B C D Figure 6. Distribution of dwellings in multi-dwelling buildings for diﬀerent regions and years, as Figure 6. Distribution of dwellings in multi-dwelling buildings for different regions and years, as Buildingsdetermined 2019, 9, x FOR by statePEER loans:REVIEW (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region,8 of 23 determined by state loans: (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) (D) Stockholm region. Stockholm region. 80 000 During the million homes70 000 programme, more than 80% of the dwellings were slab block buildings 60 000 (see Figure 7). After the mid-1970s, the number of dwellings in slab block buildings remained 50 000 constant at a low level for a40 long 000 period of time, with a small increase at the end of the 1980s. Instead, Buildings 2019, 9, x FOR PEER REVIEW 8 of 23 other building types became30 000more common. The share of point block buildings and balcony access buildings increased, but there20 000 were also increases in other types of buildings.

The distribution of different Number of dwellings 8010 000 types of buildings in different regions was rather equal in1966–1975 (there are no specific data 70for 000 different0 regions for 1960–1965) with the exception of the Stockholm 60 0001960 1965 1970 1975 1980 1985 1990 region, where the shares of 50balcony 000 access buildings andYear point block buildings dwellings were higher compared to the rest of Sweden (see Figure 8). However, when the share of dwellings in slab block 40 Slab000 block Balcony access Point block Other buildings decreased from 30the 000 mid-1970s, it did not decrease as much in the Stockholm region Figure 7. Distribution of dwellings in multi-dwelling buildings by type of building and year of comparedFigure to 7. inDistribution the rest of of Sweden20 dwellings 000 (see in multi-dwellingFigure 8). buildings by type of building and year of state

state loan. Number of dwellings loan. 10 000 0 The distribution of different1960 types 1965 of buildings 1970 in1975 different 1980 regions 1985 was 1990 rather equal in1966–1975 (there are no specific data100 for different regions for 1960–1965)Year with the exception of the Stockholm region, where the shares of balcony75Slab block accessBalcony buildings access andPoint point block block buildingsOther dwellings were higher compared to the rest of Sweden (see Figure8). However, when the share of dwellings in slab block Figure 7. Distribution of dwellings50 in multi-dwelling buildings by type of building and year of state buildings decreased from the mid-1970s, it did not decrease as much in the Stockholm region compared loan. 25 to in the rest of Sweden (see Figure8). 0

Share Share of dwellings [%] 100 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 75 ABCD 50 Slab block Balcony access Point block Other 25 Figure 8. Share of dwellings0 in multi-dwelling buildings by type of building for different periods and

regions: (A) non-metropolitan Share of dwellings [%] regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region. 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 ABCD Regarding the number of storeys, data are available for 1962–1993 (see Figure 9). Regional data for the number of storeys combinedSlab block withBalcony region access is availablePoint block fromOther 1968. Up until the late 1960s, dwellingsFigure in 8. multi-dwellingShare of dwellings buildings in multi-dwelling with three buildings or four by storeys type of buildingrepresented for di moreﬀerent than periods 50% and of the total Figuredwellings.regions: 8. (A)Share There non-metropolitan of dwellings is a clear in tendency multi-dwelling regions, (B) for Malmö more buildings region,high-rise by (C) type buildings Göteborg of building region, to befor built (D)different Stockholm in the periods metropolitan region. and regions,regions: especially (A) non-metropolitan in Stockholm regions, (see Figure (B) Malmö 10). As region, much (C) as Göteborg 81% of theregion, dwellings (D) Stockholm in multi-dwelling region. buildings built outside metropolitan regions are four storeys high or lower. However, in the StockholmRegarding region, the only number 35% ofof storeys,dwellings data fit areinto availa this category.ble for 1962–1993 (see Figure 9). Regional data for the number of storeys combined with region is available from 1968. Up until the late 1960s, 80 000 dwellings in multi-dwelling70 buildings000 with three or four storeys represented more than 50% of the total dwellings. There is a clear60 000 tendency for more high-rise buildings to be built in the metropolitan regions, especially in Stockholm50 000 (see Figure 10). As much as 81% of the dwellings in multi-dwelling buildings built outside metropolitan40 000 regions are four storeys high or lower. However, in the Stockholm region, only 35%30 of 000 dwellings fit into this category. 20 000

Number of dwellings 8010 000 70 0000 60 0001962 1967 1972 1977 1982 1987 50 000 Year 40 000 1-2 3-4 5-8 ≥9 30 000 Figure 9. Distribution of20 dwellings 000 in multi-dwelling buildings by the number of storeys and year of

state loan. Number of dwellings 10 000 0 1962 1967 1972 1977 1982 1987 Year 1-2 3-4 5-8 ≥9 Figure 9. Distribution of dwellings in multi-dwelling buildings by the number of storeys and year of state loan. Buildings 2019, 9, x FOR PEER REVIEW 8 of 23

80 000 70 000 60 000 50 000 40 000 30 000 20 000

Number Number of dwellings 10 000 0 1960 1965 1970 1975 1980 1985 1990 Year Slab block Balcony access Point block Other Figure 7. Distribution of dwellings in multi-dwelling buildings by type of building and year of state loan.

100

0 Share Share of dwellings [%] 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 66-75 76-85 86-90 ABCD

Slab block Balcony access Point block Other

BuildingsFigure2019, 8.9, Share 99 of dwellings in multi-dwelling buildings by type of building for different periods and8 of 21 regions: (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

RegardingRegarding the the number number of of storeys, storeys, data data are are available available for for 1962–1993 1962–1993 (see (see Figure Figure9). Regional 9). Regional data data for thefor numberthe number of storeys of storeys combined combined with region with isregion available is available from 1968. from Up 1968. until theUp lateuntil 1960s, the late dwellings 1960s, indwellings multi-dwelling in multi-dwelling buildings with buildings three or with four three storeys or represented four storeys more represented than 50% more of the than total 50% dwellings. of the Theretotal dwellings. is a clear tendency There is fora clear more tendency high-rise for buildings more high-rise to be built buildings in the metropolitan to be built in regions, the metropolitan especially inregions, Stockholm especially (see Figurein Stockholm 10). As (s muchee Figure as 81% 10). ofAs the much dwellings as 81% of in the multi-dwelling dwellings in buildingsmulti-dwelling built outsidebuildings metropolitan built outside regions metropolitan are four storeys regions high are or four lower. storeys However, high in or the lower. Stockholm However, region, in only the 35%Stockholm of dwellings region, ﬁt only into 35% this category.of dwellings fit into this category.

80 000 70 000 60 000 50 000 40 000 30 000 20 000

Number of dwellings 10 000 0 1962 1967 1972 1977 1982 1987 Year 1-2 3-4 5-8 ≥9 Figure 9. Distribution of dwellings in multi-dwelling buildings by the number of storeys and year of BuildingsFigure 2019, 9.9, xDistribution FOR PEER REVIEW of dwellings in multi-dwelling buildings by the number of storeys and year 9of of 23 state loan. state loan. 100

0 Share Share of dwellings [%] 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 ABCD

1-2 3-4 5-6 7-8 ≥9

Figure 10. Share of dwellings in multi-dwelling buildings by the number of storeys for diﬀerent Figureperiods 10. and Share regions: of dwellings (A) non-metropolitan in multi-dwelling regions, buildings (B) by Malmö the number region, of (C) storeys Göteborg for different region, periods(D) Stockholm and regions: region. (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region. The type of superstructure was only presented by SCB for 1968–1972. However, this period is duringThe the type peak of ofsuperstructure multi-dwelling was building only presented production—the by SCB for million 1968–1972. homes However, programme. this Therefore, period is duringit is interesting the peak to of analyse multi-dwelling these data building (see Figure production—the 11). During million this period, homes there programme. was roughly Therefore, a 50/50 itdistribution is interesting of longitudinalto analyse these load data bearing (see andFigure transverse 11). During load this bearing period, superstructures there was roughly in the Malmöa 50/50 distributionregion and non-metropolitan of longitudinal load regions. bearing The and use tran of transversesverse load load bearing bearing superstructures was roughly 10%in the greater Malmö in regionthe Göteborg and non-metropolitan region and 10% regions. lower in The the Stockholmuse of transverse region. load bearing was roughly 10% greater in the GöteborgIn the beginning region and of the 10% 1960s, lower residential in the Stockholm buildings region. were almost exclusively designed with rendered façades or clay brick façades.100 The use of rendered façades reduced during the late 1960s, and concrete façades were rather frequently used during this period (see Figure 12). Throughout the analysed period, clay brick façades were75 most commonly used except for in dwellings with state loans from 1966 when rendered façades were used slightly more often and in dwellings with state loans from 1972 50 when concrete was used slightly more often. 25

Share Share of dwellings [%] 0 ABCD Longitudinal load bearing Transverse load bearing Pillar construction Other No record Figure 11. Share of dwellings in multi-dwelling buildings by type of superstructure for different regions (1968–1972): (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

100

0 Share Share of dwellings [%] 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 ABCD

1-2 3-4 5-6 7-8 ≥9

Figure 10. Share of dwellings in multi-dwelling buildings by the number of storeys for different periods and regions: (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

The type of superstructure was only presented by SCB for 1968–1972. However, this period is during the peak of multi-dwelling building production—the million homes programme. Therefore, it is interesting to analyse these data (see Figure 11). During this period, there was roughly a 50/50 distribution of longitudinal load bearing and transverse load bearing superstructures in the Malmö regionBuildings and2019 non-metropolitan, 9, 99 regions. The use of transverse load bearing was roughly 10% greater9 of in 21 the Göteborg region and 10% lower in the Stockholm region.

100

Share Share of dwellings [%] 0 ABCD Longitudinal load bearing Transverse load bearing Pillar construction Other No record FigureFigure 11. 11. ShareShare of of dwellings dwellings in in multi-dwelling multi-dwelling building buildingss by by type type of of superstructure superstructure for for different diﬀerent Buildingsregionsregions 2019, 9(1968–1972):, (1968–1972): x FOR PEER REVIEW(A) (A) non-metropolitan non-metropolitan regions, regions, (B) (B)Malmö Malmö region, region, (C) Göteborg (C) Göteborg region, region, (D)10 of 23 Stockholm(D) Stockholm region. region.

In the beginning of the 1960s, residential buildings were almost exclusively designed with rendered façades or clay brick façades. The use of rendered façades reduced during the late 1960s, and concrete façades were rather frequently used during this period (see Figure 12). Throughout the Buildingsanalysed 2019 period,, 9, x FOR clay PEER brick REVIEW façades were most commonly used except for in dwellings with10 stateof 23 loans from 1966 when rendered façades were used slightly more often and in dwellings with state loans from 1972 when concrete was used slightly more often. The façade materials used in different regions are presented in Figure 13. The data show that clay brick façades were not the most commonly used façade throughout Sweden for the whole analysed period. From the late 1960s to mid-1970s clay brick façades were common in non- metropolitan regions and the Malmö region, but not in the Göteborg and Stockholm regions. In the Stockholm region, rendered façades were the most common type of façade. In the Göteborg region, clay brick façades were the most common type of façade, but they were only used slightly more often FigureFigure 12. 12. ofof dwellings dwellings in in multi-dwelling multi-dwelling buildings buildings by by façade façade material material and and year year of of state state loan. loan. than concrete façades. The façade materials used in diﬀerent regions are presented in Figure 13. The data show that clay brick façades were not the most commonly used façade throughout Sweden for the whole analysed period. From the late 1960s to mid-1970s clay brick façades were common in non-metropolitan regions and the Malmö region, but not in the Göteborg and Stockholm regions. In the Stockholm region, rendered façades were the most common type of façade. In the Göteborg region, clay brick façades were Figure 12. of dwellings in multi-dwelling buildings by façade material and year of state loan. the most common type of façade, but they were only used slightly more often than concrete façades.

Figure 13. of dwellings in multi-dwelling buildings by façade material for different periods and regions: (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

In 1963–1979, SCB also published the combinations of façade material and inner material used in exterior walls, the data are shown in Figure 14. For the most common façade material, clay brick, the mostFigure common 13. of dwellings inner inmaterial multi-dwelling was wood, buildings follow by façadeed by material lightweight for diﬀ erentconcrete, periods clay and regions:bricks and concrete.Figure(A) non-metropolitanThe 13. secondof dwellings most regions, incommon multi-dwelling (B) Malmö façade region, buildings materi (C)al, Göteborgby render, façade region, wasmaterial usually (D) Stockholmfor different applied region. periods on lightweight and concreteregions: or concrete.(A) non-metropolitan Concrete façadesregions, (B)were Malmö almost region, exclusively (C) Göteborg constructed region, (D) with Stockholm concrete region. as their inner Inmaterial, 1963–1979, except SCB for also some published examples the combinationswith wood and of façade lightweight material concrete. and inner Façades material of usedwood, in sandlimeexteriorIn 1963–1979, walls, brick theor sheet SCB data alsometal are shown published were inmostly Figure the designedcombinations 14. For thein combination most of façade common material with façade wood and material, as inner the innermaterial clay brick,material used the inwithinmost exterior common the walls.walls, inner the data material are shown was wood, in Figure followed 14. For by the lightweight most common concrete, façade clay material, bricks and clay concrete. brick, the most common inner material was wood, followed by lightweight concrete, clay bricks and concrete. The second most common façade material, render, was usually applied on lightweight concrete or concrete. Concrete façades were almost exclusively constructed with concrete as their inner material, except for some examples with wood and lightweight concrete. Façades of wood, sandlime brick or sheet metal were mostly designed in combination with wood as the inner material within the walls.

Figure 14. Share of dwellings by different inner material in exterior walls for different façade materials (1963–1979).

Figure 14. Share of dwellings by different inner material in exterior walls for different façade materials (1963–1979). Buildings 2019, 9, x FOR PEER REVIEW 10 of 23

Figure 12. of dwellings in multi-dwelling buildings by façade material and year of state loan.

Figure 13. of dwellings in multi-dwelling buildings by façade material for different periods and regions: (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

In 1963–1979, SCB also published the combinations of façade material and inner material used in exteriorBuildings 2019walls,, 9, 99the data are shown in Figure 14. For the most common façade material, clay brick,10 of 21 the most common inner material was wood, followed by lightweight concrete, clay bricks and concrete. The second most common façade material, render, was usually applied on lightweight The second most common façade material, render, was usually applied on lightweight concrete or concrete or concrete. Concrete façades were almost exclusively constructed with concrete as their concrete. Concrete façades were almost exclusively constructed with concrete as their inner material, inner material, except for some examples with wood and lightweight concrete. Façades of wood, except for some examples with wood and lightweight concrete. Façades of wood, sandlime brick or sandlime brick or sheet metal were mostly designed in combination with wood as the inner material sheet metal were mostly designed in combination with wood as the inner material within the walls. within the walls.

Figure 14. Share of dwellings by different inner material in exterior walls for different façade materials (1963–1979). Figure 14. Share of dwellings by different inner material in exterior walls for different façade materials 5.2.(1963–1979). One- and Two-Dwelling Buildings Regarding one- and two-dwelling buildings, the process for attaining a state loans differs Buildings 2019, 9, x FOR PEER REVIEW 11 of 23 depending on whether the applicant of the state loan is the final resident or not. If the applicant is not thethe finalfinal resident,resident, the the applicantapplicant is is firstfirst givengiven aa preliminarypreliminary decisiondecision before before the the startstart ofof thethe constructionconstruction work.work. AA second second and and final final decision decision regarding regarding state loansstate isloans given is once given the once building the hasbuilding been completed. has been Ifcompleted. the applicant If the is the applicant final resident, is the thefinal process resident, is simpler, the process with oneis simpler, decision, with and one the applicantdecision, receivesand the theapplicant decision receives about thethe statedecision loan about before the the state start loan of the before construction the start workof the [construction27]. work [27]. ForFor buildingsbuildings withwith twotwo decisions,decisions, moremore datadata areare gathered.gathered. ThroughoutThroughout thethe periodperiod wherewhere datadata fromfrom bothboth oneone andand twotwo decisionsdecisions werewere gatheredgathered (1966–1987),(1966–1987), dwellingsdwellings withwith twotwo decisionsdecisions representrepresent 53%53% ofof thethe totaltotal datadata (see(see FigureFigure 15 15).).

25 000

20 000

15 000

10 000

5 000 Number of dwellings 0 1966 1970 1974 1978 1982 1986 1990 Year One decision Two decisions Figure 15. Distribution of dwellings in one- or two-dwelling buildings by one or two-decision state Figure 15. Distribution of dwellings in one- or two-dwelling buildings by one or two-decision state loans and year of state loan. loans and year of state loan. The distribution of dwellings in diﬀerent regions is based on data from dwellings with two The distribution of dwellings in different regions is based on data from dwellings with two decisions (see Figure 16). In 1968–1980, 70% of the dwellings in one- or two-dwelling buildings that decisions (see Figure 16). In 1968–1980, 70% of the dwellings in one- or two-dwelling buildings that were given state loans following two decisions were produced in non-metropolitan regions. This were given state loans following two decisions were produced in non-metropolitan regions. This means that they were not produced in the regions of Malmö, Göteborg or Stockholm. The production means that they were not produced in the regions of Malmö, Göteborg or Stockholm. The production of one- or two-dwelling buildings dropped in the late-1970s. In the mid-1980s, an increase occurred in of one- or two-dwelling buildings dropped in the late-1970s. In the mid-1980s, an increase occurred the non-metropolitan regions. In the metropolitan regions, no signiﬁcant increase occurred. in the non-metropolitan regions. In the metropolitan regions, no significant increase occurred.

25 000

20 000

15 000

10 000

5 000 Number Number of dwellings 0 1968 1973 1978 1983 1988 Year A B C D Figure 16. of dwellings in one- or two-dwelling buildings per region for different regions and years of state loan (with two decisions for state loans): (A) non- metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

Regarding different types of one- and two-dwelling buildings, data for different types of buildings with one decision were only gathered in 1966–1967. However, 99% of the dwellings with one decision during that period were one-dwelling buildings. Based on this, it can be assumed that more than 95% of the dwellings with one decision are one-dwelling buildings. In Figure 17, different types of buildings with two decision loans are presented together with the quantity of dwellings with one decision. One dwelling buildings together with one decision dwellings contributed to the largest share of dwellings. Together they made up between 60% and 70% of the dwellings. The largest portion of the dwellings with two decisions were terraced buildings, whose development increased significantly at the end of the 1980s. Linked buildings were rather Buildings 2019, 9, x FOR PEER REVIEW 11 of 23

the final resident, the applicant is first given a preliminary decision before the start of the construction work. A second and final decision regarding state loans is given once the building has been completed. If the applicant is the final resident, the process is simpler, with one decision, and the applicant receives the decision about the state loan before the start of the construction work [27]. For buildings with two decisions, more data are gathered. Throughout the period where data from both one and two decisions were gathered (1966–1987), dwellings with two decisions represent 53% of the total data (see Figure 15).

25 000

20 000

15 000

10 000

The distribution of dwellings in different regions is based on data from dwellings with two decisions (see Figure 16). In 1968–1980, 70% of the dwellings in one- or two-dwelling buildings that were given state loans following two decisions were produced in non-metropolitan regions. This means that they were not produced in the regions of Malmö, Göteborg or Stockholm. The production Buildingsof one-2019 or ,two-dwelling9, 99 buildings dropped in the late-1970s. In the mid-1980s, an increase occurred11 of 21 in the non-metropolitan regions. In the metropolitan regions, no significant increase occurred.

25 000

20 000

15 000

10 000

5 000 Number Number of dwellings 0 1968 1973 1978 1983 1988 Year A B C D Figure 16. of dwellings in one- or two-dwelling buildings per region for different regions and years Figure 16. of dwellings in one- or two-dwelling buildings per region for different regions and years of state loan (with two decisions for state loans): (A) non- metropolitan regions, (B) Malmö region, of state loan (with two decisions for state loans): (A) non- metropolitan regions, (B) Malmö region, (C) (C) Göteborg region, (D) Stockholm region. Göteborg region, (D) Stockholm region. Regarding different types of one- and two-dwelling buildings, data for different types of buildings with oneRegarding decision different were only types gathered of on ine- 1966–1967. and two-dwelling However, 99%buildings, of the dwellingsdata for withdifferent one decisiontypes of duringbuildings that with period one were decision one-dwelling were only buildings. gathered Based in 1966–1967. on this, it However, can be assumed 99% of that the moredwellings than 95%with ofone the decision dwellings during with onethat decisionperiod were are one-dwellingone-dwelling buildings. buildings. Based on this, it can be assumed that moreIn than Figure 95% 17 ,of di thefferent dwellings types ofwith buildings one decision with two are decisionone-dwelling loans buildings. are presented together with the quantityIn Figure of dwellings 17, different with one types decision. of buildings One dwelling with two buildings decision together loans withare presented one decision together dwellings with contributedthe quantity to of the dwellings largest share with of one dwellings. decision. Together One dwelling they made buildings up between together 60% with and one 70% decision of the dwellings contributed to the largest share of dwellings. Together they made up between 60% and Buildingsdwellings. 2019, The9, x FOR largest PEER portion REVIEW of the dwellings with two decisions were terraced buildings,12 whose of 23 development70% of the dwellings. increased The significantly largest portion at the of end the ofdw theellings 1980s. with Linked two decision buildingss were were terraced rather buildings, common commonfromwhose the development latefrom 1960s the late until 1960sincreased the mid-1970s,until significantly the mid-1970s, but their at constructionbutthe theirend ofconstruction the dropped 1980s. indroppedLinked the late buildings in 1970s the late and were 1970s remained rather and remainedrather uncommon rather uncommon throughout throughout the 1980s. the 1980s.

20 000

15 000

10 000

5 000 Number Number of dwellings 0 1966 1970 1974 1978 1982 1986 1990 Year One decision buildings Two decisions: One-dwelling buildings Two decisions: Linked buildings Two decisions: Terraced buildings Two decisions: Other Figure 17. of dwellings in one- and two-dwelling buildings by type of building and year of state loan. Figure 17. of dwellings in one- and two-dwelling buildings by type of building and year of state loan. More than 95% of the dwellings with one decision may be assumed to be one-dwelling buildings. More than 95% of the dwellings with one decision may be assumed to be one-dwelling buildings. Compared to the Göteborg and Stockholm regions, in the Malmö region and in non-metropolitan regions,Compared the share to of the dwellings Göteborg built and as one-dwelling Stockholm buildingsregions, in was the rather Malmö high region from the and late in 1960s non- to metropolitanmid-1970s. Hence, regions, the the increase share of of terraced dwellings buildings built as had one-dwelling a greater eff ectbuildings on thedistribution was rather ofhigh diff fromerent thebuildings late 1960s in the to Malmö mid-1970s. region Hence, and in the non-metropolitan increase of terraced regions buildings (see Figure had 18 a). greater effect on the distributionRegarding of different the number buildings of storeys, in the data Malmö are available region and for 1970–1987in non-metropolitan (see Figure regions19). The (see number Figure of 18).storeys combined with regions is not available. At the beginning of the 1970s, dwellings in one- and two-dwelling buildings with100 one storey contributed to more than 60% of the total number of dwellings. However, as the production of dwellings with one storey was rather constant, the number of 1.5-storey 75 buildings increased significantly, and in the mid-1970s, most of the state loans were given to dwellings 50

0 Share Share of dwellings [%] 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 ABCD One-dwelling building Linked building Terraced building Other

Figure 18. of dwellings in one- and two-dwelling buildings by type of building, for different periods and regions (with two decisions for state loans): (A) non-metropolitan regions, (B) Malmö region, (C) Göteborg region, (D) Stockholm region.

Regarding the number of storeys, data are available for 1970–1987 (see Figure 19). The number of storeys combined with regions is not available. At the beginning of the 1970s, dwellings in one- and two-dwelling buildings with one storey contributed to more than 60% of the total number of dwellings. However, as the production of dwellings with one storey was rather constant, the number of 1.5-storey buildings increased significantly, and in the mid-1970s, most of the state loans were given to dwellings built with one and a half storeys. The number of dwellings built as hillside buildings and buildings with two storeys roughly varied between 2000 and 4000 dwellings/year in the 1970s. The production dropped in the 1980s and the corresponding interval was then 1000–2000 dwellings/year. Buildings 2019, 9, x FOR PEER REVIEW 12 of 23

common from the late 1960s until the mid-1970s, but their construction dropped in the late 1970s and remained rather uncommon throughout the 1980s.

20 000

15 000

10 000

Figure 17. of dwellings in one- and two-dwelling buildings by type of building and year of state loan. More than 95% of the dwellings with one decision may be assumed to be one-dwelling buildings. Buildings 2019, 9, 99 12 of 21 Compared to the Göteborg and Stockholm regions, in the Malmö region and in non- metropolitan regions, the share of dwellings built as one-dwelling buildings was rather high from built with one and a half storeys. The number of dwellings built as hillside buildings and buildings the late 1960s to mid-1970s. Hence, the increase of terraced buildings had a greater effect on the with two storeys roughly varied between 2000 and 4000 dwellings/year in the 1970s. The production distribution of different buildings in the Malmö region and in non-metropolitan regions (see Figure dropped in the 1980s and the corresponding interval was then 1000–2000 dwellings/year. 18).

100

0 Share Share of dwellings [%] 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 68-75 76-85 86-90 ABCD One-dwelling building Linked building Terraced building Other

Figure 18. of dwellings in one- and two-dwelling buildings by type of building, for diﬀerent periods BuildingsFigure 2019, 18.9, x ofFOR dwellings PEER REVIEW in one- and two-dwelling buildings by type of building, for different periods13 of 23 Buildings 2019and, regions9, x FOR PEER (with REVIEW two decisions for state loans): (A) non-metropolitan regions, (B) Malmö region,13 of 23 and regions (with two decisions for state loans): (A) non-metropolitan regions, (B) Malmö region, (C) (C) Göteborg region, (D) Stockholm region. Göteborg region, (D) Stockholm region.

Figure 19. of dwellings in one- and two-dwelling buildings by number of storeys and year of state FigureFigure 19. 19. of ofdwellings dwellings in inone- one- and and two-dwelling two-dwelling buildings buildings by by number number of ofstoreys storeys and and year year of ofstate state loan. Missing data refers to dwellings for which data were provided to Statistics Sweden (SCB), but the loan.loan. Missing Missing data data refers refers to todwellings dwellings for for which which data data were were provided provided to toStatistics Statistics Sweden Sweden (SCB), (SCB), but but number of storeys was not speciﬁed. thethe number number of ofstoreys storeys was was not not specified. specified. Buildings with cellars were rather common at the beginning of the 1970s (Figure 20), almost 50% Buildings with cellars were rather common at the beginning of the 1970s (Figure 20), almost 50% ofBuildings the dwellings with cellars in one- were and rather two-dwelling common buildings at the beginning had cellars. of the However, 1970s (Figure dwellings 20), almost with 50% cellars of the dwellings in one- and two-dwelling buildings had cellars. However, dwellings with cellars of thedecreased dwellings during in one- the 1970s and two-dwelling and 1980s. In thebuildi latengs 1980s, had almost cellars. 90% However, of the dwellings dwellings were with built cellars without decreased during the 1970s and 1980s. In the late 1980s, almost 90% of the dwellings were built decreaseda cellar. during the 1970s and 1980s. In the late 1980s, almost 90% of the dwellings were built withoutwithout a cellar.a cellar.

Figure 20. of dwellings in one- and two-dwelling buildings based on whether they had a cellar or not. Missing data refers to dwellings for which data were provided to the SCB, but information regarding FigureFigure 20. 20. of ofdwellings dwellings in inone- one- and and two-dwelling two-dwelling buildings buildings based based on on whether whether they they had had a cellar a cellar or ornot. not. cellars was not speciﬁed. MissingMissing data data refers refers to todwellings dwellings for for which which data data we were reprovided provided to tothe the SCB, SCB, but but information information regarding regarding cellarscellars was was not not specified. specified.

InformationInformation regarding regarding material material us useded for for load-bearing load-bearing structure structure in inexterior exterior walls walls and and façade façade materialmaterial was was gathered gathered for for 1966–1987 1966–1987 and and 1966–1990, 1966–1990, respectively. respectively. Regarding Regarding the the material material used used for for load-bearingload-bearing in inexterior exterior walls, walls, wood wood was was the the domi dominantnant material material throughout throughout the the period period (see (see Figure Figure 21).21). RegardingRegarding façade façade material, material, wood wood and and clay clay brick brick façades façades were were the the most most commonly commonly used used materials.materials. Together, Together, their their share share made made up up between between 70% 70% and and 95% 95% of ofthe the dwellings dwellings in in1966–1990. 1966–1990. In Inthe the mid-1960s,mid-1960s, façades façades with with clay clay bricks bricks were were mo most stcommon common and and accounted accounted for for almost almost 70% 70% of ofthe the dwellings.dwellings. The The use use of ofwood wood became became more more and and more more common, common, and and at atth eth beginninge beginning of ofthe the 1980s, 1980s, wood wood waswas used used for for more more than than 70% 70% of ofthe the dwellings, dwellings, see see Figure Figure 22. 22. Buildings 2019, 9, 99 13 of 21

Information regarding material used for load-bearing structure in exterior walls and façade Buildingsmaterial 2019 was, 9, xgathered FOR PEER REVIEW for 1966–1987 and 1966–1990, respectively. Regarding the material used14 of for23 load-bearing in exterior walls, wood was the dominant material throughout the period (see Figure 21).

Buildings 2019, 9, x FOR PEER REVIEW 14 of 23

Figure 21. of dwellings in one- and two-dwelling buildings based on the material used for load-bearing Figurein the exterior21. of dwellings walls. Missing in one- data and refers two-dwelling to dwellings buildings for which based data wereon the provided material to used the SCB, for butload- the bearingload-bearing in the materialexterior waswalls. not Missing speciﬁed. data refers to dwellings for which data were provided to the SCB, but the load-bearing material was not specified. Regarding façade material, wood and clay brick façades were the most commonly used materials. Together,Figure their21. of share dwellings made in up one- between and two-dwelling 70% and 95% buildings of the dwellingsbased on the in 1966–1990.material used In for the load- mid-1960s, façadesbearing with in claythe exterior bricks werewalls. mostMissing common data refers and to accounted dwellings forfor almostwhich data 70% were of the provided dwellings. to the The use of woodSCB, but became the load-bearing more and morematerial common, was not andspecified. at the beginning of the 1980s, wood was used for more than 70% of the dwellings, see Figure 22.

Figure 22. Dwellings in one- and two-dwelling buildings based on façade material. Missing data refers to dwellings for which data were provided to SCB, but the façade material was not specified.

6. Discussion FigureFigure 22. 22. DwellingsDwellings in in one- andand two-dwellingtwo-dwelling buildings buildings based based on on façade façade material. material. Missing Missing data data refers The data from Statistics Sweden based on Swedish state loans covers a limited period of time in refersto dwellings to dwellings for which for which data data were were provided provided to SCB, to SCB, but the but façade the façade material material was notwas specified. not specified. Swedish history and does not cover all dwellings built during this period. However, the information 6. Discussion 6.is Discussionextensive and covers, to a large extent, the peak of dwelling production in Sweden, enabling a bottom-upThe data analysis. from StatisticsHence, it Swedenis interesting based to on co Swedishmpare these state results loans covers with previous a limited research period of related time in The data from Statistics Sweden based on Swedish state loans covers a limited period of time in toSwedish building history typology and doesand to not discuss cover alldifferences. dwellings builtIt should during be thisnoticed period. that However, previous the research information that Swedish history and does not cover all dwellings built during this period. However, the information involvedis extensive the andcreation covers, of building to a large typologies extent, the may peak have of dwelling had a different production purpose in Sweden, to this enabling research a is extensive and covers, to a large extent, the peak of dwelling production in Sweden, enabling a (gathering,bottom-up describing analysis. Hence, and sharing it is interesting data to enable to compare further these studies). results For with example, previous if researchthe purpose related of a to bottom-up analysis. Hence, it is interesting to compare these results with previous research related studybuilding is to typology make a andrough to discussassessment differences. of the energy It should performance be noticed thatof a previousbuilding researchstock, not that to involveddiscuss to building typology and to discuss differences. It should be noticed that previous research that applicablethe creation refurbishment of building typologiesmeasures in may detail, have deta hadiled a diinformationfferent purpose regarding to this materials research used (gathering, is not involved the creation of building typologies may have had a different purpose to this research required.describing and sharing data to enable further studies). For example, if the purpose of a study is to (gathering, describing and sharing data to enable further studies). For example, if the purpose of a makeThe a roughlarge share assessment of dwellings of the built energy during performance the million of homes a building programme stock, not hasto also discuss been identified applicable study is to make a rough assessment of the energy performance of a building stock, not to discuss byrefurbishment previous studies measures as an in important detail, detailed part of information the Swedish regarding building materials stock to usedfocus is on not [3,11,18]. required. The applicable refurbishment measures in detail, detailed information regarding materials used is not distributionThe large of regions share of corresponds dwellings built rather during well thewith million previous homes findings programme [23] that has 65% also of beenthe dwellings identified required. builtby previous during the studies million as homes an important programme part ofwere the built Swedish in non-metropolitan building stock to regions. focus on However, [3,11,18]. after The The large share of dwellings built during the million homes programme has also been identified separatingdistribution all ofthe regions dwellings corresponds into multi-dwelling rather well withbuildings previous and findingsone- and[ two-dwelling23] that 65% of buildings, the dwellings the by previous studies as an important part of the Swedish building stock to focus on [3,11,18]. The databuilt from during state the loans million show homes that programme 59% of the were multi-dwelling built in non-metropolitan buildings were regions. in non-metropolitan However, after distributionregions and of70% regions of the corresponds one- and two-dwelling rather well buildingswith previous were findings built in non-metropolitan[23] that 65% of the regions. dwellings It is builtimportant during to the highlight million the homes rather programme large share wereof dwel builtlings in builtnon-metropolitan in non-metropolitan regions. regions, However, since after the separatingeconomic conditionsall the dwellings are likely into tomulti-dwelling be different inbu theseildings regions and one- compared and two-dwelling to those in buildings, metropolitan the dataregions. from state loans show that 59% of the multi-dwelling buildings were in non-metropolitan regions and 70% of the one- and two-dwelling buildings were built in non-metropolitan regions. It is important to highlight the rather large share of dwellings built in non-metropolitan regions, since the economic conditions are likely to be different in these regions compared to those in metropolitan regions. Buildings 2019, 9, 99 14 of 21 separating all the dwellings into multi-dwelling buildings and one- and two-dwelling buildings, the data from state loans show that 59% of the multi-dwelling buildings were in non-metropolitan regions and 70% of the one- and two-dwelling buildings were built in non-metropolitan regions. It is important to highlight the rather large share of dwellings built in non-metropolitan regions, since the economic conditions are likely to be different in these regions compared to those in metropolitan regions.

6.1. Multi-Dwelling Buildings The overall findings about the distribution of multi-dwelling building types (slab block, point block and balcony access buildings) correspond well to previous studies [11,18]. Furthermore, findings regarding the number of storeys also correspond rather well with previous studies [11,18,23,24]. However, it is important to highlight that even though the largest portion of the dwellings in multi-dwelling buildings from the million homes programme were to be found in slab block buildings with three or four storeys, roughly 50% of the dwellings were designed in another way. Still, many studies have based their work on a single reference building. Furthermore, it is common for multi-dwelling buildings in the metropolitan regions to have five storeys or more. One of the most recently published studies describes 46 typical buildings of the Swedish building stock [46]. Considering multi-dwelling buildings built during the 1970s, the study defined three different buildings, all with six storeys or more. Two of these have a building footprint, which is typical of point block buildings. Looking at the available data, the point block buildings represent roughly 5% of the multi-dwelling buildings built during the 1970s. The third type of building (a slab block building with nine storeys) represents roughly 4% of the dwellings. Altogether, the defined typical buildings for multi-dwelling buildings built during the 1970s represent less than 10% of the dwellings in the building stock. The fact that there is a rough 50/50 distribution of superstructure types is important because it provides different possibilities for energy renovation. Buildings with transverse load bearing systems can undergo major renovations to their exterior walls without major effects on the superstructure, but this is not the case for buildings with longitudinal load bearing systems. Based on statistics regarding the frequency of slab block buildings, previous studies have concluded that such buildings all use the same building technique including a transverse load bearing system and light infill walls [18,30,32]. This conclusion is wrong as the data presented here show a rough 50/50 distribution for the superstructure type. Regarding the use of façade materials and inner material in the exterior walls, the results show that although certain materials are predominant, there is still a diverse range of materials used. For example, the most common material in walls behind clay brick façade is wood (43%). However, almost 25% of the dwellings with clay brick façades have an inner material of lightweight concrete. Clay bricks (15%) and concrete (12%) also make up for more than 25%. There are also rather large regional differences regarding common façade materials. The TABULA study, which investigated potential energy savings in the Swedish building stock in 2012 [30], defined all existing multi-dwelling buildings as three-storey buildings and did not define the exterior wall constructions. The reason for this definition may have been an assumption that the energy-saving measures could be applied regardless of wall construction. This is a simplification that is likely not true. In 2010, Boverket, the Swedish National Board of Housing, Building and Planning, concluded that the average multi-dwelling building in Sweden was built in 1959 [31]. This is likely to be the mean value of all multi-dwelling buildings. This is an incorrect description of the most common building in Sweden. If the analysis was based on median values instead of mean values, it would show that the most common multi-dwelling building was built during the million homes programme. The most detailed study from Boverket [32] analysed the cost-optimal energy performance requirements for existing and new buildings. Regarding existing multi-dwelling buildings, Boverket based their analysis on two reference buildings: a three-storey building from the 1950s with lightweight Buildings 2019, 9, 99 15 of 21 concrete exterior walls covered with render and a nine-storey building from the 1970s with concrete sandwich walls. The chosen reference exterior walls cover only approximately 25% of the existing buildings built during the million homes programme in Sweden. Furthermore, they did not include the most common exterior wall construction: wooden infill walls with clay brick façades. It should be noted that the data used by Boverket from 1400 buildings to define their reference buildings is of high quality. The reason for the inadequate choice of reference buildings is due to poor use of the data.

6.2. One- and Two-Dwelling Buildings The data for one- and two-dwelling buildings show that the production of dwellings with two decisions where the applicant was not the final resident was higher at its highest point and lower at its lowest point, compared to dwellings with one decision. This indicates that residents who build their own home are not as sensitive to the market as construction companies may be. The fact that almost all dwellings in one- and two dwelling buildings were built with wooden constructions makes further work easier, because it means that mainly variations of façade material need to be considered in future work. For one- and two-dwelling buildings, it will be important to study both the roof and wall constructions. Hence, it is interesting to know whether buildings are 1.5-storey buildings or not, since these buildings have very different conditions, from the perspective of adding insulation to the roof construction. Regarding one- and two-dwelling buildings, the TABULA study based their work on two reference buildings which were both one-storey buildings with lightweight exterior walls covered with render and horizontal insulation in the roof [30]. Thus, the study does not cover any 1.5-storey buildings. This is a rather strange reference building. The chosen wall construction is likely to cover less than 5% of the existing buildings and the choice to not include 1.5-storey buildings excludes roughly 30–40% of the existing one- and two-dwelling buildings. The study from Boverket in 2010 [31], concluded that the average one- and two-dwelling building in Sweden is a building built in 1953. As previously mentioned, this is likely to be a mean value; an analysis based on median values would show that the most common one- and two-dwelling building was built during the million homes programme. The most detailed study from Boverket [32] based its analysis on two reference buildings: a 1.5-storey building and a two-storey building. Both reference buildings have wooden constructions for the exterior walls and roof, and also the façades are wooden. None of the reference buildings include a cellar. By including both a 1.5-storey building and a two-storey building, the study included almost all existing roof constructions. However, by only including wooden façades, roughly 50% of the façade constructions were excluded. Cellars, which may be found in roughly 30% of the existing one- and two-dwelling buildings, were not included in the study. A recent study that included renovation measures [21] included two different types of exterior walls: wood and lightweight concrete. Further studies on one- and two-dwelling buildings could exclude exterior walls with lightweight concrete and increase their focus on different types of roof constructions and/or façade materials, as they vary more. The study that suggested 46 different typical buildings to represent the existing building stock [46] argued that a 1.5-storey building was the most common building type to be built during the 1970s and should be the choice of a typical building as it would represent 65% of the buildings from that decade. However, looking at the available data, the most common building has one storey. Roughly 40% of the buildings from the 1970s have one storey.

7. Conclusions This study shows the importance of studying differences among building constructions in the existing building stock when studying renovation measures and analysing the renovation potential. It Buildings 2019, 9, 99 16 of 21 shows that there is a set of constructions and building techniques that were commonly used in the existing building stock in Sweden during the million homes programme and in the decades before and after. It is important to underline that there is no single construction type or building that has been predominant. Furthermore, regional differences exist. Previous studies have often assumed a rather large homogeneity and did not always include the most common constructions and building types in their studies. This concerns studies regarding the development of prefabricated building elements and studies regarding cost calculations for renovation and energy efficiency measures. Furthermore, studies that used a large set of buildings to create a building typology created typologies that cannot be confirmed by the data in this study. If assumptions of large homogeneity are misjudged, they may cause higher costs for renovation measures than predicted, and developed prefabricated building elements may apply on fewer buildings than expected. This may limit the reliability of potential studies and slow down the renovation pace or limit the actual renovation measures. To speed-up the renovation of the existing building stock in the EU and in other regions, further studies are needed to form a basis for making well-informed decisions regarding political directives and incentives and regarding actual renovation measures. Furthermore, as buildings will always be unique, the development of prefabricated building systems needs to have flexibility to enable their use on a larger scale. Based on the available data, it is possible to draw some conclusions regarding construction types, which should be prioritised in further research regarding the Swedish building stock.

7.1. Multi-Dwelling Buildings The most commonly used façade materials in multi-dwelling buildings are clay bricks, render and concrete. Façades with clay bricks are common throughout Sweden. The most common inner material used for clay brick façades is wood, which indicates that it is most likely found in light infill walls. Rendered façades are most common in the Stockholm region and are also rather common in non-metropolitan regions. The rendered façades are almost exclusively paired with lightweight concrete as the inner material. Concrete façades are common in Malmö, Göteborg and Stockholm regions. Concrete façades are almost exclusively paired with concrete as the inner material. Based on these findings, the most common constructions that should be investigated in future studies are summarised in Table1. The constructions include both load-bearing walls and light infill walls in the existing building stock.

Table 1. Summary of common constructions for exterior walls in multi-dwelling buildings in Sweden produced in the post-war period.

Region Type of Construction Non-Metropolitan Malmö-Region Göteborg-Region Stockholm-Region Insulated wood inﬁll walls with clay XXXX brick façades Lightweight concrete walls with XX rendered façades Concrete sandwich walls X X X

7.2. One- and Two-Dwelling Buildings The most common façade materials used in one- and two-dwelling buildings are clay bricks and wood. Together, these two materials represent more than 80% of the dwellings from the studied period. Almost all exterior walls are constructed with wood as the inner material. Furthermore, roof constructions with an insulated tie beam and roof constructions where the tie beam is also part of an interior floor slab (in 1.5-storey buildings) need to be studied. Based on these findings, the most Buildings 2019, 9, 99 17 of 21 common constructions for further studies are summarised in Table2. As can be seen, there are no regional differences regarding the most common constructions.

Table 2. Summary of common constructions for exterior walls and roofs in one- and two-dwelling buildings in Sweden constructed in the post-war period.

Region Type of Construction Non-Metropolitan Malmö-Region Göteborg-Region Stockholm-Region Insulated wood walls with clay brick façades XXXX Insulated wood walls with wood façades X X X X Roof constructions with insulated tie beam X X X X Roof constructions for 1.5-storey buildings X X X X

Supplementary Materials: The following are available online at http://www.mdpi.com/2075-5309/9/4/99/s1. Author Contributions: data curation, B.B.; visualization, B.B.; writing—original draft, reviewing and editing, B.B. and M.W. Funding: This study was funded by The Development Fund of the Swedish Construction Industry (SBUF) and Skanska Sverige AB as part of the project; “Klimatskal 2019”, which aims to develop robust renovation measures for existing building envelopes in Sweden. Conﬂicts of Interest: The authors declare no conﬂict of interest.

Appendix A. —Multi-Dwelling Buildings

Table A1. Available data for multi-dwelling buildings, via supplementary data ﬁle.

Available Type of Data Period for Available Data Parameters Balcony access building; Point Type of building block building; Slab block 1960–1993 * building; Terraced building; Other Göteborg region; Malmö region; Area of construction Stockholm region; Sweden 1966–1993 ** excluding metropolitan regions Storeys 1–2; 3; 4; 5–8; 9 1962–1993 ≥ Transverse load bearing; Type of superstructure Longitudinal load bearing; Pillar 1968–1972 construction; Other Autoclaved aerated concrete; Clay Material for superstructure 1963–1987 bricks; Concrete; Wood; Other Method of production for superstructure On site; Prefabricated 1968–1979 Asbestos; Autoclaved aerated concrete; Clay bricks; Concrete; Façade material 1963–1993 *** Render; Sandlime bricks; Sheet metal; Wood; Other Autoclaved aerated concrete; Clay Inner material in exterior wall 1963–1979 bricks; Concrete; Wood; Other Method of production for exterior wall On site; Prefabricated 1968–1979 Asbestos; Clay tiles; Concrete tiles; Rooﬁng 1969–1993 Roof felt; Sheet metal; Other * Terrace buildings were reported separately from 1979. Before 1979; they are included in “Other”; ** From 1987, data regarding façade material were reported by western Sweden (expanded Göteborg region), southern Sweden (expanded Malmö region) and eastern Sweden (expanded Stockholm region); *** Up until 1980, the main material for façades was reported if a mix of diﬀerent façade materials were used. 1980–1993, mixed façades were reported including the two main materials. Buildings 2019, 9, 99 18 of 21

Appendix B. —One- and Two-Dwelling Buildings

Table A2. Available data for one- and two-dwelling buildings, via supplementary data ﬁle.

Available Type of Data Period for Available Data Parameters One-dwelling building; Type of building Two-dwelling building; Linked 1966–1994 building; Terraced building; Other Göteborg region; Malmö region; Area of construction Stockholm region; Sweden 1968–1994 excluding metropolitan regions Storeys 1; 1.5; 2; >2 1970–1987 Autoclaved aerated concrete; Clay Material for superstructure 1966-1987 bricks; Concrete; Wood; Other On site: Prefabricated; Partly Method of production for superstructure 1968–1993 prefabricated Asbestos; Autoclaved aerated concrete; Clay bricks; Concrete; Façade material 1966–1993 * Render; Sandlime bricks; Sheet metal; Wood; Other Expanded polystyrene; Wood insulation/wood wool/wood Insulation in exterior wall 1966–1972 shavings; Autoclaved aerated concrete; Mineral wool Asbestos; Clay tiles; Concrete tiles; Rooﬁng 1966–1993 Roof felt; Sheet metal; Other * Up until 1973, the main material for façades was reported if a mix of diﬀerent façade materials was used. 1973–1993, mixed façades were reported including the two main materials.

References

1. International Energy Agency (IEA). SHC Task 40/ECBCS Annex 52 Towards Net Zero Energy Solar Buildings, Fact Sheet; IEA: Paris, France, 2011; p. 2. Available online: http://task40.iea-shc.org/ (accessed on 7 March 2019). 2. Economidou, M.; Atanasiu, B.; Despert, C.; Maio, J.; Nolte, I.; Rapf, O. Europe’s Buildings under the Microscope—A Country-by-Country Review of the Energy Performance of Buildings. 2011, p. 132. Available online: http://bpie.eu/publication/europes-buildings-under-the-microscope/ (accessed on 7 March 2019). 3. Boverket, Energimyndigheten, Förslag Till Nationell Strategi för Energieffektiviserande Renovering av Byggnader. 2013, p. 244. Available online: https://www.boverket.se/sv/om-boverket/publicerat-av-boverket/ publikationer/2013/forslag-till-nationell-strategi-for-energieffektiviserande-renovering-av-byggnader/ (accessed on 7 March 2019). 4. Berggren, B.; Janson, U.; Wall, M. Nationella Och Internationella Erfarenheter Från Energirenovering Med Stor Energibesparing, Bygg och Teknik. 2011, p. 5. Available online: https://issuu.com/byggteknikforlaget/ docs/2-11/46 (accessed on 7 March 2019). 5. Hastings, R. Lessons from Exemplary Housing Renovations. 2010, p. 33. Available online: http://mojo.iea- shc.org/Data/Sites/1/publications/Lessons_from_Case_Studies.pdf (accessed on 7 March 2019). 6. Risholt, B.; Time, B.; Hestnes, A.G. Sustainability assessment of nearly zero energy renovation of dwellings based on energy, economy and home quality indicators. Energy Build. 2013, 60, 217–224. [CrossRef] 7. Morelli, M.; Rønby, L.; Mikkelsen, S.E.; Minzari, M.G.; Kildemoes, T.; Tommerup, H.M. Energy retrofitting of a typical old Danish multi-family building to a “nearly-zero” energy building based on experiences from a test apartment. Energy Build. 2012, 54, 395–406. [CrossRef] 8. Tommerup, H.; Svendsen, S. Energy savings in Danish residential building stock. Energy Build. 2006, 38, 618–626. [CrossRef] 9. Liu, L.; Moshfegh, B.; Akander, J.; Cehlin, M. Comprehensive investigation on energy retrofits in eleven multi-family buildings in Sweden. Energy Build. 2014, 84, 704–715. [CrossRef] Buildings 2019, 9, 99 19 of 21

10. International Energy Agency (IEA). Advanced Housing Renovation with Solar and Conservation. 2010. Available online: http://task37.iea-shc.org/ (accessed on 7 March 2019). 11. Janson, U.; Berggren, B.; Sundqvist, H. Energieffektivisering vid Renovering av Rekordårens Flerbostadshus. 2008, p. 105. Available online: http://vpp.sbuf.se/Public/Documents/ProjectDocuments/3ce58210-ddae- 4c70-8761-b62ee696a4b3/FinalReport/SBUF%2011936%20Slutrapport%20Energieffektivisering%20vid% 20renovering%20av%20rekordårens%20flerbostadshus.pdf (accessed on 7 March 2019). 12. Nik, V.M.; Mata, E.; Kalagasidis, A.S.; Scartezzini, J.-L. Effective and robust energy retrofitting measures for future climatic conditions—Reduced heating demand of Swedish households. Energy Build. 2016, 121, 176–187. [CrossRef] 13. Österbring, M. Spatial Analysis of Urban Housing Stocks. Licentiate Thesis, Department of Civil and Environmental Engineering, Chalmers University of Technology, Gothenburg, Sweden, 2016. 14. Caputo, P.; Costa, G.; Ferrari, S. A supporting method for defining energy strategies in the building sector at urban scale. Energy Policy 2013, 55, 261–270. [CrossRef] 15. Ascione, F.; de Masi, R.F.; de Rossi, F.; Fistola, R.; Sasso, M.; Vanoli, G.P. Analysis and diagnosis of the energy performance of buildings and districts: Methodology, validation and development of Urban Energy Maps. Cities 2013, 35, 270–283. [CrossRef] 16. Wang, Q.; Holmberg, S. A methodology to assess energy-demand savings and cost effectiveness of retrofitting in existing Swedish residential buildings. Sustain. Cities Soc. 2015, 14, 254–266. [CrossRef] 17. Zimmermann, M. Prefabricated Systems for Low Energy Renovation of Residential Buildings. 2012, p. 24. Available online: http://www.iea-ebc.org/Data/publications/EBC_PSR_Annex50.pdf (accessed on 7 March 2019). 18. Mjörnell, K.; Werner, G. Teknikupphandling: Rationell isolering av klimatskärmen på befintliga flerbostadshus. 2011, p. 52. Available online: http://www.bebostad.se/library/1858/slutrapport-turik- etapp-1.pdf (accessed on 7 March 2019). 19. International Energy Agency (IEA). EBC Annex 50 Prefabricated Systems for Low Energy Renovation of Residential Buildings. 2012. Available online: http://www.iea-ebc.org/projects/project?AnnexID=50 (accessed on 7 March 2019). 20. Mata, É.; Kalagasidis, A.S.; Johnsson, F. Energy usage and technical potential for energy saving measures in the Swedish residential building stock. Energy Policy 2013, 55, 404–414. [CrossRef] 21. Ekström, T.; Bernardo, R.; Blomsterberg, Å. Cost-effective passive house renovation packages for Swedish single-family houses from the 1960s and 1970s. Energy Build. 2018, 161, 89–102. [CrossRef] 22. Niemelä, T.; Kosonen, R.; Jokisalo, J. Energy performance and environmental impact analysis of cost-optimal renovation solutions of large panel apartment buildings in Finland. Sustain. Cities Soc. 2017, 32, 9–30. [CrossRef] 23. Niemelä, T.; Kosonen, R.; Jokisalo, J. Cost-effectiveness of energy performance renovation measures in Finnish brick apartment buildings. Energy Build. 2017, 137, 60–75. [CrossRef] 24. Nardi, I.; Lucchi, E.; de Rubeis, T.; Ambrosini, D. Quantification of heat energy losses through the building envelope: A state-of-the-art analysis with critical and comprehensive review on infrared thermography. Build. Environ. 2018, 146, 190–205. [CrossRef] 25. Sandberg, K.; Orskaug, T.; Andersson, A. Prefabricated wood elements for sustainable renovation of residential building façades. Energy Procedia 2016, 96, 756–767. [CrossRef] 26. Hall, T.; Vidén, S. The Million Homes Programme: A review of the great Swedish planning project. Plan. Perspect. 2005, 20, 301–328. [CrossRef] 27. Wittchen, K.B.; Mortensen, L.; Holøs, S.B.; Björk, N.F.; Vares, S.; Malmqvist, T. Building Typologies in the Nordic Countries; SBi: København, Denmark, 2012; p. 83. 28. Boverket, Bostadspolitiken—Svensk Politik för Boende, Planering och Byggande under 130 år. 2007, p. 133. Available online: https://www.boverket.se/sv/om-boverket/publicerat-av-boverket/publikationer/ 2007/bostadspolitiken/ (accessed on 7 March 2019). 29. Statiscis Sweden. 2018. Available online: http://www.scb.se/ (accessed on 7 March 2019). 30. Statistics Sweden Statistiska Meddelanden. Flerfamiljshus och Småhus med Beslut om Statliga Bostadslån 1967–1994; Statistics Sweden: Örebro, Sweden, 1967–1994. 31. European Parliament, Directive 2010/31/EU of the European Parliament and the Council of 19 May 2010 on the energy perfromance of buildings. Off. J. Eur. Union 2010, 153, 13–15. Buildings 2019, 9, 99 20 of 21

32. Intelligent Energy Europe, TABULA—Typology Approach for Building Stock Energy Assessment 2009–2012. Available online: http://episcope.eu/iee-project/tabula (accessed on 7 March 2019). 33. Spets, K. Byggnadstypologier—Sverige, TABULA 2009–2012. 2012, p. 95. Available online: http: //episcope.eu/fileadmin/tabula/public/docs/brochure/SE_TABULA_TypologyBrochure_Mdh.pdf (accessed on 7 March 2019). 34. Boverket, Teknisk Status i den Svenska Bebyggelsen—Resultat Från Projektet BETSI. 2010, p. 184. Available online: http://www.boverket.se/sv/om-boverket/publicerat-av-boverket/publikationer/2011/teknisk-status- i-den-svenska-bebyggelsen/ (accessed on 7 March 2019). 35. Boverket, Optimala Kostnader för Energieffektivisering—Underlag Enligt Europaparlamentets och Rådets direktiv 2010/31/EU om Byggnaders Energiprestanda. 2013, p. 332. Available online: https://www.boverket.se/ sv/om-boverket/publicerat-av-boverket/publikationer/2013/optimala-kostnader-for-energieffektivisering/ (accessed on 7 March 2019). 36. Corrado, V.; Ballarini, I. Refurbishment trends of the residential building stock: Analysis of a regional pilot case in Italy. Energy Build. 2016, 132, 91–106. [CrossRef] 37. Csoknyai, T.; Hrabovszky-Horváth, S.; Georgiev, Z.; Jovanovic-Popovic, M.; Stankovic, B.; Villatoro, O.; Szendr˝o, G. Building stock characteristics and energy performance of residential buildings in Eastern-European countries. Energy Build. 2016, 132, 39–52. [CrossRef] 38. Dascalaki, E.G.; Balaras, C.A.; Kontoyiannidis, S.; Droutsa, K.G. Modeling energy refurbishment scenarios for the Hellenic residential building stock towards the 2020 & 2030 targets. Energy Build. 2016, 132, 74–90. 39. Diefenbach, N.; Loga, T.; Stein, B. Reaching the climate protection targets for the heat supply of the German residential building stock: How and how fast? Energy Build. 2016, 132, 53–73. [CrossRef] 40. Filippidou, F.; Nieboer, N.; Visscher, H. Energy efficiency measures implemented in the Dutch non-profit housing sector. Energy Build. 2016, 132, 107–116. [CrossRef] 41. Loga, T.; Stein, B.; Diefenbach, N. TABULA building typologies in 20 European countries—Making energy-related features of residential building stocks comparable. Energy Build. 2016, 132, 4–12. [CrossRef] 42. Sandberg, N.H.; Sartori, I.; Heidrich, O.; Dawson, R.; Dascalaki, E.; Dimitriou, S.; Vimm-r, T.; Filippidou, F.; Stegnar, G.; Zavrl, M.Š.; et al. Dynamic building stock modelling: Application to 11 European countries to support the energy efficiency and retrofit ambitions of the EU. Energy Build. 2016, 132, 26–38. [CrossRef] 43. Sandberg, N.H.; Sartori, I.; Vestrum, M.I.; Brattebø, H. Explaining the historical energy use in dwelling stocks with a segmented dynamic model: Case study of Norway 1960–2015. Energy Build. 2016, 132, 141–153. [CrossRef] 44. Sartori, I.; Sandberg, N.H.; Brattebø, H. Dynamic building stock modelling: General algorithm and exemplification for Norway. Energy Build. 2016, 132, 13–25. [CrossRef] 45. Serghides, D.K.; Dimitriou, S.; Katafygiotou, M.C. Towards European targets by monitoring the energy profile of the Cyprus housing stock. Energy Build. 2016, 132, 130–140. [CrossRef] 46. Serrano-Lanzarote, B.; Ortega-Madrigal, L.; García-Prieto-Ruiz, A.; Soto-Francés, L.; Soto-Francés, V.-M. Strategy for the energy renovation of the housing stock in Comunitat Valenciana (Spain). Energy Build. 2016, 132, 117–129. [CrossRef] 47. Brown, N.W.O.; Olsson, S.; Malmqvist, T. Embodied greenhouse gas emissions from refurbishment of residential building stock to achieve a 50% operational energy reduction. Build. Environ. 2014, 79, 46–56. [CrossRef] 48. Brown, N.W.O.; Malmqvist, T.; Bai, W.; Molinari, M. Sustainability assessment of renovation packages for increased energy efficiency for multi-family buildings in Sweden. Build. Environ. 2013, 61, 140–148. [CrossRef] 49. Gontia, P.; Nägeli, C.; Rosado, L.; Kalmykova, Y.; Österbring, M. Material-intensity database of residential buildings: A case-study of Sweden in the international context. Resour. Conserv. Recycl. 2018, 130, 228–239. [CrossRef] 50. Björk, C.; Kallstenius, P.; Reppen, L. Så Byggdes Husen 1880–2000: Arkitektur, Konstruktion och Material i Våra Flerbostadshus; Svensk Byggtjänst: Stockholm, Sweden, 2013. 51. Björk, C.; Nordling, L.; Reppen, L. Så Byggdes Villan: Svensk Villaarkitektur Från 1890 Till 2010; Formas: Stockholm, Sweden, 2009. Buildings 2019, 9, 99 21 of 21

52. Nägeli, C.; Camarasa, C.; Jakob, M.; Catenazzi, G.; Ostermeyer, Y. Synthetic building stocks as a way to assess the energy demand and greenhouse gas emissions of national building stocks. Energy Build. 2018, 173, 443–460. [CrossRef] 53. Statistics Sweden. Number of Completed Dwellings. 2017. Available online: http://www.scb.se/en/ finding-statistics/statistics-by-subject-area/housing-construction-and-building/housing-construction-and- conversion/new-construction-of-residential-buildings/pong/tables-and-graphs/number-of-completed- dwellings/ (accessed on 7 March 2019). 54. Microsoft, Excel 2013, Redmond, Washington, USA. 2013. Available online: www.microsoft.com (accessed on 17 April 2019). 55. Statistics Sweden. Antal Lägenheter Efter Region, Hustyp och Byggnadsperiod, År 2013–2016. 2017. Available online: http://www.statistikdatabasen.scb.se/pxweb/sv/ssd/START__BO__BO0104/BO0104T02/ ?rxid=5fd6c9a0-a705-41c0-86a1-e69915f0e8d1 (accessed on 7 March 2019).

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).