August 2023

DOI: http://dx.doi.org/10.7488/era/3776

Exectutive Summary

Aims

The Scottish Government aims to reduce car kilometres by 20% by 2030 from a 2019 baseline. Parking policy has been acknowledged as having the potential to play an important role in supporting this reduction target. In response to this, ClimateXChange commissioned an evidence review of the impact of parking policies on car use. This report contains the results of that review.

This research has gathered evidence on the effectiveness of different parking management interventions in reducing car use. Its purpose is to inform the development of parking policies which support the joint commitment by Scottish Government and the Convention of Scottish Local Authorities (COSLA) to reduce car use by 20% by 2030.

Findings

The analysis of the literature led to the following key findings.

Impact on car use

Five parking intervention types were identified as having an impact on one or more of the following elements: car kilometre reduction, modal split and car ownership.

Parking standards, off-site or non-adjacent provision of residential parking, and low-car and car-free housing is linked to positive impacts on car kilometres travelled, car ownership and modal split. Parking availability and location can influence car use; car-free developments have been found to have car use levels at less than half of city-wide averages, while parking located at 50m or more from dwellings was associated with 25% fewer car trips.

Parking pricing can contribute to car kilometre reductions and modal shift, particularly when combined with capacity reduction measures. Car parking costs have also been found to significantly influence car ownership levels.

Workplace parking levies (WPL) were found to have a positive impact on mode share. Public transport improvements implemented in tandem with WPL schemes were identified as significant contributing factors in encouraging modal shift.

Park and ride was found to generally increase vehicle kilometres travelled (VKT) when located close to destinations. However, when located close to journey origins, it was associated with reductions in vehicle kilometres of the order of 1.5km per park and ride user.

There is evidence that parking capacity reductions at city or neighbourhood level have an impact on car kilometres travelled and modal split. In the context of workplaces, there is strong evidence that the provision of parking is linked to an increase in car mode share.

Equity and equality issues

There is extremely limited evidence regarding whether the reductions in car km and changes in modal split achieved by parking interventions are shared across social groups.

Evidence is not available to draw conclusions on how different intervention types may align to inequality reduction goals focussing on island communities and remote rural and rural areas, as per the Scottish Government’s Urban Rural Classifications (Scottish Government, 2020).

Alignment to national policies and strategies

Four of the five intervention types for which strong evidence was found are associated with positive impact on car kilometre reduction, modal shift or car ownership. The exception to this is park and ride. Therefore, these interventions broadly align to climate change mitigation goals in the Scotland’s National Transport Strategy (NTS2) and National Planning Framework 4 (NPF4).

Among the four, there are also broad alignments found between individual intervention types and some other aspects of the NTS2 and NPF4. These aspects include encouraging active travel, reducing levels of car dominance, reducing congestion and air pollution, and supporting sustainable investments.

There is a lack of evidence in the reviewed literature on the application of a place-based approach in parking interventions. However, some of the intervention types may be better positioned to support the use of a place-based approach, given their potential to help reduce the space dedicated to car parking in a particular locality.

Recommendations

In considering the merit of implementing the intervention types with the greatest impact on car use we recommend:

• Testing temporary changes to parking spaces in order to gather further data on the different measures
• Promoting parking management in the context of workplace travel plans
• Considering the significance of site in decisions around park and ride

Contents

Executive summary

Glossary and acronyms

Introduction

Aims and scope

Policy context

Findings

Conclusions and recommendations

References

Appendix A. Methodology

Appendix B. Summary of papers providing evidence of impact on car use

Appendix C. Bibliography

Glossary and acronyms

Introduction

The Scottish Government aims to reduce car kilometres by 20% by 2030 from a 2019 baseline. Parking policy has been recognised as having the potential to play an important role in supporting this reduction target. In response to this, ClimateXChange commissioned an evidence review of the impact of parking policies on car use.

In addition to providing evidence on the impacts of different types of parking intervention on vehicle kilometres travelled (VKT), gathered evidence on factors influencing the implementation of parking measures was gathered to inform debates around the public and political acceptability of different intervention types.

Aims and scope

Aims

The main aim with this research is to gather evidence on the effectiveness of different parking management interventions in reducing car use. Its purpose is to inform the development of parking policies which support the joint commitment by Scottish Government and Convention of Scottish Local Authorities (COSLA) to reduce car use by 20% by 2030.

More specifically, the research set out to:

• review available international evidence on the impact of different parking policies on reducing car kilometres;
• identify success factors and describe successful policies and good practices in order to draw lessons in policy design and implementation applicable to Scotland; and
• identify unintended consequences of the reviewed policies, including for protected groups and local businesses, and barriers to intervention.

These objectives led to the formulation of the following research questions, which informed the nature and extent of our research:

1. What is the impact of the intervention with regard to its contribution to reducing car kilometres and how does the actual impact compare to any ex-ante (modelled) prediction?
2. To what extent do the interventions align with national and regional transport strategies and the National Planning Framework 4, and support a place-based approach?
3. What success factors can be identified related to the interventions?
4. What lessons can be drawn for Scotland across policy design and implementation?
5. What unintended consequences resulted from the interventions (e.g., with regard to protected groups and local businesses)? How do these consequences compare to the predictions made prior to the implementation of the interventions. If there was a difference, why did this occur?
6. (How) did the interventions contribute to an equitable reduction in car kilometres?
7. What barriers to interventions were identified (e.g., related to cost efficiency, sustainability, legislation, public acceptability)? Were any recommendations made and/or experience gained in how to overcome them?

Scope

The review targeted academic and grey literature produced within a 15-year period prior to the research, focusing on interventions in jurisdictions similar to Scotland. For this reason, smaller northwest European countries, particularly the Nordic countries, were of key interest. The Nordic countries are particularly of interest because of their generally low population densities, extensive rural areas and large numbers of small towns. In many of these areas, population decline is a significant concern. At the same time, they offer excellent transport and planning data. The analysis was to include both transport and planning policy and cover site-level and design-based interventions, including parking standards, as well as the location and layout within developments.

Prior to beginning the research, an initial list of parking intervention types was developed, to inform and guide the search for relevant literature. It included the following terms:

• Parking time limits, permits and pricing, both on- and off-street
• Parking pricing in relation to vehicle characteristics
• Parking pricing in relation to household or user characteristics
• Levies/taxes on off-street parking
• Parking standards, off-site or non-adjacent provision of residential parking, low-car and car-free housing
• Park and ride (P&R)
• Parking capacity reductions at city or neighbourhood level and use of resulting space
• Residential parking in historic areas that are pedestrianised
• Shared-use parking
• Effective and fair parking enforcement
• Parking for EVs
• Mobility hubs
• 20-minute neighbourhood and 15-minute cities
• Parking and road space transformation/reallocation.

Methodology

The main phases of the research were a systematic literature search, followed by a literature review and the analysis of collected data. A systematic literature search on Google Scholar resulted in over 4,700 results. The results were sifted by title and then by abstract to identify relevant papers for review.

Searches using the regular Google search engine were also conducted in in Catalan, Dutch, English, German, Norwegian, Spanish and Swedish. This was done to identify papers, particularly in the grey literature, which may have been missed in the systematic search, in addition to non-English language papers which may not have appeared when searched for in English. The languages used in the search reflected the linguistic abilities of the research team and considered the countries of interest established during the inception phase. In addition, contacts working in relevant academic fields were asked to recommend papers.

Ultimately, 139 papers were selected for a detailed review. A full explanation of the methodology is provided in Appendix A, while a full list of papers included in the review is provided in Appendix C.

Policy context

Research Question 2 considered the extent to which different intervention types align to transport strategies and the NPF4, and to what extent they support place-based approaches. This section provides a brief overview of relevant strategies, policies, and the place-based approach to provide context for the findings on this question.

A route map to achieve a 20 per cent reduction in car kilometres by 2030

The route map for a 20% reduction in car kilometres by 2030 (Transport Scotland, 2022) responds to the Scottish Government’s Climate Change Plan commitment to reduce car kilometres by 20 percent by 2030, which in turn forms part of the country’s statutory obligations for greenhouse gas emissions reductions by 2045. The route map, which was developed in partnership with COSLA, builds on the vision set out in NTS2 (Transport Scotland, 2020) and aims to help reduce overreliance on car use through four key behaviours: making use of sustainable online options to reduce the need to travel; choosing local destinations to reduce the distance travelled; switching to walking, wheeling, cycling or public transport where possible; and combining a trip or share a journey to reduce the number of individual car trips made.

In describing interventions to support modal shift, the route map highlights that the Scottish Government will provide support to local authorities to ensure that, in their approach to parking, local transport strategies show how parking measures will contribute to meeting emissions reductions targets while also considering their impact on different travellers, including pedestrians, cyclists, public transport users and disabled car users.

The route map also highlights the development of workplace parking levy (WPL) regulations and guidance and the discretionary powers resulting from the Transport (Scotland) Act 2019, which enable local authorities to incorporate WPLs in their local transport strategies to disincentivise private car use. Funds generated from WPL schemes will have to be directed towards initiatives to help the travelling public, including public and sustainable transport provision and infrastructure.

The route map outlines the role of parking in interventions that can help people live well locally. It notes that car dependency has led to environments where the movement and parking of cars are prioritised, and that this encourages car use and discourages the use of other modes. It calls for a reduction in car dominance in local places and emphasises the importance of being able to access opportunities, including those for active travel, locally.

2020 National Transport Strategy

NTS2 (Transport Scotland, 2020) is structured around four main priorities: reducing inequalities, taking climate action, helping deliver inclusive economic growth, and improving health and wellbeing. The strategy’s Sustainable Travel Hierarchy provides an overarching framework for its policies and informs the Scottish Government’s decision-making around transport. The hierarchy prioritises walking and wheeling, followed by cycling. Public transport follows, then taxis and shared transport, while private car use is a found at the bottom. The Sustainable Investment Hierarchy, meanwhile, prioritises reducing the need to travel unsustainably. Following this are maintaining and safely operating existing assets, and capitalising on existing capacity. Targeted infrastructure improvements should only be carried out once the other steps have been taken. The strategy refers to parking explicitly in reference to the above-mentioned Transport (Scotland) Act 2019, which will enable local authorities to introduce workplace parking levies with the aim of supporting efforts to reduce private car use.

Two of NTS2’s four priorities align with the focuses of the specific research questions. The priority ‘Taking climate action’ identifies the outcomes of helping deliver net zero, adapting to the effects of climate change and promoting greener, cleaner choices, and is closely related to the core focus of this research; measures which contribute to reductions in car kilometres, as covered by Research Question 1^[2]. A second priority, ‘reducing inequalities’, aims to support ease of use, affordability and fair access to transport, and therefore is closely related to Research Question 6^[3],[4]. The remaining priorities are ‘helping to deliver inclusive economic growth’ and ‘improving health and wellbeing’.

No weighting is applied to any of NTS2’s four priorities, giving them equal importance. The priorities are interdependent and cross cutting in nature; for example, improving health and wellbeing through promoting active travel contributes to taking climate action, while reducing inequalities is linked to achieving inclusive economic growth. Individual interventions may contribute to multiple priorities. For example, in addition to reductions in VKT, and therefore contributions to climate change goals, an intervention may have other positive outcomes, such as encouraging the use of active travel options, or contributing to congestion reductions in urban areas. These outcomes can in turn benefit people’s health and wellbeing.

The strategy also underlines that transport-related efforts to transition towards a net zero economy should be carried out in accordance with the Scottish Government’s Just Transition principles^[5].

National Planning Framework 4

NPF4 (Scottish Government, 2023) is structured around three overarching goals – sustainable places, liveable places and productive places – and informed by six spatial principles: just transitions, conserving and recycling assets, local living, compact urban growth, rebalanced development and rural revitalisation. Organised under the framework’s goals are 33 policies.

Policy 13 on sustainable transport aims to ‘encourage, promote and facilitate developments that prioritise walking, wheeling, cycling and public transport for everyday travel and reduce the need to travel unsustainably’ (Scottish Government, 2023, p.57). Its policy outcomes are:

• Investment in transport infrastructure that supports connectivity and reflects place-based approaches and local living.
• More, better, safer and more inclusive active and sustainable travel opportunities.
• Developments are in locations which support sustainable travel.

In reference to local development plans (LDPs), Policy 13 states that they should promote a place-based approach that considers how to reduce car dominance, including minimising the space dedicated to car parking. It highlights the importance of considering elements such as local living and 20-minute neighbourhoods, car ownership levels and accessibility for users of all abilities.

With specific regard to levels of car parking provision, paragraph e) states that ‘proposals which are ambitious in terms of low/no car parking will be supported, particularly in urban locations that are well served by sustainable transport modes and where they do not create barriers to access by disabled people’ (Scottish Government, 2023, p.58).

When considering the alignment of intervention types to NPF4, it should be noted that some of the research questions have a clear focus on key elements of the framework. Of its six overarching goals, ‘Sustainable places’ covers policies on climate change mitigation and adaptations, and sustainable transport. This goal is also linked to the spatial principle ‘Just Transition’, which is concerned with ensuring the transition to net zero is fair and inclusive. These elements are closely linked to Research Questions 1 and 6, respectively.

NPF4 highlights instances where certain policies are relevant to cross-cutting policy outcomes. In the case of Policy 13, links are drawn to outcomes falling under all three goals: sustainable places, liveable places and productive places. Policy 13 clearly contributes to a transition towards more sustainable, lower emissions travel including active travel and public transport, as aimed at under Sustainable Places. Under Liveable Places, it is indicated that Policy 13 is linked to achieving ‘homes that meet our diverse needs’. Specifically, the policy makes clear that the views of disabled people must be sought when seeking to reduce reliance on the car, including when this is done by managing parking provision. Finally, under productive places, Policy 13 is linked to ‘rural revitalisation’ in that it ensures, through the assessment of transport impacts of developments, that an area’s needs and characteristics are considered, while the policy also contributes to ‘lifelong health and wellbeing’ by encouraging active travel.

20 minute neighbourhoods, in addition to being referred to in Policy 13, are the specific focus of NPF4’s Policy 15, which states that ‘development proposals will include, where relevant, 20 minute neighbourhoods’ (Scottish Government, 2023, p.61).

NPF4 encourages LDPs to promote a place-based approach. It states that LDPs should promote a place-based approach that considers how to reduce car dominance, including minimising the space dedicated to car parking. It highlights the importance of considering elements such as local living and 20-minute neighbourhoods, car ownership levels and accessibility for users of all abilities.

Place-based approach

Place-based approaches involve understanding the issues, interconnections and relationships in a place and coordinating action and investment to improve the quality of life for that community. Importantly, place-based approaches are not about understanding an issue or policy context in a particular geographical area.  Rather, they aim to understand the place and then plan policy responses that are coordinated with co-benefits across a range of outcomes. They are collaborative processes that take a long-term approach. Place-based working requires the formation of partnerships across the public, private and third sectors and with communities directly (ourplace.scot, 2023).

Summary

Overall, the Scottish policy context for transport is one in which there is a place for using parking as a means of reducing demand for travel by car and for planning parking into new developments. In this way, the dominance of parked vehicles may be reduced, potentially enabling road space reallocation to support and encourage active travel and public transport.

Findings

In this section of the report, we summarise the evidence that responds to each of our research questions but, rather than responding question by question, we have grouped them into three overarching categories, reflecting both the availability of evidence for all research questions, and the fact that certain research questions are closely related. In addition, we disaggregate the results by intervention type to present the evidence of the effectiveness of each intervention, but also the barriers to their implementation and lessons that the literature provides about how to implement these measures.

Impact on car use

In this section we consider the quantity and quality of the evidence on impact of different parking-related interventions on reducing car kilometres. It is important to note that, as well as papers reporting changes in car km, we reviewed an equal number of reporting changes in modal split (i.e. the proportions of trips by different modes). These were included because a change in modal split is also likely to lead to a change in car km, although to what degree is unknown due to a lack of data on trip distances. There are also papers that report an association between parking and car ownership. This is significant as if people are less likely to own a car, they are much less likely to generate km travelled by car. It should be noted, however, that people who do not own a car may be more reliant on lift-giving by others and that this may lead to additional car trips being undertaken by lift providers.

In the following sub-sections, the findings from some papers that are classified as being of high-quality, relevant to each topic, are described in more detail. These papers have been selected because of the robustness of their methodology (for example, they may include a control group or use statistical methods^[6] to control/explain the impact of other variables that influence car travel, such as socio-economic factors); and because they show strong evidence that the parking interventions that they study have an association with reduced car use or ownership. A full explanation of how strength of evidence and robustness of methodology was assessed is provided in Appendix A, while tables containing headline figures on strength of evidence and robustness of methodology are found in Appendix B.

Car kilometres

This section reviews the evidence that parking interventions have an impact on car kilometres travelled.

A total of 37 of the reviewed papers were found to provide evidence of parking interventions having an impact on car kilometres travelled. Most show a moderate or high impact, though in seven cases little impact was observed from the parking intervention. With regard to which of the intervention types identified in Section 4.2 were included in these papers, parking pricing featured most highly (11 papers), followed by park and ride (eight), parking levies (four), parking standards, off-site or non-adjacent provision of residential parking, low-car and car-free housing (three), parking capacity reductions (two) and effective and fair enforcement (one)^[7]. The number of papers scored at each level for robustness of methodology and strength of evidence is shown in Appendix B.

Some seven papers cited evidence of increased (six papers) or no change (one) in car kilometres as a result of parking policy and all were on P&R. All of the papers on other intervention types found that parking policy reduced vehicle km.

Average changes in car kilometres were found to be difficult to compare, as many studies report results in different ways. For example, city centre traffic levels, city traffic levels, traffic levels on main roads, nationwide VKT, nationwide VKT from commute traffic and search traffic in different areas are all used as measures. The comparability of the data does not affect the strength of evidence observed in individual studies, but it does mean that the number of studies showing exactly the same type of impact on car km is very limited.

The evidence is clear that parking pricing reduces car km, but the studies that quantified the reduction gave a wide range to the associated reductions, between 0.3% and 16%. This wide range is due to the impact being highly sensitive to the proportion of drivers who are affected by the charges or (for example, those who have a space provided by their employer are not affected by on-street charges); and because a few studies also reported on the impact of pricing measures implemented alongside overall parking capacity reductions (e.g. Pfaffenbichler and Schopf, 2011).

Modelled results from parking pricing in two studies indicate a 0.3 to 3% reduction in VKT at entire city level or nationwide level from a fee of, typically, €1 to €3 a day^[8]. (Palmer and Ferris, 2010; Ecorys, 2022). A similar level of reduction is also reported in empirical studies such as Ostermeijer et al. (2022). On the other hand, two modelling studies considered the impact of much higher prices, those that the market would charge for providing off-street parking. Here, the reduction in car km was much larger, at 6% (Netherlands, nationwide, commute traffic only) and 16% respectively (all travel, nationwide, Switzerland) (CE Delft, 2018; Swiss Federal Office for Spatial Planning, 2021).

Real time parking occupancy monitoring and demand-responsive pricing at a city-centre scale can reduce parking search km travelled in the affected area by 20% or more. Thorwaldson et al. (2021) report the results of an experimental project in downtown San Francisco (SFPark), whereby off-street and on-street parking occupancy was measured (via sensors) and monitored, and prices adjusted by +/- USD 0.50 depending on occupancy, all in real time. In this way, less busy parking spaces became more attractive, while price increases in the busiest locations reduced demand and hence also search traffic kilometres travelled; and search traffic was directed to off-street car parks. This was communicated to drivers online and via an app (Zimmerman, Klein and Schroeder, 2014)

Empirical studies of parking pricing and capacity reduction on-street in Paris (early 2000s) and of low-traffic neighbourhoods (LTNs) in London in 2010s show VKT area-wide (all in Paris; just residents for LTNs) reduced by 6.4% to 16% (Pfaffenbichler and Schopf, 2011; Kodransky and Hermann, 2011) In Paris, the policy in this period both reduced the number of on-street parking spaces available from 172,800 in 2003 to 158,700 in 2007; and, in addition, made more of these spaces subject to a charge, so that whilst in 2003, of the total on-street parking available, 49,600 spaces had no charge or other restriction, this number had been reduced to only 2,700 by 2007. The change in parking price over this time is not reported (Pfaffenbichler and Schopf, 2011).

On the price elasticity, papers dealing with this topic were mostly in agreement that short term parking price elasticity is around –0.3, meaning that a 10% rise in price will lead to a 3% fall in demand, although one paper found elasticities greater than 1, indicating that demand falls more in percentage terms than the percentage rise in price (Milosavljević and Simićević, 2016). If anything, modelled impacts tend to be smaller than empirically observed ones – this is a finding from Milosavljević, and Simićević’s work, but also from comparing other papers reviewed for this study.

Park and ride was found to generally increase VKT when the P&R sites are located close to destinations, for example on the edge of a city. Hanssen et al. (2016) conducted a systematic analysis of evaluations of P&R sites and found that P&R is associated with reduced VKT where it is located close to traveller origins, while the opposite is true where it is located close to destinations. Similar findings came from an analysis of data for over 180 P&R sites presented by Zijlstra et al. (2015). These latter authors reported that, within their sample, P&R sites close to home origins intercept 21 cars (whose destination is a major city centre) per 100 parking spaces, while private car kms reduce (by 1 to 4 km per user, with an average of 1.5km) and public transport increases. P&R on urban fringes close to a final destination city centre intercept about 47 cars per 100 parking spaces provided, and car travel increases by 1 to 4 km per user. The reason for the higher number of cars intercepted at the destination P&R is because the site is closer to the destination, but the reason for the increase in car km is because a quite high proportion of these car trips would have used public transport for the whole trip in the absence of the P&R site.

The following is a summary of key findings from this section:

• Parking pricing of the order of €1 to €3 per day can reduce car km by between 0.3% and 3% at entire city or nationwide scale, while other studies identified reductions of between 6% and 16%. The reduction level is highly sensitive to the proportion of drivers who are affected by the charge and whether it is implemented alongside overall parking capacity reductions.
• Real time parking occupancy monitoring and demand-responsive pricing at a city-centre scale can reduce parking search kilometres travelled in the affected area by 20% or more.
• Park and ride was found to generally increase VKT when located close to destinations. However, when located close to journey origins, it was associated with reductions in vehicle kilometres.

Modal split

This section reviews the evidence that parking interventions have an impact on mode choice for trips. This is different from vehicle km in that the evidence here does not include measurement of the change in distance travelled as a result of the parking intervention.

Forty-seven papers were found to contain evidence of parking interventions’ impact on modal split. Of these, 17 covered parking standards, off-site or non-adjacent provision of residential parking, or low-car and car-free housing, while 14 included parking time limits, permits and pricing and on- and off-street, and 10 included parking levies. Other interventions for which an impact on modal split was found were P&R (six papers), parking capacity reductions at city or neighbourhood level (three), parking and road space transformation/reallocation (two), parking pricing in relation to household or user characteristics (one) and shared use parking (one)^[10]. For a breakdown of the number of papers ratings at different levels for robustness of methodology and strength of evidence, see Appendix B.

There is strong evidence based on papers with robust methodology that modal split is influenced by parking pricing and parking availability. These factors can reduce car modal split by 25-50% compared to a baseline where there is ample free parking. There is also strong evidence that travel plans have a greater impact on modal split when they include some form of parking management.

National travel survey data from Norway (Christiansen, Engebretsen and Hanssen, 2015; Christiansen, Hanssen, and Skollerud, 2015) shows that when employers provide parking, car mode share is at 58%, while when parking is not provided, it sits at 20% (Christiansen, Engebretsen and Hanssen, 2015). Since this work draws on a national survey, it provides some of the most robust evidence found on modal split; the studies have a large sample size due to being based on a large-scale household travel survey. Furthermore, since the survey also enquires about the location and cost of respondents’ parking place at work and at home, it is possible to draw inferences at the level of different cities about the relationship between parking availability (and whether it is charged for) and travel behaviour, and also car ownership (see Section 6.1.3)^[11]. In the Geneva region, a significant difference was also observed between car use by workers whose employer provides parking (39% mode share) and those that do not (24%) (Swiss Federal Office for Spatial Planning, 2021).

There is also evidence that the availability of parking spaces at a city scale influences mode split. McCahill et al. (2016) found that an increase in supply of between 0.1 and 0.4 parking spaces per person across US cities was associated with a 30% increase in car mode share.

Cars were also found to have a lower modal share in car-free and low-car residential areas. Car-free developments in Germany and the Netherlands were found to have car use levels at less than half the city average in their respective settlements (Foletta and Field, 2011; Melia, 2014), while in Norway, a study of four major cities found that people whose residential parking spot is 50m or more from their house make 25% fewer car trips than those whose parking spot is immediately outside their home (Christiansen, Hanssen and Skollerud, 2015).

Cairns et al. (2010) reviewed in detail the content and impact of travel plans at 21 large workplaces (13 of them private sector) in Britain in the early 2000s. The difference in impact of the travel plan depending on whether or not some form of parking management in place was stark. At organisations with parking management in place, the travel plans achieved an average reduction in drive-alone commuting of 24%, whilst those without parking management achieved an average reduction of 10%, from a much higher original level of staff driving alone. As the authors state “organisations that had addressed parking in some way had achieved more than double the reduction in car use of those that had not, and had car driver levels which were, on average, 25% lower” (Cairns et al., 2010, p481). Some of the success factors underpinning the measures in the travel plan are discussed in Section 6.2.

Only two of the organisations without parking management achieved reductions of more than 10% in drive-alone commuting. Perhaps unsurprisingly, the organisations which did introduce parking management were those where there were many more staff than parking spaces available: parking management is a driver of travel plan success, but parking management becomes necessary where there is a shortage of parking. The paper did not control for other variables, such as public transport accessibility, which could also affect the impact of the travel plans.

Workplace parking levies (WPL) were also found to have an impact on mode share. In Nottingham the introduction of the WPL led to 8.6% of car commuters switching to other modes (Dale, et al., 2019). According to Richardson (no date), the car mode share for trips to Perth (Australia) city centre dropped from 50% in the mid-90s to 35% in 2010, in part due to the introduction of a levy on off-street parking in the city which led to an increase in price and reduction in supply, but also due to improved public transport and new high density inner-urban residential development within walking distance of the city centre.

The following is a summary of key findings from this section:

• There is strong evidence that modal split is influenced by parking pricing and parking availability. These factors can reduce car modal split by 25-50% compared to a baseline where there is ample free parking.
• There is also strong evidence that travel plans have a greater impact on modal split when they include some form of parking management.

Car ownership

There is strong evidence of the effect of parking price, location and availability on car ownership. A total of 17 of the reviewed papers provided evidence on the impact of parking interventions on car ownership, with 13 of these including findings on parking standards, off-site or non-adjacent provision of residential parking, low-car and car-free housing; and seven providing findings on parking time limits, permits and on- and off-street pricing^[12].

Comparing suburbs of the four largest cities in the Netherlands with their city centres, car parking costs (including search time, related to availability) accounted for 30% of the variation in car ownership (Ostermeijer et al., 2019). An increase in on-street residential permit prices in the centre of Amsterdam from the current €500 to €3,600^[13] a year would cut car ownership by 24%, based on a regression analysis of existing car ownership levels in different districts of the city, which finds that a €100 increase in annual parking permit costs is currently associated with a 1.7% decline in car ownership. Based on this observed relationship between permit price and car ownership, the authors use modelling to predict the impact of a much larger, hypothetical price increase.

Also in central Amsterdam, the price of a resident’s permit rose from about €180 in 2001 to €530 in 2018, and available spaces per permit fell by 20%. Car ownership fell by 16% in Amsterdam during this period but rose by 17% across the Netherlands as a whole inline with general economic growth (Strategy Development Partners and Martens, 2019). Therefore, it is reasonable to assume that residents’ permit prices and availability may have played a role in influencing car ownership in Amsterdam. However, the study did not control for other variables such as changes in socio-economic factors including, for example, an increase in single-person households in the city.

Analysis of Norwegian travel survey data for its four largest cities found that only 19% of households with a parking space on their own property did not own a car. By comparison, among households who only have on-street parking, 53% did not own a car. This effect is independent of other socio-economic factors. For example, regression analysis of the associated data found that having reserved car parking had twice the impact on the probability of owning a car compared to the effect of being a family with small children (Christiansen, Hanssen, and Skollerud, 2015).

Even greater differences in ownership were found in the case of the Freiburg Vauban car-reduced development, which has one third of the average level of car ownership of the Federal German State in which it is located (Kirschner and Lanzendorf, 2020). This is without, however, controlling for other factors, such as demographics and public transport accessibility.

Not all the reviewed literature found a strong relationship between parking availability and car ownership. A study in London (Leibling, 2014) found very little relationship between lower parking availability (both on- and off-street) and car ownership, but also states that the methodology used to draw this conclusion was not particularly robust and that the case for or against a relationship between the two remains to be proven. Leibling makes considerable reference to outer London and its lower density, lower public transport accessibility and higher car dependence than inner and central London. Outer London is not dissimilar in this way to other UK cities outside London, whilst inner and central London are very different.

The following is a summary of key findings from this section:

• car parking costs have been found to account for up to 30% of the variation in car ownership; and
• access to on-site, as opposed to on-street, parking has been linked in some studies to significantly higher levels of car ownership.

The influence of parking compared to other factors that affect car use and ownership

There is obviously a relationship between parking availability, cost and accessibility by other modes: locations that tend to have limited and/or priced parking are often located in places that have high public transport, walking and cycling accessibility. Therefore, these factors must be controlled for to isolate the effect of priced or restrictive parking since it is to be expected that locations with very good accessibility by other modes would have a higher share by these modes even with unlimited parking. Controlling for other influences on mode choice and/or car ownership was, however, done in four of the reviewed papers (Ostermeijer et al., 2019; Thorwaldson et al., 2021; Christiansen, Engebretsen and Hanssen, 2015; Christiansen, Hanssen, and Skollerud, 2015). That said, these papers still found a statistically significant impact from pricing or restricting parking.

Equitable reduction in car kilometres

In this section, we consider the quantity and quality of the evidence on the extent to which different intervention types contribute to increasing equality and achieving an equitable reduction in car kilometres.

There is extremely limited evidence in the reviewed literature regarding whether the reductions in car kilometres and changes in modal split achieved by parking interventions are shared across social groups. In total only five papers discussed the issue. One, Gonzalez et al., 2022 (on parking regulations in Madrid) found evidence that change in travel mode in response to parking charging was greater amongst drivers on lower incomes, whilst wealthier drivers simply paid the charge and maintained their existing travel habits.

Other papers discussed other equity issues of parking:

• Impact on the use of streetspace, especially for older people and families with small children (Kirschner, 2021).
• Inequitable distribution of streetspace – for example, in Berlin, 10 times as much public space is devoted to parking as to children’s playgrounds (Agora Verkehrswende, 2018).
• The cost of parking in residential developments bundled with the cost of housing, so that car-free households (which tend to be poorer) also pay parking costs (Marsden, 2014).

A report by Parking Brussels (2020), devotes an entire chapter to the issue of parking management and equity. It starts from the assumption of Gonzalez et al. (2022) but then presents ways in which equity in parking could be enhanced and finds evidence of places that have done so, stating:

“Measures in favour of disadvantaged populations can be taken to address the limitations and challenges they face in terms of mobility and parking.

• To address the fact that the affordability of housing for low-income households can be made more difficult by minimum requirements for residential parking, housing and parking can be disconnected from each other (unbundling), resulting in more affordable homes without a parking space.
• Social rates can be applied for parking, based on the users’ income. The examples mentioned in this context concern parking tickets for residents (we do not have any cases of socially priced short-term parking).
• In Parking Benefit Districts in the United States, parking revenues are used to support public services and urban planning, creating a redistributive effect. That offers the opportunity to tackle two problems at the same time: the pressure on parking availability and the need for financial resources for the community.
• Because a P&R enables people living on the outskirts of the city to reach the city centre (and so their work) more easily while space for cars in the city centre remains limited (or at least more expensive), such a facility can play a part in redistributive territorial justice.
• The characteristics of neighbourhoods with social housing (particularly the highly urban appearance of large complexes) require modifications to deal with problems such as, for example, the under-occupancy of underground parking garages” (Parking Brussels, 2020, p24).

In addition, parking cashout at work (paying all employees a monthly payment equivalent to the cost of providing parking at the workplace, rather than providing parking automatically) is more equitable than providing parking free to all employees (but not providing anything for those who choose not to drive to work) and has been found to lead to significant reductions in VMT (Thorwaldson et al, 2021).

Overall, there is a lack of research on parking in relation to issues of equity and equality. We can conclude that there is limited evidence (two papers) that parking pricing will affect lower income drivers more than higher income drivers, though other papers did discuss equity issues related to parking more broadly, as explained above.

Impactful interventions

Here we present a list of interventions showing the relationship between each intervention and the research questions, including their impact on travel and car ownership, but also lessons for and barriers to implementation. These interventions have been selected on the basis of the number of papers that show evidence that they influence car kilometres, modal split and/or car ownership. They are listed in table 1, which summarises the impact they were found to have on car ownership, km’s and modal split.

Table 1. Summary of intervention types and their impact

i Six out of eight of the papers providing evidence of a link between P&R and car use found that P&R was associated with increases in VKT when the site is located close to the journey destination. One further paper found no impact on VKT. Only when P&R sites were located closer to journey origins did the literature confirm evidence of reductions in VKT.

The majority of papers considered evidence that parking standards, off-site or non-adjacent provision of residential parking, low-car and car-free housing have an association with reduced car use or ownership, so this intervention is dealt with first.

Parking standards, off-site or non-adjacent provision of residential parking, low-car and car-free housing

There was strong evidence of the impact of these interventions on both car use (modal split) and car ownership. As noted earlier, these factors can reduce car modal split by 25-50% compared to a baseline where there is ample free parking; and the probability of owning a car also reduced by a similar order of magnitude for residents of car-reduced developments, or for those whose car parking space is on-street or 50m or more from their home, compared to those who have a car parking space directly at home. The most comprehensive study we reviewed on this topic was Foletta and Field (2011), which looks in detail at eight car-reduced or car-free developments in the UK (Greenwich Millennium Village), Netherlands, Germany and Sweden, providing typically 0.5 parking spaces per residential unit (on a range of zero to 1.1), almost always provided in separate parking structures at a distance of 100m to 400m from dwellings. A few developments ban vehicle access completely, whereas others have some roadways where access is allowed for loading and unloading, and a separate network of fully accessible pedestrian and cycle paths. Where residents are disabled, parking spaces may be reserved for them, but these spaces may still not be directly outside their home.

Studies of the Norwegian context (Christiansen, Engebretsen and Hanssen, 2015; Christiansen, Hanssen, and Skollerud, 2015) do not look at specific developments, but the parking conditions they report—where parking is not adjacent to dwellings—are delivered through the design of developments (particularly of housing cooperatives, which are widespread in Norway, Sweden and Denmark) in a similar way to those reported by Foletta and Field (2011; vehicle access to dwellings is permitted, but not parking directly outside. Parking standards in Scandinavian new-build dwellings are of the order of 1 to 1.5 spaces per unit, but the studies also include data from people who live in older dwellings (for example, flats built before 1945) with no or more limited off-street parking. Data pertaining to these older developments can still be considered relevant, as they are similar in parking provision terms to newer developments with little or no off-street parking.

With regard to destinations, Cairns et al (2010) in their study of UK travel plans note that the most effective workplace travel plans were those at employers that had parking management in place; but these were also those that had limited amounts of off-street parking available—32 spaces per 100 FTE staff compared to around 80 spaces at the employers without parking management.

Based on their review of eight sites around Europe, Foletta and Field (2011) came up with a number of key factors that can help the success of car-reduced development. These points were echoed in the other literature reviewed:

• Developments should be close to town and city centres, and well-connected to them by high quality public transport.
• Within the developments, there should be safe, direct, accessible and comfortable cycling, wheeling and walking routes, and green space. If these are shared with vehicles, they should be heavily traffic-calmed and clearly marked as streets with pedestrian/bike priority.
• The developments should be more accessible on foot, by bike and by public transport than by private vehicle.
• Parking should be provided away from dwellings, and sold or rented separately from dwellings. Residents should not be eligible for on-street parking permits in the surrounding areas.
• Car-sharing and in some cases cargo-bike sharing should be available.
• Certain developments had binding or non-binding agreements for residents to sign, committing them to either not owning a car at all, or to parking it off street away from their dwelling.

The principle unintended consequence of parking standards that limit the amount of parking in new developments is overspill parking. It is to be expected that if a development with limited parking is located in an area with unrestricted on-street parking, or where residents can get a permit for on-street parking, then car use and ownership will be higher than where there is no available on-street parking around the development and users of the development will put pressure on that parking.

A further unintended consequence reported by Foletta and Field (2011) is that in some developments where access is permitted only for loading and unloading, residents in some streets nonetheless park outside their homes. The degree to which this occurs depends considerably on peer pressure (or the lack of it), but these developments, including Scandinavian housing cooperatives, normally have some form of elected management committee which can exert pressure on residents who do not comply with such rules.

Parking time limits, permits and pricing and on- and off-street

Our literature review did not identify any evidence on the effect of parking time limits or permits on car km or modal split. There was however evidence of the impact of parking pricing on car km and modal split, from both modelled and observational studies such as Ecorys, 2022; Pfaffenbichler and Schopf, 2011; Kodransky and Hermann, 2011; Swiss Federal Office for Spatial Planning, 2021; and McCahill et al., 2016. The reported changes in travel behaviour from these studies are summarised in Section 6.1, above.

In terms of lessons for the smooth(er) implementation of parking pricing, several papers made some helpful points. The key barrier to implementation identified was (lack of) public acceptability (Palmer and Ferris, 2010). It was argued that parking pricing should be part of an integrated transport policy rather than standalone so that, for example, reduction in numbers of parking spaces goes hand-in-hand with improvements in the public realm (Tully et al., 2022). Khandokar (2016) pointed out that public consultation that clearly informs people about the planned changes in parking prices and availability and how these will affect them personally will help to build acceptance, as will ringfencing of the money raised for environmental and transport improvements (Kodransky et al., 2011). Certain papers conducted attitudinal research and found that the rationing of parking through permits and/or making these permits tradeable was seen to be fairer than pricing parking (e.g. Brands et al., 2021). Better information about available parking that already exists is also an important part of reducing the negative impacts and therefore increasing the acceptability of new on-street pricing schemes (Albalate and Gragera, 2018).

Few papers mentioned unintended consequences in relation to parking pricing, with the exception of the issue of overspill parking. Several pointed out that overspill parking can occur if workplace or other off-street parking is priced in a location where there is available, free on-street parking (e.g. Melia and Clark, 2016 looked at the impact of parking pricing and parking space reductions on the University of West of England campus in north Bristol). However, the most detailed evaluation of the introduction of new on-street parking controls, including pricing, in Vienna noted that overspill parking outside the new zones was less than had been expected, because the zones were large enough to make a shift of mode the lowest-cost response for people from outside Vienna who had previously driven into the now-priced zones (City of Vienna, 2020).

Parking levies

There are very few examples of parking levies in the world. For this review we found examples from Nottingham, England as well as Perth, Sydney and Melbourne in Australia. The parking levy is a tax levied by a local or regional authority on certain types of off-street parking. In Nottingham, it is levied on parking provided by employers (with the exception of some, such as the NHS) that provide 10 or more parking spaces for their staff. In 2023, it was set at £522 per space per year. In the Australian cases, the levy applies to all long-stay off-street parking, public or private, in the city centre and inner city, except off-street residential parking.

In both cases, the tax is levied on the owner and/or operator of the parking space and there is no obligation to pass it on to the driver who parks there; in 2009 the charge in Melbourne reached AS$800 per year and then increased with inflation, whilst in Sydney in 2012 it was over AS$2,000. The stated purpose of the levies is to reduce congestion (although of course some revenue is also raised), but the degree to which congestion is reduced is obviously related to whether or not the final user of the parking space has to pay. Young, Currie and Hamer (2014) suggest that the proportion of users who actually pay the charge in the Melbourne case is low. In Nottingham, the charge is passed on to users for about 53% of spaces, concentrated at the largest employers. In both cases, the levy has seen a fall in the number of liable (levied) parking spaces as employers and operators took them out of use or changed them to other forms of parking (visitor, short-stay) and, in Melbourne, it has led to a slowing in the growth of private non-residential parking supply.

Research by Loughborough University (Dale, 2017) shows a statistically significant link between the introduction of the levy in Nottingham and a fall in congestion. There is also some evidence of modal shift away from car as a result (see Section 6.1 above). However, even when passed on, the levy is a small proportion of total travel costs and many other factors affect congestion (roadworks, changes in economic activity and so on) (Nottingham City Council, 2019). In Melbourne the levy has contributed to a reduction in car mode share for trips to the city centre but as part of a package of measures including much improved public transport (Young et al., 2014). There is no such evidence in Sydney (Ison et al., 2014). The authors suggest this is because, even when the cost is passed on to users, they are not aware of it (it is a salary deduction, for example).

Implementation appears to have been relatively straightforward in the Australian examples, perhaps because the levy represents a small part of the cost of providing parking. In Nottingham, it was introduced in 2012 after approximately 10 years of preparation, a key part of which was working with employers on travel planning. In addition, the first line of a new tram and improvements in local bus services, cycling infrastructure and a new station were all delivered before the levy went live, and the promise of the revenue raised by the levy has helped to lever in central government funds. The levy was designed to be simple to administer and by only including employers with 10 parking spaces or more, it is not a burden on the smallest businesses. There is no evidence that companies have relocated because of the levy, and economic growth in Nottingham has continued to be higher than in competitor English cities (Nottingham City Council, 2019).

The main unintended consequence of the Australian schemes was a big growth in discounted “early-bird” commercial parking offers, giving those who arrived early at public off-street parking a discount because of the operating hours specified in the parking levy legislation. The literature we reviewed does not reference any unintended consequences associated with the Nottingham scheme.

Park and ride

Park and ride is intended to reduce car use, particularly for travel to city centres, by enabling drivers to park before reaching their destination and change to public transport for the final part of the trip. To make them attractive to users, fast and frequent public transport to the final destination is required, which may require subsidy as well as priority over other traffic. Park and ride will attract more users where city centre parking availability is reduced and/or the price is increased over time. For example, Oxford’s P&R system benefits from a long-standing policy of very limited new off-street public and private parking in the city centre (Parkhurst, 2011). Nonetheless, even if the P&R is well-used, if it is located close to the destination, it is unlikely to have reduced vehicle km, due to the unintended consequences outlined below (Zijlstra et al., 2015).

Park and ride sites located close to the final destination, for example on the fringes of a city, may not reduce vehicle km because:

• People who previously made their whole trip by public transport now drive most of the way to take advantage of the better and cheaper public transport from the P&R.
• Some users are people who walk in from nearby residential areas.
• The P&R may be used for parking for other land uses nearby.
• If bus-based, the bus trips themselves generate additional traffic.
• In the longer term it may stimulate more car-based development in the origin location that otherwise would have been in the city centre.

The literature (e.g. Hanssen et al., 2016) is therefore clear that P&R should be located close to trip origins so that public transport is used for the majority of the trip length. Nonetheless, such P&R facilities are typically rail-based and therefore raise the problem of whether valuable land around stations—especially in town centres—should be used for car parking; and whether these P&R also stimulate more car-dependent lifestyles by enabling long-distance commuting. Hanssen et al. (2016) therefore include in their paper some guidance on how to locate and design P&R to minimise risks that it will have these unintended consequences.

Parking capacity reductions at city or neighbourhood level

This topic relates to the idea of taking away parking spaces (on-street in particular) and devoting the space to other uses: pedestrianisation, wider footways, bike lanes, green space and so on. When carried out on a large scale, this intervention is clearly associated with lower car use (see Pfaffenbichler and Schopf, 2011). However, only nine papers were found dealing directly with this topic and, of these, only four provided information on implementation lessons, barriers and unintended consequences. The most comprehensive guidance on implementation is contained in Rye et al. (2022), where the experience of several cities across Europe in reducing on-street parking supply is described. To quote directly from pp 47-48 of that publication:

“There are ways, however, to address these concerns and the almost inevitable complaints that will be heard when new parking management measures are proposed (although bear in mind that once the new measures are implemented, experience shows that almost all these complaints will die away as people realise that the measures work). To be prepared, however, the following points need to be taken into account:

• The phrase “there is not enough parking” will come up. It is important therefore to have carried out some basic surveys to measure parking occupancy in the busiest streets and off-street car parks at different times of day; but also in the general vicinity (within 5-10 minutes’ walk) of these busiest areas. Invariably this shows that whilst in the busiest areas there are times of day when demand exceeds supply, it also shows that there is almost always spare capacity (including off-street car parks that few people are aware of). It can also show that long term parkers occupy space that could be used for shoppers and visitors.
• It is crucial to communicate the changes in parking management fully, including the reasons for them and the expected benefits.
• The planned parking management measures need to be easy to understand. If they are not, misunderstandings will occur and these will create myths about the new scheme which will make them more difficult to implement.
• For the two reasons above, cities may wish to consider contracting in some specialised marketing and public relations assistance – people who know how to “sell” a message, and also who know how to deal with negative reactions, particularly on social media.”

Specifically with regard to the removal of on-street parking, the literature also suggests that experimentation can be useful, with temporary changes to parking spaces to other uses; and that it is crucial to demonstrate what the space has been used for instead of parking.

The most-feared unintended consequence of parking capacity reductions (and parking pricing) is negative impacts on retailing – that customers will no longer travel to the businesses where on-street parking has been removed and changed to other uses. However, the literature that was found on this topic was unanimous in its findings that parking availability is much less important to retail vitality than factors such as quality of the urban environment and retail offer (e.g. Witte and Mingardo, 2017; Olimstad, and Gjellebæk, 2015); and that retailers usually overestimate the importance of parking for their custom.

Conclusions and recommendations

Conclusions

The research has gathered evidence on the effectiveness of different parking management interventions in reducing car use. The findings can support development of parking policies by Scottish Government and Local Government partners with a view to achieving the Scottish Government’s and COSLA’s joint target of reducing car kilometres. Following the discussion, table 2 summarises the findings on impactful interventions.

Impact on car use

Overall, strong evidence was found for five parking intervention types of impact on one or more of the following: car kilometres travelled, modal split or car ownership. These were parking standards, off-site or non-adjacent provision of residential parking, and low-car and car-free housing; on- and off-street parking pricing; parking levies; park and ride; and parking capacity reductions at city or neighbourhood level. In the case of all but P&R, the impact was found to be positive. In the following paragraphs, specific conclusions are provided for each of the five intervention types.

There is evidence linking parking standards, off-site or non-adjacent provision of residential parking, and low-car and car-free housing to positive impacts on car kilometres travelled, car ownership and modal split. Parking availability and location can influence car use; car-free developments have been found to have car use levels at less than half of city-wide averages, while parking located at 50m or more from dwellings was associated with 25% fewer car trips.

There is also strong evidence of the effect of parking availability and location and price on car ownership. Having a parking space on one’s own property, as opposed to only having access to on-street parking, has been linked with car ownership levels that are 34 percentage points higher in nationwide Norwegian studies. Marked differences in levels of car ownership have also been found in the context of car-free developments when compared to city-wide levels. Cars were also found to have a lower modal share in car-free areas; in the Netherlands and Germany, car-free development were found to have car use levels at less than half the city averages.

Parking pricing can contribute to car kilometre reductions and modal shift, particularly when combined with capacity reduction measures. Parking pricing of the order of €1 to €3 per day can reduce car kilometres by between 0.3% and 3% at entire city or nationwide scale, while other modelled studies identified city-wide reductions of between 6% and 16% when parking pricing changes were accompanied by on-street capacity reductions or implemented in low-traffic neighbourhoods. The reduction level associated with pricing measures is highly sensitive to the proportion of drivers who are affected by the charge and whether or not it is implemented alongside overall parking capacity reductions.

Parking pricing has also been linked to car ownership. Car parking costs, including search time, were found to account for 30% of the variation in car ownership across four Dutch cities.

Workplace parking levies were found to have a positive impact on mode share. Public transport improvements implemented in tandem with WPL schemes were identified as significant contributing factors in encouraging modal shift.

Park and ride was found to generally increase VKT when located close to destinations. However, when located close to journey origins, it was associated with reductions in vehicle kilometres of the order of 1.5km per park and ride user.

There is evidence that parking capacity reductions at city or neighbourhood level have an impact on car kilometres travelled and modal split. As explained in conclusions on parking pricing, when implemented in tandem with on-street pricing measures, on-street parking capacity reductions have been linked to area wide VKT reductions of between 6.4% to 16%. There is also evidence that the availability of parking spaces at a city scale influences mode split, with an increase in supply of between 0.1 and 0.4 parking spaces per person across US cities being associated with a 30% increase in car mode share.

In the context of workplaces, there is strong evidence that the provision of parking is linked to an increase in car mode share. In nationwide Norwegian studies, car mode share stood at 58% when parking was available, versus 20% when it was not, while a 15 percentage point difference was identified by research conducted in Switzerland. It is also notable that in the context of employee provided parking, there is strong evidence that workplace travel plans have a greater impact on modal split when they include some form of parking management.

It should be noted that all of the reviewed literature reported on studies conducted in urban settings, which correspond to the categories of Large Urban Areas and Other Urban Areas in the Scottish Government’s Urban Rural Classifications (Scottish Government, 2020)^[14]. There is therefore a lack of evidence on whether or not the intervention types identified in this section would have similar impacts on car use in rural areas.

Equity and equality issues

There is extremely limited evidence on whether the reductions in car km and changes in modal split achieved by parking interventions are shared across social groups, and it is not possible to draw conclusions on this topic. In total, only five papers discussed such issues, (including the impacts of parking on accessibility and housing costs), while only one study included findings on the impact of parking pricing on lower income groups.

Given the lack of evidence on the impact of parking interventions outside urban contexts it also not possible to draw conclusions on how different intervention types may align to inequality reduction goals focussing on island communities and remote rural and rural areas, as per the Scottish Government’s Urban Rural Classifications (Scottish Government, 2020).

Alignment to policies and strategies

Parking standards, offsite or non-adjacent provision of residential parking, and low-car and car-free housing interventions align to goals set out in NPF4 on the basis that they help encourage walking, wheeling and cycling, and play an integral role in low-car and car-free developments. Parking capacity reductions at a city and neighbourhood level also align with the goals of NPF4, which seek to reduce levels of car dominance.

Interventions contributing to reduced traffic volume in city centres, such as parking pricing and, in some cases, P&R will have positive effects on congestion and air pollution levels, aligning to NTS2 goals around improving health and wellbeing. Furthermore, workplace parking levies have the potential to contribute to goals set out in NTS2’s Sustainable Travel and Investment Hierarchies, for example if funds from the levies are invested in local public transport improvements that correspond to the hierarchies’ aims.

Support for place-based approaches

There is a lack of explicit evidence in the reviewed literature on the application of place-based approaches in parking interventions. However, some of the impactful intervention types, such as parking standards, off-site and non-adjacent residential parking, and low-car and car-free housing; and parking capacity reductions at city or neighbourhood level, may be better positioned to support the use of a place-based approach based on their potential contributions to minimising the space dedicated to car parking in a particular locality.

Table 2. Summary of findings on impactful interventions

i Six out of eight of the papers providing evidence of a link between park and ride and car use found that park and ride was associated with increases in VKT when the site is located close to the journey destination. One further paper found no impact on VKT. Only when park and ride sites were located closer to journey origins did the literature find evidence of reductions in VKT.

Recommendations

In considering the merit of implementing the intervention types with the greatest impact on car use we recommend:

• In removal of on-street parking to first consider implementing temporary changes to parking spaces where infrastructure allows. This would allow the approach to be tested and for gathering of further practical data on the implementation of measures.
• To promote parking management in the context of workplace travel plans given the significant difference that including parking measures in this type of plan has been found to make to modal shift outcomes.
• Future decision-making around park and ride should consider the significance of site location relative to journey origins and destinations to optimise the contribution of this intervention type to car use reduction targets.

Research gaps

During the project we have identified the following areas needing further research:

• The intervention types lacking strong evidence (positive or negative) around impact on car use reduction. Specifically, these intervention types are: parking pricing in relation to vehicle characteristics; parking pricing in relation to household or user characteristics; residential parking in historic areas that are pedestrianised; shared-use parking; effective and fair parking enforcement; parking for electric vehicles; mobility hubs; 20-minute neighbourhoods and 15-minute cities; and parking and road space transformation/reallocation.
• The impact of different types of parking intervention on equality and the equitable reduction in car kilometres.
• The implementation of parking interventions in rural contexts.
• How, and to what extent, different intervention types support place-based approaches.

Research recommendations

• It is recommended that consistent before-and-after studies should be conducted when trialling parking interventions. This would help provide an evidence base to inform decision making. The studies should also aim to collect data on aspects, including barriers to intervention, success factors, unintended consequences and lessons for policy, and implementation. As noted above, the reviewed literature often did not provide data on these areas of research. Such before-and-after studies could also collect data on public attitudes towards the interventions. Recent research (van Wee, Annema and van Barneveld, 2023) has found that support for controversial policies in the area of transport often increases following implementation. Further research around parking measures could provide additional insights into public attitudes and levels of support.
• Further research should also be conducted into current parking policies in Scotland to provide a detailed baseline to inform future policies. Apart from a recent unpublished report commissioned by Transport Scotland (Systra, 2021), which provides some brief summaries of parking interventions in local authority areas, there is a lack of up-to-date research in this area.
• Research should be commissioned into the impact of different types of parking intervention on equality and the equitable reduction of car kilometres. This should consider whether reductions in car use are shared across social different social groups and whether the interventions contribute to broader equality goals.

Related to this last point, research should be commissioned into the impact of parking interventions on car use in rural settings. As explained above, all of the reviewed literature had an urban focus. Research conducted in rural contexts should help confirm whether the same measures would be applicable in non-urban settings from the perspective of car use reduction. It should also provide data on how and to what extent the interventions align to Scottish Government goals of reducing urban-rural inequalities.

Finally, research should be commissioned into how different types of parking intervention support place-based working. This could begin with research on measures identified in this report as having potential to minimise space dedicated to car parking in particular localities, namely parking standards, off-site and non-adjacent residential parking, and low-car and car-free housing; and parking capacity reductions at city or neighbourhood level.

References

The following is a list of sources cited in the report.

Agora Verkehrswende., 2018. Umparken – den öffentlichen Raum gerechter verteilen Zahlen und Fakten zum Parkraummanagement. Agora Verkehrswende, Berlin

Albalate, D. and Gragera Lladó, A., 2018. Misinformation and Misperception in the market for parking. IREA–Working Papers, 2018, IR18/12.

Brands, D., Verhoef, E. and Knockaert, J., 2021. Pcoins for parking: a field experiment with tradable mobility permits. Tinbergen Institute Discussion Paper

Cairns, S., Newson, C. and Davis, A., 2010. Understanding successful workplace travel initiatives in the UK. Transportation Research Part A: Policy and Practice, 44(7), pp.473-494

CE Delft., 2018. The CO2 impacts of travelling differently. Anders Reizen. NS, Delft

Christiansen, P., Engebretsen, Ø. and Hanssen, J.U., 2015. Parkeringstilbud ved bolig og arbeidsplass. Fordelingsffekter på bilbruk og bilhold i byer og bydeler. TØI rapport, 1439

Christiansen, P., Hanssen, J.U. and Skollerud, K., 2015. Boligparkering i store norske byer-parkeringstilbudets effekt på bilhold og bilbruk. TØI rapport, 1425

City of Vienna., 2020. Ausweitung der Parkraum-bewirtschaftung in Wien Nachher-Untersuchung 11. City of Vienna. Vienna, Austria

Dale, S., 2017. Evaluating the impacts on traffic congestion and business investment following the introduction of a Workplace Parking Levy and associated transport improvements (Doctoral dissertation, Loughborough University).

Dale, S., Frost, M., Gooding, J., Ison, S. and Warren, P., 2014a. A case study of the introduction of a workplace parking levy in Nottingham. In: Ison, S.G. and Mulley, C. eds., 2014. Parking: issues and policies. Emerald Group Publishing

Dale, S., Frost, M., Ison, S. and Budd, L., 2019. The impact of the Nottingham Workplace Parking Levy on travel to work mode share. Case Studies on Transport Policy, 7(4), pp.749-760

Ecorys., 2022. Exploring the quantitative impact of behavioral measures on mobility. Report for Ministry of Transport, The Hague, Netherlands

Foletta, N. and Field, S., 2011. Europe’s vibrant new low car(bon) communities. Institute for Transportation and Development Policy, New York.

Gonzalez, J.N., Gomez, J. and Vassallo, J.M., 2022. Do urban parking restrictions and Low Emission Zones encourage a greener mobility?. Transportation Research Part D: Transport and Environment, 107, p.103319

Gunnarsson-Östling, U., 2021. Housing design and mobility convenience—The case of Sweden. Sustainability, 13(2), p.474.

Hanssen, J.U., Tennøy, A., Christiansen, P. and Øksenholt, K.V., 2016. How can P & R facilities contribute to reduced emissions of greenhouse gases. In European Transport Conference

Kansen, M., Waard, J. and Savelberg, F., 2018. Sturen in parkeren. Kennisinstituut voor Mobiliteitsbeleidj. KiM

Katoshevski-Cavari, R., Bak, N. and Shiftan, Y., 2018. Would free park-and-ride with a free shuttle service attract car drivers? Case studies on transport policy, 6(2), pp.206-213

Khandokar, F., 2016. Determinants for intention to change travel mode choice behaviour of NHS hospital staff (Doctoral dissertation, Loughborough University).

Kirschner, F. and Lanzendorf, M., 2020. Parking management for promoting sustainable transport in urban neighbourhoods. A review of existing policies and challenges from a German perspective. Transport Reviews, 40(1), pp.54-75

Kirschner, F., 2021. Parking and competition for space in urban neighborhoods. Journal of Transport and Land Use, 14(1), pp.603-623

Kodransky, M. and Hermann, G., 2011. Europe’s parking u-turn: from accommodation to regulation. Institute for Transportation and Development Policy, New York.

Leibling, D., 2014. Parking supply and demand in London. In: Ison, S.G. and Mulley, C. eds., 2014. Parking: issues and policies. Emerald Group Publishing

McCahill, C.T., Garrick, N., Atkinson-Palombo, C. and Polinski, A., 2016. Effects of parking provision on automobile use in cities: Inferring causality. Transportation Research Record, 2543(1), pp.159-165

Melia, S. and Clark, B., 2016. Evaluation of the change in parking policy on Frenchay campus. Centre for Transport and Society, University of the West of England.

Milosavljević, N. and Simićević, J., 2016. User response to parking policy change: A comparison of stated and revealed preference data. Transport Policy, 46, pp.40-45

Nottingham City Council (2019). Transport Scotland Bill: Workplace Parking Levy Amendments. Submission from Nottingham City Council

Olimstad, M. and Gjellebæk, I., 2015. Hva betyr gateparkering for handelen? Oppsummering av norske og internasjonale studier. SVV, Oslo.

Oost, T., 2022. How to make car-free neighbourhoods work: the factors that contribute to the success of a car-free neighbourhood (Masters dissertation, University of Groningen).

Ostermeijer, F., Koster, H., Nunes, L. and van Ommeren, J., 2022. Citywide parking policy and traffic: Evidence from Amsterdam. Journal of Urban Economics, 128, p.103418

Ostermeijer, F., Koster, H.R. and van Ommeren, J., 2019. Residential parking costs and car ownership: Implications for parking policy and automated vehicles. Regional Science and Urban Economics, 77, pp.276-288

ourplace.scot (2023). ‘About Our Place’. Available at: https://www.ourplace.scot/about-us#:~:text=Our%20Place%20is%20a%20site,Service%20and%20Glasgow%20City%20Council

Palmer, D. and Ferris, C., 2010. Parking measures and policies research review. Wokingham: Transport Research Laboratory

Parkhurst, G. and Meek, S., 2014. The effectiveness of park-and-ride as a policy measure for more sustainable mobility. In: Ison, S.G. and Mulley, C. eds., 2014. Parking: issues and policies. Emerald Group Publishing

Parking Brussels., 2020. Parking Policy in the Brussels Capital Region – Benchmarking Report

Pfaffenbichler, P. and Schopf, J.M., 2011. Einfluss der Parkraumorganisation und der Anzahl der Stellplätze auf die Nutzung des motorisierten Individualverkehrs und die Erreichung verkehrs-, umwelt-und siedlungspolitischer Ziele (PAN). Austrian Energy Agency, Vienna

Richardson, E., no date. The role of parking in limiting traffic growth and congestion. Unknown

Rye, T., Tully, S., Godin, G., Schmalholz, N. and Hertel, M., 2022. Parking and SUMP. Using parking management to achieve SUMP objectives effectively and sustainably. European Platform on Sustainable Urban Mobility Plans.

Scottish Government, 2020. Scottish Government Urban Rural Classification 2020. Available at: https://www.gov.scot/publications/scottish-government-urban-rural-classification-2020/

Scottish Government, 2023. National Planning Framework 4. Available at: https://www.gov.scot/publications/national-planning-framework-4/

Strategy Development Partners and Martens, M., 2019. Parkeerbeleid als stuurmiddel voor woon-werkverkeer. Ministry of Infrastructure and Water Management, The Hague

Swiss Federal Office for Spatial Planning., 2021. Shaping Mobility in Agglomerations: Parking Management

Systra, 2021. Scottish Parking Policy Review: Summary of key findings.

Thorwaldson, L., Thomas, F. and Carran-Fletcher, A., 2021. Evaluating the Greenhouse Gas Emission Reduction Benefits from Land Transport Mode Shift Programmes and Projects: A Research Note. Waka Kotahi, NZ Transport Agency, 4.

Transport Scotland, 2020. National Transport Strategy. Available at: https://www.transport.gov.scot/our-approach/national-transport-strategy/

Transport Scotland, 2022. Reducing car use for a healthier, fairer and greener Scotland: A route map to achieve a 20 per cent reduction in car kilometres by 2030. Available at: https://www.transport.gov.scot/media/50872/a-route-map-to-achieve-a-20-per-cent-reduction-in-car-kms-by-2030.pdf

Witte JJ and Mingardo G., 2017. Parking policy, parking duration and spend of shoppers in the Netherland. Erasmus University Rotterdam, working paper

Young, W., Currie, G. and Hamer, P., 2014. Exploring the impact of the Melbourne CBD parking levy on who pays the levy, parking supply and mode use. In: Ison, S.G. and Mulley, C. eds., 2014. Parking: issues and policies. Emerald Group Publishing

Zijlstra, T., Vanoutrive, T. and Verhetsel, A., 2015. A meta-analysis of the effectiveness of park-and-ride facilities. European Journal of Transport and Infrastructure Research, 15(4)

Zimmerman, C., Klein, R. and Schroeder, J., 2014. San Francisco urban partnership agreement : national evaluation report . Available at: https://www.sfmta.com/getting-around/drive-park/demand-responsive-pricing/sfpark-evaluation

Appendices

Appendix A. Methodology

Literature search and sift

The main phases of research included a systematic literature search, a sifting process to identify relevant papers among the search results and a review of the selected papers. Forty individual searches were conducted on Google Scholar, using strings of search terms containing Boolean operators and created to reflect the 14 intervention types identified during the project’s inception phase (see Section 4.2). The searches yielded over 4,700 results.

The papers were then sifted by title with the aim of selecting the most relevant papers. The title sift resulted in 927 papers. Once duplicates, which had been found in multiple searches, were removed, a total of 386 papers remained. These papers were then sifted again based on the contents of their abstract. The criteria applied during the abstract sift were:

• country of study (with specific interest in OECD countries, particularly smaller northwest European ones);
• references to parking intervention types of interest to the study;
• relevance to research questions; and
• indications of methodological rigour (see Section 9.1.2).

The papers were assigned a priority rating from a four-point scale based on the level of relevance to the study. Seventy-eight Grade 4 papers (include in the review) and 85 Grade 3 papers (possibly include in the review) were identified.

In addition, searches were conducted using Google’s regular search engine in English, Catalan, Dutch, German, Norwegian, Spanish, and Swedish with the aim of identifying papers, particularly in the grey literature, which may have been missed during the systematic search. This resulted in 34 additional sources being identified.

Requests were also made to contacts of the research team working in relevant academic fields for recommendations on papers which may be of interest to the study. This resulted in 31 papers being recommended, 25 of which were subsequently reviewed.

Literature review

The review included the 78 Grade 4 papers, in addition to 25 contact-recommended papers and the 34 papers identified during Google searches. Two Grade 3 papers were also reviewed following a more in-depth sifting of the Grade 3 sources carried out with the aim of identifying those that were most relevant to the research. This gave a total of 139 papers.

During the review, data relevant to each of the seven research questions was extracted from each source. The papers were also assessed on the basis of robustness of methodology and strength of evidence in order to provide an overall indication of the strength of the research that the review’s findings are based on. A qualitative robustness of methodology scale was applied as shown in the following table.

The strength of evidence assessment was based on the extent to which individual papers included findings on the impact of a given intervention on car kilometres travelled, modal split or car ownership, either providing evidence of positive or negative findings. Papers that contained clear findings linking interventions to either positive impacts, (this is to say reductions in VKT, modal shift away from car use or reductions in car ownership) or negative impacts (for example increases in VKT) were considered to have provide strong evidence of impact. On the other hand, those containing inconclusive findings or findings indicating a weak link between interventions and car kilometres travelled, modal split or car ownership were considered to provide weak evidence.

The review also provided a final opportunity to sift out papers that were not considered sufficiently relevant. In the case that a source did not provide any relevant data related to the research, it was discarded and not included in the subsequent analysis. Ultimately, 39 papers were discarded, leaving 100 to be analysed.

Limitations

Time and resource constraints placed some limitations on the literature search. The decision was made to conduct the systematic search using only Google Scholar as it is the largest academic database. This meant that the search process did not have to be repeated for multiple databases. Google Scholar was considered to offer the broadest reach, in terms of the size of the database, but it is possible that some literature that may have appeared during searches in other databases and search engines was not present in the results generated by Google Scholar.

It should also be noted that our searches were informed by the list of intervention types finalised during the inception phase, which include 14 different categories of measure (see Section 4.2). The way that intervention types were described and grouped at this stage influenced the subsequent formulation of the Boolean search terms that were used in the search. It is possible that different descriptions and groupings of intervention types may have generated a greater number of relevant results.

Time constraints also influenced how the sifting of search results was conducted. As we were required to sift a large number of results, sifting was first conducted by title. During the title sift, some papers may have been incorrectly excluded due to limited amount of information informing the selection. During the subsequent abstract sift, the likelihood of this happening was reduced due to the larger amount of information available. This said, some papers may still have been incorrectly excluded during this process.

Limited amounts of data was available on the impact of some parking interventions on car kilometres travelled, and on the impact of parking measures on equality. This has limited the responses to some of the research questions and allowed for the identification of gaps in the literature.

Appendix B. Summary of papers providing evidence of impact on car use

The following tables provides a breakdown of the papers that provide evidence of an impact on car kilometres travelled, modal split and car ownership, based on strength of evidence, robustness of methodology and intervention type.

Car kilometres travelled

i The total can sum to more than the total number of papers as some of the reviewed papers deal with more than one intervention.

Modal split

ⁱ The total includes papers mentioning traffic reduction and congestion reduction.

ⁱⁱ One paper may cover more than one intervention hence totals may not sum to 47.

Car ownership

i One paper may cover more than one intervention so totals may not sum to 17.

Appendix C. Bibliography

The following table contains a full list of papers included in the literature review. It indicates the intervention types described by each paper and the area of impact (car kilometre reduction, modal split, car ownership), if any, that the paper provides evidence on.

© Published by Ansons Consulting, 2023 on behalf of ClimateXChange. All rights reserved.

While every effort is made to ensure the information in this report is accurate, no legal responsibility is accepted for any errors, omissions or misleading statements. The views expressed represent those of the author(s), and do not necessarily represent those of the host institutions or funders.

info@climatexchange.org.uk

+44(0)131 651 4783

@climatexchange_

www.climatexchange.org.uk

1. Six out of eight of the papers providing evidence of a link between park and ride and car use found that park and ride was associated with increases in VKT when the site is located close to the journey destination. One further paper found no impact on VKT. Only when park and ride sites were located closer to journey origins did the literature find evidence of reductions in VKT. ↑
2. Research Question 1 is:‘What is the impact of the intervention with regard to its contribution to reducing car kilometres and how does the actual impact compare to any ex-ante (modelled) prediction?’ ↑
3. According to NTS2, the assessment of effects around equalities should consider poverty and in particular child poverty, gender inequalities, social isolation, transport needs of young and older people and disabled people. In addition, Scotland’s regional differences and the differing needs of cities and towns should also be considered. In cities and towns, congestion and detrimental effects on air quality are important. In rural area, more limited public transport options and longer commuting distances are important factors affecting people’s choices and opportunities. Island communities can face particular challenges related to these factors and often greater levels of isolation. ↑
4. Research Question 6 is ‘How did the interventions contribute to an equitable reduction in car kilometres?’ ↑
5. The Just Transition to net zero includes the following principles: supporting environmentally and socially sustainable jobs; supporting low-carbon investment and infrastructure; develop and maintain social consensus through meaningful engagement with workers, communities, non-governmental organisations, businesses, industry bodies and any other relevant groups; making all possible efforts to create decent, fair and high-value work in a way that does not negatively affect the current workforce and overall economy; and contributing to resource-efficient and sustainable economic approaches which help address inequalities and poverty (Transport Scotland, 2020). ↑
6. Regression analysis is a statistical method allowing the quantification of the relationship between one or more independent variables and a dependant variable. In contexts where multiple independent variables influence the outcome, regression analysis enables the extent of the impact of each independent variable on the dependant variable to be identified. ↑
7. Note that the total for the number of interventions may differ from the total for the number of papers since some papers included references to multiple intervention types. ↑
8. At the time of writing this is equivalent to between £0.86 and £2.58. ↑
9. The figures for the total reduction in car kilometres travelled and percentage reduction in traffic volume were generated by multiplying the average car journey length by the number of car journeys avoided following the interventions. As such, the estimated reduction in traffic volumes is a proxy for the percentage reduction in car kilometres travelled. ↑
10. Note that the total for the number of interventions may differ from the total for the number of papers since some papers included references to multiple intervention types. ↑
11. There is, however, no information on the level of charge that people surveyed pay. It should also be noted that the studies do not consider a planned intervention but rather the impact of parking conditions experienced in different parts of the city or country (e.g. distance of parking spaces from home). ↑
12. Note that the total for the number of interventions differs from the total for the number of papers since some papers included references to multiple intervention types. ↑
13. At the time of writing, €500 is equivalent to £427 and €3,600 is equivalent to £3,074. ↑
14. The document classes ‘Large Urban Areas’ as having populations of over 125,000 plus and ‘Other Urban Areas’ as having populations of between 10,000 and 124,999 inhabitants (Scottish Government, 2020). ↑

Research completed March 2023

DOI: http://dx.doi.org/10.7488/era/3231

Executive summary

Aims

This report looks at different options for conducting whole building assessments of multi-owner and mixed-use buildings, through a literature review and structured conversations with stakeholders. These assessments are needed to plan the improvement of building fabric efficiency and installation of zero direct emissions heating systems.

This work is useful because most current building assessment methods in Scotland are used for single dwellings and not at a whole building level. Additionally, the most commonly used assessment methods for both domestic and non-domestic properties are intended to measure compliance with building regulations rather than retrofit design.

Findings

Current assessment methods cannot be used for the purpose of retrofit design because they are designed for comparison rather than absolute calculations of building performance. The assessment of multi-owner and mixed-use buildings requires two methods, which cannot currently be combined to produce a single assessment.

Based on analysis of whole building energy assessment approaches internationally, we found that:

• none of the examples have been developed specifically for multi-owner and mixed-use buildings
• several assessment approaches can co-exist and fulfil different functions (i.e., compliance and design), eg Denmark and France have additional assessment approaches beyond energy performance
• best practice assessment approaches go beyond energy modelling and use holistic frameworks; for example, PAS 2035 is a British framework for delivery of quality retrofits of domestic buildings.

We outline three options for how a whole building assessment methodology could be developed in Scotland. The options cover a range of costs both for conducting the assessment and for method development. For this reason, they range in the level of detail and accuracy.

Option 1 is a low-cost option, based primarily on assumed data rather than measured data. It involves updating existing methods to complete a whole building assessment.

Advantages:

• Low-cost option, for both assessment and development
• Utilises the existing workforce and the management arrangements associated with producing Energy Performance Certificates.

Limitations:

• The existing methodology must be modified in order to assess communal and non-domestic spaces
• An assumption-based assessment cannot adequately consider the risk associated with the retrofit of multi-owner or multi-use buildings.

Option 2 is a detailed assessment approach with PAS 2035.

Advantages:

• A holistic retrofit assessment rather than an energy performance assessment, designed to mitigate the risk of unintended consequences of retrofit
• Infrastructure for PAS 2035 training, qualifications and certification is already being put in place, although this must be scaled up.

Limitations:

• The supply chain is not yet capable of delivering PAS 2035 retrofit
• Changes to modelling approaches are required for this option, meaning higher development costs.

Option 3 presents an assessment approach that draws on best practice from the international examples we found.

Advantages:

• Like option 2, option 3 is designed as a holistic assessment approach, which aims to mitigate the risks of retrofitting multi-owner and mixed-use buildings

Limitations

• It has the highest associated development costs as it does not build on existing approaches already used in Scotland
• The Scottish context differs from the examples we have reviewed, particularly in terms of the lack of formal management structures in multi-owner buildings, low prevalence of communal heating and the current practice of assessing only individual flats.

Glossary

Introduction

The Scottish Government aims to reduce emissions from buildings by 68% compared to 2020 levels by 2030 and reach net zero emissions in 2045, according to the Heat in Buildings Strategy. Achieving these targets requires retrofitting Scotland’s existing building stock, reducing energy demand by improving fabric efficiency and installing zero direct emissions heating – systems.

The retrofit of multi-owner and mixed-use buildings is challenging for a number of reasons. These include a lack of building management structures such as owners’ associations, endemic disrepair of tenement blocks (Robertson, 2019) and current property law arrangements. The Scottish Law Commission (Scottish Law Commission, 2022) is considering the law around compulsory owners’ associations and a draft Tenement Maintenance Bill which could introduce new provisions to facilitate common works by Spring 2026. One key challenge to the retrofit of these buildings is that currently we cannot undertake energy assessments at a whole building level. Whole Building Assessments will be required to assess and make appropriate recommendations for communal works for both energy efficiency work and communal heating systems.

This report presents results from a desk-based scoping exercise to identify options for developing an assessment method of multi-owner and mixed-use buildings in Scotland. The research findings are primarily based on published literature. This was supplemented by input from 14 stakeholders from 11 organisations (see Section 10).

The report contains:

• a summary of assessment methods that are currently used in the UK and their suitability for assessing multi-owner and mixed-use buildings
• an outline, analysis and comparison of eight international examples of whole building assessment methods
• three possible options for the development of an assessment method of multi-owner and mixed-use buildings in Scotland.

Existing assessment methods in Scotland

Two methodologies are currently used to produce Energy Performance Certificates (EPCs) in the UK. This section provides a brief overview of these energy models, as well as two other types of energy model currently in use in Scotland: Passivhaus Planning Package (PHPP) and Dynamic Simulation Models (DSM).

Standard Assessment Procedure (SAP)

Overview

SAP is a steady-state modelling method. This means it uses a simple energy balance calculation (heat loss vs gains) for each individual day of the year. This type of energy modelling is commonly used for domestic buildings. The calculation models heat loss, internal gains, solar gains, energy balance, carbon emissions, heating, ventilation, internal lighting, cooling and renewable energy sources.

SAP is used for both new and existing residential buildings. Full SAP is primarily used to generate an EPC for new dwellings whereas RdSAP (Reduced Data SAP) is used to generate an EPC for existing dwellings. RdSAP uses the same calculation as full SAP but uses a simplified data collection process. This enables the calculation to take place where a complete data set for a property is unavailable, and for a lower cost than full SAP.

Accuracy

Comparability rather than accuracy is the primary objective of an RdSAP assessment. In order to make fair comparisons between buildings it ignores factors such as local climate conditions and makes assumptions about the number of occupants.

There is evidence that steady-state tools such as SAP and RdSAP are inaccurate in predicting the energy consumption of dwellings (Sierra, et al., 2018). However, while SAP as a model is considered accurate, inaccuracies in its outputs are caused by the assumptions and default values used (AECB, 2008), such as for occupancy. Full SAP calculations rely on fewer inferred values than RdSAP. Tests conducted by Passivhaus Trust (Passivhaus Trust, 2020) found that SAP can accurately calculate space heating and hot water demand, although it is less accurate for aspects such as internal heat gains and the efficiency of Mechanical Ventilation with Heat Recovery systems. This is because the more a dwelling deviates from the standard assumptions within SAP, the less accurate the modelling is.

RdSAP contains too many assumptions to be of value when designing or installing retrofit measures. To aid its simplicity, RdSAP relies on a number of inferred values. These include assumptions for airtightness, thermal bridging, area of windows, wall thickness, wall u-value (based on age), ventilation type and heating efficiency. There is also evidence of low accuracy (in terms of reproducibility), errors and variable quality in RdSAP assessments (Hardy & Glew, 2019), (Jenkins, et al., 2017).

Simplified Building Energy Model (SBEM)

Overview

SBEM is used to produce EPCs for non-domestic buildings. SBEM utilises a different calculation methodology to SAP. For the generation of an EPC, the SBEM calculation utilises standardised information for several factors to allow comparability between similar building types. In an SBEM calculation, the actual building geometry is entered into the software and zones are defined for each of the spaces (e.g. swimming pool, small shop unit). SBEM then assigns a standardised occupancy profile to each zone based on figures derived from the Chartered Institute of Building Services Engineers. These occupancy profiles are contained within ‘locked databases’ which the user cannot change (BRE, 2015).

Accuracy

Like SAP, SBEM requires a certain amount of standardisation to enable comparability between buildings for benchmarking purposes. However, and as acknowledged by the standard, this is at the expense of accuracy. SBEM is more flexible than SAP. Some parameters are held in ‘accessible databases’ which allow the assessor to override default parameters and use their own measured or observed data. There are also a series of ‘locked databases’ within SBEM, the purpose of which is to enable allow fair and consistent comparison between buildings. This means that measuring and interrogating actual energy performance is not possible.

Limitations of SAP and SBEM

The background calculations used by SAP and SBEM could be used to assess multi-owner or mixed-use buildings for retrofit. However, both tools would need to be adapted to fulfil this purpose. In their current form neither methodology can be used for assessing multi-owner and mixed-use buildings for the purpose of retrofit design for two main reasons:

• “multi-owner and mixed-use”

In a multi-owner or mixed-use building, flats will be assessed using SAP. Non-domestic and communal circulation spaces will be assessed using SBEM. The results and outputs of the two calculations cannot (currently) be directly compared or aggregated.

To produce a single calculation for a whole building, a unified calculation methodology is required. Alternatively, the tools could be adapted so that they can be combined. The core calculations behind SAP and SBEM are very similar, and stakeholders explained that theoretically the two tools could be mixed if required.

• “retrofit design”

Both SAP and SBEM were designed as tools to demonstrate compliance with energy efficiency aspects of the building regulations. Therefore, both methods are intended for comparison rather than absolute calculations of building performance. SAP in particular, is now used in a range of other unintended ways, including as an assessment tool to inform retrofits, and as a design tool. These issues are discussed in more detail in (Etude, et al., 2021). However, as design is not the intended purpose of these tools, there are naturally limitations in how well they perform for this purpose.

They are also both carbon assessment tools, rather than energy performance tools. These factors limit their usefulness as a tool for assessing buildings for the purpose of retrofit design.

Despite their limitations, the ubiquity of these methods and their existing integration in legislation means there may be advantages in adapting one of them to be used for the purpose of assessing multi-owner and mixed-use buildings for retrofit.

Passivhaus Planning Package (PHPP)

Overview

PHPP is a similar calculation to SAP and is used to determine heat demand in buildings. However, PHPP can be used to calculate both domestic and non-domestic parts of buildings under the same methodology (albeit with separate calculations).

Accuracy

PHPP has a high level of flexibility as it allows users to alter parameters which are locked in RdSAP, full SAP and SBEM approved software. These include:

• More detailed inputs for window U-value calculations
• The installed efficiency of building services, ventilation, and domestic hot water
• The effect of shade on heat gains
• Occupancy characteristics, such as hours of use and internal heat gains (Essential for overheating assessment)
• Insertion and calculation of the energy consumption of small power and white goods

This greater flexibility and lower reliance on assumptions means that PHPP is a more appropriate design tool than SAP. The flexibility of the software means that greater accuracy can be obtained when modelling a building by inputting data from in-situ testing which may be required by a whole building assessment^[1].

For example, heat demand is based on several inputs, which can be modelled using the software. A heat demand calculation can be performed without in-situ testing, but monitoring of actual heating costs, temperature and relative humidity can be used to better predict the impact of measures on heating costs.

Dynamic Simulation Models (DSM)

SAP, SBEM and PHPP are steady-state models. Dynamic simulation models (DSM) are an alternative approach to energy modelling. DSM is typically used in large and complex buildings. It is more accurate than steady-state approaches because it can account for more physical processes, such as energy flows between spaces of the building. DSM is more complex than the equivalent steady-state approaches as it requires more information inputs, and a greater amount of time to conduct an assessment.

The results of DSM are highly accurate. DSM produces a 3D model of a building and is used to simulate the impact of various retrofit upgrades.

There are various tools for DSM that are currently used in the UK such as IES, EnergyPlus and Sefaira, all suitable for both domestic, non-domestic and mixed-use buildings. No formal qualifications are required for DSM, although it is generally conducted by Mechanical Engineers and some Architects. In comparison to a specialist tool such as PHPP, there is already a large workforce in Scotland undertaking DSM assessments.

International examples

Eight international examples of whole building assessments of multi-owner and mixed-use buildings are summarised in Table 1. Full details on our findings can be found in Section 11. The summary table allows for comparison between the assessment approaches and includes PHPP and SAP for comparison. International asessment approaches are further discussed in relation to options for whole building assessment for multi-owner and mixed use buildings in Section 6.

Table 1. Overview of international examples of energy frameworks and models

Considerations for developing an assessment approach

Scalability

There is a risk that data gathering for assessment purposes becomes disproportionate to the scale of the retrofit. A whole building assessment should be scalable. This means it should have the ability to identify relevant tests to be conducted prior to work commencing. It should not require all multi-owner and mixed-use buildings to undertake the same level of testing regardless of the complexity of the building and proposed improvements. This would ensure that information is collected when required and relevant. It would also gradually increase industry understanding of the Scottish multi-owner and mixed-use building stock over time.

Site visit and in-situ testing

A solely desk-based assessment is not appropriate for this purpose. A site visit is necessary to gather information such as:

• Detailed measurements of the building including the sizes and characteristics of all external openings.
• Where information exists on previous energy efficiency upgrades or building work, it should be possible and straightforward to include this information in the calculations.
• A visual check that building services have been properly commissioned and are operating as intended (such as trickle vents, boiler, extract vents). Signs of inadequate ventilation, such as mould, odours and condensation, should be recorded.

In-situ testing could also be carried out, determined by the complexity of the building. Tests may include:

• Thermography using a thermal imaging camera to identify areas of concentrated heat loss and building defects.
• Temperature, relative humidity, consumption monitoring and CO₂ monitoring. This is to determine how occupants currently use the building and the adequacy of ventilation.
• In-situ U-value monitoring to determine the actual performance of the building fabric.
• Moisture analysis of the existing fabric to determine the suitability of the building for certain types of insulation.

The tests above aim to reduce the risk of unintended consequences when upgrading a building. They can also identify maintenance issues and problems with the existing building services.

Occupant behaviour and consumption data

Energy assessments for benchmarking or compliance purposes intentionally exclude data on consumption or occupancy. This is to allow for meaningful comparisons between buildings. An assessment for the purpose of retrofit design will be more accurate if it considers how energy is used across the whole building. This is important because the sizing and design of ventilation systems to avoid summer overheating is dependent on developing an accurate picture of occupancy. Inferred occupancy data is likely to underestimate the number of occupants in small properties, particularly social housing.

Moreover, assessments should consider both regulated and unregulated energy use. In a multi-owner or mixed-use building this would require gathering data from each flat or non-domestic space. This would allow advisors to give recommendations on the efficiency of appliances, which make up a greater proportion of overall energy use in buildings with a high level of fabric efficiency.

The alternative to collecting operational data is to develop a series of profiles for different building types and household types, as is currently done with SBEM.

Development of retrofit recommendations

Based on the assessment, recommended retrofit measures and plans are either generated through software or specified by a professional (or a combination of the two). Due to the complexity of multi-owner and mixed-use buildings, software-generated measures alone are not appropriate for retrofit design and should be either checked or specifically recommended by a qualified professional.

Policy and regulatory landscape

A whole building assessment approach could fulfil both regulatory compliance and retrofit design functions. Alternatively, it could operate alongside the EPC system. Several internal examples (Denmark, Flanders, France, and Germany) have both a mandatory EPC-style assessment (though more advanced than RdSAP) as well as an optional and more tailored scheme focused on long-term retrofit design (sometimes called Building Renovation Roadmaps).

In Canada the EnerGuide assessment is used for both compliance and retrofit design. The tool uses normalised data for compliance, and inputs can be changed for design purposes.

Decision-making structures in multi-owner buildings

In the design of a whole building assessment approach for Scotland, it is important to consider the practical context of decision-making and organisation of multi-owner buildings. Most of the international examples exist in a context where multi-owner buildings have different decision-making structures to Scotland. For example, in Denmark, Sweden and Germany, it is a legal requirement for flat owners to be part of a building owner association. In some cases, this association acts as the legal owner of the building.

Whole building assessments and long-term retrofit plans are easier to implement into such structures where there are stricter regulations on building management and a stronger tradition of collective organising. In Scotland, the design of a whole building assessment approach will have to consider the current building management practice and policy, such as the role of property factors and need for owners’ associations.

Existing heat infrastructure

It is also important to note that the reviewed countries have different heat infrastructure compared to Scotland. For example, most of the reviewed European assessment approaches are designed for a context where communal heating is the norm, in the form of communal oil or gas boilers in each building, or where district heating is more widespread.

Options for developing an assessment approach

The following section outlines three options for the development of an assessment approach for multi-owner and mixed-use buildings. The options provide a range of costs (both cost of assessment and method development costs) and a range in the level of detail and accuracy. The differences between the three options are illustrated in Table 2.

Option 1: Assumption-based

Overview

This option explores updating RdSAP to complete a whole building assessment. In its current form an RdSAP calculation excludes common areas and cannot be used for non-domestic spaces. This is because RdSAP includes background assumptions that only apply in residential properties which cannot be changed. To assess non-domestic or common areas, the background assumptions would need to be altered. This would allow assessment of all areas of a multi-owner and mixed-use building.

Data collection

Data collection for RdSAP involves a site visit. The assessor will:

• Take dimensional details of the entire property
• Look at heating and hot water systems
• Assess the building fabric (external walls, glazing, exposed floors and accessible roof space)

The survey normally takes 0.5-1 hour during which time photographs are also taken.

Assessor qualifications

The survey is conducted by a qualified Domestic Energy Assessor. No specific experience is required to undertake the 3-day course and test to become a DEA (Elmhurst Energy, 2023). It is uncommon for building professionals to qualify as DEAs as the assessments are not profitable for those with specific experience in construction.

Data input

The assessor inputs the data to an accredited online RdSAP software provided by the organisation that the DEA is accredited to (e.g. Elmhurst, Stroma). All software is approved by Building Research Establishment (BRE). The software generates an EPC and a report.

Data input involves inserting overall building dimensions for the gross internal area, external walls, roof and floors and information about the installed building services. Many inputs are assumed, for example wall insulation depth is assumed based on the construction date. U-values for wall insulation, loft or roof insulation, floor insulation and glazing are inferred, but can also be over-written by the assessor if actual values are known and evidenced. In practice it is rare to over-write inferred data with actual data. This is because there are no direct benefits to the client, it may often be beyond the competency of the assessor and incurs a higher cost for the assessment.

RdSAP has limited flexibility, for example values for airtightness, thermal bridging, ventilation, and occupancy cannot be changed. This means that the impact of these aspects on energy performance cannot be assessed. Ventilation and thermal bridging contribute significantly to a building’s overall heat loss.

Output

RdSAP assessments produce an EPC which gives the property an energy rating. A report is also generated which details recommended steps to improve the property’s SAP rating and the typical financial savings from each improvement. The recommended measures are algorithmically generated, and the software does not consider building condition or the interaction between different retrofit measures.

Cost of assessment estimate

Currently an RdSAP assessment can cost between £50 to £100^[2], dependent on the scope and scale of the property. If a significant amount of work is required to modify SAP for the purpose of whole building assessments, this may increase the cost to building owners.

Estimated cost for a block with 6 flats:

• RdSAP costs for one flat @ £50 to £100 x 6 = £300 to £600

Limitations

Without modifications, SAP cannot be used to assess communal and non-domestic spaces. This means that it is unsuitable for the assessment of multi-owner and mixed-use buildings. SAP is owned by the UK Government. Development of SAP, including any modifications, is currently contracted to BRE, who collaborate with two advisory groups SAPIF (the SAP Industry Forum) and SAPSIG (the SAP Scientific Integrity Group).

The RdSAP assessment is based on an unobtrusive survey based on what the assessor can see at the time of survey. Assessors rarely have access to additional information such as detailed drawings. As such, RdSAP relies on assumptions at both input and output stage.

The recommended retrofit measures are intended to increase the SAP score of a building, rather than to reduce its overall energy use. SAP scores are an energy cost metric, not an energy efficiency or carbon metric. Therefore, an RdSAP assessment will not necessarily recommend measures which target an end goal of a net zero retrofit.

An assumption-based assessment such as RdSAP cannot accurately assess the interaction between recommended retrofit measures. Identifying potential interactions between measures is necessary to understand and predict the impact of retrofit on the movement of heat, air and moisture within a building. The use of an assumption-based tool creates the risk of poor outcomes such as damp or mould, and expensive remediation work. For example, insulation measures to reduce heat loss, air infiltration and air leakage may have unintended consequences for internal air quality and for the movement of moisture through the building fabric. This is a particular risk in pre-1919 buildings, and in non-traditional buildings constructed in the 1960s and 1970s (BSI, 2022).

An assumption-based tool also risks a ‘performance gap’ where predicted energy savings are not delivered in practice.

Currently RdSAP cannot recommend connection to communal or district heating, even when connection to a local heat network is possible (Alembic Research Ltd, 2019).

Option 2: Enhanced energy modelling with PAS 2035

Overview

Option 2 looks at a more detailed assessment approach using either full SAP or Passivhaus Planning Package (PHPP) modelling in conjunction with PAS 2035.

The UK Government has already invested in PAS 2035 as a holistic retrofit approach. By ‘holistic’ we mean an approach that considers a range of building performance issues such as comfort, maintenance, heritage and air quality, rather than energy performance only.

PAS 2035 permits the use of both full SAP and PHPP modelling methodologies when assessing the impact of retrofit measures on existing buildings. PAS 2035 is useful for whole building assessments because it identifies where more detail is required to mitigate the risk of unintended consequences in retrofit. The risk is primarily created by the interaction between existing and new building fabric and services. PAS 2035 tailors the requirement for an assessment based on building characteristics such as age, historic significance of the property, or the number and technical complexity of measures being installed. There is much in common with international examples of frameworks such as the Global Technical Diagnosis (GTD) in France.

PAS 2035 is currently used in the UK on retrofit schemes including ECO4 and is likely to be included in Scottish delivery programmes as part of the Heat in Buildings Strategy (Scottish Government, 2022)^[3]. PAS 2035 has been developed with funding from the Department for Energy Security and Net Zero (DESNZ, formerly BEIS), meaning there is infrastructure in place for training, qualifications and certification in the UK. Additionally, there is a national body, Trustmark, which has been set up to oversee the certification and quality assurance of projects completed in accordance with PAS 2035. Certification is through an online portal which retains information from the building assessment and improvements. This data can inform future works and energy calculations.

Data collection

Using a PAS 2035 approach requires more data collection than an RdSAP assessment. A site visit may take 2-4 hours and is tailored to the construction type and planned measures. The assessment aims to develop a detailed understanding of the construction of the building, any previously installed energy efficiency measures and any defects.

Assessor qualifications

This assessment is carried out by a Retrofit Assessor (RA). This role is a Domestic Energy Assessor (DEA) with an additional RA qualification. The RA is overseen by a Retrofit Coordinator (RC). The RC is a construction professional who uses their judgement and the PAS 2035 framework to determine a suitable level of assessment for any given building.

For SAP modelling there is an existing workforce of qualified On Construction Domestic Energy Assessors (OCDEA). An additional training course for multi-owner and mixed-use buildings could be introduced, as is the case in Canada for EnerGuide assessments.

For PHPP modelling a professional background in buildings and a two-week training course is required. Currently, there are around 600 professionals (Passivhaus designers) qualified to use PHPP in the UK (Passive House Institute, 2023). Significant upskilling of the workforce is required to make PHPP the default calculation methodology.

Data input

Unlike RdSAP, full SAP and PHPP allow assessors to input a greater amount of information about the building, increasing the accuracy of the calculation. Examples of the additional information include details on calculated thermal bridging, measured air permeability and a requirement for thermal calculations for all elements of the building fabric. The two software packages are similar steady-state calculations, however, the user interface for most SAP software is more limited than PHPP. PHPP provides greater flexibility than SAP as it allows for additional data including occupancy, commissioned performance of building services, domestic hot water, appliances, internal heat gains, shading and components. This increased flexibility makes PHPP more suitable as a design tool which can be used by professionals, whereas currently SAP is primarily used for compliance in new buildings.

Output

The output of this approach is an ‘improvement option evaluation’ and a ‘medium term plan’. The improvement option evaluation under PAS 2035 is a report by a RC. The report outlines the current condition of the building, its suitability for receiving retrofit measures and a recommended package of measures to achieve an ‘intended outcome’. If the ‘intended outcome’ was to install a low temperature heating technology, such as a heat pump, the package would likely include significant fabric upgrades. It can be used to assess suitability for communal heating systems, but this is based on the experience of the assessor rather than an automatically generated results.

PAS 2035 requires assessments to present the cost implications of retrofit measures in a simple way, such as a payback calculation for each individual measure.

The medium-term plan sets out the sequencing of installation to ensure that retrofit measures do not impair critical functions of the building such as ventilation, moisture management or heating. If a heating system was proposed for a building, the medium-term plan would set out what preparatory work (such as insulation improvements) would need to occur prior to installation.

Cost of assessment estimate

A PAS 2035 retrofit assessment for a multi-owner or mixed-use building will require several days of time from a Retrofit Assessor, Retrofit Coordinator and possibly other professionals, for example for an air pressure test. It is likely to cost several thousand pounds. An indicative cost of for a block with six flats using PHPP is detailed below^[4]:

• Desktop PHPP costs for one flat @ £500 x 6 = £3000
• Air pressure test for one flat @ £350 x 6 = £2100
• Thermography (all six flats) = £1250
• Temperature and relative humidity monitoring = £500
• In situ U-value monitoring = £500
• Moisture Analysis = varies

Approximate costs = £5000 to £7350

Limitations of PAS 2035 framework

The PAS 2035 approach was developed to improve the piecemeal approach to retrofit which has resulted from RdSAP assessments. However, a key limitation of this option is that the supply chain has not fully matured to deliver the requirements of the PAS 2035 standard. In comparison to DEAs there are relatively few RAs, RCs or installers. However, it is likely that this is a short-term challenge that will be alleviated as the approach becomes more mainstream.

The more detailed assessment is more expensive than an RdSAP assessment. However, it is well documented that increased assessment, design and quality assurance is critical to the success of retrofit measures (Bonfield, 2016).

Limitations of SAP modelling

Without modifications, SAP cannot be used to assess communal and non-domestic spaces. This is because SAP includes background assumptions that only apply in residential properties which cannot be changed.

If SAP were modified for these purposes, it would require more detailed inputs to be used as a design tool. Improvements such as accurate measurement of thermal bridges and better measurement of airtightness have been recommended to DESNZ (formerly BEIS) for SAP 11 (Etude, et al., 2021). However, fully assessing energy performance in buildings for design rather than compliance purposes would require modifications that allow experienced users to edit the default assumptions used by SAP. The changes required to make SAP compatible with whole building assessment calculation of multi-owner and mixed-use building buildings are significant. It would require extensive industry consultation.

Limitations of PHPP modelling

Unlike SAP and SBEM, PHPP cannot currently be used to generate an EPC for benchmarking or compliance. Therefore, buildings would require two separate assessments (one for retrofit design and one for compliance), unless PHPP was permitted as an approved methodology for EPCs. There is currently a collaboration between the Association for Environment Conscious Building (AECB), Passivhaus Trust and Elmhurst Energy to develop a common energy reporting process capable of using either PHPP or SAP.

PHPP software cannot generate capital costs for retrofit measures. Costs could be generated as part of an assessment process if provided by the assessor. This would mean that costs would not be consistent across assessments, but they may be more realistic for the building owners.

Option 3: Best practice from international examples

Overview

Unlike options 1 and 2, this option does not present one specific approach to whole building assessments. Instead, it draws out best practice examples from the countries that were reviewed. Based on these suggestions, it would be possible to design an assessment approach which is tailored to the needs of multi-owner and mixed-use buildings in Scotland, and which would fulfil the identified requirements.

Most of the international examples assess the building as a whole with no distinction between communal and private areas (Table 1). As a result, there are limited examples of best practice for the assessment of individual flats within a whole building assessment.

Data collection

Data collection follows an approach similar to PAS 2035 and is more detailed than for RdSAP. In the examples we found (Section 11), site visits typically take between 1 and 4 hours, during which the assessor gathers information on the type, material, and condition of the building envelope and heating system. This option consists of a risk-based scalable approach to data collection, which ensures that the process is proportionate to the scale of the retrofit being undertaken. This is exemplified by the GTD approach in France, which consists of certain mandatory steps and a number of optional steps. The assessor determines which optional steps are required, based on the complexity of the building and whether it has communal heating. In old buildings an assessor may include the optional step of visiting all flats, rather than the standard approach of visiting a sample of flats. Another optional step is to undertake a week of temperature and humidity monitoring.

Following the approach used in Denmark’s EPC and BetterHome plan, the assessor can also be required to collect secondary off-site data such as building plans or data on conservation areas from the local authorities.

An assessor could visit all flats or make general recommendation for flats on the assumptions that they are similar. Most international examples do not calculate heat loss of individual flats, though some assessments will include visits to a sample of flats. Both TH-C-ex and 3CL in France contain an option for calculations for flats in certain circumstances. For example, within TH-C-ex an individual flat would be assessed if it is arranged differently to the rest of the building. However, the focus of the assessment is communal improvements. Within 3CL there is an option for individual flats to pay extra for a specific assessment of the individual flat.

With the exception of Canada, none of the international examples we found appeared to have a post-installation assessment. Canada’s EnerGuide assessment requires a post-installation assessment to validate that the work has been completed and provide a measure of energy saved and greenhouse gas emissions reduced as a result of the retrofit. The post-occupancy assessment is required before homeowners receive the Canada Greener Homes Grant (funding is provided as a reimbursement). Germany’s Individueller Sanierungsfahrplan (iSFP) and Denmark’s BetterHome plan can both include additional meetings as part of the retrofit development process, but only one data collection audit takes place.

Assessor qualifications

For a detailed approach to data collection and analysis, assessors will need to have experience to meet the complexity of the building. This would need to be regulated.

The requisite skills for a whole building assessment can be provided and assessed in several ways. The GTD in France is normally conducted by a multidisciplinary team consisting of both an architect and a heating engineer since the assessment requires a wide range of skills.

In countries where only one assessor is required, the training typically reflects the requirements for the role, for example by requiring an engineering degree or professional experience in a similar field (Denmark, Germany, Sweden). Canada has a robust training scheme for EnerGuide assessors, and early assessments are audited (which, if failed, can result in the assessor re-training) (Etude, et al., 2021).

Data input

Calculation software for this option is designed to allow assessors to input the relevant variables impacting on energy performance. An example of this is EnerGuide in Canada, which allows assessors to change standard assumptions around occupancy, hot water use or appliance use.

The assessment could also make use of data collected off-site through building drawings, as is done for BetterHome in Denmark.

In terms of assessing communal areas, Flanders stood out in this regard with the EPC Common Parts which is used to assess communal areas alone. A building assessment can also be designed to include considerations for multi-use buildings as exemplified by the Danish EPC assessment which divides multi-use buildings into three zones depending on use: Domestic, Office, and Storage. These zones are used to distinguish different temperature requirements and times of use. The assessment produces a single output based on all zones.

Output

Most of the reviewed international examples provide recommendations at building level rather than for individual flats. On this basis it is possible to make general recommendation for flats on the assumptions that they are similar. Recommendations for individual flats require more thorough assessment, as outlined under ‘Data Collection’.

Option 3 will produce a highly detailed retrofit plan similar to the Building Renovation Roadmap produced in Denmark, Germany, Flanders and France. The assessment provides a staged plan for retrofit leading to a final low-carbon outcome for the building. It considers the interaction between recommended measures. We have not assessed the comparative risk mitigation of the different international models.

Several cost metrics can be included in the assessment output: capital costs of installations, running costs, cost savings and energy savings, cost-benefit over time. Running costs will depend largely on other potential measures that are installed. Cost savings can be less accurate than energy savings due to fluctuating energy prices.

It was not possible to make an assessment of the relative accuracy levels of the models due to lack of access to the specific costing methodologies. As an example, the Danish EPC provides costed recommendations, as well as annual cost savings. Germany’s iSFP contains cost data for each package of measures and includes a comparison of current and future energy costs, as well as CO₂ emissions, energy demand and energy consumption.

EnerGuide in Canada includes energy savings rather than cost savings. Similarly, in France the focus of the GTD is energy savings rather than cost savings. Return on investment figures are provided to building owners, but at a later stage in the assessment process so that energy gains are the primary focus of retrofit decision-making (CoachCopro, 2020).

The examined models did not compare costs of different heating systems. This is likely because cost is not the most important factor in recommending appropriate heating systems. Most of the reviewed countries have more extensive heat network and communal heating infrastructure than Scotland. Therefore factors such as proximity to an existing network may influence recommendations.

Option 3 offers tailored guidance to the building owners. The iSFP assessment in Germany provides two documents for building owners: a renovation roadmap and an implementation guide for measures. BetterHome in Denmark can include an optional retrofit implementation (‘project’) phase where the adviser is responsible for coordinating installation works.

This option requires a framework to allow the assessment to be linked to the building in a national database. This could be achieved through a building passport scheme such as those in Germany, France and Flanders, or the UK’s TrustMark scheme. None of the examples we reviewed appear to have all assessment data inputs available to the building owner. However, some models include more available data than others; for example, Flanders’ building renovation roadmap, Woningpas, includes an online logbook featuring energy performance, renovation advice, and various housing data (BPIE, 2018).

Guidance for building owners could be designed to include information on permissions and warrants required. It is likely that assessors would input this information (as with BetterHome in Denmark), rather than an automated process. Whether a building is listed or in a conservation area could be flagged through an automated process, however a building professional would be required to detail which consents are needed for the proposed retrofit work. Similarly, the requirement for planning permission or building warrants are subjective. The assessment process could flag where these may be required.

Cost of assessment estimate

An assessment will likely cost several thousand pounds per building. Some typical costs for the international examples of assessments are provided below for reference^[5]:

• Typical figures for an energy audit in France in 2011 were €2,500 to €6,000 (£2,219 to £5,326) for a building with 50 or fewer flats (ADEME, 2016). We would anticipate that these costs are now higher.
• The simplified DPE (French EPC) costs between €1,000 to €4,000 (£888 to £3,551) for the whole building (2016 figures) (ADEME, 2016).
• One assessor in Germany advertises the approximate cost of €2,300 (£2,042) for iSFP for a multi-owner building with four flats (Baupal, 2022).

Limitations

The specific methodology for option 3 would need to be designed, alongside work to identify or develop suitable software for the assessment. As a result, the cost of developing the assessment approach will be the highest of the three options. This option also requires significant upskilling of the workforce, similar to option 2.

As with PHPP, it unclear how option 3 would fit within the existing EPC framework. It is possible to draw inspiration from the reviewed countries to design a whole building assessment which is used either alongside an EPC to fulfil a separate purpose (this would require two building assessments) or to replace the existing EPC framework.

Finally, the reviewed international examples are designed for countries with different heating system landscapes. The installation of communal heating systems is less of a policy priority than in Scotland, and therefore there are limited lessons that can be drawn from the reviewed examples.

Summary of options

The differences between the three options are illustrated in Table 2. Explanatory notes are numbered and listed below.

We were unable to obtain any figures relating to the cost of developing an assessment approach. Table 2 indicates whether each option would have a high, medium or low cost of development based on the amount of work required to either modify an existing process or develop a process from scratch.

Table . Performance of the three options against parameters of interest

Explanatory notes for Table 2

Option 1

1. This is an output of an RdSAP assessment but it has too many assumptions to be considered accurate (Etude, et al., 2021).
2. Running cost estimates are provided but are based on outdated energy costs.
3. Capital cost estimates are provided but are based on standardised figures which are only updated with a new EPC.
4. Guidance is for fabric and services only. It does not consider risk, interaction between measures or the adequacy of ventilation. This cannot be considered as a whole building approach.

Option 2

1. Limited workforce for such assessments as PAS 2035 is not yet a requirement in Scotland. There is a large workforce with transferrable skills (to become a Retrofit Coordinator) but this pool of professionals is smaller than the number of current DEAs.
2. Guidance is provided but is not standardised. Guidance is based on the experience of the Retrofit Coordinator and their professional judgement.
3. Communal or individual heating systems would be identified if specified as an ‘intended outcome’ of the assessment.
4. Energy efficiency measures would be recommended for both flats and commonly owned areas if this is specificed as an ‘intended outcome’ of the assessment.
5. Investment is required to facilitate PAS 2035. If PHPP were to be used for the assessment of multi-owner and mixed-use buildings, it would require national oversight in a similar way to SAP.

Option 3

1. We cannot specify an example of best practice as we did not review the data input and data processing of international energy models as part of this scoping research.
2. The only example of post-occupancy assessments we identified was the EnerGuide assessment in Canada.
3. All examples that include more detailed assessment are more likely to be low-regret.
4. Option 3 is designed to enable this. This could be achieved through a building passport scheme such as those in Germany, France, and Flanders (BPIE, 2016), or the UK’s TrustMark scheme.
5. Option 3 is designed to enable this; however we have not identified any examples where the raw input data is stored and available for future use.
6. We did not have access to specific costing methodologies and therefore cannot assess which example is the most accurate. Generally, cost or energy savings were more common than running costs.
7. Estimated capital costs are provided by most international examples. We have not examined the methodologies for calculating capital costs in the international examples.
8. We did not identify examples of this, but it would be possible to include.
9. The different models provide different level of guidance. Best practice examples include Danish BetterHome and German iSFP as these are centred around the customer experience.
10. In the Danish BetterHome, this is included in the practical design of energy efficiency measures, as the assessor investigates relevant legal requirements. It is not automated.
11. GTD in France contains an additional optional study for switching from individual heating systems to communal.
12. This is possible although not common in the reviewed examples.
13. This is possible depending on assessor skills/knowledge and requirements. For example, a heating engineer is involved in GTD assessment in France.
14. GTD in France is a good example of this.

Conclusions

Current assessment approaches

There are two main limitations that prevent SAP and SBEM from assessing multi-owner and mixed-use buildings for the purpose of retrofit design:

• Outputs from SAP and SBEM cannot currently be combined to produce a single calculation for whole multi-owner or mixed-use buildings
• Both SAP and SBEM were designed as tools to demonstrate compliance with energy efficiency aspects of the building regulations. Therefore, both methods are intended for comparative purposes rather than absolute calculations of building performance.

Lessons from international examples

The international examples we reviewed are not approaches that have been developed specifically for multi-owner and mixed-use buildings. In most cases, the approaches are used across all building types (single- and multi-owner). Some, such as EnerGuide in Canada, have additional assessments and training requirements for multi-owner and mixed-use buildings.

The international examples demonstrate that several assessment approaches can co-exist and fulfil different functions (i.e., compliance and design). This can be seen in Denmark and France, which have additional assessment approaches beyond EPCs.

Best practice assessment approaches go beyond energy modelling to consider aspects such as building condition, comfort and air quality. These additional elements should be considered if an approach is to be useful in informing retrofit to achieve net zero. Frameworks such as PAS 2035, iSFP (Germany) and GTD (France) encourage a holistic approach to retrofit planning in this way.

Key considerations for an assessment approach

A whole building assessment should be scalable. This removes the risk of data gathering for assessment purposes becoming disproportionate to the scale of the retrofit.

A solely desk-based assessment is not appropriate for this purpose. A site visit is necessary to gather information. In-situ testing could also be carried out, determined by the complexity of the building.

An assessment for the purpose of retrofit design cannot rely on generic inferred data on occupancy and energy consumption. The assessment could include operational data or use a series of detailed profiles for different building types and household types, similar to those currently used by SBEM.

Due to the complexity of multi-owner and mixed-use buildings, software-generated improvement measures alone are not appropriate for retrofit design. These should either be checked or specifically recommended by a qualified professional.

Developing an approach for Scotland

We outlined three options for how a whole building assessment methodology could be developed in Scotland. The options illustrate a range of costs, both of assessment and development, and a range in the level of detail and accuracy.

Option 1 is a low-cost option, based primarily on assumed data rather than measured data. It involves updating RdSAP to complete a whole building assessment. The advantages of this option are that it can use the existing DEA workforce and the management arrangements associated with producing EPCs.

However, SAP must be modified in order to assess communal and non-domestic spaces. Additionally, an assumption-based assessment such as RdSAP cannot adequately consider the risk associated with the retrofit of multi-owner or multi-use buildings. This may lead to defects, a ‘performance gap’ and unintended consequences such as damp or moud.

Option 2 is a detailed assessment approach using either full SAP or PHPP modelling alongside PAS 2035. This option is a holistic retrofit assessment, rather than an energy performance assessment, and is designed to mitigate the risk of unintended consequences of retrofit.

Infrastructure for PAS 2035 training, qualifications and certification is already being put in place by DESNZ (formerly BEIS) but must be scaled up. Changes to SAP and PHPP modelling approaches are also required for the development of option 2. SAP requires more detailed inputs to be used as a design tool and modifications to assess communal and non-domestic spaces. PHPP requires national oversight arrangements and significant upskilling of the workforce.

Option 3 draws on examples of best practice from the international examples that were reviewed. It has the highest associated development costs as it does not build on any approaches already used in Scotland. Like option 2, option 3 is designed as a holistic assessment approach, which aims to mitigate the risks of retrofitting multi-owner and mixed-use buildings. The main limitation of this option is that direct comparisons cannot be made between the international examples and Scotland, particularly in terms of the management of multi-owner buildings, the prevalence of communal heating and the need to assess individual flats as well as whole buildings. However, there are other areas of best practice that could be incorporated into a whole building assessment approach for Scotland.

References

AC Environnement, 2021. Nouveau DPE : Quelles évolutions pour les logements collectifs?. [Online]
Available at: https://www.ac-environnement.com/actualite/nouveau-dpe-les-evolutions-en-logements-collectifs

Acceo, 2022. DPE Collectif2021 : Décryptage. La nouvelle obligation énergétique en copropriété. [Online]
Available at: https://www.acceo.eu/fr/actualite/115/dpe-collectif-2021-decryptage.html

ADEME, 2016. Copropriétés: viser la sobriété énergétique, France: Agence de l’Environnement det de la Maitrise de l’Energie.

AECB, 2008. Projecting Energy Use and CO2 Emissions from Low Energy Buildings, Barnoldswick, Lancashire: Association for Environment Conscious Building.

Afnor competences, 2023. DIAGNOSTIC TECHNIQUE GLOBAL (DTG). [Online]
Available at: https://competences.afnor.org/formations/diagnostic-technique-global-dtg

Alembic Research Ltd, 2019. A Review of Domestic and Non-Domestic Energy Performance Certificates in Scotland, Edinburgh: Scottish Government.

Arobiz, 2023. Etudes thermiques TH-C-E EX. [Online]
Available at: https://www.arobiz.com/diagnostic/etudes-thermiques.html

Bank of England, 2022. Bank of England Database: Daily spot exchange rates against Sterling. [Online]
Available at: https://www.bankofengland.co.uk/boeapps/database/Rates.asp?Travel=NIxIRx&into=GBP
[Accessed 9 December 2022].

Baupal, 2022. iSFP: individueller Sanierungsfahrplan – Kosten, Nutzen & Förderung. [Online]
Available at: https://www.enter.de/blog/isfp-individueller-sanierungsfahrplan-kosten-nutzen-forderung#was-kostet-ein-sanierungsfahrplan-isfp-mit-bafa-frderung

BDEW, 2019. Wie heizt Deutschland 2019?. [Online]
Available at: https://www.bdew.de/media/documents/Pub_20191031_Wie-heizt-Deutschland-2019.pdf

Beuth, 2018. DIN V 18599-1:2018-09, Berlin: Beuth Verlag.

Bonfield, P., 2016. Each Home Counts: An Independent Review of Consumer Advice, Protection, Standards and Enforcement for Energy Efficiency and Renewable Energy, London: BEIS.

Boverket, 2021. Implementation of the EPBD: Sweden, EU: Concerted Action Energy Performance of Buildings.

Boverket, 2022. Primärenergital och byggnadens energiprestanda. [Online]
Available at: https://www.boverket.se/sv/byggande/bygg-och-renovera-energieffektivt/energihushallningskrav/primarenergital-och-byggnadens-energiprestanda/

BPIE, 2016. Building Renovation Passports: Customised roadmaps towards deep renovation and better homes, Brussels: Buildings Performance Institute Europe.

BPIE, 2018. The Concept of the Individual Building Renovation Roadmap: An in-depth case study of four frontrunner projects, EU: iBRoad.

BRE, 2015. A Technical Manual for SBEM: UK Volume, London: DCLG.

BSI, 2022. PAS 2035/2030:2019 Retrofitting dwellings for improved energy efficiency. Specification and guidance., London: BSI.

Bundesministerium für Wirtschaft und Klimaschutz, 2020. Checkliste Persönliches Gespräch und Datenaufnahme beim ersten Vor-Ort-Termin, Berlin: Bundesministerium für Wirtschaft und Klimaschutz.

Byggahus.se, 2019. Vad kostar en energideklaration?. [Online]
Available at: https://www.byggahus.se/vad-kostar-en-energideklaration

Canada Energy Audit, 2022. ASHRAE Energy Audit Levels 1, 2 and 3. [Online]
Available at: https://www.canadaenergyaudit.ca/commercialbuildingea

Canada Mortgage and Housing Corporation, 2016. Occupied Housing Stock by Structure Type and Tenure. [Online]
Available at: https://www.cmhc-schl.gc.ca/en/professionals/housing-markets-data-and-research/housing-data/data-tables/housing-market-data/occupied-housing-stock-structure-type-tenure

CoachCopro, 2020. Realization of the Global Technical Diagnosis (DTG) Co-ownership, Paris: Agence Parisienne du Climat.

Danmarks Statistik, 2020. Boligbestanden. [Online]
Available at: https://www.dst.dk/da/Statistik/emner/borgere/boligforhold/boligbestanden

Danmarks Statistik, 2022. Bestanden af bygninger. [Online]
Available at: https://www.dst.dk/da/Statistik/emner/erhvervsliv/byggeri-og-anlaeg/bestanden-af-bygninger

DeLanghe, 2019. Renovation of the Belgian property co-ownership law. [Online]
Available at: https://de-langhe.be/en/renovation-of-the-belgian-property-co-ownership-law/

Elmhurst Energy, 2023. ABBE Domestic Energy Assessor (Online Live). [Online]
Available at: https://www.elmhurstenergy.co.uk/product/abbe-domestic-energy-assessor-live-online/

Energimyndigheten, 2020. Ny statistik över Energianvändningen i småhus, flerbostadshus och lokaler. [Online]
Available at: https://www.energimyndigheten.se/nyhetsarkiv/2020/ny-statistik-over-energianvandningen-i-smahus-flerbostadshus-och-lokaler/

Energistyrelsen, 2022. Håndbog for Energikonsulenter (HB2021). [Online]
Available at: https://hbemo.dk/haandbogen

Energistyrelsen, 2023. Hvad koster et energimærke?. [Online]
Available at: https://ens.dk/ansvarsomraader/energimaerkning-af-bygninger/hvad-koster-et-energimaerke

Engie, 2018. Prix du DPE : connaître le budget d’un diagnostic de performance énergétique pour votre logement. [Online]
Available at: https://particuliers.engie.fr/economies-energie/conseils-economies-energie/conseils-travaux-renovation/tarif-dpe.html

Entranze, 2008. Average number of dwellings per building. [Online]
Available at: https://entranze.enerdata.net/average-number-of-dwellings-per-building.html

Entranze, 2008. Share of multi-family dwellings in total stock. [Online]
Available at: https://entranze.enerdata.net/share-of-multi-family-dwellings-in-total-stock.html

EPCInvest.be, 2023. Is your energy expert recognized?. [Online]
Available at: https://www.epcinvest.be/erkend-energiedeskundige/

Etude, et al., 2021. Making SAP and RdSAP 11 fit for Net Zero, London: Department for Business, Energy and Industrial Strategy.

European Commission, 2017. NEEAP 2017 Flanders Annex B Roadmap for the Renovation of Buildings, Brussels: European Commission.

European Commission, 2021. Preliminary analysis of the long-term renovation strategies of 13 Member States. Commission Staff Working Document, Brussels: Council of the European Union.

European Commission, 2023. Building Passport Flanders (Woningpas). [Online]
Available at: https://joinup.ec.europa.eu/collection/egovernment/solution/building-passport-flanders-woningpas/about

European Education Area, 2023. European Credit Transfer and Accumulation System (ECTS). [Online]
Available at: https://education.ec.europa.eu/education-levels/higher-education/inclusive-and-connected-higher-education/european-credit-transfer-and-accumulation-system

Exacompare, 2020. Qu’est-ce que la méthode 3CL-DPE?. [Online]
Available at: https://blog.exacompare.fr/diagnostic-immobilier/quest-ce-que-la-methode-3cl-dpe/

Exacompare, 2021. Combien coûte un diagnostic technique global (DTG)?. [Online]
Available at: https://blog.exacompare.fr/diagnostic-immobilier/combien-coute-un-diagnostic-technique-global-dtg/

Federal Ministry For Economic Affairs And Climate Action , 2023. Energy Efficiency Strategy for Buildings. [Online]
Available at: https://www.bmwk.de/Redaktion/EN/Artikel/Energy/energy-efficiency-strategy-for-buildings.html

Flemish Energy and Climate Agency, 2023. Plaatsbezoek energiedeskundige. [Online]
Available at: https://www.vlaanderen.be/epc-pedia/overzicht-epc-informatie/plaatsbezoek-energiedeskundige

Flemish Energy and Climate Agency, 2023. Taken en verantwoordelijkheden. [Online]
Available at: https://www.vlaanderen.be/epc-pedia/overzicht-epc-informatie/taken-en-verantwoordelijkheden

Flemish Government, 2023. Energy performance certificate (EPC) when selling or renting a residential unit. [Online]
Available at: https://www.vlaanderen.be/energieprestatiecertificaat-epc-bij-verkoop-of-verhuur-van-een-wooneenheid

Flemish Government, 2023. Erkenning tot energiedeskundige type A. [Online]
Available at: https://productencatalogus.vlaanderen.be/fiche/571

Flemish Government, 2023. Explanation of the components of the EPC Common Parts. [Online]
Available at: https://www.vlaanderen.be/epc-van-de-gemeenschappelijke-delen-van-een-appartementsgebouw/uitleg-bij-de-onderdelen-van-het-epc-gemeenschappelijke-delen

Gebaudeforum Klimaneutral, 2022. Antworten auf häufig gestellte Fachfragen (FAQ). [Online]
Available at: https://www.gebaeudeforum.de/service/faq/

Gebaudeforum Klimaneutral, 2022. Grundlegendes zum iSFP, Berlin: Bundesministerium für Wirtschaft und Klimaschutz.

Gebaudeforum Klimaneutral, 2023. DIN V 18599. [Online]
Available at: https://www.gebaeudeforum.de/ordnungsrecht/bilanzierungsnormen/din-v-18599/

Green Home, 2022. Refurbishing residential buildings step by step with an individual refurbishment roadmap (iSFP). [Online]
Available at: https://www.green-home.org/wp-content/uploads/2022/06/8_GREEN-Home_GP_iSFP_EN.pdf

Greener Homes, 2022. Greener Homes Inc – Canada Greener Homes Program. [Online]
Available at: https://greenerhomesbc.ca/pdf/fee-schedule.pdf

Hardy, A. & Glew, D., 2019. An analysis of errors in the Energy Performance certificate database. Energy Policy, Volume 129, pp. 1168-1178.

Hellio, 2023. DPE collectif: mode d’emploi du diagnostic en copropriété. [Online]
Available at: https://copropriete.hellio.com/blog/conseils/dpe-collectif#:~:text=Selon%20les%20chiffres%20de%20l,le%20syst%C3%A8me%20de%20chauffage%20utilis%C3%A9.

Jenkins, D., Simpson, S. & Peacock, A., 2017. Investigating the consistency and quality of EPC ratings and assessments. Energy, Volume 138, pp. 480-489.

Lartigue, B. et al., 2022. Energy performance certificates in the USA and in France—a case study of multifamily housing. Energy Efficiency, Volume 15.

Legifrance, 2021. https://www.ac-environnement.com/actualite/nouveau-dpe-les-evolutions-en-logements-collectifs. [Online]
Available at: https://www.legifrance.gouv.fr/download/pdf?id=7hpbVyq228foxHzNM7WleDImAyXlPNb9zULelSY01V8=

Ministère de la Transition énergétique, 2023. Diagnostic de performance énergétique – DPE. [Online]
Available at: https://www.ecologie.gouv.fr/diagnostic-performance-energetique-dpe

Ministry of Infrastructure, 2019. Sweden’s Third National Strategy for Energy Efficient Renovation, Stockholm: Ministry of Infrastucture.

Natural Resources Canada, 2015. EnerGuide Rating System Technical Procedures Version 15.1, Ottawa: Natural Resources Canada.

Natural Resources Canada, 2023. Sample EnerGuide Assessment Report. [Online]
Available at: https://vancouver.ca/files/cov/sample-energuide-assessment-report.pdf

NIRAS, 2016. Evaluering af Bedrebolig: Indikatorer på effekt, virkning og spredning, Copenhagen: Energistyrelsen.

Passive House Institute, 2023. Find a Passive House Professional. [Online]
Available at: https://cms.passivehouse.com/en/training/find-professional/

Passivhaus Trust, 2020. EPCs as Efficiency Targets, London: Passivhaus Trust.

Public Service France, 2023. Diagnostic technique global (DTG) de la copropriété. [Online]
Available at: https://www.service-public.fr/particuliers/vosdroits/F32059

Quartz+Co, 2015. Energisektorens historiske omstilling og betydning for Danmark, Copenhagen: Quartz+Co.

Robertson, D., 2019. WHY FLATS FALL DOWN, Edinburgh: Built Environment Forum Scotland.

Scottish Government, 2020. Scottish House Condition Survey: 2019 Key Findings, Edinburgh: Scottish Government.

Scottish Government, 2022. Heat in Buildings strategy – quality assurance: policy statement, Edinburgh: Scottish Government.

Scottish Government, 2022. Housing Statistics 2020 & 2021: Key Trends Summary, Edinburgh: Scottish Government .

Scottish Law Commission, 2022. Tenement law: compulsory owners’ associations. [Online]
Available at: https://www.scotlawcom.gov.uk/law-reform/law-reform-projects/tenement-law-compulsory-owners-associations/

Senova, 2016. Quelle simulation thermique pour les audits de copropriété?. [Online]
Available at: https://coproprietes.senova.fr/conseils-techniques/quelle-simulation-thermique-audits-copropriete/

Sierra, F. et al., 2018. Comparison of prediction tools to determine their reliability on calculating operational heating consumption by monitoring no-fines concrete dwellings. Energy and Buildings, 176(1), pp. 78-94.

Smith, N., 2019. Condominium regulation in France. Sofia, CLGE General Assembly 2019.

Solvari, 2023. EPC gemeenschappelijke delen. [Online]
Available at: https://www.epccertificaat.be/epc-gemeenschappelijke-delen

Statistics Canada, 2017. Census in Brief: Dwellings in Canada. [Online]
Available at: https://www12.statcan.gc.ca/census-recensement/2016/as-sa/98-200-x/2016005/98-200-x2016005-eng.cfm

Wohngluck, 2022. Restructuring roadmap: What are the benefits of an individual restructuring roadmap (iSFP)?. [Online]
Available at: https://wohnglueck.de/artikel/individueller-sanierungsfahrplan-isfp-71179

X-tendo, 2022. Innovative EPCs features in Denmark: testing results and replication potential. [Online]
Available at: https://x-tendo.eu/blog/innovative-epcs-features-in-denmark-testing-results-and-replication-potential/

Research Methodology

1. Research Framework

We developed a research framework which contained specific research questions, appropriate search terms and evidence inclusion and exclusion criteria.

A large part of this task was identifying the different terminology for “multi-owner” and “mixed-use” buildings, as well as different terms for “whole building assessment”. Search terms were developed to ensure that our searches would pick up as many international examples as possible, regardless of the different terminology used.

As part of the research framework we also defined “Whole Building Assessment” and identified the component parts and tools which might be used during an assessment (see Table 3).

Table . Research framework: parts of a whole building assessment

1. Scoping

A long list of 22 ‘countries of interest’ was developed. Based on relevance to the Scottish context and the availability of information on their building assessment process this list was distilled down to nine international examples (including PHPP).

‘Similarity to the Scottish context’ was defined in terms of:

• Age profile of the country’s building stock.
• Multi-owner and mixed-use buildings proportion of building stock.
• Ownership and management structures in multi-owner and mixed-use buildings.

1. Data gathering

Data gathering was conducted as a desk-based review of literature available online. The aims of the data gathering exercise were to:

• Gain an understanding of the assessment approaches in the international examples identified.
• Understand potential costs of developing and managing methodologies.
• Gather stakeholder opinions on the options for developing a whole building. assessment approach in Scotland for multi-owner and mixed-use buildings.

We also sought input from expert stakeholders through calls and email correspondence. A total of 42 stakeholders, identified as industry experts, were contacted for input into the research, and 14 provided contributions.

10 Stakeholders from the following organisations contributed through a video call:

• Boverket (Swedish National Board of Housing, Building and Planning)
• Building Research Establishment (BRE)
• Building Research Solutions (BRS)
• Carbon Futures
• GreenGeneration
• Sustenic
• Royal Institution of Chartered Surveyors (RICS)

Stakeholders from the following organisations contributed via email correspondence:

• Danish Energy Agency
• EALA Impacts
• Frankfurt Energy Department
• The Paris Climate Agency

1. Analysis of international examples

Following the data gathering we conductive a comparative analysis of the international examples of assessment approaches. The analysis focussed on the following aspects of the approaches:

• Accuracy
• Reliability
• Cost
• Ease of use
• Necessary qualifications and workforce
• Adaptability (for multiple building types, or the ability to conduct more detailed assessments for complex cases)
• Detail of output for householders

1. Development of options for whole building assessment approach

Based on the research findings we developed three options for how a whole building assessment methodology could be developed. For each option we outlined:

• The types of skills and workforce required
• An estimate of the cost of the assessment for property owners
• The steps needed to develop and manage the assessment approach
• The limitations of the approach

Information on the four factors listed above was not available for every option.

International examples

To outline their relevance to the Scottish context we have highlighted the percentage of multi-owner buildings, and the age of the building stock. The age of building stock gives an indication of the construction type. For example, pre-1919 is used to determine ‘Traditional’ buildings which are associated with solid wall construction methods and materials such as wood and stone.

Different countries use different proxy dates to categorise national housing stock. Some of the reviewed countries had data available for pre-1919 and others pre-1945. In Scotland, 19% of all occupied dwellings (not buildings) were built pre-1919, and 30% pre-1945 (Scottish Government, 2020).

Without further analysis we were not able to access data on multi-owner and mixed-use buildings in Scotland. The available Scottish data relates to individual dwellings. Of all occupied dwellings in Scotland, 37% are tenements or other flats (i.e. the dwelling is one of multiple within one building), and 14% of these are pre-1945 (Scottish Government, 2020) (Table 4).

Of Scotland’s 2.6 million dwellings, 15% are private rented (or the household is living rent free), and 23% are social rented properties (Scottish Government, 2022).

We have also highlighted any relevant similarities or differences in terms of building management and ownership structures, and the prevalence of communal heating systems.

Table . Proportion of Occupied Dwellings in Scotland by age and type. Data from SHCS 2019

Canada

Framework: EnerGuide

EnerGuide Ratings (expressed in gigajoules per year) are used for national benchmarking of building performance. There is an additional level of service which is used to produce Renovation Upgrade Reports (Natural Resources Canada, 2023) for the purpose of retrofit.

Energy model: HOT2000

Strengths:

• Based on an energy usage metric rather than cost, which is more useful for the purpose of retrofit design.
• Considers both regulated and unregulated energy use.
• Uses specific efficiency values for systems, which improves the accuracy of energy calculations (in comparison to using assumed values).
• Includes consideration of thermal bridging (inferred from wall, floor and roof construction types) and overheating.
• Assessors provide comments and guidance specific to the property and based on building condition.

Limitations:

• Assessors’ recommendations lack detail when compared to fuller energy audits (Wohngluck, 2022), (BPIE, 2018).
• Flats do not get an individual assessment.
• Prices for retrofit measures are not provided.
• Used for low rise multi-owner buildings only (three or fewer storeys). High rise buildings are assessed as commercial buildings using an ASHRAE energy audit.

Building stock and heating infrastructure: A similar number of multi-owner buildings to Scotland (34%) (Statistics Canada, 2017). The majority (70%) of multi-owner buildings are rented (Canada Mortgage and Housing Corporation, 2016). Most multi-owner buildings have individual heating and cooling systems in each flat, although some older buildings have communal gas boilers.

Building Management: Condominium associations are a legal requirement and are used to manage the repairs and maintenance of all common areas. They are managed by an elected board of directors (‘Condo Board’). In practice most Condo Boards employ a management company.

Regulatory Context: There are no national energy requirements for buildings. City and provincial authorities may implement their own regulations. EnerGuide assessment is a requirement for accessing the national Greener Homes Grant. EnerGuide and HOT2000 are also used to assess compliance with voluntary standards such as BC Energy Step Code (a standard in British Columbia which seeks to go beyond legal Building Standards).

Assessment Process: The entire building structure, including all units and common areas, is assessed in a single assessment. Flats do not get an individual assessment. EnerGuide is only used for blocks with fewer than 6 apartments. Larger blocks are treated as commercial buildings and are assessed using ASHRAE Energy Audits (Canada Energy Audit, 2022).

The assessment includes:

• Visual inspection and measurement to determine surface area and insulation levels.
• Manufacturer efficiency values (from appliance manuals) for mechanical systems (heating system, air conditioning system, ventilation system, and domestic hot water). If manufacturer figures are not available default values from HOT2000 are used.
• Blower door test to detect air leakage and measure air changes per hour.
  • Depressurisation test where required. This is to test for combustion spillage which is the flow of harmful combustion gases (such as carbon monoxide) back into the home. This is a risk in buildings with high levels of air-tightness and inadequate ventilation.
• The EnerGuide rating (used for national benchmarking) rates the house independent of occupant behaviour. However, the occupancy and energy usage of households is incorporated for calculating the ‘Estimated Household Energy Use’ figures on the report for householders. These figures are also considered as part of the recommendations of measures.

Accuracy: Heat demand is modelled rather than measured. Where possible efficiency values of mechanical systems are used as inputs, rather than assumed values. However, the efficiency stated by manufacturers will differ from the actual in-use efficiency of appliances.

Intrusiveness: A pre- and post- assessment are carried out, each lasting 2-3 hours. Air blower test requires prior preparation from the householder, and for fuel-fired heating or water systems to be switched off. Householders are asked to provide appliance manuals and to comment on any existing problems and planned renovations.

Improvement Recommendations: A roadmap of improvements is provided. These are a combination of recommendations generated and prioritised by HOT2000, and some suggested by the assessor. Recommendations are prioritised to be fabric first based on house-as-a-system concept.

Costs for improvements are not provided in the Renovation Upgrade Reports. Potential energy reduction figures are given.

EnerGuide does not flag any legal requirements as part of its recommendations because building codes and by-laws differ by province in Canada.

Heating system considerations: EnerGuide is closely linked to the Greener Homes grant programme. Under the grant programme multi-owner buildings (known as MURBs) with three or more units are not eligible for heating upgrades, but all other measures are eligible. To make an assessment for communal heating a more in-depth survey would be required^[6].

Adaptability: EnerGuide is used for both multi-owner and mixed-use buildings. In mixed-use building an additional risk assessment is required. This ensures that the non-domestic space can be appropriately assessed alongside domestic spaces. It highlights any precautions that may need to be taken in the assessment, for example to account for processes and equipment that generate a large amount of heat or building configurations which may impact on blower door tests (Natural Resources Canada, 2015)^[7].

Costs: Not regulated and vary dependent on the assessment organisation. Based on the costs from one provider (Greener Homes, 2022) the pre and post assessment for a block with 5 flats would cost $2500 (+ tax) (approximately £1500).

Qualifications and Training: Assessments are completed by government registered energy advisors, who must pass two exams and be affiliated with a licensed Service Organisation. There is an additional exam for MURBs. There are no particular pre-requisites for training.

Denmark

In Denmark, the national energy calculation programme Be18 is used to produce two different outcomes: an EPC, and a BetterHome plan. These two frameworks each address different aspects of the energy efficiency improvement process.

Building stock and heating infrastructure: The building stock is similar to that of Scotland. 41% of dwellings are in multi-owner buildings (Entranze, 2008), and 21% of these dwellings were built before 1919 (Danmarks Statistik, 2020). Unlike Scotland, most multi-owner buildings are supplied by district heating (66% of the entire building stock) (Danmarks Statistik, 2022), (Quartz+Co, 2015). Two thirds of this heat come from combined heat and power plants.

Building management: In multi-owner buildings with individual ownership of the units, all owners are legally required to be part of an owners’ association which is responsible for the maintenance of common spaces and heating systems. In urban areas it is also common for buildings to be owned by a housing cooperative, which in those cases are responsible for common spaces and heating systems.

Regulatory context: EPCs have been issued in Denmark since 1997. The current scheme has been in place since 2006 following the implementation of EPBD. Unlike Scotland, the whole building is assessed. EPCs are a legal requirement when a building is rented, sold, or built. An EPC is valid for 10 years. Access to and guidance on the calculation programme Be18 is delivered through SBi instruction 213.

Denmark: EPC

Framework: EPC (energimærkeordningen)

Energy model: Be18

Strengths:

• The calculation software is flexible and allows for data from in-situ testing to be included where it is available.
• Emphasis on heat-loss through thermal bridges results in a realistic account of the building’s energy efficiency (similar to PHPP).
• Highly specific regulations ensure uniformity across buildings.
• Calculation programme allows detailed descriptions of individual building parts, such as distinct U-values for eight different window types (similar to PHPP).
• Recent framework changes have made the recommendations more specific and easier to implement.

Limitations:

• Recommended improvements and cost-savings may not match the experience of the householder.

Assessment process: The assessment covers the whole building and requires access to common spaces such as loft, basement, and stairwell. It is not necessary for the assessor to visit the individual flats if sufficient information is provided by the owner. The calculations account for a large number of parameters. The energy consultant must follow a strict set of rules outlined in ‘Handbook for Energy Consultants’ (Energistyrelsen, 2022).

Accuracy: The heat demand of the building is based on a visual inspection and standardised assumptions, rather than in-situ tests and consumption data. The calculation takes into account a large number of inputs relating to the material and condition of the building envelope. The consultant is required to assess whether the observed data matches the building drawings and registered data.

Intrusiveness: The assessment is carried out during one visit lasting one hour or more, depending on the size of the building. If there is not enough data available, and if the owner consents, a ‘destructive’ assessment may be carried out. This could include drilling into the wall to determine insulation type and thickness.

Improvement recommendations: Improvement recommendations are divided into ‘cost-effective’ and other improvements. ‘Cost-effective’ recommendations are defined as those where the associated savings cover the cost of the investment before the component must be replaced. As such, the definition includes an estimate of the lifespan of the components of the energy saving investments. Recommendations in this category are costed, and annual savings are estimated. The recommendations are considered to be highly tailored, with distinct recommendation for flooring, walls, loft, insulation, heating system, and electricity (X-tendo, 2022). Additionally, an EPC must include if there is potential to install solar PV and heat pumps. The assessor is not required to advise on potential required planning permissions.

Heating system considerations: If the building is not already part of a district heat network, the assessor must consider if it is possible to achieve energy improvements by upgrading the boiler, changing the boiler type, installing solar PV or heat pump, changing the heating system, or connecting to an existing district heat network. EPCs are not currently required to include considerations for other communal heating systems.

Adaptability: The same framework and calculation programme is used for residential, commercial, and mixed-use buildings. For mixed-use buildings, the EPC calculation can include different ‘zones’ to account for the different energy needs of the building. There are three zones: residential, office, and storage. EPCs are issued for both existing and new buildings.

Costs: The Danish authorities sets a maximum assessment cost for smaller buildings within different size brackets (Energistyrelsen, 2023). The figures below are converted based on Bank of England Exchange rates (Bank of England, 2022):

• <100 m²: £724
• 100-199 m²: £797
• 200-299 m²: £869

There is no upper cost limit for larger residential and commercial buildings.

Qualifications and training: EPCs are carried out by a certified energy consultant who must be employed by a certified energy certification company. Energy consultant certification is achieved by attending course worth 10 ECTS (European Credit Transfer and Accumulation System) (European Education Area, 2023). Two years of experience in a relevant field is required.

Denmark: BetterHome

BetterHome is a one-stop-shop initiative with the aim to make it easier for building owners (such as home-owners, housing cooperatives, and owner associations) to retrofit of their home. It was initially designed for single-family buildings but has been expanded to include multi-owner buildings as of 2017. It is a voluntary, market-based scheme developed by four companies and the Danish Energy Agency. The scheme contains two parts: a plan, which provides an overview of potential improvements and costs, and a project, where the adviser coordinates the installations from start to finish (X-tendo, 2022).

Framework: BetterHome (BedreBolig)

Energy model: Be18

Strengths:

• Provides a very high quality of recommendations that are tailored to both the needs of both the householders and the building.
• The project phase of BetterHome makes the installation of retrofit measures more achievable and successful.
• Supports householders to achieve a higher EPC rating.

Limitations:

• BetterHome is more expensive and time-consuming than an EPC, and it requires a high level of engagement from the building owner.
• It is unclear how the process is carried out in larger multi-owner buildings with several owners with different priorities and energy usage.

Assessment process: The BetterHome assessment uses the same calculation programme (Be18) as an EPC assessment, and an existing EPC is commonly used as part of the assessment input. The process is longer and with a greater focus on the specific needs and interest of the building owner. If the BetterHome plan is approved by the owner, the advisor goes on to develop and coordinate the retrofit project from start to finish (NIRAS, 2016).

Accuracy: The assessment uses similar input to EPC (see above).

Intrusiveness: The duration of the assessment varies widely depending on the size of the building and the needs of the owner. Because a BetterHome plan is optional, the assessment is not required to consider all parts of the building unless requested by the building owner. Multiple meetings will follow the assessment if the building owner decides to pursue the project stage.

Improvement recommendations: The main output of the BetterHome plan is a list of recommended actions, such as insulation of pipes or certain parts of the building fabric. These are tailored to the priorities of the building owners and, unlike the EPC, can include recommendations related to energy use and habits.

Heating system considerations: There are no legal requirements for the heating system to be considered, but a BetterHome plan will typically include suggestions relating to the heating system. As noted above, most multi-owner buildings are already connected to a district heating network. If connection is not possible, ground- or air-source heat pumps may be recommended.

Adaptability: BetterHome was originally developed for single-family homes only. In 2017 it was expanded to include multi-owner buildings. It was difficult to find information about how larger multi-owner building owners and owner associations are engaging with the initiative. It appears that there is an option to request a BetterHome plan for only part of a building.

Cost: The cost of a BetterHome plan is around £700, though it may be higher for larger buildings (NIRAS, 2016).

Qualifications and training: The BetterHome plan and project is carried out by a certified BetterHome advisor. Certification is achieved by taking a specific BetterHome course, which requires at least two years of experience in the field.

Flanders, Belgium

Framework: EPC

Energy model: Software for EPC common parts

The Flemish Government and Climate Agency have an approved certification software programme for the calculation of the EPC common parts (Flemish Energy and Climate Agency, 2023). It takes into consideration the building envelope and any communal space heating, hot water, ventilation, lighting or solar (Solvari, 2023). The method is primarily used to encourage staged retrofit at the communal level.

Strengths:

• Supports co-owners to consider upgrades at the building level.
• Options to use known input data or standardised assumptions.
• Relatively inexpensive.
• Optional additional testing available.
• Woningpas (Building Passport) digital file stored on a government database (European Commission, 2023).

Limitations:

• Energy experts cannot amend the recommendations that are produced by the software.
• Prices are not provided.
• Does not consider how individual dwellings interact with communal areas.

Building stock and heating infrastructure: In both Scotland and Flanders approximately 14% of flats were built pre-1945 (European Commission, 2017)^[8]. The average number of dwellings per multi-owner building in Flanders is 6.5 (European Commission, 2017). Most buildings are heated with natural gas boilers. Like in most other of the European countries we consulted, multi-owner buildings are usually heated by one boiler located in the basement (unless connected to a district heat network) – this system is called ‘central heating’ outside of the UK.

Building management: Owners of an apartment within a multi-owner building own their individual dwelling and a share of the common areas. An owners’ assembly takes place at least once per year.It is common for the owners’ association to contribute to a reserve fund to assist with large one-off works(DeLanghe, 2019).

Regulatory context: Upon selling or renting a property, both an individual apartment EPC and an ‘EPC common parts’ are required (Flemish Government, 2023). EPC common parts was introduced in 2022 to inform the owners of each residential unit about the collective steps they can take to make the building more energy efficient. An energy efficiency ‘grade’ or ‘class’ is not supplied.

Assessment processes: The heat demand and heat cost efficiency are calculated using the approved software. The energy expert must follow strict set of rules and working methods outlined in an ‘inspection protocol’ (Flemish Energy and Climate Agency, 2023).

Accuracy: Heat consumption is based on standardised assumptions and bills are not required from the householders. Heating demand is calculated from several inputs based on the actual condition and observed information from the building. For example, specific boiler types, window makes, insulation thickness etc. can be recorded or looked up using the software and included in the calculation. No physical testing takes place.

Intrusiveness: One visit takes place and although no physical testing is required detailed information on the characteristics of the insulating envelope of the building (walls, floors, ceilings etc) and building condition (year of construction, type of building, etc) are collected. Energy experts therefore request that, where possible, as much of this information is provided to them prior to the visit (Flemish Energy and Climate Agency, 2023). This data collection exercise could be time consuming for householders.

Improvement recommendations: Although no physical testing is required, due to the large number of inputs, recommendations are relatively detailed. However, they are generated automatically, and the energy expert cannot adjust, remove or change the order of the recommendations (Flemish Government, 2023). Capital costs are not included. The recommendations therefore provide a first level of information to owners regarding communal works, but it is recommended that as a next step owners get a construction professional or architect to use the EPC to assist them in planning the most logical execution of the necessary works.

Heating system considerations: It was unclear if the EPC Common Parts framework requires that the energy expert makes specific recommendation relating to the heating system and, in that case, what those recommendations could look like.

Adaptability: The model can be used for new and existing residential multi-owner buildings. Owners can request that additional measurements are taken (e.g., checking the type and thickness of internal wall insulation) but this is not a requirement and standard assumptions can be used instead.

Costs: Estimated costs for the EPC common parts assessment are as follows (Solvari, 2023):

• >16 units € 600 + € 10 per apartment
• 5-15 units € 300 + € 20 per apartment
• 2-4 units € 300 + € 15 per apartment

The certificate is valid for 10 years but expires if major retrofit is conducted (e.g., 15% of the building envelope is insulated, any collective heating is replaced). The cost of the assessment is not regulated.

Qualifications and training: only an energy expert ‘type A’ can carry out a residential EPC inspection (Flemish Government, 2023). The course and exam can be taken by anyone. Those with a ‘relevant’ background degree e.g., an engineer or architect can take an accelerated course (EPCInvest.be, 2023).

France

We identified two energy models being used for whole building assessments in France: 3CL and TH-C-ex, which are discussed in turn below.

Building stock and heating infrastructure: A similar proportion of buildings are multi-owner (44% of dwellings are multi-owner in France, and 37% of buildings in Scotland.) Multi-owner buildings are generally older in France than they are in Scotland. The average number of dwellings per multi-owner building is 7.6 (Entranze, 2008). Communal heating (known as collective heating) is widely used in multi-owner buildings.

Building management: The ownership structure differs to Scotland in that there is a legal requirement to form a ‘copropriete’ (co-ownership) (Smith, 2019). For buildings in co-ownership, retrofit measures such as heating upgrades must be voted on unanimously.

Regulatory context: Under the Energy Performance of Buildings Directive (EPBD) France developed the Diagnostic de Performance Energétique (DPE), the French equivalent of an EPC. Unlike Scotland, all DPEs are conducted at building scale (Ministère de la Transition énergétique, 2023). Additional regulations for multi-owner buildings are also in place. For example, a Global Technical Diagnosis (GTD) is mandatory if:

• The building presents a danger to the health or safety of occupants.
• A building is more than ten years old and is newly divided into flats.
• Co-owners vote for a GTD by a simple majority during their general meeting (Public Service France, 2023).

France: 3CL

The 3CL calculation is a steady-state energy model used for the DPE. It calculates the energy consumption of the building using a large amount of input data (Lartigue, et al., 2022). To make the calculation of the energy consumption easier, the French Government has certified several software packages which are based on the 3CL calculation.

Under the DPE, the calculation method is primarily used for benchmarking purposes. It can also be used for retrofit design under the GTD^[9].

Framework: EPC (known as DPE)

Strengths:

• Provides a first level of information to owners for a low modelling cost (Senova, 2016).
• Offers the possibility to create an individual DPE from the DPE-collective but this is only possible where similar heating, cooling, domestic hot water and ventilation systems are in place (e.g., communal heating, or where all flats have the same heating systems).
• Includes a temperature profile based on geographical location. This means that calculations will be more accurate than models that rely on a standardised temperature profile for a large region or country.

Limitations:

• The method is based on standardised assumptions of occupancy and occupants’ behaviour (Exacompare, 2020). This is only a limitation if the assumptions cannot be edited to inform further analysis.
• Does not provide sufficient accuracy to effectively improve the energy performance of buildings.

Assessment processes: To carry out a DPE at the building scale there are minimum requirements around the number and location of individual units in a building that must be visited as part of the assessment (AC Environnement, 2021). Heat demand is calculated for the whole building, rather than individual flats. This calculation accounts for a large variety of parameters including heat losses through the building envelope; heat transfer due to air exchange; energy consumption of the ventilation auxiliaries; solar gains; and thermal inertia. Heat consumption is calculated from the heating demand by considering the power efficiency of the heating system(s) and the heating degree hours based on geographical location (Lartigue, et al., 2022), (Legifrance, 2021).

Accuracy: The 3CL method does not require in-situ testing or consumption monitoring. It uses a mix of measured and modelled data as observed information can be input in the software (e.g., the type of heating system and its age). The inclusion of geographical location is also included. This means that calculations will be more accurate than models that rely on a standardised temperature profile for a large region or country.

Intrusiveness: One visit takes place, and the assessment is relatively unintrusive as no in-situ tests are required. The accompanying building inspection is moderately detailed and considers characteristics of the insulating envelope of the building (walls, floors, thermal bridges etc.) and building properties (region, altitude, orientation etc). Approximately one hour per 100m² is required by energy experts (Engie, 2018).

Improvement recommendations: Recommendations are made by the assessor based on the outputs from the software. They include estimated costs; cost savings and priority works (Ministère de la Transition énergétique, 2023). The recommendations provide a first level of information to owners but are not detailed enough to support significant improvements in the energy performance of the building (Acceo, 2022).

Heating system considerations: 3CL is used to assess buildings with communal heating. We could not ascertain whether 3CL would be used to recommend communal heating in a building where flats currently have individual heating systems.

Adaptability: The model is adaptable as it is used for all buildings receiving a DPE. This means new and existing domestic (including multi-owner) and non-domestic (including mixed-use) buildings are assessed using the same method. It can also be adapted for different heating set ups – both communal and individual systems.

Costs: The DPE costs between €1,000 and €4,000 for the whole co-ownership (ADEME, 2016). Assuming the average number of dwellings to be 7.6 this equates to €132 – €526 per owner. DPE is valid for 10 years and the price is not regulated.

Qualifications and training: The collective DPE must be carried out by a certified diagnostician. This expert must have professional liability insurance and hold a DPE certification “all types of buildings”. The level of qualification required is higher than that of an individual apartment DPE (Hellio, 2023)^[10]. In addition, diagnosticians must use the approved calculation software.

France: TH-C-ex

TH-C-ex was developed and defined by the Centre Scientifique et Technique du Bâtiment (CSTB) and is used for energy audits of existing buildings. The calculation is completed by using approved software (Arobiz, 2023) and TH-C-ex is used for several types of energy audit including the Global Technical Diagnosis (GTD) (Public Service France, 2023)^[11].

Under the GTD, TH-C-ex is used to support multi-owner property upgrades and retrofit. TH-C-ex is also used to assess if renovations meet the Passivhaus Standard or the BBC Label (low consumption building label).

Framework: Global Technical Diagnosis (GTD)

Strengths:

• The GTD is a more thorough process than that used to produce a DPE as behavioural calculations and physical measurements are considered (including an architectural audit).
• Recommendations are detailed and organised into staged packages of works.
• The framework outlines base and optional assessments and either steady-state or dynamic modelling can be used.
• It is a good example of targeting expertise in the most complex buildings to give retrofit advice.

Limitations:

• Building usage scenarios (e.g. assumed heating temperatures) are standardised. The values used do not reflect commonly observed heating practices (Senova, 2016).

Assessment processes: The assessment to conduct the GTD uses precise calculation formulas. These determine the primary energy consumption of an existing building accounting for heating, cooling, lighting, ventilation, auxiliaries and the preparation and storage of DHW. The heat demand and heat cost efficiency of the whole building are calculated using the software. Consumption is presented in both delivered and primary energy.

Accuracy: The GTD requires a complete analysis of the existing property and the software uses a mix of modelled and measured data (similar to PAS 2035). Physical tests and observed information are gathered to provide data on:

• indoor temperatures
• wall temperatures
• indoor humidity
• masonry thickness of exterior walls
• thickness of visible insulation
• height under ceilings
• measurement of ventilation flows (if ventilation has ducts)
• wall humidity level

Where possible, actual consumption is included in the software and owners supply energy bills from the past three years (CoachCopro, 2020). The model is more accurate than 3CL due to the ability to include data from physical tests.

Intrusiveness: One visit takes place (an energy and full architectural audit), and physical tests are required. Details from the architectural audit are not included in the TH-C-ex calculation however they are used to inform the recommendations made to householders. Owners can also opt for a more detailed weeklong ‘measurement campaign’ which would require two visits.

Improvement recommendations: Recommendations are automatically produced by the software. They include costs, return on investment and priority works. Potential financial aid is also indicated in a general manner without quantification. The recommendations are then tailored by the auditor to support significant improvements in the energy performance of the building over a ten-year period. They are organised into three categories (CoachCopro, 2020):

• Priority 1 – short term or urgent works
• Priority 2 – medium term
• Priority 3 – long term

Auditors are required to produce at least two different work plans for the ten-year period (CoachCopro, 2020).

Heating system considerations: TH-C-ex uses static thermal modelling as standard. There is an option of dynamic thermal modelling if required by the complexity of the building or intended outcomes. The methodology refers to an additional optional study for switching from individual heating systems to communal (and vice versa) (CoachCopro, 2020).

Adaptability: TH-C-ex is appropriate for existing buildings only, and a separate model is used for new buildings. This means that existing mixed-use and multi-owner buildings can be assessed using the same method. However, the GTD framework is specifically aimed at multi-owner residential buildings so it is unclear whether this could also be used for mixed-use or non-domestic properties.

Costs: Estimated costs for the full GTD assessment are extremely varied and depend on the number of dwellings as well as the type of firm or professional hired. The adaptability of the method with base and optional assessments also leads to variation in costs. The GTD was reported as more expensive than the DPE assessment. For smaller buildings (i.e., 4-7 units) the cost is approximately €1,200 and for larger buildings (i.e., 16-19 units) €2,700. Estimates include auditor fees, travel, and the submission of the report (Exacompare, 2021).

Qualifications and training: the TH-C-ex calculation must be carried out by a certified professional. The full GTD course is approximately 35 hours over five days, which is similar to the PAS 2035 Retrofit Coordinator course. A three-year diploma in a ‘relevant’ field (e.g., certified real estate diagnostician or thermal engineering qualification) is required as a pre-requisite for the course (Afnor competences, 2023).

Germany

Framework: Individueller Sanierungsfahrplan (iSFP)

Energy model: DIN V 18599 in iSFP software

DIN V 18599 is the national calculation methodology used in Germany to assess the energy performance of buildings^[12]. It calculates the useful, final and primary energy requirements for heating, cooling, ventilation, domestic hot water and lighting (energy balance) of buildings (Beuth, 2018). Since 2017, it has also been used for iSFP software to produce an individual building renovation roadmap (iSFP).

Strengths

• Accessible and easy to understand documents (Wohngluck, 2022).
• Offers options for staged or one-off energy renovations.
• Financing available for the audits (BPIE, 2018).
• Strong focus on consultation with co-owners and designing for their needs.

Limitations:

• Lacks depth in comparison to more complete energy audits (which provide 150-page reports) (Wohngluck, 2022), (BPIE, 2018).
• It is unclear how many inputs are required for the calculation and how in-depth the measured data needs to be.
• It provides a snapshot that may become redundant over a short period of time.

Building stock and heating infrastructure: Around a fifth of multi-owner buildings are pre-1945, similar to Scotland (14%) (Entranze, 2008). The average number of dwellings per building is 7.7 (Entranze, 2008). Like in Scotland, natural gas boilers are the most common heating system (48%). 14% of dwellings are supplied by district heating (BDEW, 2019). Most buildings have a building-wide heating system and communal boiler, rather than individual boilers in each dwelling. This heating system is called ‘central heating’ (Zentralheizung) and can use either a gas or oil boiler.

Building management: Owners form a legally-required condominium association which is a self-governing body that meet on an annual basis to vote on any building related issues.

Regulatory context: In 2017 Germany introduced a new software-based tool for retrofit called an individual building renovation roadmap (“Individueller Sanierungsfahrplan” – iSFP). This is an optional assessment that can be used to improve the energy efficiency of buildings (Federal Ministry For Economic Affairs And Climate Action , 2023).

Assessment processes: the iSFP assessment follows a seven-step process.

1. Initial consultation with owners, data on the building condition and services are recorded and user requirements are discussed.

2. The energy performance of the building is calculated using balancing software.

3. Based on the meeting with owners, the data and calculations, refurbishment proposals are drafted.

4. The proposals are discussed with the owners to agree on the final refurbishment concept.

5. The iSFP and the implementation instructions are elaborated in detail.

6. The iSFP is printed and handed over to the owners.

7. The iSFP and the individual documents are explained, and questions clarified in a final meeting with the owners. Two documents are provided – a renovation roadmap and guidelines for renovation measures (Gebaudeforum Klimaneutral, 2022).

Accuracy: The inputs are based on a mixture of modelled and measured data. Where actual data is available it can be inserted into the calculation. For example standard or specific indoor temperatures can be included in the model (i.e., room temperature set on the thermostat) (Bundesministerium für Wirtschaft und Klimaschutz, 2020). Detailed information in relation to the building condition is also gathered. It considers:

• the envelope (walls, roof, windows, floors)
• systems (heating, hot water preparation, heat and hot water distribution, including storage and transmission, ventilation)
• ‘quality assurance’ (thermal bridges and airtightness)

Where possible, bills are required from the householders; however, these are only used to calculate energy costs. Where bills are unavailable, “typical consumption” values are used (Gebaudeforum Klimaneutral, 2022)^[13].

Intrusiveness: Two visits take place. One is for data collection and initial consultation, and the second is a meeting at handover stage. It is recommended that half a day is needed for the first visit. The assessment is relatively unintrusive.

Improvement recommendations: The iSFP has been designed to be a user-friendly tool that includes both short and long-term measures and suggests ways to avoid lock-in effects. Recommendations include estimated costs, cost savings and priority works. Auditors design a comprehensive package of measures to achieve deep renovation considering the owners’ specific needs with the aim of successfully nudging them to initiate deep renovations. The recommendations are tailored to support significant improvements in the energy performance of the building over a fifteen-year period (Green Home, 2022).

Heating system considerations: It was unclear if there are specific requirements relating to heating system upgrades. Common heating system upgrades are from communal oil to communal gas boiler, or from communal gas boiler to district heating, if available (BDEW, 2019).

Adaptability: The calculation method (DIN V 18599) is adaptable as it is used for all buildings in Germany (Gebaudeforum Klimaneutral, 2023). However, the iSFP software is only used for residential buildings as it was developed to support the residential market to retrofit their properties (BPIE, 2018).

Costs: Estimated costs for the full iSFP assessment are extremely varied and depend on whether it is a single or multi-owner home. The German government currently offer up to 80% funding support for the assessment which is capped at €1,700 for multi-owner buildings. An assessment of a multi-owner building with four dwellings is around € 2,300 (Baupal, 2022).

Qualifications and training: The iSFP must be carried out by a building energy consultant who has completed specific training (Baupal, 2022). Those from a ‘relevant’ background e.g., electrical engineer, construction specialist, architect, real estate expert can apply to the course (this is similar to PAS 2035).

Sweden

Like in Denmark, EPCs are issued for the whole building. Unlike many other European countries, the Swedish EPC is based on measured delivered energy, rather than modelled heat demand.

Framework: EPC (energideklaration)

Energy model: Unspecified

Strengths:

• Reflects how energy is actually used in the property.
• Improvement recommendations are likely to have a clear impact on the householder.
• Includes a geographical adjustment factor to account for temperature differences across the country.

Limitations:

• The quality of the recommendations can vary widely depending on who carried out the EPC assessment.
• Relies on normalisation of values to enable comparisons between buildings.

Building stock and heating /infrastructure: Sweden has a large share of multi-owner dwellings (58%) (Entranze, 2008). 20% of multi-owner buildings were built before 1930 (European Commission, 2021). Most multi-owner buildings are already connected to district heat networks (90%). Electric heating including heat pumps account for 8% (Energimyndigheten, 2020).

Building management: Sweden’s housing stock is characterised by a large degree of public and co-operative ownership (68%). Only 32% are under private ownership (Ministry of Infrastructure, 2019). Properties in Sweden typically have ‘warm rent’, where heating and hot water costs are included in the rent. This puts the incentive of energy efficiency improvements on the owner rather than the tenant^[14].

Regulatory context: A valid EPC must be available when a building is sold, rented, or built. Guidance and regulations are set out by the National Board of Housing, Building, and Planning (Boverket). Boverket sets out a baseline of inputs required in an EPC assessment, but the full extent is up to the individual energy export and depends on the information provided by the building owner.

Assessment process: An assessment of the building’s heat demand is carried out by an independent, certified energy expert. An in-situ inspection of the building is required and must consider the building’s orientation, passive solar radiation, and the climate of the location (as outdoor temperatures vary greatly across Sweden). Thermal properties are taken into account, including U-values of roofs, walls, windows, and outer doors; cold bridges; and the airtightness of the building envelope. The energy expert also notes the type and condition of the heating, hot water, and ventilation systems. The general condition of the building is not considered.

Accuracy: The energy expert must verify that the information provided by the owner aligns with the visual inspection. There is no national calculation programme; it is up to the energy expert which programme is used. The assessment includes a geographical adjustment factor to account for different outdoor temperatures across Sweden. This means that buildings in some regions are allowed to have a higher heat demand compared to the requirements (Boverket, 2022). Additionally, different fuel types are weighted differently, with electricity being weighted higher. In 2020, the energy performance calculation methods were changed to make buildings more comparable regardless of the fuel type used for heating (Boverket, 2021).

Intrusiveness: The assessment is carried out during one visit, lasting from 30 minutes and up to a few hours depending on the size of the building. The data collected depends on where was accessible on the day, but the energy expert is not required to access individual units.

Improvement recommendations: Recommendations are made for the building as a whole, though some may be relevant to individual flats, such as recommendations regarding taps and radiators. As the regulations are limited, Swedish EPCs can contain either few or highly detailed recommendations. This is reflected in the cost of the EPC (Byggahus.se, 2019). A benefit of using actual consumption data in the assessment is that the recommended improvements have a clearer connection to the householder’s energy use.

Heating system considerations: Under the EPBD a building’s heating system must be inspected if it has a space heating output of more than 70 kilowatts. This inspection must include an assessment of the system’s efficiency and recommendation of cost-effective measures to improve the system’s efficiency. Only new buildings are required to consider alternative heating systems.

Adaptability: The Swedish EPC does not distinguish between building use, so it is also used for mixed-use buildings. It is used both for existing and new buildings.

Costs: The cost of an EPC is not regulated and generally increases with the size of the building. An EPC for a multi-owner building typically costs between £900 and £1400^[15].

Qualifications and training: Energy experts are certified through a certification body and work as independent contractors. Certification requires a relevant background (education or experience) and passing an exam. There are separate exam preparation courses, but they are not a requirement.

© Published by Changeworks, 2023 on behalf of ClimateXChange. All rights reserved.

While every effort is made to ensure the information in this report is accurate, no legal responsibility is accepted for any errors, omissions or misleading statements. The views expressed represent those of the author(s), and do not necessarily represent those of the host institutions or funders.

This work was supported by the Rural and Environment Science and Analytical Services Division of the Scottish Government (CoE – CXC).

ClimateXChange
Edinburgh Climate Change Institute
High School Yards
Edinburgh EH1 1LZ

+44 (0) 131 651 4783

info@climatexchange.org.uk

www.climatexchange.org.uk

If you require the report in an alternative format such as a Word document, please contact info@climatexchange.org.uk or the following postal address: ClimateXChange, Edinburgh Climate Change Institute, High School Yards, Edinburgh EH1 1LZ

1. 
   

   Johnston, D. et al (2020) Are the energy savings of the passive house standard reliable? A review of the as-built thermal and space heating performance of passive house dwellings from 1990 to 2018 ↑

   
   
2. 
   

   The costs in this estimate are based on the experience of a certified Domestic Energy Assessor in Scotland. ↑

   
   
3. 
   

   Concerns have been raised about the applicability of some aspects of the PAS 2035/30 standards in Scotland, Scottish Government will set up a technical group to work with BSI to develop the standards. ↑

   
   
4. 
   

   The costs in this estimate are based on the experience of a certified Passivhaus Designer in Scotland ↑

   
   
5. 
   

   All currency conversions are using xe currency converter (February 2023) ↑

   
   
6. 
   

   Such as a commercial building energy analysis ASHRAE ↑

   
   
7. 
   

   see Appendix A: Risk Assessment of Mixed-Use Buildings ↑

   
   
8. 
   

   stats based on 2014 data ↑

   
   
9. 
   

   3CL can also be used for a “simplified version” of the Global Technical Diagnosis (when apartments have individual heating, less than 20 units, or collective heating with less than 50 units). However, it doesn’t have to be used in these circumstances, the “complete” energy model (see TH-C-ex) can be used instead. ↑

   
   
10. 
    

    Individual owners can request a separate apartment DPE if they want one, but this does not remove the requirement to have a building scale DPE ↑

    
    
11. 
    

    The GTD is a holistic tool designed to inform condominium owners in France about all key technical and thermal aspects in their buildings. The aim is to encourage owners to implement a programme of works, with a particular focus on energy efficiency. It includes a list of works necessary for the conservation of the building, their cost and summary of measures to be carried out over the next ten years. The GTD uses a methodology that is adaptable for all sizes of multi-owner building, including those with collective and individual heating. ↑

    
    
12. 
    

    DIN 4108-6 and DIN 4701-10 are alternative methodologies however these are due to expire. ↑

    
    
13. 
    

    “Typical consumption” was derived by the Institute for Housing and Environment from a sample of 1,700 buildings. It is based on the average heating energy consumption that a building of the same size and the same energy standards. ↑

    
    
14. 
    

    Personal communication with a representative from Boverket ↑

    
    
15. 
    

    Estimated based on personal communication with a representative from Boverket, and costs from three EPC companies. ↑

    
    

Please note that a small change has been made to Table 13, OPEX for Lined Rock Caverns, in May 2024.