Net zero retrofit of the building stock

1. Introduction

Buildings account for a quarter of global annual emissions through operation, with a further 8% associated with the construction industry (UNEP & Global ABC 2025). Most of the world now acknowledges the need for significant reductions in emissions, including improvements both to the performance of the existing stock and more efficient new construction. Current mitigation policies, however, are not predicted to be sufficient to meet the Paris Agreement target (UNEP 2024), with recent studies suggesting that limiting warming to 1.5°C may require cumulative global emissions to be kept to six years’ worth of current annual levels (Lamboll et al. 2023). While building energy intensity (kWh/m²) has fallen by almost 10% over the past decade, this is only around half of that estimated to be necessary to meet the long-term decarbonisation goals. A tripling of energy efficiency retrofits and on-site renewable technologies by the end of the decade are identified as:

While the scale of the challenge and the lack of sufficient progress are sobering, it is also a time where the availability of data about the building stock, and the knowledge and computing power to assess that data, are more powerful and readily available than ever (Batty 2016; Dahlström et al. 2022). This includes ongoing building data collection for certification schemes such as Energy Performance Certificates (EPCs) in Europe (Li et al. 2019) or the Leadership in Energy and Environmental Design (LEED) rating system in the US (USGBC 2025); the greater availability of detailed high-resolution energy use data through smart metering (Webborn et al. 2019); and the increasing availability of large-scale datasets beneficial to building analyses such as built form data (Steadman et al. 2020; Google Research 2025). Consequently, despite the size and complexity of the undertaking, researchers are arguably better placed than ever to understand the current state of the building stock, and evaluate the retrofit options. Indeed, the benefits of these improvements in data and computing power have already impacted greatly on research. For example, historically, stock modelling for retrofit assessment has typically been carried out using simplified models for small numbers of archetypes. However, researchers are now moving towards running complex dynamic simulations at a disaggregate (individual buildings) level for large urban areas, or entire building sectors (Dong et al. 2023).

Naturally, despite these technical improvements, considerable uncertainty remains about the theoretical and practical factors that will determine both the overall potential for emissions reductions (Gillich et al. 2019), and also the transition process and cumulative emissions and costs associated with that transition (CCC 2020):

• What is the overall potential for reducing energy use?

• How is this affected by features such as hard-to-treat elements?

• What might be the cost?

• How should architectural heritage be dealt with?

• What part do newer technologies (such as the use of hydrogen for heating) have to play?

• How can it be ensured that emissions reductions are achieved while also accommodating related concerns such as overheating or fuel poverty?

• And so on ….

Alongside these concerns around quantifying the technical potential, there also remain significant uncertainties around how improvements to such a large proportion of the stock should be encouraged and met. Considering the UK context, for example, a recent target of 600,000 domestic heat pump sales per year by 2028 as part of the transition to net zero (rising to 1.6 million per year by 2035) represents an 11-fold increase compared with national sales in 2022 (NAO 2024). Similarly, an analysis of London estimated that peak wall insulation retrofits at a rate about 150,000 per year would be required to achieve the city’s net zero goals (Arup 2018), more than twice the 2024 UK total wall insulation retrofits delivered through government schemes between solid and cavity wall improvements (DESNZ 2025). And in the English education sector, it has been shown that even when improving schools at a rate of 650 per year (considerably higher than the current rate), the interim emissions reduction targets may still be difficult to meet (Godoy-Shimizu et al. 2022). These concerns are further amplified by the fact that delaying action on the delivery of retrofits is predicted to have significant implications on the emissions reduction potential (CCC 2020), as well as the overall costs and national debt (OBR 2021). Of course, places projecting significant growth in their building stock (generally expected in emerging and developing economies; UNEP & Global ABC 2025) will need to place a greater emphasis on efficient new construction, but retrofits of the existing stock will still be important:

• How might the aforementioned three-times increases in building retrofits and renewable uptake be achieved?

• How can greater engagement from the general public be encouraged?

• Are the skills and worker capacity in place to undertake the necessary works?

• And so on ….

As guest editors, coming with backgrounds predominantly in quantitative research, we initially envisioned this special issue to focus primarily on the technical questions around retrofit potential. Therefore, it is exciting, instead, to have found that many of the papers have examined the transition of the built environment towards net zero through a socio-economic lens, exploring the drivers and barriers to greater uptake of retrofits, and how policy, knowledge-sharing and more concerted engagement between stakeholders could be used to improve the rate of uptake needed to meet the long-term emissions reduction targets.

2. Context

While there are numerous barriers to a greater rollout of retrofits identified across the literature, globally a critical government-level issue is a lack of clear policy and detailed roadmaps to achieving building sector decarbonisation (UNEP & Global ABC 2025). Significantly, while much of the world has produced nationally determined contributions (NDCs) outlining their overall emissions reduction goals as part of the Paris Agreement (UNEP & Global ABC 2025), ‘few countries address the building sector comprehensively in their NDCs’. A related key barrier that ideally requires local or national government intervention is a lack of clear and consistent guidance from trusted sources; ideally linked to strong consumer protection schemes able to ensure that retrofits are designed and undertaken correctly. Both are crucial where, as is the case for the built environment, the goal is to encourage a large number of small stakeholders to engage with building retrofits, many of whom may have very limited prior knowledge about the topic. Indeed, a study of UK domestic retrofits found that even amongst more knowledgeable and engaged homeowners, the lack of clear, precise, concise and consistent guidance was still a key barrier (Wise et al. 2025); as was the lack of a single, consistent source of information.

Currently, information often comes from multiple sources that may not paint the entire picture and may contradict each other. Similar points around the need for greater access to credible expert guidance—crucially, with technical information tailored to be suitable for, and understandable by individual building users—have been identified elsewhere, such as in China (Jia et al. 2021), South Africa (Amoah & Smith 2024) and Australia (Alam et al. 2019). One possible approach to dealing with this is through the use of the one-stop-shops model, where instead of dealing with multiple advisors and contractors, a building owner has a single point of contact that provides the entire retrofit service, from design to installation to quality assurance (Boza-Kiss et al. 2018).

Given the sheer number of building retrofits that are projected to be carried out and the very short period of time over which this will hopefully occur, the issue of how (or if) quality of work can be maintained will be crucial. The need to ensure works are carried out correctly, plant is sized and installed appropriately, etc. is especially important if retrofits move away from the perspective of fabric-first, as recent research has suggested may be necessary for cost and timing reasons (Eyre et al. 2023). A recent study of the impact of PAS 2035 (the UK’s best-practice guidance, including some mandatory requirements, for domestic retrofits) found that their introduction did not necessarily correspond with a material improvement in works’ quality (Fylan & Glew 2022). More worryingly, a survey of work carried out under the UK’s flagship Energy Company Obligation (ECO) scheme has found that almost all (98%) external wall insulation installations and almost a third of internal wall insulation installations since mid-2023 have major problems that require repair (NAO 2025). Several factors are at play here, including an under-skilled workforce, and lack of oversight or clear and strong monitoring. A similar study for heating electrification noted how much more complex and time-consuming the ‘customer journey’ was for heat pump installations than for gas boilers (DESNZ 2024). Given the common refrain that people typically do not touch the thermostat settings for their gas boilers (Peffer et al. 2011)—and the fact that heat pumps are typically considerably more complex pieces of kit—it is unsurprising that the study highlighted a need for post-installation monitoring to be a key part of any rollout.

At the building level, as expected, finances remain a key barrier to greater retrofit uptake, with high upfront costs for energy efficiency measures alongside uncertainties about payback periods and how energy efficiency improvements may be reflected in building sale prices or rent (Brounen et al. 2020; Gholamzadehmir et al. 2025). A recent survey of 10,698 UK households found that of those interested in undertaking energy efficiency improvements, the majority were concerned about costs and the long-term savings that could be achieved (Citizens Advice 2025). The knock-on is that domestic energy efficiency investments can often be dominated by higher income households (IEA 2024). For society as a whole, this is doubly unfortunate, since energy efficiency improvements to low-income households can represent dual benefits of improving the building stock and reducing fuel poverty (Tozer et al. 2023). Thus, a common recommendation is the need for greater policies and funding schemes providing financial support to low-income households.

A related issue is that of split incentives, affecting rental and multi-occupant properties generally and low-income households especially (Bird & Hernández 2012), where different motivations between who pays for (and has control over) improvements and who benefits can result in under-investment in retrofits. The impact of these factors depends greatly on local context, and while numerous studies have explored this issue using data from Europe and North America, a recent paper was:

Analysing data from housing in Mexico, the study found that renters typically have less energy-efficient homes, pay higher bills compared with owner-occupiers and, importantly, are less knowledgeable about energy efficiency policies. To combat this issue of information asymmetry, policies such as minimum energy performance standards (MEPS) for rental properties, financial incentives (Castellazzi et al. 2017) and greater energy labelling/certification are recommended.

Uncertainties around post-retrofit performance and the impact on economic viability are another key concern. The performance gap is a well-documented phenomenon, related to factors including occupancy behaviour (and issues such as the rebound effect), as well as work quality, plant commissioning and closely tied to the underlying building models used to predict potential performance (Donn 2025). However, alongside these ‘smaller’ building-scale factors are large-scale external factors that can quickly disrupt things in unpredictable ways. In recent years, events such as Brexit, the COVID-19 lockdowns and the cost-of-living crisis have affected fuel prices, construction costs and retrofit uptake rates (Oyegoke et al. 2024), as well as typical building-use patterns (Gaspar et al. 2022; Hollick et al. 2024). Alongside these factors that directly affect financial viability are related concerns around potential indirect negative impacts of retrofit measures; most significantly the potential for rising fuel prices linked to the relative costs of electricity and fossil fuels or increased risks of overheating, damp and poor indoor environmental quality (Grassie et al. 2022; Hao et al. 2022; Ortiz et al. 2020).

Naturally, these issues are less distinct (and more numerous) than is suggested by the examples above; and policy instruments often tackle many of them together rather than in isolation. Nonetheless, these issues will need to be addressed in order to have any chance to meet the ambitious retrofit goals for net zero, with far greater consideration paid to what motivates building owners to actually undertake improvements (Brown et al. 2025). Given that these issues can be very context specific, studies have also highlighted the need for further research with wider global coverage (Madushika & Lu 2023). With these challenges in mind, the following section introduces the contributions received for this special issue, summarising their key insights.

3. Contributions to this special issue

Following the call for papers, a total of 24 abstracts were submitted for the special issue. From these, nine peer-reviewed articles have been published here (Table 1).

Reflecting the variety of papers received, there are various lenses through which they could be considered. The papers could be grouped together, e.g. around their scope (in terms of the types of buildings assessed, or their geographical or temporal coverage); methodology (from technical, modelling-based to qualitative desktop studies and interview-based papers); or the project stages covered (from preliminary assessments of viability to reflections on long-completed projects). However, a key point raised in the initial call was that effective large-scale building retrofits will require engagement and buy-in at all scales: from regional and national levels, for effective regulation and policy, to individual building owners and users onto whose buildings any improvements will be made. As such, the contributions to this special issue are discussed here in terms of the stakeholders that form the focus of each paper.

Gupta et al. present an application of LEMAP, a new mapping tool that combines technical and socio-economic analyses as part of urban-scale assessment of retrofit suitability. The study demonstrates the potential for using various datasets to guide domestic heat pump deployment, identifying the houses that are technically most suitable for heat pumps, but also the households that are financially and digitally ready for the technological shift. Using an area of Oxford as a case study, the paper shows that these two indicators of retrofit readiness do not always align, highlighting the need for greater consideration of occupancy factors such as digital literacy, especially when considering technological changes that require different user behaviour.

Where Gupta et al. considered technical potential as part of an overall retrofit potential assessment, Adebisi and McArthur focus on improving the building modelling process itself. Their paper presents the development of nine archetypal heating, ventilation and air-conditioning (HVAC) arrangements for Canada, associated with different building uses, built forms and construction periods. In each case, the current performance has been modelled, along with several potential post-improvement performances, based around different retrofit goals. The paper fills a key gap in large-scale building simulation: the need for open and clear assumptions of building systems informed by existing data and expert knowledge.

While Gupta et al. used socio-economic data as broad indicators of household readiness for retrofits, Schunemann et al. examine household decision-making directly. Using a Rubicon model to produce a causal loop diagram (CLD) of domestic energy retrofit decision-making, they attempt to identify key areas where policy might be best suited to improve uptake. Of particular interest are those key moments that define where otherwise uninterested homeowners might decide to go ahead with improvement works. Their research, focused on Germany, illustrates the need for careful consideration not only of the factors that currently influence public perception but also, crucially, how these factors may change over the coming years as retrofit rollouts ramp up.

While the long-term ambitions for building improvement represent an unprecedented scale of work, Gillich identifies lessons that can be learned from another hugely ambitious building retrofit scheme from recent history. The paper examines the UK programme during the 1960s and 1970s to transition 13.5 million UK homes from town gas to natural gas. Through the lens of market transformation theory, the paper identifies shortcomings in several areas of modern net zero policy that could benefit from the policies that underpinned the successful gas-ready programme. One key issue identified is the need for greater effort spent in engaging with and informing the general public on the benefits of such a retrofit programme, importantly, taking the discussion beyond simple considerations such as costs.

Another paper also looks to the past: Charles’s study examines how the current capacity for retrofits in the UK social housing sector has been shaped by history, and explores ways effective net zero policy could be better defined. The paper shows how a lack of clear and consistent policy, alongside chronic underfunding, has resulted in boom/bust cycles which, in turn, have reduced the pool of expertise necessary to commission and undertake building improvements, particularly in the social housing sector. The paper calls for stable, long-term retrofit policies and funding mechanisms if the transition towards net zero is going to be achieved in an equitable manner. Highlighting the risk that social housing—often with particularly vulnerable tenants—can be treated as test subjects for domestic retrofits, the paper calls for a far greater emphasis to be placed on stakeholder engagement not just before and during any works but, crucially, post-retrofit as well.

Bordass et al. focus on building users in further detail, presenting the benefits of their ‘people-first’ approach to retrofits. Presenting two recent case studies, the paper is a call to arms, demonstrating the benefits of placing a greater emphasis in engaging building users with the retrofit process. Through the use of open workshops, continued post-retrofit support and shared access to building performance data, the projects aimed to improve the technical literacy of the building users, empowering them to identify operational issues and passively operate their buildings. Most reassuringly, the paper notes that this continued engagement, while requiring greater commitment from building professionals, was found to be a positive experience from them as well.

While Bordass et al. examined what can be achieved through small, grassroots projects, King and Jemtrud focus on retrofit providers working at a much larger scale. The paper presents a desktop study carried out on three established initiatives that provide large-scale, deep energy retrofits (DER); Energiesprong, based in the Netherlands, and RetrofitNY and REALIZE-MA, based in the US. Using the five steps of a strategic niche management (SNM) model, the study explores how each initiative addresses local barriers, and is shifting from early experiments towards the goal of providing financially independent mass deployment of retrofits. The paper notes that even across these well-known examples, DERs remains a fledgling market still reliant on protection mechanisms. Given the scale at which retrofits will need to be carried out in the near future, the paper highlights the need for ongoing research and data collection as these initiatives continue to grow and others like them appear.

Considering the South African context, Terblanche et al. examine the drivers and barriers to greater uptake of net zero building design in the country. Presenting the results of interviews with five building designers, they note that, alongside financial concerns, a key challenge is a general lack of awareness and technical understanding compounded by a lack of strong government support, which has resulted in a reluctance to adopt new design techniques. Critically, the paper notes that net zero building policy remains inconsistent across the country, with regions lagging behind the major cities ‘due to regulatory and infrastructural deficiencies’, and highlighting a need for clear and unified regulations and policy.

While much of the discussion above focuses on the impact of retrofits on operational performance (annual emissions, fuel costs, etc.), the final paper by Alaux et al. instead examines the issue of embodied carbon. Using life-cycle assessment (LCA), the paper models the impact of different future pathways for the Austrian building stock towards 2050, from a business-as-usual scenario to scenarios that prioritise building reuse or construction material recycling. The analyses reveal that a considerable reduction in total building embodied carbon can be achieved by embracing circular economy principles. These results, taken in conjunction with the note that as much as 70% of building demolition in economies like Austria’s is primarily driven by aesthetic or financial reasons (rather than technical or regulatory needs), emphasises a need for greater consideration of embodied carbon within policy. An issue that, no doubt, will only become more important as operational emissions fall.

4. What’s next?

Taken as a whole, the papers that make up this special issue illustrate what a monumental task it will be to turn the long-term building retrofit goals from ambitious words into successful actions. Tripling energy retrofits and on-site renewables by 2030 will require collaboration across a wide range of institutions and disciplines. Given the scale of the task, the issues raised in this special issue will all need to be addressed, through research, policy and practice, in order to efficiently transition the global built environment towards net zero.

Historically, however, research has found that successful interdisciplinary collaboration can be particularly difficult in the context of urban sustainability, especially in acting on research (Evans & Marvin 2004, 2006). This reflects both the complexity of individual fields and the base knowledge required for each, but also the need to produce policy or guidance that crosses between academic disciplines that may have radically different approaches, world views, and underlying models and methods.

Reassuringly, however, across the papers in this special issue—despite a wide range of academic backgrounds, timeframes, geographical contexts, and types and scales of retrofits—common threads arise around the need for consistent long-term policy, with a greater focus on public engagement and a collaborative approach across stakeholders, and increased provision of reliable, open, consistent advice based on trustworthy data. These clearly point to key areas for further research:

• What form might improved stakeholder engagement take, and how could the long-term impacts be quantified?

• Can the ‘softer’ grassroots interventions proposed by Bordass et al. be successfully scaled up or applied to different contexts?

• What other lessons can be learned from major historical retrofit programmes, such as the gas-ready scheme outlined by Gillich?

• And so on ….

As researchers, we can fall into our disciplinary niches, or we can look outwards. To date, urban stock models have primarily focused on evaluating the energy or emissions pathways towards net zero, often treating occupant behaviour using simple, fixed assumptions or rules. Instead, the challenges for research will be to consider the complex dynamics that influence how building users choose to install retrofits, for example, or modelling how public engagement might work and change over time. Given the modern capabilities of high-end computing power, and the increasing availability of highly disaggregate information covering buildings, systems, households and individual behaviour, perhaps a next step could be modelling that further bridges the gaps between some of the socio-economic ideas and technical building modelling.

Competing interests

The guest editors have no competing interests to declare. Philip Steadman is an Editorial Board member of Buildings & Cities, and both he and Daniel Godoy-Shimizu are guest editors of this special issue. Neither author was involved in the review and decision process for this editorial.