Domestic Energy Efﬁciency Scenarios for Northern Ireland

: Building fabric retroﬁtting is an important ﬁrst step in improving building energy efﬁciency. The United Kingdom’s (UK) housing stock is one of the most inefﬁcient in Europe, and Northern Ireland has the second-highest level of fuel poverty in the UK. This Northern Irish case study developed three fabric retroﬁt scenarios that estimate potential demand reductions, CO 2 emissions removals and retroﬁt costs. The ﬁrst scenario reduces domestic demand by 10% and removes 6% of domestic emissions. The second scenario is more ambitious than the ﬁrst, and results in an 18% reduction in demand and 12% of emissions removed. The third scenario proposes fabric retroﬁtting to PassivHaus standard and results in a 42% reduction in demand and 27% of emissions removed. Furthermore, retroﬁt schemes can provide up to approximately 350,000 jobs annually between 2022 and 2050 for the Northern Irish population. This study demonstrates how fabric retroﬁt scenarios can be streamlined to the unique features of a housing stock. It shows that fabric retroﬁt research is important for the formulation of energy efﬁciency policy and emphasises that domestic sector retroﬁtting will yield socioeconomic and environmental beneﬁts locally and internationally.


Introduction
Energy efficiency is one of the key measures needed to achieve the net zero targets announced by the UK and many other governments worldwide [1]. Domestic emissions account for two-thirds of the total emissions and the UK's housing stock is one of the most inefficient in Europe [2].
The UK's Standard Assessment Procedure (SAP) rates the energy and environmental performance of houses from bands A to G (with A and G being the most and least efficient, respectively) [3]. The creation of energy efficiency strategy is a devolved power, but the UK government publishes general guidance for all regions [2,4,5]. The UK government sets energy efficiency targets focusing on improving the efficiency, and consequently averages the SAP ratings of the housing stocks across all regions. In 2017, the Clean Growth Strategy stipulated that all UK homes reach an SAP rating of C by 2035, and all fuel-poor and rented households reach an average SAP rating of C by 2030 [2].
Presently, the UK is targeting net zero greenhouse emissions by 2050, with current building regulations reflecting fabric and heating system standards [6]. Thus, advancing the energy efficiency of UK buildings has been listed as a national infrastructure priority, with the highest possible standards implemented for proposed energy efficiency upgrades of existing buildings and new-build constructions [4,7].
Existing studies on the impact of retrofit on existing housing show that retrofitting can yield considerable reductions in domestic demand and carbon emissions, can provide socioeconomic benefits and can provide insights for existing building regulations updates. In 2010, approximately 4 million jobs were saved and created due to the construction of 125,000 new energy-efficient homes in the United States through a scheme facilitated by the American Recovery and Reinvestment Tax Act [1]. There were significant savings in energy has been set [26]. In Northern Ireland, the most recent fuel poverty strategy was published in 2011 [27]. The strategy states that improved energy efficiency is one of the tools needed to eradicate fuel poverty. However, no clear targets towards fuel poverty mitigation are indicated, and the NI fuel-poor strategy has not been updated since 2011.
Nevertheless, existing energy efficiency programmes in NI are mainly targeted at fuel poverty; examples include the Warm Homes Scheme, the Affordable Warmth Scheme and the Northern Ireland Sustainable Energy Programme (NISEP). Through recent programmes, 9000 cavity walls were fully insulated and 66,000 lofts were topped up with insulation between 2016 and 2018 [23]. Energy efficiency schemes are run by local authorities and government departments such as the Northern Ireland Housing Executive (NIHE) and the Department of Communities [17]. Still, more efforts are needed towards creating energy efficiency programmes for NI.
In 2020, the Zero-In on NI Heat networking project initiated conversations on barriers and opportunities for heat decarbonisation in NI among market experts and consumers [28]. The project revealed that consumers are willing to facilitate heat-sector decarbonisation, and government leadership is important in this transition. Currently, all energy efficiency schemes in NI implement retrofit measures for about 16,500 buildings per year [29]. Still, existing studies indicate that more houses would need to be retrofitted annually to achieve significant domestic energy efficiency improvements that align with national UK targets [12,29,30]. As such, there should be more emphasis on developing more energy efficiency schemes and producing a strategy that can drive improvements in the energy efficiency of existing and new domestic fabrics.

Energy Efficiency Potential across NI
The Department for the Economy (DfE) released the latest energy strategy for NI in December 2021 [30]. The strategy states that up to GBP 2.4 billion will be spent on reducing energy use via the introduction of low carbon heating and investing in energy efficiency measures in the domestic sector between 2021 and 2030. During this period, 50,000 buildings (approximately thrice the current number of buildings retrofitted annually in NI) will need to be retrofitted annually. Additionally, a large pilot domestic retrofit scheme will be launched in 2022; this scheme will consider the requirements for heat pumps and be aligned with current and other pilot retrofit programmes. While this new strategy gives some general direction for energy efficiency improvements in NI, it does not provide a much-needed comprehensive framework on domestic energy efficiency implementation.
In 2014, the Department for Communities (DfC) opened a consultation for a new fuel poverty strategy for NI [31]. The new strategy would address the Warm Home Scheme delivery model, qualification criteria and available energy efficiency measures. This strategy is yet to be published. National Energy Action (NEA) and Energy Action Scotland (EAS) have recently published a report with some recommendations for fuel poverty mitigation in NI [32]. The report states that all NI fuel-poor homes can be upgraded to SAP rating C between 2025 and 2030 at a cost of GBP 440 million. However, there are no details on how this can be achieved.
The DfE's new energy strategy aims to tackle fuel poverty by implementing measures that will cause carbon emission and energy bill reductions [30]. Still, there is insufficient detail on what measures would be applied, what percentage of the fuel-poor housing stock would receive these measures, a target SAP rating and a retrofit timeline.
The CCC's Sixth Carbon Budget (6CB) advice suggests an 11% reduction in total domestic demand and 4% removal of emissions (based on 2016 levels) in NI by 2050. This can be achieved if fabric retrofit measures are implemented in 410,000 existing NI dwellings (including fuel-poor households) by 2050 [12,33]. However, the 6CB dataset does not provide a comprehensive breakdown of the number of NI houses (in terms of age or type) to receive the recommended retrofit measures. New builds will be zero-carbon; nevertheless, there is insufficient detail on what measures would be applied to the building envelopes of new buildings.
The NIHE's 2019 report on the cost of improving the SAP rating of NI dwellings indicates that 23,200 F-and G-rated occupied and vacant houses can be improved to rating E via fabric retrofit at a total cost of approximately GBP 87 million, with an average cost of GBP 3700 per house [34]. Although this study presents the costs of improving the efficiency of NI houses with the lowest SAP ratings, it does not investigate the impact of energy efficiency programmes for houses of EPC bands D and E. Furthermore, this study does not consider the impact of its recommended improvements on the SAP rating of the entire NI housing stock and there is no analysis for future housing stock.
Another 2019 NIHE study outlines eight scenarios specifying fabric and heating system upgrades [35]. A total of 632,000 houses received at least one retrofit measure, at a total cost of GBP 2.4 million with an average cost of GBP 3133 per house. The average SAP rating for the domestic stock increased from 65.8 to 73.8 (from band D to band C). However, this study did not specify the impacts of these scenarios on heat demand or carbon emissions. Additionally, there is no consideration of future domestic heat demand, carbon emissions and new builds.
ARUP's most high-reaching scenario for NI aligns with UK net zero targets. It suggests a 55% reduction in energy demand by 2050 and states that NI policy must promote the retrofitting of a maximum of 50,000 buildings per year [29]. Although ARUP presents an analysis that considers the effect of potential NI energy efficiency policy outlines on domestic retrofitting, the study does not give a detailed breakdown of how retrofitting can be achieved. The number of houses to be retrofitted and changes in demand per measure are not clearly represented, and this limits the feasibility of the study. The 2021 NIHE report on the cost of carbon savings in Northern Ireland's housing stock states that 390,000 dwellings can be improved to band C at a total cost of GBP 2.4 billion, with an average cost of GBP 6200 per dwelling [36]. Moreover, 3.2 tonnes/year of CO 2 emissions can be removed with improvement to band C. Additionally, 586,000 houses can be improved to band B at a cost of GBP 9.2 billion and at an average cost of GBP 15,600 per house. However, this study is not sufficiently detailed to suggest a defined pathway for NI domestic energy efficiency. Moreover, there is no breakdown of the archetypes recommended for retrofit or impacts of the suggested scenarios on domestic heat demand.
The novelty of this study is the initiation of a comprehensive framework for achieving fabric energy efficiency in NI through the implementation of measures and prescribing specific numbers of existing and new households. It outlines three varying levels of retrofit with corresponding changes in demand and carbon emissions that consider government guidance for NI housing stock (including fuel-poor households). Additionally, this study provides a retrofit cost breakdown per scenario and presents the socioeconomic impacts.

NI Housing Stock Model
For the purposes of this study, an NI building stock model was developed. It is a datadriven model that features the 2018 NI housing stock, disaggregated into 25 different domestic archetypes (houses by age and type). The model's inputs were actual data obtained from the Land and Property Services (LPS), the NIHE and the Northern Ireland Statistics and Research Agency (NISRA) [18,[37][38][39][40]. Figure 1 shows a diagrammatic representation of the methodology. The input data were based on estimations of the number and percentages of the following parameters: 1.

4.
Wall, loft, door and solid floor insulation levels per archetype.
Total domestic heat demand in 2018. 7.
Total emissions due to domestic heating. 8.
Change in demand per measure. 9.
Carbon emissions factors per heating fuel. 10. Retrofit costs per measure per house type. 11. Central heating types per archetype in alignment with existing levels of wall, loft, door and solid floor insulation as well as glazing and airtightness.
The model outputs were generated by applying relevant derivation formulas based on retrofit scenario objectives. Outputs include: 1.
2050 projections on the uptake of insulation, glazing and airtightness improvements per archetype under three different levels of retrofit (or retrofit scenarios).

2.
2050 demand projections (for entire existing and future housing stock and per archetype).

3.
Carbon emissions removed due to retrofit improvements by 2050 (for entire housing stock and per archetype).
It is estimated that 84% of the entire NI housing stock of 786,362 houses requires one or more retrofit measures (see Table 1). Limited and no insulation 22%

Potential for door insulation 65%
No potential for door insulation 35% Note: * indicates where average energy efficiency levels have been assumed for the entire stock due to unavailability of data.
The model was calibrated and validated by accounting for the actual heat demand estimations for the base year derived from NI energy consumption data from the Department for the Economy [41]. Additional calibration was performed using data derived from the change in demand per measure from the National Energy Efficiency Data Framework (NEED), as updated in the 6CB analysis [33]. The estimated changes in demand per measure were derived for the NI context by deriving the change in demand per measure per square metre of domestic floor area. These derivations were applied to the NI housing stock by accounting for floor areas per archetype in the NI housing stock.

Estimations of Existing Levels of Energy Efficiency in NI 2.2.1. Wall Insulation
Wall insulation retrofit has been termed a critical first step in preparing houses for low-carbon heating technologies [42]. As wall insulation can potentially result in a 9% reduction in domestic heat consumption (see Table 2), this measure can bring one of the most significant energy efficiency improvements and fuel poverty reduction [33,43]. Generally, the state of partially filled and uninsulated cavity walls in Northern Ireland is substandard; approximately 90% have issues with mould and damp [45]. The leading cause of mould growth on walls is excess humidity, and inadequate insulation catalyses mould growth [45,46]. Therefore, partially filled and uninsulated cavity walls recommended for retrofit are assumed hard to treat in the NI housing stock model [16]. If properly installed, insulation can help reduce the occurrence of mould in addition to facilitating heating demand and cost reductions (which can help with tackling fuel poverty) [47,48].
Of the entire 2018 NI housing stock, 69% of houses have full cavity wall insulation, and 17% houses have no wall insulation (8% and 9% have solid walls and cavity walls, respectively) [18,39,40]. The percentages of houses with partial cavity wall insulation and external/internal solid wall insulation are 8% and 7%, respectively [40]. Furthermore, 1919-1965 houses with cavity walls are classed as having uninsulated cavities, as pre-1965 cavity wall insulation was substandard [18].

Loft Insulation
A total of 52% of houses have lofts with insulation greater than 150 mm; 29% of houses have lofts with insulation less than 150 mm. 1% of houses have no loft insulation and another 1% have loft insulation of unknown thickness. The remaining 17% of houses have no lofts because they have been converted to a room with stairs or the pitch of the roof was too shallow for insulation installation [18]. Flats/apartments on the top floor would need loft insulation top-up, and not a complete installation.
Houses requiring a complete loft insulation installation have the highest numbers without central heating among all houses that need to be retrofitted (see Figure 2). Houses without central heating are most likely vacant [18]. Nevertheless, vacant houses should not be excluded from retrofit schemes. Unoccupied houses are essential to the operation of the housing market; when they have been empty for over six months, they are termed vacant and can become problematic [49]. In 2019-2020, there were 16,802 applicants seeking housing from the Housing Executive [50]. Vacant homes should be included in retrofit schemes as they can converted into social homes beneficial to homeless persons [49,50].

Glazing and Door Insulation
Of the NI housing stock, 26% has post-2006 double glazing, 61% has pre-2006 double glazing, 9% has partial double glazing and 3% has single glazing. Higher standards of window construction were not implemented in NI till 2006; however, in other UK regions, these standards were implemented in 2002 [16,51].
Most single-glazed houses are heated by oil, gas and solid fuels (see Figure 2). Roughly 60-65% of households heated with solid fuels are low-income [18]. From April 2021 to March 2022, 14 NISEP schemes will be implemented to improve the energy efficiency of low-income houses, but none of them considers glazing [52]. Glazing can afford a modest but appreciable reduction (2-4%) in heat demand (see Table 2) and should be included in similar schemes [33]. There is limited/no data on the level of door insulation in NI. Therefore, this study recommends that all houses with single and pre-2006 double glazing should have their external doors insulated.

Airtightness
Prior to 2006, the Northern Ireland Building Regulations did not specify detailed standards for residential airtightness [53]. The Air Tightness Testing and Measurement Association (ATTMA) air permeability and pressure testing procedure states that for every site development, the pressure test must be performed on three units of each house type or 50% of all units of each house type, whichever is less [13,54]. The maximum air permeability for a building should be 10 m 3 /(h·m 2 ) at 50 Pa [13]. Consequently, the infiltration rate in air changes per hour (ac/h) is estimated as 0.15 ac/h for houses that have an air permeability of >5 m 3 /(h·m 2 ) [54].
Each house type should be tested, but the current trend is such that not all houses are tested despite the stipulated compulsory procedure [55]. By 2006, air permeability and pressure standards and testing became mandatory in NI [13]. Therefore, all pre-2006 houses are estimated to need improved airtightness. Approximately 84% of the NI housing stock (pre-1919-2006 houses) has the potential for improved airtightness; the other 16% (post-2006 houses) is estimated to have infiltration rates according to current NI building standards [18,56]. The ATTMA air permeability test should be carried out on pre-2006 housing stock before retrofit.

Solid Floor Insulation
The potential for solid floor insulation lies with pre-2000 houses, where pre-1991 houses have uninsulated solid floors, and 1992-1999 houses have solid floors with limited insulation [57]. As such, 22% of existing houses (pre-2000 houses) have solid floors with no/limited insulation, and 78% of houses built between 2000 and post-2007 have insulated solid floors.
Most houses requiring solid floor insulation are heated by oil and solid fuels (see Figure 2). As roughly 60-65% of households heated with solid fuels are low-income, a substantial percentage of low-income houses have the potential for solid floor insulation [18]. Current NISEP schemes do not consider improving floor insulation [52].
Solid floor insulation can afford an 8% reduction in heat demand and should be included in NI retrofit programmes, especially for low-income households (see Table 2) [33]. At present, data on NI houses with suspended timber floors are unavailable.

Domestic Demand and Carbon Emissions
NI's annual domestic heat energy consumption in 2018 was 12,764 GWh for oil, gas, coal, electricity and renewable fuels [58]. The annual space heating demand was derived as 6932 GWh with an average demand per household of 8815 kWh. Additionally, the NI domestic sector produced 2.63 MtCO 2 e in 2018 [59]. Oil is the most-used heating fuel (64%), and gas is the second most-used domestic heating fuel (24%). With a fuel share of 24% for gas fuel, 76% of domestic stock can be estimated as being off-gas grid houses (see Figure 3) [18]. Thus, only approximately 196,000 houses are on the gas grid, which is a 5% increase in the number of houses connected to the gas grid in 2016 [18]. A third gas extension is underway in NI, and it is projected that 60% of the entire NI housing stock would be within the gas grid area. However, new investments in distribution network developments and connections of houses to the grid will be needed if the gas grid will be used for heating supply by 2050 [28].   According to Gillich et al., there are usually deviations between the modelled and actual changes in demand resulting from energy efficiency measures [62]. Such deviations were investigated in the change in demand per domestic energy efficiency measure for the CCC's Fifth Carbon Budget Advice [62]. However, this study derived changes in demand per measure using the National Energy Efficiency Data Framework (NEED) as updated in the 6CB analysis [33]. As there is currently no NEED for NI, relative estimations based on the 6CB's NEED dataset were applied in this study.
In every scenario, the average new build demand was estimated by accounting for demand reductions due to the installation of fundamental measures (cavity wall insulation and roof insulation top-up, post-2006/triple glazing and airtightness), on the current average demand per house. Demand reductions align with the level of retrofit of each scenario and it is assumed that these measures are built into the new house during construction.
Flats/apartments with electric heating were excluded from the demand and emissions reductions across all scenarios, as the average demand of this subcategory is approximately 5000 kWh [41]. Therefore, the heating demand for this archetype is likely to not reduce due to the rebound effect (see Tables A1 and A3 for change in demand per measure). Annual demolition rates were considered but were discovered to have insignificant impacts on the results. This study estimated NI net stock losses (conversions, closures and demolitions) over the recommended retrofit period (2022-2050) to be −2% [63].

Central Scenario
The Central Scenario aligns with the CCC's 6CB Balanced Pathway Scenario in terms of measures and percentages of houses proposed for retrofit. It presents a modest and balanced pathway for domestic retrofit in NI, recommending different percentages of the existing housing stock for retrofit to current building regulations (see Tables 3, 4, A1 and A2).  Additionally, new buildings constructed in the period of retrofit are built to current building regulations. The uptake of five basic retrofit measures-wall, floor and loft insulation, double glazing and airtightness-are recommended in the Central Scenario (see Figure 4). Over half of houses with uninsulated solid walls, and all houses with partially filled cavity walls and uninsulated cavity walls, are fully insulated by 2050. In addition, 1919-1980 houses with uninsulated solid walls are recommended for external insulation. Differently, pre-1919 houses with uninsulated solid walls are recommended for internal insulation, as owners of pre-1919 houses prefer to preserve the external appearance of their houses [16]. All pre-1919-1964 houses with partial cavity walls and pre-1919-1980 uninsulated cavity walls are fully insulated. There are no post-1980 houses with solid walls, as cavity wall construction was mandated for new builds by regulations in the 1980s [45].
All as there is no flat/apartment with single glazing in this category. As recommended by Element Energy's supplementary workbook, the model assumes that window units would be replaced every 20 years to preserve their capacity to conserve energy [66].
A total of 25% of all houses with the potential for solid floor insulation receive solid floor insulation between 2022 and 2050. Most houses retrofitted with solid floor insulation present the highest percentage (97%) of houses off the gas grid retrofitted by 2050. All houses with the potential for improved airtightness are retrofitted by 2050. The distribution of houses with the potential for improved airtightness is similar to that of houses receiving wall insulation retrofits, with 74% off the gas grid. Table 4 represents the annual rate of retrofit (houses/year) for each measure if this scenario is evenly implemented between 2022 and 2050.

High-Performance Scenarios: High-Level and Passive Scenarios
The High-Level Scenario is more ambitious than the Central Scenario, recommending that all of the 2018 housing stock requiring any of the five basic measures be retrofitted by 2050 (see Figure 4). However, all single and pre-2006 double glazing is improved to triple glazing, which is a high-performance measure suitable for a more advanced retrofit (see Tables A1 and A2) [33]. Additionally, houses built between 2018 and 2020 that have post-2006 double glazing are recommended for a triple glazing retrofit. The Passive House Scenario is the most ambitious of all three scenarios and investigates the impact of retrofitting the stock to passive house standards by 2050 (see Tables 3 and A1-A4, Figure 4). All measures, with an additional measure of external door insulation, are installed as needed in all houses requiring a retrofit. In these high-performance scenarios, new builds are built to the retrofit standard of each scenario.

Fuel Poverty Scenario
In 2018, there were 131,000 fuel-poor houses in NI [23]. The relative poverty threshold in NI is GBP 328 per week, indicating an annual income of approximately GBP 16,000 [67]. There are three variations of fuel poverty based on the percentage of its income that a household spends on heating fuel: marginal (10-15%), severe (15-20%) and extreme (greater than 20%) [68].
This study considers all fuel-poor houses; only 15% of them are on the gas grid (74% are heated by oil) [18]. The current state of the energy efficiency of fuel-poor houses was estimated by considering 2016 NIHCS data on the different levels of wall insulation, loft insulation and glazing data (see Table 1). As the 2016 NIHCS has no information on fuel-poor households and floor insulation, airtightness and door insulation, the same assumptions used to disaggregate the energy efficiency measures of the general housing stock were used to disaggregate the energy efficiency measures for fuel-poor houses. The average SAP rating of fuel-poor houses is 58.65 (band D). This study recommends that all fuel-poor houses are retrofitted to the level of retrofit of each scenario by 2030, in line with general UK guidance.

Change in Demand and Emissions across All Scenarios
There is a 10% reduction in the heat demand of the existing housing stock by 2050 (see Figures 5 and 6). This reduction is slightly lower than the 6CB's estimated 11% reduction in NI's existing domestic demand by 2050 from 2018 levels due to retrofitting only [66]. The highest changes in demand come from improved airtightness, because almost all of the existing housing stock is recommended for airtightness retrofitting. Other significant changes in demand come from cavity wall and solid floor insulation, which are recommended for approximately all houses needing these measures.  The highest percentage reduction in demand comes from solid wall insulation (see Table 2). However, houses with solid walls are the least efficient in the stock; they are also often heritage houses [69,70]. This makes them difficult and expensive to insulate. Therefore, a low uptake of solid wall insulation is recommended, and this limits the potential of this measure. Around 6% of domestic carbon emissions are removed by 2050. However, the 6CB advice suggests that 4% of domestic emissions are removed by 2050 (see Figure 7). The CCC's percentage of emissions removed might be less due to dissimilarities between the breakdown of heating types in houses recommended for retrofit by this study and the CCC's study. The highest removals of carbon emissions come from improving airtightness, cavity wall insulation and loft insulation top-up. In the High-Level Scenario, there is an 18% reduction in the heat demand of the existing housing stock by 2050 (see Figures 5 and 6), which is significantly above the 6CB's estimated 11% reduction in NI's existing domestic demand by 2050 [66]. The highest changes in demand come from houses retrofitted with solid floor insulation. Solid floor insulation incurs the third-highest reductions in demand and is applied to about half of the total number of houses requiring retrofits. Other significant changes in demand come from pre-2006 double glazing to triple glazing improvements and improved airtightness. Around 12% of domestic emissions can be removed in this scenario; the highest removals of carbon emissions come from improving pre-2006 double glazing to triple glazing, solid floor insulation and improving airtightness (see Figure 7).
The Passive Scenario results in a 42% reduction in the heat demand of the existing housing stock by 2050 (see Figures 5 and 6). In the Passive Scenario, the highest changes in demand come from houses retrofitted with solid floor insulation. Other significant changes in demand come from improved airtightness and cavity wall insulation. Furthermore, this scenario estimates that 27% of the current emissions can be removed by 2050; the highest removals in carbon emissions come from solid floor insulation, improved airtightness uptake and cavity wall insulation in this scenario (see Figure 7).

General Cost Impacts and Potential Improvements in SAP Rating
The Central Scenario costs approximately GBP 2 billion (see Table 5). Overall, cavity wall insulation, upgrading pre-2006 to post-2006 double glazing and solid floor insulation are the costliest measures to implement. Furthermore, the bungalow and the flat/apartment are the most and least expensive house types to retrofit, respectively. Although the detached house has the largest floor area, the bungalow has the highest building cost per square metre of all house types because it occupies a high value of land for its floor area [71]. The average cost of retrofit across all house types is GBP 3700, with bungalows and flats showing the highest and lowest average costs of GBP 5500 and GBP 1900, respectively. These costs are broadly in line with other studies and assumptions mentioned earlier. Source: [33,72].
However, the 6CB Balanced Pathway Scenario shows that the average household will need a GBP 9000 investment in retrofit [66]. Houses are recommended for a range of retrofit packages costing between GBP 1000 and 10,000, depending on their retrofit needs [66]. The CCC considers additional energy efficiency improvements such as heating system upgrades, which accounts for a higher average cost of retrofit per house than those in this study. This study's Central Scenario only proposes fabric retrofit measures with average costs of GBP 50-7400 per house.
The implemented measures and accompanying SAP points in the Central Scenario are outlined by building regulations [21]. In the Central Scenario, 23,000 existing houses receive at least one retrofit measure annually by 2050. Thus, the average SAP rating for existing and new NI housing stock increases to 73 (band C) by 2050 (the average SAP points for a new build is estimated as 86 (band B)) [22]. If Northern Ireland aims at improving domestic energy efficiency in line with other regions, the Central Scenario must be delivered through accelerated schemes implementing the retrofit of 75,000 houses annually with at least one measure by 2030.
The NIHE's cost of improving 23,200 occupied and vacant F-and G-rated dwellings to band E through envelope and heating system retrofitting is GBP 87 million, with an average cost of approximately GBP 3700 per dwelling [34]. This study estimates a similar average cost of retrofit (GBP 3700) with approximately 670,000 houses being recommended for retrofitting with measures including costly and potent measures such as solid wall and solid floor insulation.
Additionally, the NIHE suggests the average SAP rating for the domestic stock will increase from band D to band C, with oil and gas heating installations and fabric (easy-totreat cavity wall insulation, loft insulation and double glazing) upgrades of 632,000 homes at total and average costs of GBP 2.4 million and GBP 3133, respectively [35]. However, this present study recommends that the NI stock can achieve an average SAP rating of C by 2050 due to fabric retrofitting only, at an average cost of GBP 3700.
The High-Level Scenario costs GBP 5.9 billion; the measures that are most expensive to implement between 2022 and 2050 are upgrading pre-2006 to triple glazing, solid floor insulation and cavity wall insulation (see Table 6). The individual costs of applying all measures to all house types rise due to the increased costs of improving glazing specifications.
Consequently, the average cost of retrofit across all house types rises to approximately GBP 4500, with bungalows and flats showing the highest and lowest average costs of GBP 7000 and GBP 2000, respectively. Source: [33,72].
The Passive Scenario costs GBP 10.7 billion, and the measures that are most expensive to implement between 2022 and 2050 are upgrading pre-2006 double glazing to triple glazing, solid floor insulation and external solid wall insulation (see Table 7). Wall insulation is the most expensive retrofit measure to implement per house (with an average cost of approximately GBP 38,000). This high cost is due to the extremely airtight construction, minimal thermal bridging and significant thickness of the insulation required (the thickness of a passive wall would range between 300 and 500 mm) [73]. The detached house and the terraced house are the most and least expensive house types to retrofit, respectively. The detached house has the highest number of windows and largest floor area among other house types [74]. Therefore, the cost of installing triple glazing and loft installation to passive house standard in the detached house is impacted by its size, making it more expensive than other house types. The terraced house is the cheapest house type to retrofit to passive house standards of all five house types, as it has the lowest average wall area of all house types (see Table 8). The average cost of retrofit across all house types rises to approximately GBP 11,000, with detached houses and flats showing the highest and lowest respective average costs of GBP 15,000 and GBP 6000.   Across all scenarios, the cheapest overall costs are attributed to improved airtightness and loft insulation installations, while the cheapest measures per house are loft insulation top-up and improved airtightness (see Tables 5-7). Cheaper measures such as improved airtightness and loft insulation top-ups can be installed by homeowners themselves; the airtightness of a flat/apartment can be improved at costs as low as GBP 12 (see Table 8). Additionally, external solid wall insulation remains the most expensive measure to implement per house.

Cost Impacts and Improvement in SAP Rating for Fuel-Poor Households
In the Central Scenario, the total cost of retrofitting all fuel-poor households by 2030 is GBP 930 million, which is 45% of the scenario's overall cost (see Table 5). In the High-Level Scenario, the cost of retrofitting all fuel-poor houses by 2030 rises to GBP 1.1 billion due to pre-2006 double glazing to triple glazing retrofitting (see Table 6). This cost accounts for 18% of the total cost of the scenario. In the Central and High-Level Scenarios, cavity wall insulation, solid floor insulation and pre-2006 double glazing to post-2006 double glazing retrofitting maintain the highest gross retrofit costs among all measures applied to the fuel-poor housing stock.
In the Passive Scenario, the cost of retrofitting all fuel-poor houses rises to approximately GBP 1.8 billion due to the scenario's ambitious retrofit standard (see Table 7). This accounts for 17% of the total cost of the Passive Scenario. In this scenario, the overall costs of implementing pre-2006 double glazing to triple glazing, solid floor and external solid wall insulation retrofit in the fuel-poor housing stock are the highest among all of the costs of the passive retrofit measures. In the High-Level and Passive Scenarios, the percentage costs of retrofitting all fuel-poor houses based on the total scenario cost decreases due to the higher numbers of non-fuel-poor houses retrofitted by 2050. The average retrofit costs per house and measure for the general housing stock remain the same for the fuel-poor housing stock.
The average SAP rating for the NI fuel housing stock can improve from band D to band C (an average SAP rating of 72) by 2030. To achieve this, at least 12,300 fuel-poor houses would need to receive at least one retrofit measure between 2022 and 2030. This number is slightly less than the average number of houses currently retrofitted annually in NI, indicating that NI currently has the potential to eradicate fuel poverty over the next 10 years.

Economic Impacts
The retrofit costs across all scenarios were analysed according to the costs of labour and materials used in the construction industry [74]. In the Central and High-Level Scenarios, loft insulation costs may be broken down into 47% and 53% for materials and labour/transport, respectively, as one square metre of 100 mm insulation thickness costs GBP 1.
In the Passive Scenario, 93% of the loft insulation retrofit cost is claimed by labour. The labour costs for loft insulation are significantly higher than in the other two scenarios, due to the specialist level of skill required to install this measure to the passive house standard [75]. For insulated doors, it is estimated that 81% of the retrofit costs are allocated to materials, as the addition of insulation increases the unit price of the door [33].
Across all scenarios, 31% and 69% of the glazing retrofit costs are allocated to labour and materials, respectively, according to Building Cost Information Service (BCIS) index specifications. The airtightness improvement costs are only associated with materials [74]. Approximately half of solid wall insulation costs are claimed by labour costs, while the remainder is allocated to materials (53%, 48% and 49% in the Central, High-Level and Passive Scenarios, respectively). For internal wall insulation, 70% of the cost per house is spent on labour; for external wall insulation, the BCIS index specifies that 38% is spent on labour [76].
For cavity wall insulation, most of the retrofit cost is spent on labour (97% in the Central and High-Level scenarios, and 98% in the Passive Scenario), as insulation materials cost GBP 2/m 2 . Similarly, most of the solid floor insulation costs are allocated to labour (97% in the Central and High-Level Scenarios, and 99% in the Passive Scenario). Labour costs for solid floor insulation are considerable due to the rigorous installation process. Typically, solid floors are either overlaid with expensive hi-tech insulation materials, or the concrete must be broken up and reinstated with new rigid insulation below the screed [77].
Overall, in the Central Scenario, circa 70% and 30% is attributed to labour and materials, respectively. In the High-Level Scenario, the retrofit costs are split roughly halfway between labour (54%) and materials (46%). In the Passive Scenario, approximately 60% and 40% of the retrofit costs are assigned to labour and materials, respectively. A cumulative 5%, 3% and 2% saving due to economies of scale may be applied to the Central, High-Level and Passive Scenarios. Savings due to economies of scale are attributed to 5-10% savings from the labour and transport required for installing wall insulation in over 10 houses [74].
Fabric retrofit solutions facilitate job creation and employment opportunities [2]. The potential number of construction-related jobs that could be generated annually between 2022 and 2050 is 72,000, 170,000 and 348,000 from the Central, High-Level and Passive Scenarios, respectively. This general estimation is based on the mean weekly earnings and the number of weekly and annual working hours in the NI construction industry. In the NI construction sector, the mean weekly earnings and working hours are GBP 590 and 43 h (including overtime), respectively; the average number of hours worked in a year is estimated as 1367 [78][79][80].
In NI, the rate of unemployment for people over 16 has increased and the proportion of people aged 16-64 who were not working and not looking for or available to work decreased in the last year [81]. The Central, High-Level and Passive Scenarios can contribute total labour costs of GBP 1.4 billion, GBP 3.2 billion and GBP 6.6 billion, respectively, to NI's local economy. Therefore, there is substantial potential for job creation through these massive retrofit programmes in NI, implying economic benefits worth exploring. Furthermore, the demand for retrofit materials can lead to further employment in the manufacturing sector. Materials may be sourced internationally or locally, but labour costs will be of economic benefit to the local NI context. Table 5 shows that across all scenarios, the cost of implementing all retrofit measures to each house type in each scenario is generally cheaper than its corresponding new build cost. However, the cost of retrofitting a terraced house to passive house standards is more expensive than the cost of rebuilding it. In this case, it would be better to rebuild a highly energy-inefficient terraced house than retrofit it to passive house standards.

Socio-Environmental Impacts
Retrofitting will reduce the heat demand and energy bills per household, making heat more affordable for all income classes, especially those in fuel poverty. This study shows that if all measures are applied across all house types, the demand required to heat a bungalow, terraced house, semi-detached house, detached house and flat can be reduced by up to 11%, 11%, 9%, 8% and 8%, respectively, in the Central Scenario (see Table 9). The more ambitious scenarios can yield greater reductions in demand per house type. The highest percentage of house types requiring retrofits is attributed to the terraced houses. The most ambitious Passive Scenario can result in up to a 48% reduction in the heat demand of a terraced house, if all passive measures are applied.
Fuel-poor people live in cold, damp homes during winter and this can be fatal [82]. If more homes are well heated, the rate of deaths from extremely cold indoor conditions will reduce. With better levels of retrofit, domestic building envelopes would be better insulated, and thermal comfort levels would improve. Higher levels of thermal comfort and, hence, better quality of indoor air can reduce rates of cardiovascular disease, asthma and cancer [83].
Retrofitting the building envelope eases the transition to a low-carbon heating system. Building envelope retrofitting may help initiate positive consumer attitudes towards adopting low-carbon technologies [84]. Currently, NI consumers do not have adequate education and support to invest in energy efficiency measures [28]. This study provides information on energy savings and the cost of retrofit, which consumers will find useful.
Some relatively affordable measures such as glazing and airtightness improvements, which have some appreciable impacts on heating demand, can be included in existing and future government retrofit schemes. This would encourage NI homeowners to invest in improving the energy efficiency of their dwellings. Improved domestic energy efficiency would lead to reduced carbon emissions, contributing towards climate change mitigation in NI. Although the 6CB advice specifies that 4% of NI's residential emissions can be removed from 2018 levels by 2050, this study shows that higher amounts can be removed (6-27%) by improving the efficiency of the building fabric. Additionally, with well-insulated houses in NI, a lessened reliance on high-carbon heating fuels, especially oil, is feasible.

Conclusions and Recommendations
This study investigated the socioeconomic and environmental impacts of different levels of domestic fabric retrofit in the context of Northern Ireland. The study makes the following recommendations, which are applicable in other contexts and were illustrated by this Northern Irish case study.
Firstly, different archetypes will have techno-economic variations, which should be considered in the application of fabric retrofit when retrofit programmes are developed and executed. In the NI context, flats and terraced houses are the cheapest to retrofit to passive house standards among the different archetypes. However, it is cheaper to rebuild a highly energy-inefficient terraced house than to retrofit it to passive house standards. Terraced houses and flats/apartments have the highest and lowest numbers, respectively, requiring most retrofit measures such as wall insulation, loft insulation top-ups, solid floor insulation, glazing, and airtightness improvements. These two house types have the lowest costs of retrofit across all scenarios, and this can inform the creation of current and future energy efficiency grants.
Secondly, the installation costs of retrofit measures should be an important consideration for the creation of energy efficiency schemes, especially those targeted at fuel poverty. In NI, cavity wall insulation, solid floor insulation and upgrading pre-2006 double glazing to post-2006 double glazing/triple glazing are the costlier measures. Policy can encourage the creation of grants/schemes to offset some of the cost of these measures for owners of house types (bungalows, detached houses and semi-detached houses) with high retrofit costs. Current and future energy efficiency programmes for low-income households can make provision for solid floor insulation and glazing improvements. Similar considerations can be made for households in fuel poverty. The following categories of fuel-poor households show the highest numbers needing the assigned measure: Approximately 85% of all fuel-poor houses recommended for retrofitting are off the gas grid.
Similarly, some measures that have greater environmental impacts than others can be easily implemented by homeowners due to cheap material and installation costs and ease of installation. Local policy can encourage the implementation of such measures first and at a higher rate than others. Airtightness improvements, cavity wall insulation, loft insulation top-up, solid floor insulation and glazing improvements yield greater potential for reductions in demand and emissions than other measures in the NI context. A total of 23,000 houses can be retrofitted annually from 2022 to 2050 with a range of measures with average costs of GBP 60 to GBP 8000.
Airtightness improvement and loft insulation top-up have the cheapest costs per house among these potent measures and can be easily implemented without professional help [72,74]. Loft insulation installation is required by an evident percentage of vacant houses, which can be retrofitted to provide high-quality living spaces. Policy should facilitate incentives encouraging homeowners to invest in airtightness upgrades and loft insulation top-ups. An airtightness test should be carried out on every house built pre-2006 before airtightness upgrades are carried out.
Finally, local authorities can create and accomplish high-reaching energy efficiency programmes and targets, while using national guidance as a springboard. The development of energy efficiency frameworks can inform building regulations for existing and new buildings. This would encourage the uptake of optimally energy-efficient domestic fabrics, which can improve quality of life and aid an easy transition to low-carbon heating. This study's NI scenarios demonstrate that the heat demand and resulting emissions of the existing housing stock can be reduced further than UK government advice shows by 2050. However, if NI must achieve this, 23,000 houses will have to be retrofitted with at least one measure from now till 2050. Consequently, more efforts towards the creation of energy efficiency schemes in NI are needed. NI currently has the potential to abate fuel poverty between 2022 and 2030. The average SAP rating of NI's fuel-poor housing stock can improve from band D to band C by 2030, if 12,300 fuel-poor houses are retrofitted with at least one measure between 2022 and 2030.
Around 74% of the houses recommended for retrofitting are off the gas grid, indicating that retrofitting is a critical first step in preparing the NI housing stock for the uptake of lowcarbon heating systems. The improvement of the energy efficiency of domestic envelopes should be prioritised as the first step towards achieving net zero targets in NI. The scenarios' cost estimations suggest a guide to the amount of public funds that can be assigned to energy efficiency programmes for all households. Finally, this study's findings can inform revisions of NI building regulations towards local and UK-wide decarbonisation targets.