Publication - Research and analysis

Energy Efficiency Standard for Social Housing: peer review

Published: 22 Oct 2013
Energy and Climate Change Directorate
Part of:
Environment and climate change

Peer review scrutinising the example dwellings in the Energy Efficiency Standard for Social Housing consultation document.

Energy Efficiency Standard for Social Housing: peer review
3 Commentary on Assumptions

3 Commentary on Assumptions

Each retrofit example in the EESSH consultation document uses a baseline energy efficiency standard from 1990. Three improvement stages are used to consider the cumulative application of efficiency measures to improve the energy performance of the dwelling ('2015 Scottish Housing Quality Standard' ( SHQS), '2020 Further Measures', and '2050 Advanced Measures').

The purpose of these retrofit examples is to illustrate possible routes by which social landlords can achieve compliance with the standard through application of typical measures. It is important to note that the suggested measures do not represent a definitive list, and will not be suitable for all dwellings; the examples do, however, cover a broad range of social housing types and can help social landlords to identify an approach suitable for their own stock. It should be noted throughout that social landlords can achieve the EESSH using any measures that they consider appropriate.

The first Peer Review objective was to consider the assumptions used in these retrofit examples, determine whether they provide a sufficient representation of Scotland's social housing stock, and identify any significant building types not currently represented by the initial 23 examples. This process required a comparison of the assumption data against historic information, previous building standards and the default SAP and RdSAP assumptions. Additional commentary is provided, discussing the practical application of the proposed measures based on the experiences of previous projects.

3.1 Description of the social housing stock

3.1.1 Fuel mix

The retrofit examples consider properties which feature gas and electric heating systems. Whilst information is not available describing Scotland's heating energy consumption independently prior to 2005, UK-wide data shows there has been a significant shift in the fuel mix for domestic energy since 1990: then, 8% was provided by coal, 63% by gas and 18% electricity; by 2010 this had changed to 1% coal, 66% natural gas and 21% electricity [3] .

For the purposes of the EESSH, it should be noted that while solid mineral fuel only accounts for 1% of the domestic fuel mix, it will be harder to bring properties heated by this fuel up to similar ratings set for gas and electrically heated dwellings (without changing fuel type). This is because RdSAP allocates a very poor environmental impact rating to such fuels. It should also be noted that biomass may be considered as a solid fuel and while the calculated Environmental Impact rating is very good the Energy Efficiency rating may be much poorer, as this rating takes account of the fuel cost.

3.1.2 Building stock composition

The draft EESSH outlines 23 retrofit examples representative of the most common building types identified within the social housing stock. The retrofit technologies discussed in this report are therefore considered applicable, and relevant, to a large proportion of this stock. The retrofit examples aim to demonstrate the impact of typical energy efficiency measures; however, it is important to bear in mind that the approach applied to actual dwellings will vary from building to building, subject to their individual features and characteristics. These retrofit examples are not intended as prescriptive routes to compliance across the stock, but can be used as a reference point by social landlords, particularly in relation to the target ratings outlined by the SHQS and EESSH.

In reviewing the suitability of these example dwellings, it is important to understand which aspects of the social housing stock they represent. An analysis of this data was used to inform the selection of additional example dwellings (see Section 5) to address areas not currently represented.

Appendix A: Housing Stock Summary summarises the composition of the entire Scottish domestic building stock, detailing the number of properties for each building type and period of construction. Consideration of the mean CO 2 emissions relative to these categories shows that in most cases, the older stock has higher emissions [4] . The 23 retrofit examples were cross-referenced against the categories listed in Appendix A, highlighting a number of anomalies:

  • None of the examples describe detached properties (although this could be said to be a valid exclusion, as these represent less than 1% of the social housing stock [5] )
  • No pre-1919 houses are included, which is justified as the majority of social pre-1919 dwellings owned by social landlords stock are tenements ( Figure 1 and Figure 2)
  • With the exception of group C (four-in-a-block flats constructed between 1920 and 1949) the retrofit examples do not consider any 'other flats'. Figure 2 shows a significant proportion of high-rise dwellings, constructed between 1965 and 1982. These properties are often unable to accommodate low-cost measures such as cavity and loft insulation and thus may be deemed 'harder-to-treat', requiring the installation of higher-cost systems to improve their energy performance [6] .
  • In the EESSH consultation a significant number of the retrofit examples were of post-1982 construction. This reflects the changes to the minimum building standards and in particular U-values, occurring from 1990 onwards, which need to be captured. Examples, relating to later age bands were not provided, because higher specified U-values in building standards mean that it is unlikely that enhancement will be needed to achieve the EESSH.

Figure 1 Distribution of the Scottish building stock by period of construction

Figure 1 Distribution of the Scottish building stock by period of construction

Figure 2 Distribution of the RSL Scottish building stock by building type and period of construction (numbers refer to example dwellings)

Figure 2 Distribution of the RSL Scottish building stock by building type and period of construction (numbers refer to example dwelling)

3.2 Commentary on retrofit examples

The assumptions for the first improvement stage, ' SHQS 2015', were compared against the criteria outlined in Annex C of the SHQS. Annex C requires dwellings to meet sufficient insulation and heating system criteria (elements 31 to 34), and to attain a minimum NHER or energy efficiency ( SAP) rating in relation to the fuel type (element 35 - see Appendix B). The 'Further Measures 2020' stage is compared against the proposed Environmental Impact ( EI) and Energy Efficiency ( EE) ratings detailed by the EESSH consultation, relating to the calculated CO 2 emissions and operation costs respectively (see Appendix C). These are also referred to as ' EPC ratings' within this report.

The following section summarises the key findings from the review of assumptions associated with each retrofit example, and highlights a number of factors which should be taken into consideration when using the data.

3.2.1 Building form

The example dwellings assume simple building forms. Therefore, it should be noted that any building feature which contributes to increased heat loss surface area, relative to the example dwellings, will result in higher space heating requirements. This will result in higher fuel bills and CO 2 emissions, and consequently lower ( i.e. poorer) EE and EI ratings. This increase in heat loss area may be due to a number of factors, for example:

  • An unusual floor layout resulting in a greater exposed façade area relative to the floor area (this will also prove detrimental to the ground floor U-value due to increased exposed perimeter length, contributing to further heat loss).
  • A dwelling being adjacent to an unheated or partially heated building.
  • A dwelling featuring an unusual building detail, such as a bay window or an overhang, contributing to an increase in the exposed construction area.

Section 5 of this report details a number of additional example dwellings which feature additional heat loss area ( e.g. gable-end flats). The difficulty that some of these dwellings have in reaching the EESSH is discussed in sections 7 and 8. It should also be recognised that some specific features (such as bay windows, heated porches, large glazing ratios, etc.) make the suggested retrofit targets more challenging when modelled in RdSAP. Some of these dwellings will already fall into a conventional 'harder-to-treat' category, where "harder-to-treat" is used to describe solid-wall construction or off-gas properties [7] but can also include homes with no loft or within high-rise developments [8] . It should however be noted that the EESSH consultation takes account of off-gas housing by proposing a lower rating requirement for such properties. For those that do not meet existing harder-to-treat definitions (and it is suggested the number of such dwellings could be quantified), an additional "difficult-to-treat" definition might be required to recognise homes that, although not conventionally harder-to-treat, will have similar challenges in reaching, and funding retrofit measures for the EESSH.

3.2.2 Age band

Example groups C and H cover multiple RdSAP age bands (see table 1). This has little impact for group C as the two age bands would be treated in similar ways. However the two RdSAP age bands covered by group H (1999 to 2007) assume different performance parameters, due to changes in the minimum Building Standards enforced from 2002.

The example dwellings in group H assume that systems are specified to the standards corresponding to the earlier RdSAP age band (1999-2002). The suggested improvement measures specified for example dwellings H21, H22 and H23 are relevant for both age bands within group H. However, the dwellings approved under more stringent building standards will demonstrate better energy performance characteristics. Reference could be made to Appendix D of this report to view a summary of the different performance specifications associated with the pre- and post-2002 building standards that will affect the RdSAP calculation.

Table 1: Retrofit Example dwellings and RdSAP age bands

Time Period EESSH Retrofit Examples RdSAP 9.91 Table S1: Scotland
Pre 1919 A. Pre 1919 Age band A: before 1919
1920-1924 C. Four in a block 1920 - 1949 Age band B: 1919-1929
1930-1934 B. Interwar 1930 - 1949 Age band C: 1930-1949
1950-1954 D. Post War 1950 - 1964 Age band D: 1950-1964
1965-1969 Age band E: 1965-1975
1975-1980 E. 1976 - 1983 Age band F: 1976-1983
1985-1989 F. 1984 - 1991 Age band G: 1984-1991
1990-1994 G. 1992 - 1998 Age band H: 1992-1998
2000-2004 H. 1999 - 2007 Age band I: 1999-2002
2005-2007 Age band J: 2003-2007
Post 2007 Age band K: 2008 onwards

3.2.3 Walls

Cavity wall construction can lead to heat loss through party walls, subject to air movement between lower floors and the loft space. This is taken into account within the SAP methodology, but is yet to be recognised by RdSAP. It should be noted that in such instances, the total heat loss area may be greater than that represented by the RdSAP calculation.

Solid wall, system built and timber frame properties are recognised as harder-to-treat where they cannot accommodate cavity wall insulation. Internal or external wall insulation may be appropriate for some dwellings, but this technology can be limited in application due to a number of factors. For internal solutions the reduction in internal floor area, incompatibility with internal features and tenant disruption are limitations. External wall insulation is usually unsuitable for listed buildings and conservation areas and can be difficult to achieve where the dwelling forms part of a mixed tenure block.

Section 5 presents a number of additional example dwellings which can be considered harder or expensive to treat on account of wall type (including gable end solid wall dwellings, timber frame, 'No fines' concrete and ' BISF' steel frame construction).

3.2.4 Floors

For some properties with uninsulated timber and solid concrete floors, ground floor insulation can substantially improve energy efficiency but the installation process is often very invasive (particularly for solid floors), and may incur additional expense ( e.g. decant of tenants and/or their belongings). Careful consideration needs to be given to the installation process to ensure minimal damage to lifted floorboards, and the provision of sufficient ventilation to mitigate the development of damp or rot. On this basis, the example dwellings note the installation of floor insulation as an 'advanced measure'.

The baseline U-value used by the RdSAP methodology for suspended timber floors (which are the default ground floor construction type for pre-1929 dwellings in Scotland) does not incorporate the benefits of insulation. In line with enhancements to the calculation methodology, documentary evidence applicable to the property being assessed ( e.g. manufacturer specifications) must be retained for audit purposes if overwriting the default U-value associated with the floor, to account for the benefits and the thickness of insulation entered.

Research for Historic Scotland has provided an example of where the default U-values attributed to a solid concrete floor have been observed to differ significantly from those found by examples of in situ testing [9] . This means that, relative to the default U-values associated with the baseline performance, the dwelling would be subject to greater heat loss through the floor, and consequently greater energy and CO 2 emission savings would be expected post-application of the ground floor insulation.

3.2.5 Lofts

There is limited data available to inform the 1990 baseline assumption for loft insulation. It has been observed that the age of the dwelling bears a strong correlation to whether the dwelling has any loft insulation (where there was no national requirement for new houses prior to the mid 1960's [10] ). Appendix E summarises data from a number of sources, considering the level of loft insulation present relative to the age of the dwelling, and reporting the corresponding U-values. Based on this information, two changes to the 1990 baseline assumptions have been proposed, namely:

(i) Specification of 12mm loft insulation for 1965-1975 dwellings, as represented by example group D

(ii) Specification of increased loft insulation thickness from 100mm to 150mm for the baseline dwelling in example group G.

These changes will not affect the EPC ratings achieved by the improvement measures specified for dwellings in example groups D and G, and therefore will not compromise the ability of such dwellings to meet the SHQS or EESSH. They will however alter the energy performance of the 1990 baseline, and thus impact on the percentage improvement relative to this.

In relation to the period 2006 to 2015, installers and Government regulated insulation schemes specified insulation levels of 150, 200 and 250mm considerably exceeded the original SHQS standard. However, in the early 1990s some contractors and housing associations may have installed to the lower SHQS value. Meanwhile, more latterly it is more likely that the higher standard will have been specified. The current SHQS standard specifies 100mm for existing insulation and 270mm as desirable when 'topped up'.

3.2.6 Open chimneys

Fireplaces are a common feature in older buildings. Therefore, some of the dwellings in groups A to D (up to 1964) could feature at least one fireplace for the 1990 baseline assumption. Examples without chimneys will have lower infiltration levels and space heating requirements for the 1990 baseline. As part of the validation exercise this was explored. One and two chimneys were modelled to provide social landlords with an understanding of how this would affect dwellings' energy efficiency and environmental ratings. This testing was separate from the re-modelling of the baseline. This is discussed further and the separate modelling results are detailed in the Report Annex: Review and Validation of Consultation Case Studies (now Retrofit Examples). The impact was not sufficient to change the overall banding of the property, or compliance with the standard.

3.2.7 Boiler efficiency

For pre-1998 boiler systems, the SAP methodology [11] states a winter efficiency of 66%, and a summer efficiency of 56-57% (with the exception of pre-1979 floor mounted boilers, and those that are not fan assisted). Such efficiencies are applied to the baseline dwellings in example groups A to G, and can be reduced by 5% where there are limited controls ( e.g. no room thermostats or thermostatic radiator valves ( TRVs)). The overall efficiency associated with the boiler will be dependent on the extent of the domestic hot water load, relative to the space heating load.

The building standards applicable at the start of the construction period represented by example group H (1999 to 2007) did not specify a minimum boiler efficiency. The target U-values, for exposed building elements, were determined using a seasonal boiler efficiency of 72% (this would be expected to correspond to a non-condensing system). As of 2002, amendments to building standards stipulated a minimum boiler efficiency of 78%, but this was also relative to improved U-value specifications, which have not been applied to the example dwellings. Based on the information detailed in the SAP methodology [12] , winter and summer efficiencies of 74% and 64% respectively are considered appropriate for the group H baseline dwellings.

Standard 6.3 of the Domestic Technical Handbook [13] specifies a minimum seasonal efficiency of 88% ( SEDBUK 2009) for gas boilers. It is proposed that the value used to describe the boiler efficiency for any future installations is no lower than 88% as it represents the minimum efficiency in current building standards.

For biomass boilers it should also be noted that while its EI rating is very good, RdSAP currently allocates a maximum 65% efficiency rating (and it should be noted that there are biomass boilers specified with far greater efficiencies), making it hard for biomass-fuelled properties to achieve a high EE rating.

3.2.8 Lighting

Energy efficient lighting was not prevalent on the market around 1990, therefore 100% inefficient lighting as a baseline assumption is appropriate. As legislation pushes towards the phase out of GLS (tungsten filament) fittings, the assumption of 100% energy efficient lighting for the SHQS 2015 improvement measures seems reasonable.

3.2.9 Renewables

The effective installation of renewables such as photovoltaics ( PV) or solar thermal is sensitive to a number of criteria, for example, orientation, roof space, roof angle, shading, storage space (for hot water) and ability of the existing hot water system to withstand a pressurised system. Should optimal conditions not be met for any of these criteria, additional costs may be incurred to improve the operational performance of the system (for example, investment in a supporting frame to improve the tilt angle or installation of multiple micro-inverters to overcome shading issues).

Given these considerations, renewable technologies should therefore be considered by the retrofit examples as 'advanced measures', as opposed to being included in the ' SHQS 2015' or 'Further Measures 2020' improvement stages. These may, in some cases, be considered sooner by social landlords should they be appropriate for their housing stock. Renewables may, for example, be more widely applied in dwellings located off the gas grid. For these properties the displacement of the high costs and CO 2 emissions from oil, LPG fuel or electricity can yield significant financial and environmental benefits. In some cases this will be a cost-effective way of working towards compliance with the SHQS and improved EESSH. Payment levels from the RHI and Feed in Tariff ( FIT) will be a major factor in these calculations.

3.2.10 Fuel source

Many of the dwellings surveyed for the 1991 SHCS were identified to have no central heating, particularly the older stock (almost half of pre-1919 social rented tenements). These homes featured systems different to those described by the electric- and gas-heated 1990 baseline example dwellings. It should therefore be remembered that the methodology outlines an approach to improve the current Scottish social rented stock, and does not represent actual CO 2 emissions reductions (relative to 1990 levels).

For pre 1919 flats electric heated example dwellings A-1(E) and A-2(E) and gas A-2 (G) are reliant on the application of 'advanced measures' to meet the proposed target ratings for the EESSH, but this approach may not be suitable for all dwellings. Options are more limited for older mid-floor flats like A-2. For those located on or near the gas-grid, fuel switching to gas heating may offer a cost-effective way of attaining the highest possible EPC rating band.

Fuel source needs to be considered early on in an overall plan to reduce the buildings CO 2 emissions, where it will alter the target compliance ratings and influence subsequent technologies installed. Particularly in relation to electric heating, it should be recognised that certain areas of Scotland actually promote electric heating as an affordable and environmentally sustainable heating source. In Orkney [14] , for example, the high volume of renewably-sourced electricity means that electric storage heating systems provide valuable support to the distribution network. Electric storage allows generation peaks from intermittent renewables to be stored and matched with demand peaks. Against this context, households in this area are currently being encouraged to move from high-emissions fuels such as oil and solid fuel to storage electricity and heat pumps. However, because electric-heated properties achieve low EI ratings, they could face additional challenges in meeting the EESSH, depending on what fabric improvements are possible.

Over time the EI of electric heating is likely to improve as the grid is decarbonised. This should be considered in relation to how technologies such as electric storage heaters and heat pumps will relate to the standard in future. For example, a system installed and modelled today will have a significantly better EI if still in operation in 2020. This assumes the Scottish Governments targets for renewable energy productions are achieved.

3.3 Assessment and modelling

3.3.1 Skill of the assessor

The RdSAP calculation plays a significant role in terms of demonstrating compliance for both the SHQS and EESSH. This places a high importance on the skill of the assessor. In addition, a number of the inputs to the RdSAP calculation are subject to the judgment of the assessor, and can have a significant impact on the results. For example, the ability to identify a typical glazing area relative to the age and form of the building will impact on the dwelling's average U-value, and consequently space heating requirements, fuel costs, and CO 2 emissions. A significant issue in RdSAP 2009 9.91 is the ability to input actual U-values, whereas previous versions relied solely on default values. Based on plans, a skilled assessor will be able to identify variations from default U-values for some buildings of non-traditional construction. In other cases research evidence is needed from in-situ testing. For example, solid walls in pre-1919 buildings can have much better U-values than the SAP defaults, as can more modern concrete floors; while single glazing can have significantly worse U-values than the default assumptions [15] . Documentary evidence of in-situ testing must be retained by the assessor for audit purposes if they overwrite any default U-value in RdSAP (more details on this are provided in Section 5.2).

3.3.2 Modelling packages of measures

The example dwellings outline a set of cumulative improvement measures across three stages; ' SHQS 2015', 'further measures 2020' and 'Advanced Measures 2050'. It is important to emphasise that the effect of measures applied across different improvement stages cannot simply be added together: the extent of the savings realised by any individual measure will be dependent on other efficiency measures in place. For example, installing improved glazing in a dwelling with an inefficient boiler will demonstrate greater savings compared to the same installation in a dwelling with an efficient boiler.

The example dwellings do not prescribe a route to compliance but can be used by social landlords to inform the development of an energy efficiency plan, or appreciate the extent of measures necessary to meet the target ratings. Some of the proposed technologies may not be universally suitable. For example, budgets and replacement cycles may restrict the potential to upgrade glazing in the short to medium term, especially where older double glazing has been installed in the late 1990s or early 2000s. In such cases, the measures can be factored into longer-term plans for compliance. A range of measures should therefore be used to identify improvements of sufficient scale to comply with the standard. Any measures that may not be immediately cost-effective ( e.g. triple glazing) may prove relevant to meeting future requirements.

3.4 Options for listed buildings

Older building stock can often be listed buildings or have other conservation status, which will restrict the application of measures, deemed to significantly change the appearance or character of the building.

Social landlords will need to correspond with their local planning authority to determine the likely impact and consequences of any limitations, and identify measures which are deemed acceptable. The Scottish Government may wish to accept documentary evidence from landlords in cases where properties are prevented from meeting the standard due to such planning regulations.

3.4.1 Glazing

Almost all the example dwellings in groups A to C consider the replacement of single glazing with double glazing, but this cannot always be accommodated where the building is listed or otherwise protected.

3.4.2 Wall insulation

The application of this measure may also be restricted in some older and/or listed buildings due to internal features or an excessive reduction in the internal floor area. Internal wall insulation products that are recommended by Historic Scotland for breathability (important to the health of solid masonry walls) do not attain a sufficient U-value to attract ECO support.

Costs for the 'specialist' improvement systems that are often needed in older listed buildings are generally considerably higher than for more standard systems; not only is this not accounted for in the examples' cost assumptions, but this is also likely to be a considerable barrier to uptake for some social landlords.


Email: Agnes Meany