Building regulations - new domestic buildings - modelling of proposed energy improvements: research report

Task 1: Establish Current Baseline

26. Task 1 of the analysis was to establish the current baseline situation for new buildings in Scotland. To achieve this involved three main stages of analysis:

1. Estimating a national annual build profile, based on defined building types/sub-types.

2. Energy modelling of these building sub-types, including identification of current (2015) Building Standards compliant specifications.

27. This section of the report sets out the evidence and rationale for the approach taken. It also summarises key assumptions and results.

1.4 Establishing the national annual build profile for new buildings

28. The first stage of the process was the identification of a suitable number of evidence-based building types and sub-types. 'Building types' refer to the function of the building (detached house, mid-floor flat etc.). 'Building sub-types' represent distinct combinations of building type, heating fuel/technology and ventilation strategy. The selected sub-types should be representative of the buildings constructed in Scotland over the last few years.

29. To derive the building sub-types analysis was undertaken of an extract of the Energy Performance Certificate (EPC) database for new domestic buildings provided by the Scottish Government. This included information relating to certificates issued over the period from January 2016 to December 2018. A total of 53,495 EPC records were included in the analysis.^[1] This analysis and the resulting sub-types are described below.

30. The EPC database contains information on each individual property, including building type, heating fuel/technology and total floor area (other parameters such as fabric U-values, and further information on building services are discussed later in section 0). This information was used to identify the numbers of different building sub-types and their proportions within the database extract. A summary is shown in Table 1.4a below.

Table 1.4a: Proportions of all building types in EPC database extract, split by heating fuel

Building Type	Biomass	DHN	Electric	Mains gas	Oil	LPG	LNG	Total
House	0%	0%	0%	0%	0%	0%	0%	0%
Basement flat	0%	0%	0%	0%	0%	0%	0%	0%
Detached bungalow	0%	0%	1%	1%	1%	0%	0%	3%
Detached house	0%	0%	3%	27%	2%	0%	0%	33%
Detached maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Enclosed end-terrace house	0%	0%	0%	0%	0%	0%	0%	0%
Enclosed end-terrace maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Enclosed mid-terrace maisonette	0%	0%	0%	0%	0%	0%	0%	0%
End-terrace bungalow	0%	0%	0%	0%	0%	0%	0%	0%
End-terrace flat	0%	0%	0%	0%	0%	0%	0%	0%
End-terrace house	0%	0%	0%	7%	0%	0%	0%	8%
End-terrace maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Ground-floor flat	0%	0%	1%	8%	0%	0%	0%	10%
Ground-floor maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Mid-floor flat	0%	1%	1%	9%	0%	0%	0%	11%
Mid-floor maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Mid-terrace bungalow	0%	0%	0%	0%	0%	0%	0%	0%
Mid-terrace house	0%	0%	0%	7%	0%	0%	0%	8%
Mid-terrace maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Semi-detached bungalow	0%	0%	1%	1%	0%	0%	0%	1%
Semi-detached flat	0%	0%	0%	0%	0%	0%	0%	0%
Semi-detached house	0%	0%	1%	13%	0%	0%	0%	15%
Semi-detached maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Top-floor flat	0%	1%	1%	8%	0%	0%	0%	10%
Top-floor maisonette	0%	0%	0%	0%	0%	0%	0%	0%
Total	0%	3%	10%	84%	3%	1%	0%	100%

Notes Percentages do not all add up to 100 due to rounding.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

31. The sub-types which represent more than or equal to 5% of the total building stock in the database extract are highlighted in green in Table 1.4a. Those representing more than 0.5% but less than 5% are highlighted in yellow.

32. For derivation of the national build profile for the purpose of this project, it is sufficient to focus on the most common sub-types as these will allow sufficient determination of the impact of any changes to Building Standards – additional sensitivity analyses can be undertaken later if necessary on less common, but important, sub-types.

33. As a first step, those sub-types representing over 5% of the total were proposed for definite inclusion. This resulted in the following 6 sub-types which represents around 80% of the build mix:

• Detached house, gas
• Semi-detached house, gas
• Mid-terrace house, gas
• Ground-floor flat, gas
• Mid-floor flat, gas
• Top-floor flat, gas

34. The semi-detached house sub-type represents records identified as 'semi-detached house' or 'end-terrace house' in the database, as these forms are equivalent and their average floor areas were very similar.

35. The remaining building sub-types were also considered where they represented over 0.5% of the buildings in the database extract. The most significant remaining sub-types were electric-heated; accounting for 10% of the database entries in total. Within this, detached homes were the most prevalent (3% of all records). However, as the electric-heated sub-types covered a range of built forms, a semi-detached house was chosen as being more representative of electric-heated homes overall. Whilst the average total floor area for electric-heated homes was found to be around 120m² (excluding outliers >300m2), the distribution of the floor areas was skewed towards the lower end of the range – the median value was just under 100m². Of the electric-heated homes, around 70% have an air source heat pump (ASHP) as their main heating source, with a further 5% having other types of heat pump. Therefore this sub-type is assumed to be heated by an ASHP.

36. As the total number of LPG-heated homes, oil-heated homes and homes connected to a heat network were each under 3% these were excluded from the build profile. However, heat networks are planned to be the subject of sensitivity analysis under Task 3 (see section 1.15); potentially along with either direct electric^[2] or oil-heated homes.

37. This gives a revised list of seven sub-types in total, shown in Table 1.4b. The middle column shows the adjusted proposed baseline build mix proportions, based on the EPC proportions but scaled up to total 100%. The final column uses these adjusted proportions and a total annual build figure of 18,207 homes per year to estimate baseline numbers of completed homes by sub-type per year. The total annual build figure is based on Housing Statistics for Scotland data on completions across all sectors for the calendar years 2016-2018, from which an average has been taken.

Table 1.4b: Seven building sub-types selected for analysis

Building Sub-types	Proportion of EPC database buildings	Proportion of baseline build mix	Annual build numbers
Detached house, gas	27%	30%	5,521
Semi-detached house, gas	21%	23%	4,158
Mid-terrace house, gas	7%	8%	1,467
Ground-floor flat, gas	8%	9%	1,706
Mid-floor flat, gas	9%	9%	1,706
Top-floor flat, gas	8%	9%	1,706
Semi-detached house, ASHP	10%	11%	1,943
Total	90%	100%	18,207

Notes Percentages do not add up to totals due to rounding.

The figures for semi-detached house, ASHP are based on the total number of electric-heated homes of any built form.

The 'proportion of baseline build mix' figures for the flats have been adjusted to be of equal proportions, to represent a 3-storey block of flats. The total number of flat blocks built per year is calculated as 1706*3/12, which assumes a 3-storey block with 4 flats per storey.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Government, Housing Statistics for Scotland – All sector new build (completions), September 2019 (data for calendar years 2016-2018 used to derive an average annual total build rate)

38. The ventilation strategy was also considered as a determining factor for identifying different sub-types for analysis. From the EPC records, the particular ventilation type could not be determined, as the main ventilation field ('Mechanical Ventilation') only had three options: 'mechanical, extract only', 'mechanical, supply and extract', or blank (assumed to be naturally ventilated homes). As the first option only accounted for 20 of the 53,495 analysed records, it is reasonable to assume that the 'mechanical, supply and extract' option (accounting for around 45% of all records) covers various types of mechanical ventilation, not just Mechanical Ventilation with Heat Recovery (MVHR) but also including mechanical extract ventilation (MEV).^[3] Feedback from developers via Homes for Scotland suggested that MVHR was not commonly specified, supporting this assumption. Overall, there appeared to be a downwards trend for mechanical vent across the registration years for all dwelling types (though data only covered three years), and a move to natural ventilation. Looking at 2018 registrations only, natural ventilation accounted for the following proportions of each building sub-type: mains gas detached 50%, semi-detached 57%, mid-terrace 56%, flats 76%, and ASHP semi-detached 78%.

39. The EPC database extract provided good evidence for including natural ventilation sub-types across the range of built forms and heating types. Including building sub-types for the mains gas houses cases based on MEV was additionally considered. Feedback from Homes for Scotland also suggested that where mechanical ventilation is specified, decentralised MEV (dMEV) is more commonly used in houses (and centralised MEV in flats). Therefore a dMEV case was tested using SAP 9.92 software, based on the mains gas semi-detached sub-type and using the specifications set out in section 0, but with dMEV, and the PV array size being reduced slightly to achieve compliance within 1% of the Target Emission Rate (TER). It was found that results did not differ substantially from the natural ventilation case. The adjustment to achieve compliance within 1% of the TER meant that the Dwelling Emission Rate (DER) would be the same as the natural ventilation case, using SAP 9.92 carbon emission factors. Using SAP 9.92 energy results but applying SAP 10.1 carbon emission factors and primary energy factors led to the dMEV case DER being 4% lower than the natural ventilation case, and the Dwelling Primary Energy Rate (DPER) being 2% lower for dMEV. It was also expected that capital costs for the two cases would be similar. Therefore it was agreed with Scottish Government that separate dMEV sub-types were not needed for inclusion in the baseline scenario.

40. Data on representative building floor areas for each of the building sub-types selected for analysis was required to inform energy modelling and costing. The EPC database extract was interrogated for this information, and Table 1.4c shows the average total floor areas (TFA) found for each sub-type.

Table 1.4c: Building sub-type average total floor areas ( TFA), based on EPC database extract

Building Sub-type	Average TFA in EPC database* (m²)	Proposed TFA in modelling (m²)	Basis for model
Detached house, gas	142	141	England/Wales Part L 2020 detached house, adjusted
Semi-detached house, gas or ASHP	87	84	England/Wales Part L 2020 terraced house
Mid-terrace house, gas	86
Ground-floor flat, gas	68	70	England/Wales Part L 2020 large flat
Mid-floor flat, gas	70
Top-floor flat, gas	71

NotesOutliers which might skew results have been excluded from this analysis, as follows: TFA > 300m² for detached houses (under 3% of all detached houses), TFA > 150m² for semi-detached houses (under 3% of all semi-detached and end-terrace houses). No adjustments made for mid-terrace houses or flats as excluding outliers made little difference to results.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

41. Table 1.4c also shows the TFAs used in the modelling. These have been based on existing models where possible (i.e. those available either from previous modelling to inform the 2015 revision of Building Standards for new dwellings in Scotland, or those used for the recent Part L 2020 modelling in England/Wales). Models have been selected based on their TFAs being within 5% of the averages found in the EPC database extract. Where this was not possible (i.e. in the case of the detached house) it is proposed that an existing model is scaled for use in the analysis.

42. For flats, the distribution of TFAs was also looked at in case there would be a significant advantage to modelling two different flat sizes. However, analysis did not strongly support a case for this, suggesting that the single size is reasonably representative of flats constructed in Scotland (the median flat TFA was also around 70m²). Additionally, previous work on Part L 2020 in England and Wales can be used to inform consideration of where constraints might apply in flats for particular solutions (e.g. heat pump sizing for smaller flat sizes).

43. Table 1.4d summarises key dimensions for the building forms used in the analysis.

Table 1.4d: Building type summary of key dimensions

Areas (m2)	Detached	Semi-detached	Mid-terrace	Flat - Ground	Flat - Mid	Flat - Top
Party wall	0	42	84	29	29	29
Exposed wall	172	94	52	41	41	41
Semi-exposed wall	0	0	0	12	12	12
Roof –Main	70	42	42	0	0	70
Roof – Bay window	1	1	1	0	0	0
Ground floor	71	43	43	70	0	0
TFA	141	84	84	70	70	70
Upper floor (1st)	70	42	42	0	0	0
Total window area	28.8	14.6	12.8	13.8	13.8	13.8
Window area – North	3.0	0.0	0.0	7.9	7.9	7.9
Window area – East	15.3	7.6	7.6	0.0	0.0	0.0
Window area – South	1.7	1.8	0.0	0.0	0.0	0.0
Window area – West	8.7	5.2	5.2	5.9	5.9	5.9
Total as % of TFA	20%	17%	15%	20%	20%	20%
Opaque door area	4.3	2.1	2.1	2.1	2.1	2.1
Zone 1	19	20	20	24	24	24
Zone 1 as % of TFA	13%	24%	24%	35%	35%	35%

Notes Some figures do not add up to totals due to rounding.

Zone 1 areas as proportions of TFA are similar to the averages for the equivalent dwelling types in the EPC database (where they are as follows: detached 16%, semi-detached 21%, mid-terrace 23%, flat 35%).

For the flat building types, some results will be presented in terms of a block of flats (assumed to consist of 12 flats in total; 3 storeys of 4 flats each; assumption agreed with Scottish Government).

Window areas include glazed doors. Were solid doors included in addition, the % of TFA calculations would be closer to 25% (as in the notional dwelling), though still slightly under 20% for the semi-detached and mid-terrace dwelling types.

Source Floor plans and elevations for England/Wales dwelling types, with scaled adjustments where required to meet TFAs.

1.5 Energy modelling of building sub-types

1.5.1 2015 compliant specifications – introduction

44. The EPC database extract was reviewed to find typical specifications to inform 2015 Building Standards compliant models. These models form the basis of the counterfactual/baseline assessment. The method used for the analysis and results found are set out below.

45. The same exclusions were made from the database as described in Section 1.4 above (with the same 53,495 EPC records being included in the analysis). For analysis of fabric U-values, for each element type records were additionally excluded where the average U-values in the database exceeded the maximum area-weighted average values set in the Scottish Technical Handbook – Domestic, Table 6.3, column (a) (Scottish Government, 2019).

46. For elements of the specification where numerical values were applicable (e.g. U-values), the analysis focussed on median values. It also looked at interquartile ranges in these cases. For non-numerical data (e.g. heating control types), the most common entries were focussed on. A guiding principle was that the 2015 notional building specifications set out in Table 6.1 of the Technical Handbook (Scottish Government, 2019) for the relevant fuel types (mains gas, electricity) would be adopted unless there was evidence to suggest that common practice involved significant deviation from the specification (e.g. for U-values, if the notional building value was outside the interquartile range). Where data was not available in the EPC database, notional building values were adopted.

47. Published responses to the Scottish Government's recent review of energy standards: call for evidence were also used to inform the analysis (Scottish Government, 2018). In particular, they were used to sense-check overall compliance strategies and to inform changes required to specifications where the EPC database analysis would otherwise have led to non-compliant models. The main area where this was the case was in relation to solar PV.

48. The findings are summarised in the sub-sections and tables below, covering fabric (section 1.5.2, Table 1.5a), ventilation (section 1.5.3, Table 1.5d), heating (section 1.5.4, Table 1.5e and Table 1.5f), lighting (section 1.5.5, Table 1.5g), and PV (section 1.5.6, Table 1.5h). The three data columns in each of the tables respectively show the proposed Building Standards 2015 compliant specification; the 2015 notional building specification; and the values found in the EPC database.

1.5.2 2015 compliant specifications – fabric

49. For building fabric, where data was available in the EPC database extract, the findings suggested that the specifications in the 2015 notional building are a good representation of common practice for new build homes. Therefore no changes to the notional building specification are proposed, except for window areas/orientation to reflect the building forms being modelled. Details are shown in Table 1.5a.

Table 1.5a: Building Standards 2015 compliant specifications – fabric

Element	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
External Wall U-value	0.17	0.17	0.17 is within 25th-75th percentiles. Median is 0.19 (0.18 for flats).
Corridor Wall U-value	0.17	0.17	-
Party Wall U-value	0.0	0.0	0.0 is most common value (>90% of homes).
Floor U-value	0.15	0.15	0.15 is median value.
Roof U-value	0.11	0.11	0.11 is median value.
Window U-value	1.4	1.4	Not available
Window g-value	0.63	0.63	Not available
Door U-value	1.4	1.4	Not available
y-value	0.08	0.08	Not available
Thermal Mass Parameter	As actual	As actual	Not available
Window overshading	Average	Average	Not available
Window orientation	Based on floor plans and England/Wales assumptions	E/W	Not available
Exposed window, door, rooflight area	Based on floor plans	25% TFA (or, if less, total exposed façade area)	Not available

Notes As explained above, where EPC data was missing, 2015 notional values have been taken.

For window orientation/area details, see Table 1.4d above.

Units for U-values and y-values are W/m²K.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019); Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

50. For the Thermal Mass Parameter (TMP), the values for the building types have been calculated based on the dwelling dimensions and the assumptions set out in Table 1.5b. Timber framed walls have been assumed as this is the prevalent construction in Scotland.^[4] The resulting values are shown in Table 1.5c.

Table 1.5b: TMP calculation construction assumptions

Element	Construction
Heat loss floors	Slab on ground, screed over insulation
Party floors / roofs	Timber I-joists, carpeted
Intermediate floors / roofs	Plasterboard ceiling, carpeted chipboard floor
Heat loss walls	Timber framed, one layer of plasterboard
Party walls	Twin timber frame, double plasterboard both sides
Internal walls	Plasterboard on timber frame
Heat loss roofs	Plasterboard, insulated at ceiling level

Source AECOM assumptions

Table 1.5c: TMP calculated values by building type

Building type	TMP (kJ/m²K)
Detached house	102
Semi-detached house	109
Mid-terrace house	115
Flat - Ground	142
Flat - Mid	62
Flat - Top	71

Source AECOM calculations and thermal mass assumptions for different construction elements used in SAP.

1.5.3 2015 compliant specifications – ventilation

51. For ventilation, the main proposed change compared to the 2015 notional building is to the air permeability rate, as analysis showed that the notional value was outside the interquartile range for EPC database entries. The number of sheltered sides is also proposed to be varied, based on built form. Details are shown in Table 1.5d.

Table 1.5d: Building Standards 2015 compliant specifications – ventilation

Element	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
Ventilation type	Intermittent extract fans with trickle vents	Intermittent extract fans with trickle vents	Non-mechanical is most common type
Air permeability rate (m³/m².h @50Pa)	5.0	7.0	5.0 is median value, interquartile range 4.0-5.0.
No. of extract fans	4 for houses; 3 for flats (as per 2015 notional)	4 for dwellings with TFA>80m2, 3 for smaller dwellings	Not available
No. of sheltered sides	Based on built form (0 for detached, 1 for semi-detached, 2 for mid-terrace and flats)	2	Not available
No. of chimneys/open flues	0	0	Not available

Notes As explained above, where EPC data was missing, 2015 notional values have been taken.

For more detailed analysis of data on ventilation types/strategies in the EPC database, see Section 1.4 above.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

1.5.4 2015 compliant specifications – heating

52. For heating, mains gas heating and electric (ASHP) space and water heating specifications are considered separately. Details are shown in Table 1.5e and Table 1.5f.

53. For mains gas, the main proposed change compared to the 2015 notional building is the exclusion of waste water heat recovery (WWHR), as EPC database analysis showed that this was installed in only a small proportion of cases.

54. Other more minor changes include the assumption of combi boilers for all building types except detached houses, based on EPC database analysis. The hot water cylinder size for the detached house is also proposed to be varied for costing purposes, based on assumptions made in England and Wales Part L 2020 work, but with the assumption that the declared loss factor specified in the 2015 notional is met. For showers, a new data entry field is required for SAP 10 – here the assumed value has been based on the maximum fittings consumption optional requirement level for showers in Approved Document G in England, Table 2.2 (HM Government, 2016). This is the same assumption as used in Part L 2020 modelling work for England and Wales. It was used as an equivalent value could not be found in the Scottish Technical Handbook – Domestic section 3.27.2. For heating flow temperatures, categories have also changed in SAP 10.1 and an assumption of >55°C has been made based on AECOM's understanding of common practice and the assumption used in Part L 2020 modelling work for England and Wales.

Table 1.5e: Building Standards 2015 compliant specifications – gas cases

Element	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
Space Heating Source	Condensing gas boiler	Condensing gas boiler	Condensing gas boiler most common
Emitters	Radiators (standard size)	Radiators (standard size)	Radiators in nearly all cases
Efficiencies (SEDBUK)	89.0%	89.0%	Not available
Flow temperatures	>55°C	55°C (e.g. radiators as emitters); max. category in SAP 9.92 is >45°C	Majority >45°C (max. category)
Controls	Time and temperature zone control, interlock, ErP Class V controls, delayed start	Time and temperature zone control, interlock, weather compensation, delayed start	Time and temperature zone control most common; no data on other items
FGHR?	None	None	FGHR in about 10% of mains gas cases
Pump details	2013 or later, in heated space	Not specified - assume same	2013 or later most common, no data on location
Flue type	Balanced, fan-assisted	Balanced, fan-assisted	Balanced in nearly all cases
Boiler type/size	Detached: 18kW system/regular Semi/Mid/Flat: 24kW combi	System/regular for all dwelling types	Boiler+cylinder in majority of detached houses, combi in majority of other dwelling types
Domestic Hot Water Source	As for space heating	As for space heating	As for space heating
Hot water cylinder size (where applicable)	200l, detached house only	150l, all dwelling types	Not available
Hot water cylinder declared loss factor (where applicable)	1.89 kWh/day	1.89 kWh/day	Not available
Primary circuit loss assumptions (where applicable)	Cylinder thermostat, separate timer, fully insulated primary pipework	Cylinder thermostat, separate timer, fully insulated primary pipework	Not available
Shower flow rate (l/min)	8	n/a	n/a
No. of showers	2 for all building types	n/a	n/a
No. of baths	1 for all building types	n/a	n/a
WWHR?	None	Instantaneous, 45% efficiency, 2 showers if TFA>100m², otherwise 1 shower	WWHR in only 4% of mains gas homes (only 3% in 2018)
Secondary heating	None	None	Secondary heating very uncommon
Electricity tariff	Standard	Not specified - assume standard though	Nearly all on standard tariff

Notes As explained above, where EPC data was missing, 2015 notional values have been taken.

Boiler efficiencies will be adjusted for relevant control types and flow temperatures as required. For SAP 9.92, weather compensation provides a +3% improvement (checked in example Scotland TER worksheet as not specified in 2015 notional building) resulting in a data entry of 92.0%; in SAP 10.1 Table D1 the improvement is only +0.7% for design flow temperatures over 55°C resulting in a data entry of 89.7%.

See text above table for discussion of hot water cylinder assumptions.

The numbers of showers and baths have been based on building type floor plans. The shower flow rate corresponds to the maximum value in Section 7 Aspect Silver level 4. These assumptions are required for SAP 10.1 only.

For SAP 10.1 modelling, it is assumed that water use is below 125 litres per person per day (data entry not required for Scotland in SAP 9.92).

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

55. For the ASHP case, the main proposed changes compared to the 2015 notional building are the use of the heat pump to provide water heating as well as space heating (instead of electric immersion only); exclusion of secondary space heating; and exclusion of waste water heat recovery (WWHR). These changes are based on the specifications in the majority of ASHP entries in the EPC database analysis.

Table 1.5f: Building Standards 2015 compliant specifications – ASHP case

	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
Space Heating Source	ASHP	ASHP	ASHP most common electric heating type
Emitters	Radiators (assume large size)	Radiators (assume large size)	Radiators most common emitter type
Efficiencies (Seasonal Performance Factor, SPF)	170%	175.1%	Not available
Flow temperatures	Assumption not required; use efficiency	55°C	>45°C most common temperature
Controls	Time and temperature zone control	Time and temperature zone control	Time and temperature zone control most common
Domestic Hot Water Source	As for space heating, including electric immersion	Electric immersion only	From main most common in majority of cases
Hot water cylinder size (where applicable)	150l	150l, all dwelling types	Not available
Hot water cylinder declared loss factor (where applicable)	1.89 kWh/day	1.89 kWh/day	Not available
Primary circuit loss assumptions (where applicable)	Cylinder thermostat, separate timer, fully insulated primary pipework	Cylinder thermostat, separate timer, fully insulated primary pipework	Majority have cylinder thermostat, other data not available
Shower flow rate (l/min)	8	n/a	n/a
No. of showers	2 for all building types	n/a	n/
No. of baths	1 for all building types	n/a	n/a
WWHR?	None	Instantaneous, 45% efficiency, 2 showers if TFA>100m², otherwise 1 shower	No WWHR in nearly all ASHP cases
Secondary heating	None	10% electric	No secondary heating in majority of ASHP cases (around 75%)
Electricity tariff	Standard	Not specified - assume standard	Majority on standard tariff (around 90%)

Notes As explained above, where EPC data was missing, 2015 notional values have been taken.

ASHP efficiency based on SAP default as a specific value cannot be specified in SAP 9.92 or SAP 10.1, and as value is close to notional and ASHP case over-complies. It should be noted however that the Domestic Building Services Compliance Guide (Scottish Government Building Standards Division, 2014) sets a minimum Coefficient of Performance (CoP) of 2.5 for space heating and of 2.0 for water heating (SAP efficiencies are based on the Season Performance Factor (SPF), which is not directly comparable. The CoP is a measure of instantaneous efficiency, whereas the SPF is a measure of average efficiency across the year taking account of the varying temperature conditions but importantly also considering energy use of auxiliary components such as circulation pumps and direct electric heating (where present)^[5]. Scottish Government confirmed their preference to model the default efficiency however rather than adjusting other specification elements. See discussion of overall specification in section 1.5.8 below.

The numbers of showers and baths have been based on building type floor plans.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

1.5.5 2015 compliant specifications – lighting

56. Nearly all records in the EPC database had 100% low energy lighting. However more detail is needed for SAP10; the proposed assumptions are shown in Table 1.5g.

Table 1.5g: Building Standards 2015 compliant specifications – lighting

	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
Fixed lighting capacity (lm)	185 x TFA	n/a	n/a
Efficacy (lm/W)	80	n/a - 100% low-energy lighting assumed	100% low energy lighting in nearly all cases

Notes The fixed lighting capacity is based on the SAP10.1 Appendix L default calculation where there is no fixed lighting/amount is unknown, and is what was suggested by BRE for Appendix R (notional building) in England.

The lighting efficacy is based on an AECOM suggestion for an improved minimum standard reflecting good practice.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

BRE, SAP10.1 Appendix L (BRE, 2019)

1.5.6 2015 compliant specifications – solar PV

57. Solar PV is included in the 2015 notional building, but was only found in about 30% of all records for flats and 20% of all records for houses in the EPC database across the 3-year period. However, when looking at 2018 registrations and mains gas heated homes only, PV is installed in around 50% of flats and 40% of houses.

58. The other elements of the proposed 2015 notional building specifications set out in the sections above were modelled for the building sub-types. It was found that without PV, the mains gas cases would be significantly under-compliant.

59. From this, and based on other findings (for example, the large number of cases in the EPC database extract where maximum U-values were exceeded), it is reasonable to suppose that many of the database entries are not compliant with 2015 Building Standards, but were assessed under earlier standards.^[6]

60. Due to the cases being under-compliant other sources of evidence were reviewed; in particular the published responses to the recent consultation on energy standards (Scottish Government, 2018). These did not provide strong evidence for improving other elements of the 2015 compliant specification, but provided a good evidence base for the inclusion of solar PV. Quotes from a selection of the responses are included below:

• "Generally, the industry has adapted to the 2015 Technical Standards, by a combination of improving fabric and services specification and the addition of renewable technologies." – Professional membership body (architecture)
• "most developers will be installing PV to comply with Section 6." – Warranty provider
• "the use of renewable energy has also increased, usually photovoltaic panels are recommended to aid compliance with the standards." – SAP software provider
• "To meet the current guidance we did not improve our building fabric. We merely added PV." – Housing developer
• "Major housebuilders and developers now recognise solar PV as one of the most effective and affordable methods to meet building regulation demands, as well as the multitude of benefits it brings to homeowners. Our members estimate as much as 70% of new developments now includes solar PV" – Renewable energy trade association

61. Based on the above, it is proposed to include PV in the gas-heated 2015 compliant cases. A summary of the proposed PV modelling specifications for gas-heated building sub-types is shown in Table 1.5h, where it is compared to the 2015 notional and the EPC database findings for main gas heated homes where PV was installed. The draft modelled array sizes (based on SAP 9.92 modelling of the other specification assumptions set out above) are shown in Table 1.5i.

Table 1.5h: Building Standards 2015 compliant specifications – solar PV

	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
PV calculation	Initially based on EPC database averages, but adjusted to achieve compliance within 1% of TER	kWp = 1% of building TFA	n/a
Region	UK average	UK average	Not available
Roof area required (m²/kWp)	6.5	8.3	n/a
Overshading factor	1.0 (none/very little)	1.0 (none/very little)	Majority none/very little (around 95%)
Orientation	SW	SW	Over 75% orientated between SE and SW
Pitch	30°	30°	Majority at 30° (around 80%)
% exported to grid	Calculated in SAP	Calculated in SAP	Calculated in SAP

Notes The 2015 notional building PV array size is subject to following limit: not exceeding 30% of roof area (divided by no. of storeys in block for flats), based on 30° roof pitch and 0.12kWp/m² PV area.

The proposed m²/kWp assumption has been updated from the notional building value based on AECOM's understanding of standard current performance. The proposed value is the same as that used in modelling for Part L 2020 in England and Wales. It is used to help inform costing.

Source AECOM analysis of EPC database extract 2016-18 SAP new build (provided by Scottish Government, December 2019)

Scottish Technical Handbook – Domestic, Table 6.1 (Scottish Government, 2019)

Table 1.5i: Building Standards 2015 compliant specifications – solar PV array sizes

PV array sizes (kWp)	Proposed 2015 Compliant Specification	2015 Notional	EPC Database
Detached house, mains gas	1.61	1.17	Mean 1.51 where installed, median 1.38, interquartile range 1.10-1.75
Semi-detached house, mains gas	0.95	0.84	Mean 0.97 where installed, median 0.83, interquartile range 0.75-1.10
Mid-terrace house, mains gas	0.95	0.84	Mean 0.84 where installed, median 0.80, interquartile range 0.75-1.04
Flat, mains gas	0.85	0.70	Mean 0.62 where installed, median 0.50, interquartile range 0.50-0.75
Block of flats, mains gas	10.20	8.40	-
Semi-detached house, ASHP	n/a	n/a	Majority of ASHP-heated homes did not have PV

Notes PV array sizes have been based on achieving compliance just below the TER.

The array size for the flats is at the upper limit based on the 2015 roof area limit (but this is based on 2015 assumptions of panel performance/roof area required, and it should be noted that there have been improvements in the performance of PV panels compared to 2015 assumptions).

Source AECOM modelling in SAP9.92.

62. As noted in Table 1.5i, for the ASHP-heated case no PV is assumed, as in the notional building (and in the majority of ASHP-heated homes in the EPC database).

1.5.7 Viability and cost-effectiveness considerations

63. The viability and cost-effectiveness of the specifications identified for the 2015 compliant building sub-types was considered. As the specifications were in general closely aligned with the median values or most common system configurations found in the EPC database, or otherwise with responses provided to the 2018 call for evidence consultation, this provided strong evidence for viability.

64. Specific development constraints in some situations would influence different specifications in practice from those set out above. The feasibility of mains gas for heating depends on the availability of a gas supply. In some circumstances the viability of PV generation may be limited – for example due to constraints on orientation or over-shading, or on the electricity grid. In both of these cases, the ASHP specification provides an example of an alternative compliance case, and the EPC database evidence indicates ASHP is installed in a significant proportion of new build homes (around 7% of homes analysed in database extract).

65. The use of specifications supported by the EPC database analysis and 2018 consultation responses provided evidence for cost-effectiveness as well as viability, as developers would not commonly adopt non cost-effective solutions to compliance. There are of course other cost-effective solutions available (an example of which might be waste water heat recovery), but it was considered most important to reflect current common practice in the baseline. A review of the specification items and overall strategies set out above did not flag significant concerns on their cost-effectiveness based on AECOM's understanding of cost data, impact on energy/carbon reductions, and the 2015 compliance targets, though some commentary on PV and ASHP is provided below.

66. Previous work on cost-optimal standards for the UK (MHCLG, 2019c) showed that where PV is being installed, larger array sizes are more cost-effective than smaller ones (in the absence of targets which may tend to limit sizes for compliance reasons). Where suitably-orientated roof-space is available, some developers may be choosing to put larger PV array on houses, in particular, and to relax other specification elements such as fabric – as suggested by one of the responses to the 2018 consultation cited in paragraph 60 above. However, although the sizes modelled for the houses are not large, they are still a minimum of 1kWp, and they are based on EPC database evidence. This item of the specification should be reviewed in the analysis of 2021 standards.

67. In the modelled 2015 ASHP case, the specification over-complies meaning there is potential for more cost-effective solutions from a capital cost perspective, particularly if more efficient heat pumps are installed in practice. For example, the fabric specifications could be relaxed to the maximum limits set in Table 6.5 of the Technical Handbook (Scottish Government, 2019a), though this would have negative impacts on occupants including increased fuel bills. The rationale for the specifications modelled is discussed in section 1.5.8 below. The potentially unintended implications of compliance flexibility for ASHP-heated homes should also be considered as part of the analysis of 2021 standards.

1.5.8 Modelling using SAP 9.92

68. The seven building sub-types set out in Table 1.4b were modelled with the 2015 compliant specifications set out in Table 1.5a to Table 1.5i. Initially, the current version of the National Calculation Methodology for domestic buildings, the Standard Assessment Procedure version 9.92 (SAP 9.92), was used (BRE, 2014). This is the version used for checking compliance with Building Standards 2015. Modelling was undertaken using Stroma's FSAP2012 software, which implements the SAP 9.92 methodology.

69. The main purpose of modelling the building types using SAP 9.92 was to check compliance with Building Standards 2015. The modelling informed the specifications set out above. The Scottish Government had requested that for each model the Dwelling Emission Rate (DER) would be within 1% of the Target Emission Rate (TER). In the gas-heated cases, this was achieved through the specifications set out above, and with adjusting the PV array sizes as required whilst keeping within reasonable limits (final modelled values are shown in Table 1.5i). For the flats, compliance was checked based on block-averaging rather than on individual units.

70. However the ASHP-heated case over-complies (with an improvement on the TER of around 13%). This is because the EPC database evidence suggests that ASHP-heated homes are over-complying in practice, in particular due to the following differences compared to the 2015 electric notional building in the majority of homes: omission of secondary heating, use of the heat pump for water heating as well as space heating and lower air permeability rates. Whilst waste water heat recovery was also omitted in the majority of homes (but is included in the notional building), this did not outweigh the other improvements. Analysis of the database did not suggest that other specification areas (e.g. U-values or air tightness) were relaxed for ASHP-heated homes. In practice ASHP-heated homes may over-comply even further if heat pump efficiencies are commonly better than the low value in the 2015 electric notional building, but this data was not available in the EPC database. Alternative specifications were discussed with Scottish Government Building Standards Division, but they took the view that the modelling should be based on the EPC database evidence and on the default heat pump efficiencies in SAP.

71. Key results from the modelling are set out in Table 1.5j. The columns show the TER and DER as calculated in SAP 9.92, using SAP 9.92 carbon emission factors; the DER calculated using the SAP 9.92 methodology but with SAP 10.1 carbon emission factors; and the Dwelling Primary Energy Rate (DPER) calculated using SAP 10.1 primary energy factors. Showing both SAP 9.92 and SAP 10.1 DERs allows a comparison of the impact of changes to the carbon emission factors separately from other changes to the methodology, as is discussed in section 1.5.10.

Table 1.5j: SAP 9.92 key modelling results by building sub-type – 2015 compliant cases – carbon emissions and primary energy

Building sub-type	TER - SAP 9.92 Factors	DER - SAP 9.92 Factors	DER - SAP 10.1 Factors	DPER - SAP 10.1 Factors
Detached house, gas	11.1	11.1	12.7	69.2
Semi-detached house, gas	13.5	13.5	14.5	79.8
Mid-terrace house, gas	12.3	12.3	13.3	73.4
Average flat, gas	12.6	12.6	13.8	75.9
Semi-detached house, ASHP	25.9	22.5	5.4	65.5

Source AECOM modelling using Stroma's FSAP 2012.

Results recorded and calculated in AECOM, '200122 Scotland Building Standards 2021 - SAP9.92 Results – v2.xls'

72. Further results from the modelling are set out in Table 1.5k. The columns show energy consumption by end-use, and energy generation from onsite PV (where applicable) as calculated using the SAP 9.92 methodology.

Table 1.5k: SAP 9.92 key modelling results by building sub-type – 2015 compliant cases – annual energy consumption by end-use, and annual onsite energy generation ( kWh/yr)

Building sub-type	Space heating	Water heating	Pumps and fans	Lighting	PV generation
Detached house, gas	6,376	2,691	75	491	1,326
Semi-detached house, gas	3,618	2,458	75	369	782
Mid-terrace house, gas	3,113	2,466	75	379	782
Average flat, gas	2,514	2,306	75	314	700
Semi-detached house, ASHP	1,953	1,301	30	369	0

Source AECOM modelling using Stroma's FSAP 2012.

Results recorded in AECOM, '200122 Scotland Building Standards 2021 - SAP9.92 Results – v2.xls'

1.5.9 Modelling using SAP 10.1

73. The seven building sub-types and 2015 compliant specifications set out in Table 1.5a to Table 1.5i were subsequently modelled using the latest available version of SAP, SAP 10.1 (BRE, 2019). This is the proposed version to be used for checking compliance with Building Standards 2021 in Scotland (and Part L 2020 in England and Wales). Modelling was undertaken using an offline program provided by the BRE, which implements the SAP 10.1 methodology (SAP.exe Build 7, 20/01/20).

74. Key results from the modelling are set out in Table 1.5l. The columns show the DER and DPER calculated using the SAP 10.1 methodology (including SAP 10.1 carbon emission factors and primary energy factors).

Table 1.5l: SAP 10.1 key modelling results by building sub-type – 2015 compliant cases – carbon emissions and primary energy

Building sub-type	DER - SAP 10.1	DPER - SAP 10.1
Detached house, gas	13.0	69.0
Semi-detached house, gas	15.0	80.2
Mid-terrace house, gas	13.7	73.2
Average flat, gas	14.5	77.6
Semi-detached house, ASHP	6.1	63.7

Source AECOM modelling using BRE's SAP 10.1 software SAP.exe, Build 7, 20/01/20.

Results recorded in AECOM, '210507 Scotland Building Standards 2021 - SAP10.1 Results – v10.xls'

75. Further results from the modelling are set out in Table 1.5m. The columns show energy consumption by end-use, and energy generation from onsite PV (where applicable) as calculated using the SAP 10.1 methodology.

Table 1.5m: SAP 10.1 key modelling results by building sub-type – 2015 compliant cases – annual energy consumption by end-use, and annual onsite energy generation ( kWh/yr)

Building sub-type	Space heating	Water heating	Pumps and fans	Lighting	PV generation
Detached house, gas	6,036	3,257	86	264	1,326
Semi-detached house, gas	3,315	3,007	86	198	782
Mid-terrace house, gas	2,774	3,018	86	204	782
Average flat, gas	2,259	2,849	86	169	700
Semi-detached house, ASHP	1,748	1,524	0	198	0

Source AECOM modelling using BRE's SAP 10.1 software SAP.exe, Build 7, 20/01/20.

Results recorded in AECOM, '210507 Scotland Building Standards 2021 - SAP10.1 Results – v10.xls'

1.5.10 Impact of the calculation methodology

76. A secondary purpose of the SAP 9.92 modelling was to help understand the impact of changes to the calculation methodology between SAP 9.92 and SAP 10.1. The mains gas semi-detached case has been used as an example to assess impacts on energy demands, based on the specification assumptions set out above. Results are shown in Table 1.5n and discussed in Table 1.5o.

Table 1.5n: Comparison of SAP 9.92 and SAP 10.1 energy modelling results for semi-detached 2015 compliant example ( kWk/yr)

Semi-detached house, gas	Space heating	Water heating	Pumps and fans	Lighting	Total Gas	Total Elec	PV generation
SAP 9.92 Results	3,618	2,458	75	369	6,076	444	782
SAP 10.1 Results	3,315	3,007	86	198	6,322	284	782

Source AECOM modelling, see Table 1.5k and Table 1.5m above.

Table 1.5o: Commentary on SAP 10.1 and 9.92 differences in energy modelling results, based on example case

Finding – SAP 10.1 compared to 9.92	Commentary
Space heating: energy consumed decreases by 8%.	Relates to assumed heating pattern changing in SAP 10.1 to a consistent pattern across the whole week, instead of a different pattern at the weekend (BRE, 2016a). Also relates to some increased gains from water heating (where consumption is increased), and increased gains from lighting in the consultation version of SAP 10.1 used in the modelling (as implemented by BRE) compared to SAP 9.92. Balancing the above somewhat, the heating efficiency adjustment due to weather compensation is lower in the SAP 10.1 modelling (+0.7% vs +3.0% gross points), tending to increase energy demands. In SAP 10.1 heating efficiency adjustments due to various types of controls have been differentiated by design/flow return temperatures and a new design flow/return temperature category has been added (80/60 or 70/60) which is assumed in the modelling.
Water heating: energy consumed increases by 22%.	The calculation of hot water consumption has been adjusted to account for different shower types and flow rates. In our example semi-detached model, the average hot water usage has increased (from around 100 litres/day in SAP 9.92 to around 120 litres/day in SAP 10.1), affecting water heating demands. The heating efficiency adjustment due to controls is also lower in SAP 10.1, increasing overall consumption. The changes to SAP 10.1 to allow shower types and flow rates to be specified were intended to better reflect consumption, and are particularly important for estimating savings from solar hot water or waste water heat recovery (BRE, 2016b).
Pumps & fans: energy consumed increases by 15% (but this is a small absolute increase).	Assumption for annual electricity consumed by heating circulation pump (2013 or later) has increased by 11kWh/yr in SAP 10.1 Table 4f compared to SAP 9.92 Table 4f.
Lighting: energy consumed decreases by 46%.	The calculation of lighting energy has been updated in SAP 10.1 to allow recognition of different low energy lighting types. Individual lighting efficacies can now be specified, instead of simply the proportion of 'low-energy lighting' of any type. The efficacy assumed for the SAP 10.1 modelling has been based on an AECOM suggestion for an improved minimum standard reflecting good practice.
PV: total energy generated is unchanged.	No change, but it should be noted that SAP 2012 used a fixed assumption for the proportion of electrical energy generated by PV systems which is consumed within the dwelling (50%). This has been replaced by a calculation which also includes recognition of the presence of battery storage. It will impact on primary energy and carbon calculations (as emission and energy factors vary for energy used on site vs exported), as well as on SAP cost calculations.
Gas: total regulated energy consumed increases by 4%.	The changes in space heating and water heating consumption explained above to some extent balance each other out when looking at total regulated gas consumption.
Electricity: total regulated energy consumed decreases by 36%.	This is largely due to the lighting calculation changes explained above, and will impact on primary energy, carbon and cost calculations though total figures are relatively low compared to gas consumption for gas-heated cases.

Source AECOM modelling and analysis, with reference to SAP manuals (BRE, 2014) (BRE, 2019) and supporting technical papers.

77. The comparison example shows that space heating energy consumption decreases in SAP 10.1 compared to SAP 9.92 but overall the regulated energy consumption associated with heating increases slightly due to more significant increases in water heating energy consumption. One implication of this is that water heating demands become relatively more significant (a trend already seen over time with improvements to fabric standards which have reduced space heating demands). Measures which reduce water heating energy consumption will potentially have more of an impact in SAP 10.1, which may affect choices for 2021 standards.

78. The change to lighting energy consumption is also significant but is unlikely to have such an impact on measures considered for 2021 standards as this is in part due to the specification modelled for 2015, which AECOM views as appropriate for 2021 as well.

79. For PV, whilst generation figures are unchanged the new ability to recognise the impact of battery storage in SAP 10.1 may affect 2021 measures and target setting.

80. There are other changes to the SAP 10.1 methodology not covered above which will affect modelled energy consumption in some cases. These changes have been summarised by BRE elsewhere (BRE, 2018), but are considered less consequential to the current main analysis. Where they have an impact on specific modelling areas or assumptions they are noted in other sections of the report.

81. A very significant change in SAP 10.1 is the update of the carbon emission factors (CEFs). A summary of the factors of particular relevance to the analysis is provided in Table 1.5p, where they are compared to the previous SAP 9.92 factors.

Table 1.5p: SAP 10.1 Carbon emission factor summary and comparison to SAP 9.92 ( kgCO₂ e/ kWh)

Fuel type	SAP 10.1	SAP 9.92
Electricity – standard tariff	0.136	0.519
Electricity – displaced from grid, PV	0.136	0.519
Electricity – sold to grid, PV	0.136	0.519
Gas	0.210	0.216
LPG	0.241	0.241
Heating oil	0.298	0.298

Notes The figures provided here for electricity are annual averages, for ease of comparison, but SAP 10.1 applies monthly factors in practice. These monthly factors include small variations between some electricity tariffs (for example 7-hour and 10-hour, and PV electricity sold to grid).

Source SAP 10.1 Table 12 (BRE, 2019) and SAP 9.92 Table 12 (BRE, 2014).

82. The electricity CEFs are very significantly reduced in SAP 10.1, and the mains gas CEF is also slightly lower in SAP 10.1. Some of the impacts of this can be seen by comparing the DER results shown in Table 1.5j above, where energy consumption and generation figures are not varied so the impact of emission factor changes can be seen more easily. For the modelled mains gas cases the DERs are higher using SAP 10.1 CEFs, as although carbon emissions associated with energy consumption (in particular electricity consumption) are decreased, the reduction in the electricity CEF also leads to a significant decrease in carbon savings from PV electricity generation. For the electric-heated ASHP case, the DER is much lower using SAP 10.1 (it is around 25% of the SAP 9.92 DER), reflecting the reduction in the electricity CEF. The ASHP case does not include PV so reductions in PV carbon savings are not applicable in this case.

83. These changes have implications for future standard setting where carbon is used as a metric, with the change in electricity CEF making electric-heated options much more favourable from a carbon perspective than in SAP 9.92 modelling, and reducing the carbon-saving impact of PV as noted above. In SAP 9.92 the electricity CEF was much higher than that for gas (ratio of 2.4; compared to 0.65 in SAP 10.1). The primary energy factors (PEFs) in SAP 10.1 have also been updated, and factors of particular relevance to the analysis are shown in Table 1.5q. The PEFs are particularly significant as the Scottish Government intends to change to primary energy as the main target metric in 2021.

Table 1.5q: SAP 10.1 Primary energy factor summary

Fuel type	PEF (kWh/kWh)
Electricity – standard tariff	1.501
Electricity – displaced from grid	1.501
Electricity – sold to grid, PV	0.501
Gas	1.130
LPG (for main heating)	1.141
Heating oil	1.180
Renewable heat generated on-site[7]	0
Renewable heat generated off-site	1
Renewable electricity generated and used on-site[8]	0
Renewable electricity supplied from grid (as part of grid mix) or exported to grid from on-site generation	1

Notes The figures provided here for electricity are annual averages, for ease of comparison, but SAP 10.1 applies monthly factors in practice. These monthly factors include small variations between some electricity tariffs (for example 7-hour and 10-hour).

Source SAP 10.1 Table 12 (BRE, 2019), and MHCLG for renewable energy PEFs.

84. The change to primary energy as the main target metric will have significant implications for how different specification options perform in relation to each other, compared to when applying the carbon emission metric used in 2015. In particular, it can be seen that whilst the SAP 10.1 average annual CEF for electricity (standard tariff) is significantly lower than the CEF for gas (ratio of 0.65), the average annual PEF for electricity is higher than the PEF for gas (ratio of 1.33). This impacts on comparisons between gas and electric options and will mean that electric-heated options will tend to show larger relative reductions in carbon emissions than in primary energy.

85. A further new consideration has been introduced with the change to primary energy as the main target metric; the application of PEFs for renewable heat and electricity. MHCLG has chosen factors which it is understood that the Scottish Government will also adopt. These are shown Table 1.5q. The choice of these factors has a significant impact on the relative attractiveness of different renewable technologies in primary energy terms. The choice of the PEF for renewables is also used in calculating PEFs for electricity supplied from the grid and therefore affects any technology using grid-supplied electricity (and in particular those which use electricity for heating as this accounts for a relatively high proportion of regulated energy demand). The difference between primary energy factors for on-site and off-site renewables also means that primary energy savings per kWh electricity generated from PV and exported are lower than savings where this electricity is used on site.

86. The implications of the change to primary energy as the main target metric will be illustrated and discussed more fully in section 0, including consideration of the need for a secondary carbon metric (see section 1.10).