Publication - Research and analysis

Developing regulation of energy efficiency of private sector housing (REEPS): modelling improvements to the target stock - Main Research Report

Published: 5 Nov 2015

This report describes how the least energy efficient dwellings in the private sector were identified and how their ratings could be improved by a range of improvement measures. Modelling was used to ascertain the least cost way of reaching different standards, with findings presented on capital costs, fuel cost savings, carbon and energy reductions.

Developing regulation of energy efficiency of private sector housing (REEPS): modelling improvements to the target stock - Main Research Report
Executive Summary

Executive Summary


In June 2013, the Scottish Government published the Sustainable Housing Strategy, which set out its commitment to consult on draft regulations that would set minimum energy efficiency standards for private sector houses.

Energy efficiency standards for housing in the social rented sector have been in place since 2004. In contrast there has been limited regulation of existing private sector dwellings to date. Meeting the 'Tolerable Standard[1]' does include minimum requirements for thermal insulation and heating and while there are regulations relating to the construction of new build housing and extensions, those relating to existing homes have been restricted to specific improvement works if and when they are undertaken.

The Regulation of Energy Efficiency in Private Sector homes (REEPS) working group was set up to oversee the development of the draft regulations.This research was commissioned to support the work of the REEPS technical subgroup.

The target stock

The Scottish Government's approach to assessing the energy efficiency of houses is the Standard Assessment Procedure (SAP). SAP ratings are the basis of the bandings shown on Energy Performance Certificates (EPCs) that are required to be produced when dwellings are sold or let.

Private sector stock is less energy efficient than social housing. Overall, 23% of the private sector stock falls within the lowest three SAP EPC bands of E, F and G. This equates to 400,548 dwellings. These were the focus of the research.

The target stock includes a higher prevalence of detached dwellings, older dwellings, dwellings with a solid stone wall construction, and dwellings that do not use mains gas or electricity as their main source of energy. They are also less likely to have access to a mains gas connection. Dwellings in rural Scotland are more likely than those in urban areas to be in the lowest EPC bands, bands E, F and G.

19% of REEPS target stock is privately rented, with higher proportions of private rented dwellings in the lowest EPC band ratings of G and F.

The aims and methods of the research

The research project was established to address the question: 'What are the most effective ways to increase the energy efficiency of the Scottish private sector dwelling stock in EPC bands E, F and G?'

The project modelled existing data from the Scottish House Condition Survey (SHCS) from 2010 to 2012, and comprised three broad phases. Phase 1 of the research involved:

  • developing a typology of the private sector housing stock in EPC bands EFG using data from the SHCS.
  • identifying associated archetypes in the data, e.g. dwellings that would represent each typology group.
  • identifying the appropriate potential energy efficiency improvement measures.
  • outlining principles for constructing a hierarchy of measures to create packages that would reach minimum thresholds of energy efficiency.
  • outlining methods for determining the costs of measures.

Developing the typology involved segmenting the data into similar dwellings in relation to the technical feasibility of improvement measures and likely gains in energy efficiency. Overall, 355 typology groupings were created with each representing around 1,100 dwellings on average (0.3% of the target stock).

In the second phase, the suitability, cost and impact of 38 potential improvements were assessed for each archetype. These were collated from a wide variety of sources and covered insulation, ventilation, heating, hot water, space and water heating controls, renewables, and other energy saving improvement options. Various impacts of measures were modelled including impact on energy efficiency (SAP rating); impact on CO2e; impact on primary and delivered energy; and cost effectiveness.

A key concern for the implementation of improvement measures is cost effectiveness. Indicative capital costs for each of the improvement measures were identified usually from the Product Characteristics Database File (PCDF) embedded within SAP software, and updated at regular intervals on behalf of the government. This is the standard set of reference costs used by SAP and by Green Deal assessments. Alternative sources of costs were investigated as part of the research.

As well as assessing individual measures, packages of measures to take each dwelling into the successive SAP bandings were identified and assessed. In the final phase, a number of policy scenarios were assessed.

A wide variety of technical issues were raised during the lifetime of this project in relation to methodology and assumptions used in the modelling. These included concerns relating to traditional buildings, the SAP methodology, and how costs are determined. The research also summarises these issues.


Four policy scenarios were modelled to explore the most effective ways in which to increase the energy efficiency of the private sector stock in EPC bands EFG.

  • Scenario 1: improving all dwellings to reach EPC Band F
  • Scenario 2: improving all dwellings to reach EPC Band E
  • Scenario 3: improving all dwellings to reach EPC Band D
  • Scenario 4: improving all target dwellings to move up one banding

The following table summarises the key results.

  Scenario 1 Scenario 2 Scenario 3 Scenario 4
All to reach Band F All to reach Band E All to reach Band D All up one band
Number of dwellings improved 29,676 170,708 400,548 400,548
Mean SAP score increase from improvements 13.7 13.4 16.4 9.4
Mean capital cost £627 £1,232 £2,672 £969
Overall capital cost £18.6 m £210.2 m £1,070.2 m £388.1 m
Mean reduction in fuel cost pa £542 £463 £483 £279
Overall reduction in fuel cost pa £16.1 m £79.0 m £193.0 m £111.6 m
Mean reduction in CO2 pa 2,113 kg 2,119 kg 2,617 kg 1,515 kg
Overall reduction in CO2 pa 62.7 k tons 362k tons 1046k tons 606k tons
Mean reduction in Primary Energy pa 10,313 kWh 9,894 12,583 7,424
Overall reduction in Primary Energy pa 306 m kWh 1,689 m 5,030 m 2,968 m
Mean reduction in Delivered Energy pa 6,889 kWh 7,460 8,911 5,105
Overall reduction in Delivered Energy pa 204 m kWh 1,273 m 3,569 m 2,044 m
Median payback period 1.1 years 2.3 3.8 2.5
Mean cost per SAP point increase £46 £92 £163 £103

In terms of the scale of the different packages of measures, Scenarios 3 and 4 affect more dwellings than do the other 2:

  • 29,676 dwellings would be improved by Scenario 1;
  • 170,708 dwelling would be improved by Scenario 2;
  • 400,548 dwellings would be improved by Scenarios 3 and 4.

Modelling results were further split by tenure (ie owner-occupied and private rented dwellings), by urban and rural locality and profiled over the next 30 years based on turnover of house sales and rental of properties.

Dwellings in rural areas tend to be less energy efficient than those in urban areas. The majority of dwellings affected by Scenarios 1 and 2 would be in rural areas (67% and 59% respectively) while 42% of dwellings affected by Scenarios 3 and 4 would be in rural areas.

19% of REEPS target stock is privately rented although a higher proportion of private rented dwellings are in the lowest EPC band ratings of G and F. Implementation will be linked to the letting of private rented properties and the sale of owner occupied dwellings. As turnover rates are considerably higher among privately rented dwellings, this sector will account for most of the target stock in the first few years of REEPS. After 3 years, private rented dwellings will account for 81% of dwellings in Scenario 1, 67% in Scenario 2 and 62% in Scenarios 3 and 4.

The average capital cost of improvements per dwelling reflects both how much each dwelling needs to be improved and the base position of the dwelling. Packages of improvements that make a large increase in energy efficiency are, on average, more expensive than those that make a small increase. The average cost of investments per dwelling in Scenario 1 is £627, Scenario 2 is £1,232, Scenario 3 is £2,672 and for Scenario 4 it is £969.

The more efficient dwellings are, the more expensive they are to improve further. The average cost of improving a dwelling in EPC band G by one band is £627 compared to £1,062 for a dwelling in EPC band E.. A similar pattern is seen with regard to cost per SAP point increase: £46 for G to F, to £126 for E to D.

The overall capital cost of improving all dwellings under each scenario is as follows:

  • Scenario 1 (to reach EPC band F) would require £18.6 million
  • Scenario 2 (to reach EPC band E) would require £210 million
  • Scenario 3 (to reach EPC band E) would require £1,070 million
  • Scenario 4 (all up one band ) would require £388 million

The average cost of improvements tends to be higher for dwellings in rural areas than dwellings in urban areas across all Scenarios. This difference is most marked for Scenario 3 - £4,092 compared to £1,656 - primarily because of the higher proportion of rural dwellings in EPC bands G and F.

Measures included in packages and policy scenarios

The total number and proportion of dwellings with specific measures included within their package of measures to bring properties up to different standards, varied across the scenarios. The three most common measures under each scenario were:

  • Scenario 1 (to reach EPC band F): Loft insulation including top-up (18,375 or 62%), Hot water tank jacket (4,651 or 16%), Room thermostat (4,457 or 15%),
  • Scenario 2 (to reach EPC band E): Loft insulation including top-up (74,369 or 44%), Low energy lighting (25,496 or 15%), Cavity Wall Insulation (24,611 or 14%),
  • Scenario 3 (to reach EPC band D): Loft insulation including top-up (157,256 or 39%), Low energy lighting (135,662 or 34%), Cavity Wall Insulation (103,250 or 26%),
  • Scenario 4 (all up one band): Loft insulation including top-up (133,380 or 33%), Low energy lighting (101,310 or 25%), Hot water tank jacket (62,406 or 16%).

There were some differences by rurality, particularly in relation to Scenario 3. A higher proportion of dwellings in rural areas than in urban areas have room in the roof insulation, floor insulation and the replacement Oil/LG Boilers in the package of measures to reach EPC band D. In contrast, low energy lighting was more commonly included in urban than in rural areas. There was very little difference in the measures included in the packages by tenure.

Impact of packages and policy scenarios

In terms of overall savings in relation to fuel costs, CO2e emissions and Primary and Delivered Energy consumption, Scenario 1 has the smallest impact and Scenario 3 has the largest impact due to the difference in terms of number of dwellings improved and scale of the improvements required. Generally, however, the more efficient a dwelling is pre-improvement measures; improvements will have a smaller impact.

The overall impact with regard to annual fuel cost savings of the different scenarios would be:

  • Scenario 1 would give an annual fuel cost savings of £16.1million.
  • Scenario 2, £79.0 million
  • Scenario 3, £193.0 million
  • Scenario 4, £111.6 million

For the least efficient dwellings, the capital cost of improvements is lowest, and the financial gains are highest. It follows that the payback period is the shortest for these dwellings. Scenario 1 has a mean payback period of 1.2 years, compared with 2.7 for Scenario 2, 5.5 for Scenario 3, and 3.5 year for Scenario 4.

Impact on overall annual CO2e emissions:

  • Scenario 1 would amount to annual savings of 63 thousand tons,
  • Scenario 2 would give annual savings of 362 thousand tons,
  • Scenario 3 would give annual savings of 1.05 million tons,
  • Scenario 4 would give annual savings of 606 thousand tons.

The larger the scale of the improvement measure the greater the reduction in CO2e emissions.

Impact in overall annual delivered and primary energy savings:

  • Scenario 1 would lead to an annual saving of 204m kWh per annum for delivered energy and 306m kWh for primary energy.
  • Scenario 2 leads to an annual saving of 1,273m kWh for delivered energy and 1,689m kWh for primary energy.
  • Scenario 3 leads to an annual saving of 3,569m kWh for delivered energy and 5,030m kWh for primary energy.
  • Scenario 4 leads to a savings of 2,044m kWh for delivered energy and 2,968m kWh for primary energy.

As with energy costs, the larger the scale of improvement made, the greater the impact. Furthermore, improving a dwelling by one band has a larger impact on the least efficient dwellings.

Individual improvement measures

The presence of energy saving features within a dwelling, the location of a dwelling, and whether the dwelling is on on/off the gas grid are important factors in determining the applicability of improvement measures and assessing their relative effectiveness in improving energy efficiency. Some improvement measures are cost effective but have little impact on SAP rating; certain measures are expensive and have little impact on SAP rating. There is a high degree of variability across the different improvement measures in terms of their relative impact in improving energy efficiency.

Individual measures were modelled for all dwellings where they were possible to implement and where they would potentially increase the SAP rating. Among the 38 improvement measures, nine improvement measures had a payback period of less than three years, though these varied in the size of their impact. Those with the shortest payback period were adding a hot water tank jacket, low energy light bulbs, switching electricity tariff, loft insulation, and installing thermostatic radiator valves (TRVs).


Email: Silvia Palombi