How should we measure the energy efficiency of a house?

You would think it was easy to measure the energy performance of a house. You probably think the key measure is the size of your annual energy bills. However, there are other ways. Are you interested in cost, carbon emissions or energy resources? To further complicate matters, measures such as carbon emissions and cost change over time, as prices change and our electricity supply mix changes.

There is a consultation on at the moment about changing the way we measure energy and carbon performance and standards for new housing - the Future Homes Consultation [1]. This is long overdue as the current methods have major flaws. However the proposed new standards have problems too and there are some very vocal critics.

Some bits of the proposals are excellent (in my view). For example the new approach is going to treat the energy from your solar panels differently according to whether you use it yourself or if you export it. Exported energy is worth only a third as much as energy you use. There is logic to this. The effect is to favour installations with battery storage because you can use much more of your own generation. Also the new standard is based on energy consumption rather than carbon emissions.



First of all here are some ways that we can describe the energy performance. I say describe here rather than measure because most of the numbers are predicted using a model rather than actually measuring.

1 Fabric energy efficiency
This is the rate of heat loss, either through the walls, roof, windows etc or through ventilation or air leakage. The actual heat loss depends on the outside temperature so this is measured in watts per unit temperature difference, usually per unit floor area (because large buildings lose more heat than small ones).

Fabric energy efficiency is partly driven by shape; a long thin house loses more heat than a square one because it has a larger area of walls. In UK building regulations a house is compared with a notional building of the same size and shape with standard levels of insulation and air tightness. This means that is just as easy for a long thin house to meet the standard as a square one, even though the square one is actually more efficient.

Fabric energy efficiency can genuinely be measured, though not very conveniently. The current standard way of doing this (a co-heating test) takes three weeks and you have to go and live somewhere else while this is going on. (One of the projects I am working on now is developing a way to measure heat loss while you are living in your house.)

2 Delivered energy
This is how much energy it takes to run the house. This includes the heating system, (space and hot water), ventilation systems (such as extract fans) and also energy for lighting, assuming standard patterns of use. Delivered energy is supposed to relate to the number of kWh on your energy bills but as residents you have a lot of influence, for example by changing thermostat timings and settings, and opening and closing windows, and how much hot water you use. When used as a measure of performance, the energy use is modelled based on a standard heating regime and hot water use.

Delivered energy is not used as a measure directly but it is used to calculate other measures: primary energy, carbon emissions and cost. Delivered energy is calculated for each fuel separately and the units are usually kWh/year/m2 i.e. energy per year per unit floor space. It takes into account the energy you generate yourself such as from solar panels.

Delivered energy depends on the heat loss and also the efficiency of the heating system. Gas boilers are up to 90% efficient. Electrical heating using a heat pump should be at least 200% efficient so one unit of electricity supplies two or more units of heat.

3 Primary energy
This is how much energy is required to supply that energy to you. It penalises electricity compared to gas because when electricity is generated in power stations it is the energy into the power station that is measured. A gas power station typically uses two units of gas for each unit of electricity generated while for coal or biomass it could be three or four. When primary energy is calculated, there are standard factors applied for each fuel which reflect typical or average supply.

The primary energy factors for electricity are changing rapidly. The factors currently used were set in 2012 and the factor then for electricity was 3.0, meaning three units of energy used to deliver one unit of electricity.  The actual factor for 2018 was about 2.3 (based on the electricity flow chart in DUKES [4]) but the proposed factor today is 1.501. This value is set in expectation of more renewable energy in the near future.

The factor for gas is also greater than  (1.13) although less than for electricity.  It takes into account energy used in extracting, refining and pumping the gas around the network, and also losses. The electricity value also includes losses.

4 Emissions rate
This is the GHG emissions generated annually in supplying energy to the house. The emissions rate is calculated, as for the primary energy rate, by multiplying the amount of each fuel by an emissions factor for the fuel. Once again, this has changed drastically from the 2012 value and the proposal is less than the current value as shown in the table.

Carbon emissions kg CO2e/kWh delivered

	2012 [6]	actual 2018 [4]	2020(proposed) [7]
Electricity	0.519	0.208	0.136
Gas	0.216		0.210

Carbon emissions is the main criterion in the current regulations but this clearly has to change. Ultimately the emissions from electricity should be nearly zero but that does not mean it will be OK for houses to use infinite amounts of electricity! Under the new proposal, direct electric heating (100% efficient) is already better than a gas boiler for carbon emissions but we cannot possibly supply enough electrical power to heat all our homes with direct electricity. This is why the new scheme uses primary energy instead of carbon as the main criteria.

5 Cost of energy
This is the main statistic behind EPC ratings. It is based on an annual bill, calculated using delivered energy and standard cost factors for each fuel. In practice energy prices vary on where you live as well as where you buy it and change over time. In the 2012 tables, the price factors were 3.48 and 13.19 p/kWh for gas and electricity (standard rate). The proposed values are an increase of 13% for gas and 33% for standard rate electricity. So while the carbon emissions for electricity have gone down relative to gas, the price has gone up!

6 Peak electricity demand
This is not used as a measure at all, as far as I know. However, I think it should be as this is critical for the cost of our energy supply infrastructure. Peak electricity demand is normally early evening in cold wintry weather and the more we use electricity for heating the greater this peak will be. It can be reduced using a heat store or batteries in the house to cover the peak demand periods.

The peak gas demand is not important as gas can be stored easily; the daily peaks in demand are covered by running the transmission grid at higher pressures. This is called line packing as it packs more gas into the same 'lines'. The pressure in the pipes that supply your house are kept steady.

The proposed changes have weaker fabric efficiency standards.
The new scheme has much weaker requirements than the old for insulation. The maximum U-values (a measure of heat loss) for each element are slightly reduced from before. For example the maximum U-value (heat loss) for walls is currently 0.30 W/K/m2 and under the new scheme this is reduced to 0.26. However, the current fabric efficiency standard requires that overall the building meets a fabric efficiency equivalent to a notional building with a wall U-value of 0.18. There  is nothing like this in the new proposal.

The proposed changes favour electric heat pumps over gas boilers.
As well as weakening the fabric efficiency criterion  the new proposal is for primary energy as the main criterion instead of carbon emissions. The effect is to promote electric heating with a heat pump over gas with a boiler and this could increase energy bills. However, there is also a proposal for a minimum affordability rating, which would mitigate this impact.

This is controversial:  ‘Part L is Broken’.
Weakening the fabric criteria gives the designer a lot more leeway in how they meet the standard. They can trade off fabric energy efficiency for a more efficient heating system and/or solar panels.  This is likely to lead to problems in the future because it is (almost always) harder to improve fabric later on than to improve heating efficiency. Adding insulation is costly and a huge hassle whereas upgrading to a better heating system may only involve replacing one unit. There is a huge amount of controversy over this. Critics include CIBSE (Chartered Institute of Building Surveyors) and LETI (London Energy Transformation Initiative) and AECB (Association for Environment Conscious Building)  LETI say that ‘Part L is Broken’ (part L being the bit of the building regulations).  They all want the regulations to improve the fabric energy efficiency standard. LETI also suggest a goal of whole life carbon assessment to include carbon emissions from building the house and heating systems as well as running it.

Energy you export to the grid counts for less than the energy you take from the grid.
Under the new scheme, primary energy factors are the primary statistic and this treats the energy you generate (e.g. from solar panels) differently depending on if you use it or export it. Energy you use counts with the normal electricity factor of 1.501 but when you export energy to the grid (for example from PV panels, the primary energy factor is only 0.501. This means that, from the point of view of meeting the primary energy maximum requirement, the renewable electricity you generate and use is worth much more than the energy you export.

There is logic behind this. Suppose that you export 1 unit of electricity and your neighbour uses it. The amount of energy that has been used is 1 unit, and the amount of energy required to supply it is 1 unit. However your neighbour has been ‘charged’ 1.501 units because they imported it from the grid. Your charge was -0.501 (i.e. a credit of 0.501) to give a total of 1.0.

This encourages batteries to go with PV and discourages installing more PV than you need.
This difference makes it much easier to meet the regulations if batteries are installed along with the solar PV panels as it is then possible to store electricity generated in the day and use it overnight or when the sun is not shining. It also reduces the benefit of putting on many more panels so that you generate far more than you need. You would need to export 3 units of unneeded power in the summer to offset one unit of power you use in the winter.

I think this is a huge step forward. I have long said we should be incentivising wind power, which provides more in the winter, because that is when we need the power. (See Solar electricity is like ice cream). Rather than invest your money in solar panels on your own house, you could invest in wind power elsewhere. (For example I have investments in Thrive Renewables.)

This table gives more detail on the energy efficiency measures used now, both as requirements for new buildings and for calculating the rating on your energy performance certificate (EPC). The EPC rating is based primarily on heating cost, though it also gives a secondary rating for emissions and energy use. EPC ratings are not based on the notional equivalent building but do have an adjustment for floor area.

	Now	Proposed
Fabric energy efficiency	Minimum standard for each element and for average heat loss.	Minimum standards for individual elements only.
Services	At least standard controls for heating and hot water system. Minimum efficiency standard for mechanical ventilation systems and lighting.	Similar, with some improvements.
Primary energy consumption	EPC Rating (secondary metric).	Maximum relative to the equivalent notional building.
Carbon emissions	Maximum, relative to the notional equivalent building. Also EPC rating ( secondary metric)	Maximum relative to notional building (secondary metric).
Cost	EPC Rating.	Minimum affordability rating

My view
It is definitely a good move to have a primary energy standard instead of carbon emissions. However, we must keep the fabric performance standard too. I have just signed the LETI consultation response. It is not exactly what I would say but it is pretty close. However I will put in my own response too and suggest the peak power demand limit.

[1] The Future Homes Standard: changes to Part L and Part F of the Building Regulations for new dwellings (www.gov.uk) Oct 2019

[2] Approved Document L - Conservation of fuel and power (consultation version) Oct 2019

[3] Briefing Note – Derivation and use of Primary Energy factors in SAP (BRE Group) Oct 2019

[4] Digest of UK Energy Statistics (www.gov.uk) 2019

[5] Conservation of fuel and power: Approved Document L (March 2014)

[6] Standard Assessment Procedure 2012 (BRE Group) Jan 2013

[7] SAP 10.1 (BRE Group) Sep 2019