Problems with energy performance certificates

We’re all familiar with energy performance certificates (EPCs). We need them to sell or let a property. But recently they’re coming under a lot of criticism, most recently in this Guardian feature. And this raises a good question - do they need changing?

In Europe, EPCs owe their existence to the EU’s 2002 Energy Performance of Buildings Directive. This directive was later recast, then requiring EPCs to be available for all properties on the market. EPCs have really grown and developed since. Around 60% of homes in England have an EPC.

But how do they actually work? How can we measure the energy performance of a building? And how is it presented? The final question is the easiest to answer: many of you will have this:

Buildings are given some score, which maps to a particular band. Finding this score is the challenge though. In the UK, we use the Standard Assessment Procedure (SAP) methodology. The EPC assessor takes a series of measurements about a property, and feeds these into some software which churns out the score and corresponding EPC band.

This band has a real impact on people’s decision making. According to Rightmove, a property portal, “a property’s EPC is a major or important factor for 80% of potential movers.” More interestingly, Knight Frank, the property consultancy, found a correlation between EPC and property value, even when adjusting for other factors. They claim moving from EPC F or G to C could lead to a 20% increase in property value. Considering the average UK house price currently stands at £260k, this EPC uplift immediately creates £50k of value for the owner. Knight Frank also estimate it costs on average £15k to take a home from EPC G to C. If these numbers are correct, energy retrofits can create billions in value: every retrofit from G to C immediately creates on average £35k per property. This could also be a great business idea - buy distressed properties, retrofit them, and then sell/ refinance and keep. There’s only one problem though - the numbers aren’t always accurate.

This problem with EPCs comes from several areas. Firstly, there are fundamental flaws in the SAP framework. Electric heating systems are awarded fewer points than gas heating (ie are seen as worse). And so, switching from a gas boiler to a heat pump can reduce the EPC of a property. And as we’ve seen above, this can be value destructive. This is not ideal, especially considering heat pumps are already unpopular in the UK.

Another problem is a side effect of the original EU directive. Making EPCs a requirement has indirectly led to a race to the bottom as EPC assessors outbid each other to win work. This impacts the quality of data collection. EPC assessments can cost as little as £30, and have devolved into a checkbox exercise, trying to collect data as quickly as possible. This also means 5 different assessors could give 5 different assessments. Which?, a UK consumer body, is advocating for higher standards amongst EPC assessors and a better EPC audit process. I spoke to a founder in the heat pump installation space and they told me about a recent site visit. The EPC claimed there was cavity wall insulation. They drilled a hole into the wall to check, and there was nothing at all. This is a major problem, especially if the property was acquired on the basis of this inaccurate EPC.

There’s also a broader question about the nature of EPCs. The intent was to prepare a standard to help compare the energy performance of buildings. This is different to calculating energy consumption, or finding the retrofit pathway. However, the modern day EPC has evolved into a blend of all of this. A lot of work is needed.

I don’t want to sound overly negative - in my opinion the UK is a global leader here. We have a standardised national process built on a scientific approach. Our entire EPC database is also public, which is a great dataset for people who are feeling nosy, or for startups who want to build solutions in this space. Germany has realised the difference needs for the different types of data, and accordingly has two types of certificates. They have a simple Usage certificate, showing historic energy consumption, and a complex Demand certificate, which must be issued by a certified specialist . This gives consumers better options, but there are downsides. Different German states have different rules on which certificates are needed, making it confusing. France had other concerns, as their EPC assessors didn’t have standardised approaches, leading to huge variations in assessment results. They’ve recently legislated to change this.

But if someone was to redesign the system, what should they do? A few questions come to mind:

1. What do we actually need? Should all buildings have a retrofit pathway? Or just an energy performance certificate? Or something else?

2. Should every dwelling be given a target to produce this new standard? How realistic are these mandates?

3. How can we avoid the weaknesses of EPCs? (ie be more accurate, more reliable, and less subjective)

4. What’s the role of the private sector here?

My view is that we need a nationwide standard for the energy efficiency of buildings. This should be comparable and reliable, showing how much energy a building uses, adjusted for square footage. I don’t think we need to mandate a retrofit pathway for buildings: the private sector can use retrofit assessors to address this on an ad hoc basis.

I’d like us to step away from the current approach, which has some bad design and is also complex, needing 200+ data points. However I like how it’s grounded in building physics. SAP is built on the Building Research Establishment’s Domestic Energy Model (BREDEM). This model needs the following calculations to find a dwelling’s energy performance:

1. energy consumption for lights, appliances and cooking
2. energy requirements for water heating
3. the dwelling’s specific heat loss
4. the dwelling’s thermal mass
5. the solar heat gain
6. the internal heat gain
7. the mean internal temperature
8. space heating energy consumption
9. cooling energy consumption
10. on-site energy generation

A lot of this data is already available. We have smart meters showing electricity and gas consumption. Solar incidence data and weather data can give us a view on average solar heat gain. Even datasets like Google Maps can give estimates for building internal area and footprint. This leads me to think we can create a new EPC standard with these existing datasets, and remove the need for bad manual surveys. There are also startups innovating in this sector.

SatVu can take 3.5m resolution data from space, showing thermal heat loss. This is already at a building level.

Could this dataset be used to give an idea of energy efficiency? Potentially, if the resolution gets higher and can spot localised heat leaks in real time. I spoke to a satellite engineer who designed satellites similar to SatVu’s. They claim it’s conceivable for newer launches to reach even higher resolutions (with some time and investment). However, this only shows heat loss from the top up. We’d need to see heat loss from the walls and windows to get a better picture of building performance. Enter Kestrix, which is building the “Google Maps of heat loss”. Kestrix use drones to capture thermal images, and use that data for non-invasive energy surveys. This is being used today by housing associations to help them understand how to retrofit properties.

Clearly there’s a lot happening in this space. I’m intrigued to see who’ll emerge victorious - will governments legislate a new data standard? Or will startups use proprietary data to build their own standard, which functions better than the existing EPCs? For now though, it seems like we’re stuck with the current version.