Room in roof insulation investigation update

Investigation

Publication date

Scheme name

ECO

This post was updated on 14 February 2017 to clarify that measures installed up to and including 28 July 2016 were included in the audit exercise. We have also made clear that the audit applied to measures that have been installed to 85% or more.

In the course of 2016, concerns were raised in relation to the standard of installation on some room-in-roof insulation (RIRI) measures. This prompted us to publish a clarification on 29 July 2016 stating that, where feasible, we always expect 100% of a RIRI measure to be insulated. We believe that this clarification has enhanced compliance with our requirements in the supply chain.

 As part of our compliance monitoring, we conducted an investigation of RIRI measures installed before 29 July 2016. We asked suppliers to commission on-site inspections on a sample of measures. This was to determine if these measures were installed to the level claimed in the notification to us, and if not, how great the difference was between the actual level of insulation and the level of insulation notified to us.

As a result of the investigation, we found that 52% of ECO2 RIRI measures installed before 28 July had not been installed to the level notified to us. On average, the percentage of measure installed was 17 percentage points less than claimed on the notification. This figure was arrived at excluding the party wall, i.e. by assuming that none of the RIRI measures have a party wall element to them. Within the results, we found variation both between suppliers and between installers, but the original sample was not designed to be robust at these levels, because to do so would have required a significantly higher number of inspections.

Based on the above results, we decided to take action to ensure that the notified scores for RIRI measures are an accurate reflection of the actual installations that have taken place. To do this, we presented suppliers with the following three options:

Option 1: Apply a correction to the ‘percentage of measure installed’ to all RIRI measures installed before 29 July with a notified percentage of measure installed equal to or greater than 85%. The scale of the correction would be 17% or the observed difference for the individual supplier, whichever is the lesser. A table with the percentage for each supplier is included at the bottom of this page.

Option 2: Conduct more inspections to obtain a statistically reliable result at the supplier or individual installer level.

Option 3: Suggest an alternative to the above two options that provides Ofgem with a reasonably accurate figure for the actual percentage of measure installed for RIRI measures installed before 28 July 2016, or assures Ofgem that the notified information is correct.

We have now received confirmation from all suppliers that they will apply Option 1 across the majority of their measures. One supplier will be conducting further inspections for a small number of measures. We expect that at this stage, suppliers will have contacted their delivery partners to discuss their chosen approach. We recommend that installers who delivered RIRI measures contact the supplier through which they delivered, or the relevant intermediary, with any questions about how their measures might be impacted.

Supplier Percentage Point Correction
British Gas 17%
Co-operative Energy 8%
EDF Energy 10%
E.ON 17%
First Utility 17%
Npower 14%
OVO Energy 17%
Scottish Power 9%
SSE 17%
Utilita 0%