Ed Humpherson to Tony Dent: Consumer Prices Accreditation Status

Dear Tony,

Thank you for your letter dated 8 January 2025 and your subsequent correspondence about inflation statistics on 27 January.

You made several points about the Consumer Prices Index (CPI) and the Consumer Prices Index including owner occupiers’ housing costs (CPIH), recommending their de-accreditation. As we have discussed previously, these statistics are undergoing an important transformation, including the changes to private rents data and the introduction of scanner data (now planned for March 2026).

In recognition of the significance of this transformation, and the high-profile use of these statistics, we published a regulatory programme of work for the UK’s consumer prices on 10 February.

We intend to re-assess CPI and CPIH in the coming years once the planned transformation changes have been fully implemented. Our decision to re-assess was made to ensure that we provide appropriate regulatory oversight for these key statistics following this period of significant development.

In advance of these re-assessments, we are planning a review of CPI and CPIH later in 2025 that will focus on how effectively ONS is handling the transformation, including its engagement with users to ensure the improvements meet their needs and whether it has assured itself and users about the quality of statistics produced using its new methods. It would be good to continue to discuss the issues you have raised to help inform our judgements in that review. I have set out our current thinking on these issues below, but this is very much an initial view, and I am of course open to further discussion.

A key point in your correspondence concerned what you describe as “lagged data from previous periods” in current price measures, in particular the Price Index of Private Rents (PIPR) statistics that feed into the CPI and CPIH rents data and also apply to owner occupiers’ housing costs of CPIH via the ‘rental equivalence’ mechanism. You also shared some analysis of inflation rates comparing the current rates with data using new private rents data only, employing the Homelet Index.

We are aware that there are differing opinions over whether rental prices should be based solely on newer rents, similar to a ‘flow’ measure of rents, or a combination of new and existing tenancies, more similar to a ‘stock’ measure of rents. Our current view is that the more stock-based approach used presently seems reasonable as it estimates ‘average’ rental price changes for the entire private rented sector population each month, rather than the ‘marginal’ rental price change calculated only from rents that have been newly agreed or renewed. As such, we do not currently consider that the stock-based approach provides, on its own, grounds for considering that CPI and CPIH do not comply with the Code of Practice.

Our assessment report on PIPR set out requirements for ONS to provide more information about the quality and representativeness of the private rents data and on the PIPR methods. Doing so should not only improve transparency but also help address any potential misunderstanding about the details of the calculation, for example, that the calculation uses a stock-based definition of rents data over 14 months but is not, to our knowledge, a 14-month moving average of private rents. The report also listed a range of user requests of the private rents statistics, including requests for a PIPR breakdown based on new rents only; and a regular publication of comparisons between PIPR estimates and housing industry series based on new rents, such as those published by Homelet, Zoopla and Rightmove. OSR has asked ONS to set out its development plans for these statistics, including how it will prioritise user requests.

You also mention the merit of both more-macroeconomic measures, like CPI and CPIH, and a measure more targeted at capturing the inflation impact on households, such as the Household Cost Indices (HCIs). Our aim is to review how far the UK’s inflation statistics meet diverse user needs, and we will consider this as part of our future work on prices statistics. Recognising the significant user interest in the HCIs, in our plan published on 10 February we have also set out our intention to conduct a review of these statistics once ONS has completed its annual weights update for the HCIs for the first time. The aim of the review would be to inform further developments and provide direction on steps that might be needed to achieve compliance with the Code of Practice for Statistics.

In your 8 January letter, you also ask for the status of the Retail Prices Index (RPI) to be re-considered and suggest it is the appropriate measure for a household-based measure of inflation. The RPI measure had its National Statistics status (now referred to as ‘accreditation’) removed in March 2013 as its methodology was found not to meet international standards. As there have been no changes to the methodology since this time, we currently have no plans to revisit the accreditation status.

We will of course seek your views as part of any user engagement work undertaken during our regulatory programme. As I indicated above, what I have set out here is our current and initial view, not a formal regulatory judgement.

Moreover, openness is key for ONS. For example, we set out our expectation, in our 2016–17 re-assessment of CPIH, that while the rental equivalence approach for calculating owner occupiers’ housing costs (which sits alongside the debates we have discussed above) seemed reasonable, ONS should keep the alternatives under review (for example, it should continue to review the evidence, and talk to users including methods experts). In these ways, ONS should, like OSR, be open and continue to facilitate the debate on these matters.

I am copying this letter to Kate Barker, Chair of the Advisory Panel on Consumer Prices – Stakeholder, and Grant Fitzner, Chair of the Advisory Panel on Consumer Prices – Technical.

Yours sincerely

Ed Humpherson
Director General for OSR