Dear Mr Cox,
Thank you for your letter of 12 November about the Authority’s proposals on the future of the Retail Prices Index (RPI).
We have been clear for a long time that RPI is not a good measure of inflation and its use should be discouraged. The proposals we put to the Chancellor are consistent with this long-held view.
It is now some ten years since the change to clothing price collection that began the controversies about the merits of the RPI. There has since then been extensive consultation and discussion about inflation measurement. All the statistical issues have been well aired. A notable feature of these discussions was the wide range of opinions. It is clear there is never likely to be consensus on the issue of the RPI or consumer prices more generally.
Against this background the Board took advice from the National Statistician, as laid out in the Statistics and Registration Service Act 2007, in which he weighed up the various options. In drawing up his advice, the National Statistician considered the views of the Stakeholder Advisory Panel on Consumer Prices (APCP-S). The Board accepted his advice and that was the basis of the proposals we put to the Chancellor to cease publication of RPI and in the short term to bring the methods of CPIH into RPI. The Chancellor responded that he was not minded to promote legislation to end RPI, but that the Government intended to consult on whether to bring the methods in CPIH into RPI between 2025 and 2030, effectively aligning the measures. The Act is clear that the Chancellor’s consent is needed for certain changes to RPI, and it is for the UK Statistics Authority to make proposals on the basis of statistical issues only.
In your letters to Ed Humpherson and me you highlighted the user need for an economic and household measure of inflation. The ONS has looked to meet these user needs by developing measures or families of measures with appropriate conceptual and statistical properties. This has given rise to the CPI/CPIH and Household Cost Indices respectively. We do not see RPI as meeting either of those user needs, but see it rather as a legacy measure.
You also asked for the Office for Statistics Regulation (OSR) to consider the proposed changes to the RPI with respect to the Code of Practice for Statistics. I enclose an annex from OSR to that effect.
I know colleagues from the ONS are in regular contact with the RPI/CPI user group and its members, and I hope this dialogue will be maintained. I also invite you to respond to the forthcoming consultation on the changes to the RPI. The Authority’s consultation, which will be undertaken jointly with that of HM Treasury, will begin on 11 March. It will be open to responses for six weeks, closing on 22 April. HM Treasury will consult on the appropriate timing for the proposed changes to the RPI, while the Authority will consult on the technical method of making that change to the RPI.
I am copying this letter to the Economic Affairs Committee and the Public Administration and Constitutional Affairs Committee.
Sir David Norgrove
For annex, see PDF attachment.