Dear Lord Forsyth,

Following the Committee’s recent report on Measuring Inflation, I write with the UK Statistics Authority’s response to your recommendations.

As your report made clear, the question faced by the Authority in 2012 was whether to make substantive changes to the construction of the Retail Prices Index (RPI). The decision made by the then National Statistician, one widely supported in the consultation at the time, was to leave the RPI unchanged. This decision gave rise in turn to the conclusion that the RPI should be treated as a legacy measure, with no future substantive changes to its construction and methods. That position was endorsed by an independent review of consumer prices led by Paul Johnson, which reported in 2015. In the period since, the Office for National Statistics (ONS) has developed alternative measures of inflation, and the Authority has urged users to move away from the RPI.

Nonetheless, the RPI continues in widespread use. This – along with new advice from ONS on the flaws of the RPI, new advice from the National Statistician’s Advisory Panels, and the urgings of your Committee – convinced the Board that further action was necessary. The then National Statistician put options for the future of the RPI to the UK Statistics Authority’s Board on 26 February 2019.

After receiving this advice, Sir David Norgrove, Chair of the UK Statistics Authority, wrote on behalf of the Board to the previous Chancellor of the Exchequer on 4 March 2019 with the following recommendations:

  • that the publication of the RPI be stopped at a point in future; and
  • in the interim, the shortcomings of the RPI should be addressed by bringing the methods of the CPIH into it.

Today the Chancellor has announced his intention to consult on whether to bring the methods in CPIH into RPI between 2025 and 2030, effectively aligning the measures.

The proposals made by the Authority address many of the recommendations made by the Committee in its report. More detailed responses to the other recommendations are set out in the attached Annex.
Yours sincerely,

Sir David Norgrove

Related Links:

ONS oral evidence – (2018)

UKSA oral evidence – (2018)

UKSA follow up written evidence – (2018)


Annex: Detailed Response to Specific Recommendations

  1. We heard evidence that the Carli formula, as used in the RPI, produces an upward bias. But expert opinion on the shortcomings of the RPI differs. (Paragraph 99)
  2. There is however broad agreement that the widening of the range of clothing for which prices were collected has produced price data which, when combined with the Carli formula, have led to a substantial increase in the annual rate of growth of RPI. (Paragraph 100)
  3. We are not in a position to reach a conclusion on the question of whether the Carli formula is problematic in areas other than clothing. Given the properties of the Carli formula that may lead to upward bias have long been evident, yet expert opinion still differs, it may be a perpetual debate. (Paragraph 101)

The Authority agrees that there is never likely to be unanimity on the issue of the elementary indices (e.g. Carli, Jevons or Dutot) used in inflation measurement. There is no single universally agreed set of criteria against which to judge them and there are specific examples where each index can be shown to produce either plausible or implausible results. A judgement therefore needs to be taken in the round.

Our view is that the Carli is not generally a good index. A thorough exploration of the issues related to the Carli index was set out in both Chapter 10 of the independent review of consumer prices by Paul Johnson and the 2012 review of UK consumer price statistics conducted by Erwin Diewert, a leading authority on index numbers.

This view is supported by international practice and the National Statistician’s Technical Advisory Panel for Consumer Prices. Many technical manuals and academic papers also highlight the undesirable properties of the Carli index. Regulations on the production of the Harmonised Index of Consumer Prices go further and state that the Carli should not be used unless it can be demonstrated to behave in a similar way to the Jevons or Dutot.

We agree that the interaction between the Carli index and the collection of clothing prices created an increase in the rate of RPI inflation in 2010. It was this event that led ONS and the Authority to put in place a programme of work that led to the 2012 consultation on the future of RPI.

4. Given its widespread use, it is surprising that the UK Statistics Authority is treating RPI as a ‘legacy measure’. The programme of periodic methodological improvements should be resumed. (Paragraph 116)

5. We are unconvinced by the National Statistician’s suggestion that in publishing statistics that serve the public good, the interests of those who may be affected negatively by any change should be taken into account. It is not clear from section 7 of the Statistics and Registration Service Act 2007 that this is a relevant consideration for the statistical authorities to be taking into account when they are producing and publishing statistics. (Paragraph 117)

6. What is clear from section 7 is that the UK Statistics Authority has to promote and safeguard the quality of official statistics, which includes their impartiality, accuracy and relevance, and coherence with other statistics. In publishing an index which it admits is flawed but refuses to fix, the Authority could be accused of failing in its statutory duties. (Paragraph 118)

7. We believe section 7 requires the Authority to attempt to fix the issue with clothing prices. Section 21 may require the Authority to consult the Bank of England over the change and obtain the consent of the Chancellor of the Exchequer, however this provision cannot be cited as a reason for not requesting the change in the first place. (Paragraph 119)

8. If the Authority requests the change, the Chancellor of the Exchequer should consent to it. It is untenable for an official statistic, that is used widely, to continue to be published with flaws that are admitted openly. (Paragraph 120) 

The announcements by the UK Statistics Authority and HM Treasury on 4 September deal with this substantive issue raised in these recommendations, and are summarised in the covering letter to this response.

9. While we accept the arguments that consumer price indices have different purposes, we do not believe this warrants the production of multiple indices for government use. Two different measures of inflation allow a government to engage in ‘inflation shopping’. (Paragraph 134)

10. The Government should address the imbalance in its use of consumer price indices. It risks undermining public confidence in economic statistics. It is encouraging to see that the present Government is taking some steps to address the imbalance, for example with the change to uprating business rates by CPI and recent discussions around rail fares. (Paragraph 135)

11. In future there should be one measure of general inflation that is used by the Government for all purposes. This would be simpler and easier for the public to understand. But the UK Statistics Authority should also continue to develop the Household Cost Indices, discussed below. (Paragraph 136)

We welcome the Committee’s recommendation that the Household Cost Indices should continue to be developed. On 28 June 2019, the National Statistician outlined the next steps in the development of these Indices.

12. We disagree with the UK Statistics Authority that RPI does not have the potential to become a good measure of inflation. With the improvements to RPI that we set out in the previous chapter, and a better method of capturing owner-occupier housing costs as discussed below, we believe RPI would be a viable candidate for the single general measure of inflation. (Paragraph 139)

13. We are not convinced by the use of rental equivalence in CPIH to impute owneroccupier housing costs. The UK Statistics Authority, together with its stakeholder and technical advisory panels and a consultation of a wide range of interested parties, should agree on the best method for capturing owner-occupier housing costs in a consumer price index. (Paragraph 153)

14. Once a method of capturing owner-occupier housing costs has been agreed, the UK Statistics Authority, after consulting the stakeholder and technical panels, should decide which index to recommend as the Government’s single general measure of inflation. The Government should have adopted the preferred candidate as its single general measure of inflation within five years. (Paragraph 154)

Owner occupiers’ housing (OOH) costs are one of the most challenging aspects of inflation to measure. There is also no single approach that will be correct in all circumstances, as the choice will depend on the purpose of the index and also practical issues around data availability. In light of this, ONS has spent the last 10 years developing and consulting on its approaches to owner occupiers’ housing costs.

The development of an OOH measure for CPI was first considered in 2009 by the Consumer Prices Advisory Committee (CPAC). The committee then spent the next three years investigating different approaches to measuring OOH costs. In September 2010 it narrowed down the options to two – net acquisitions and rental equivalence – which it evaluated in detail against the five dimensions of statistical quality defined by the European Statistical System. The Committee finally agreed on rental equivalence in April 2012, giving consideration to both conceptual appropriateness and how well the index could be calculated in practice.

A first consultation was launched in the summer of 2012, in which users were asked about rental equivalence. The responses were fairly evenly split between support for rental equivalence, net acquisitions and neither approach. The National Statistician chose rental equivalence reflecting the quality of the underlying data available and whether asset prices were appropriately treated. The process is described in more detail in Appendix A of the CPIH Compendium.

Paul Johnson’s review of consumer prices was published in January 2015. This looked again at CPAC’s recommendation to use the rental equivalence method. It concluded the underlying assumptions are reasonable in a UK context and that the measure is based on a large, detailed source of underlying data. Therefore, the Review recommended that ONS should continue to use the rental equivalence measure.

A further consultation was conducted on the findings of the Johnson Review. Responses to the review on CPIH and OOH were again mixed, highlighting that users are unlikely to come to an agreement on the most appropriate choice for measuring OOH costs.

The Office for Statistics Regulation’s 2016 re-assessment of CPIH as a National Statistic noted that ‘there is some disagreement among users about the concepts and methods…’ Work to address these recommendations resulted in a wide ranging process of user engagement on CPIH, and the publication of numerous supporting materials such as the CPIH Compendium, which articulates the rationale for ONS’s choice of rental equivalence alongside the pros and cons of each approach, an ongoing published comparison of alternative OOH measures, and documentation on the various users and uses of our consumer price inflation statistics.

ONS have also looked at international practice where they found widespread use of the rental equivalence measure. The approach taken by different countries is summarised in the CPIH Compendium. Of the 40 countries considered, the most common approach is rental equivalence (12 countries) if discounting those that exclude OOH altogether (15 countries). It is also worth noting that the method requires a reasonably large rental market to work, and so many countries may be constrained in their choice by the availability of data. The countries that use rental equivalence include the United States, Germany, Norway and the Netherlands.

In light of the 10 years of development and consultation, ONS are not minded to undertake any further engagement with users and experts specifically on rental equivalence and owner-occupier housing costs. There is never likely to be agreement on a single approach. ONS views rental equivalence as the correct approach conceptually for an economic measure of inflation, and one where sufficient data is available to make it practical. Of course, they remain committed to ongoing monitoring and development of the CPIH and the Household Cost Indices.

15.Our recommendations will not however solve the issue of index or inflation shopping immediately. The Government will need to take action in the interim to address this. (Paragraph 155)

16.While the single general measure is being determined, the Government should switch to CPI for uprating purposes in all areas where it is not bound by contract to use RPI (except for the interest rate on student loans which, as we recommended in our Treating Students Fairly report, should be set at the ten year gilt rate thus reflecting the Government’s cost of borrowing). (Paragraph 156)

17.The Government should begin to issue CPI-linked gilts and stop issuing RPI-linked gilts. We heard evidence to suggest there was sufficient demand to make a viable market. (Paragraph 170)

18.Once the long-term single official measure of inflation has been agreed, gilts should begin to be issued that are linked to that index. The prospectuses for new issuances of index-linked gilts should be clear that the inflation index will change to the Government’s single general measure of inflation once it has been agreed. (Paragraph 171)

Recommendations (15) to (18) are primarily directed at HM Government and the Authority has nothing to say on those issues. We continue to urge the Government to cease to use the RPI for its own purposes where practical.

19. Once the single general measure of inflation has been introduced, the UK Statistics Authority and the Government should decide whether RPI should continue to be published in its existing form for the purposes of existing RPI-linked contracts, or whether a programme of adjustments should be made to the RPI so that it converges on the single general measure. (Paragraph 194)

20. To avoid disruption, we envisage any programme of convergence would take place gradually, over a sufficiently long time, and that the plan for that should be published at the outset. (Paragraph 195)

21. We note that the consent of the Chancellor of the Exchequer to changes to RPI that cause material detriment to index-linked gilts holders is no longer required after the last issuance to which that clause relates to expires in 2030. (Paragraph 196)

We strongly agree that any changes to the RPI or stopping the publication of RPI needs to be carefully planned. The Authority and ONS have been discussing the mechanics of any changes with the Government in the run up to the 4 September announcement.