Clarification of publication arrangements for the Retail Prices Index and related indices: November 2016

In March 2016, the National Statistician wrote to the Chair of the UK Statistics Authority, giving an update on the future of consumer inflation statistics. This emphasised the shortcomings of the Retail Prices Index (RPI) as a measure of inflation and set out the intention that the Office for National Statistics (ONS) would publish only the minimum of RPI-related data necessary to ensure the critical and essential needs of existing users are met. Specifically, this referred to the aggregates RPI and RPIX and also the component indices of RPI (for example food and clothing).

Since then, some users have sought further clarification. This article details precisely which RPI-related data will continue to be published for the foreseeable future, which will be discontinued and presents the rationale for the decisions.

The RPI-related indices and data that will be continued or discontinued are summarised in this section. These changes will coincide with CPIH becoming our preferred measure of inflation, which is planned for March 2017. Sections 3 and 4 give further detail, along with the rationale for the decisions.

The following indices and data will continue to be published:

• Retail Prices Index (RPI)

• RPI excluding mortgage interest payments (RPIX)

• RPI sub-components (these are the “building blocks” that aggregate to the all-items RPI; a full list can be found at Annex A)

• Composite Price Index

• average price data

The following indices will be discontinued:

• RPIJ

• Tax and Prices Index (TPI)

• RPI excluding mortgage interest payments and indirect taxes (RPIY)

• RPI pensioners’ indices

RPI, RPIX, sub-components of RPI

Decision: Continue to publish

The RPI is used in a large number of commercial contracts, including index-linked gilts. Similarly, the RPIX is used in a range of contracts and exposure is traded in swap form. These indices, along with all sub-components of the RPI, are considered essential and will continue to be published. A full list of sub-components is available in Annex A.

Composite Price Index

Decision: Continue to publish and explore using the Consumer Prices Index (CPI) and CPIH for more recent data

This constructs a long-run series from various sources, going back to 1750. From 1947 onwards, RPI is used. There is value in maintaining a long-run series for research and comparability purposes and exploring the option of using CPIH for more recent data is reflected in our development plan for Consumer Price Indices.

Average prices

Decision: Continue to publish and explore moving away from the RPI basis used for uprating prices

We publish average prices for a range of homogeneous products in the inflation basket of goods and services. Currently, an average price is calculated using January data. For subsequent months, this price is uprated using the relevant RPI component index. There remains significant interest in these data. Developing a new method for calculating average prices using CPIH is reflected in our development plan.

RPIJ

Decision: Discontinue publication. We will adapt the current reconciliation of RPI and CPI to compare RPI and CPIH and will publish a direct estimate of the formula effect on RPI.

The RPIJ was developed in early 2013 to highlight the impact of the use of the Carli formula on the RPI. The Review of UK consumer price statistics, led by Paul Johnson, noted that the RPIJ was not widely used and appeared to lead to confusion. It recommended that ONS consults on discontinuing RPIJ and should continue to produce a reconciliation of the difference between RPI and the main measure of inflation. There was also little support for the measure from respondents to the consultation. The National Statistician’s letter published in March 2016 stated that RPIJ would be discontinued.

Although some additional uses of RPIJ have since emerged, there remain strong arguments against continuing to publish it:

• unless there is a strong justification, having multiple measures does not encourage a clear understanding of inflation in the UK

• RPIJ shares the weaknesses of the RPI that go beyond the Carli formula, such as population coverage and use of a direct measure of house prices to estimate owner occupiers’ housing costs

• whilst the RPI and thereby the RPIJ, will continue to be maintained through routine changes, we will not invest additional resource into its development

To continue with a measure of inflation that has these shortcomings will not help to create clear and coherent understanding of inflation in the UK.

We currently produce a reconciliation between RPI and CPI, which includes an estimate of the formula effect. This will be adapted to compare RPI and CPIH, at the point when CPIH becomes the preferred measure. However, it is clear that there is value in having a measure that specifically illustrates the formula effect on the RPI, a purpose that is currently served by RPIJ. For this reason, we will publish an estimate of the formula effect on RPI, but it will not form part of the family of price indices.

Tax and Prices Index (TPI)

Decision: Discontinue publication

The TPI measures how much the average person’s gross income needs to change to purchase the basket of goods, allowing for the average amount of Income Tax and National Insurance paid on earnings.

There was very little support for TPI in the consultation responses and we are not aware of wider use of the index. Producing this on a CPIH basis would require considerable resource investment, which is not justified given the level of interest.

RPI excluding mortgage interest payments and indirect taxes (RPIY)

Decision: Discontinue publication

The RPIY excludes mortgage interest payments and indirect taxes. Whilst CPIY has a policy application, this is less evident for RPIY and there was relatively little support for it in the consultation.

RPI Pensioners’ Indices

Decision: Discontinue publication

Some respondents to the consultation noted significant use of the RPI pensioner indices in monitoring the inflationary pressures on pensioners. Beyond the methodological limitations of the RPI, these indices are not well suited to this task as their coverage is limited to those pensioner households excluded from the coverage of the RPI (that is to say where the head of the household is of pensionable age and economically inactive and where at least three-quarters of the household’s income is from state benefits). This means that neither the RPI, nor the RPI pensioner indices, provide a complete picture of the inflationary pressures on pensioners. The development of measures of income, expenditure and price change for different household groups is a better starting point for this type of analysis.

Rossi index

Decision: discontinue publication

The Rossi index is a variant of RPI that excludes housing costs. It was introduced in 1983 to uprate income related benefits. In 2011, the CPI replaced the RPI and Rossi as the measure used for uprating, with the exception of the basic state pension, which continued to use RPI for 2011 only, before moving to CPI in 2012. Given the index is no longer used for the specific purpose for which it was designed, it will be discontinued from March 2017.

This is a comprehensive list of RPI indices and CDID identifier codes that ONS will continue to publish. The relevant 1 and 12-month inflation rates will also be published.

CHAW All items RPI
CHMK RPI excluding mortgage interest payments (RPIX)