The following is mostly derived from briefing papers published by the House of Commons Library.
The CPI and RPI are both consumer price indices and the calculation of each uses much of the same basic price data. There are, however, some major differences:
The CPI excludes some goods and services included in the RPI. Vehicle licence and TV licence duty are excluded but more importantly some housing costs, including mortgage interest payments, household insurance and council tax (but not utility bills and rents), are excluded from the CPI and account for much of the difference between the indices.
The CPI includes some costs that are not in the RPI, such as charges for financial services, foreign exchange commission for tourists changing their own currency into sterling, foreign students’ university tuition fees, and university accommodation fees.
The CPI covers a broader population than the RPI, which excludes high-income households and low-income pensioners.
The mathematical formulae used to calculate the indices differ. For a given set of data, the method used for the CPI (geometric means) will show smaller changes than that used for the RPI (arithmetic means).
The Institute for Fiscal Studies explains that changes in the CPI tend to be lower than the RPI both because it excludes most housing costs and because it is calculated differently.
To enlarge a part of the graph hover your cursor over that part and click and again if required. To restore click the "Zoom Out" button, top right or your reload button on your browser.
The Treasury forecasts of the CPI and RPI increases for the next 5 years were published in the June 2010:
OBR forecasts for RPI and CPI inflation
Forecast
2010-11
2011-12
2012-13
2013-14
2014-15
CPI (%)
2.8
2.6
1.9
2.0
2.0
RPI (%)
4.2
3.4
3.0
3.2
3.4
However the Office for Budget Responsibilty has questioned the assumptions and predicts that the gap between the RPI and CPI will widen to as much as 1.8% by 2016.
The Royal Statistical Society recently called for a “comprehensive review of issues relating to the measurement of inflation. A letter to the UK Statistics Authority expressed its main concerns. One of these was that the CPI had become the “headline index” even though it was “not necessarily the best index for all purposes”:
As we all know, the CPI was originally the Harmonised Index for Consumer Prices (HICP) constructed in order to compare price movements between EU countries; its methodology and coverage are on agreed EU definitions. Lack of agreement over the treatment of owner occupier housing costs, of particular importance for the UK, means that these are still excluded; it is debatable whether its exclusion of items such as council tax, vehicle excise duty, trade union subscriptions and television licenses, and its inclusion of spending in the UK by foreign residents, are appropriate for all purposes. Its methodology, specifically the comprehensive use of the geometric mean at the lowest level of aggregation, has many supporters. It is approved by some other international bodies as well as the EU. However, some would argue that it is not the best approach for products where consumers are typically slow to substitute newly cheaper outlets, brands or varieties for existing more expensive ones.
A further concern was that the rates of inflation shown by the CPI and RPI could at times differ widely, due to the “formula effect”:
But the RPI inflation rate has also been consistently higher than the CPI rate due to the “formula effect” whereby the RPI uses an arithmetic mean at the lowest level of aggregation while the CPI uses a geometric mean. Over the last five years, according to the ONS press notice, the difference in the annual inflation rate due to the formula effect has never been less than 0.43 percentage points and has been as high (recently) as 0.86 percentage points.
The RSS considered it “highly unsatisfactory that a difference in statistical treatment should generate such a substantial difference in the two indices”.
The UK Statistics Authority responded on 6 October as follows:
We continue to regard both the Consumer Prices Index (CPI) and Retail Prices Index (RPI) as important measures of consumer price inflation. We believe the CPI should become the primary measure of consumer price inflation but only when the inclusion in the index of owner occupiers' housing costs has been achieved. We note that the ONS has an active research and development programme for the CPI and the RPI, and are preparing for wider user engagement on a forward work programme, starting this autumn.
So will that mean the proposed CPIH will then include Council Tax? Don't be fooled!
"While the consumer price index (CPI) is acceptable for macroeconomic purposes and for international comparisons within the EU we do not believe its coverage is generally appropriate for inflation compensation purposes". - Royal Statistical Society
"The official price indices currently available for the United Kingdom are misleading the general public and are of doubtful relevance for policy purposes. This is not an ethical issue in the sense that government statisticians or politicians are deliberately misleading the British public about inflation. What has happened is that the statisticians have given more weight to questionable economic theory than to the public's need for a clear and transparent measure of price inflation. The end result is that politicians may make bad decisions because they are using bad statistics and the general public loses faith in the statisticians because of the gap that they see between their own daily shopping experience and the official measure of inflation." Ward, Blades and Carson, Statistical Journal of the International Association for Official Statistics, 27, 31 -37, January 2011
Updated article from Straight Statistics (with comments) 