A report commissioned by the UK Statistics Authority (UKSA) was published yesterday by Paul Johnson, Director of the Institute of Fiscal Studies, into the range of consumer price indices in the UK. The report considers the available indices, what is the ultimate aim of measuring inflation, and focuses on what price indices will best meet the nation’s needs.
Background to the review
The Office for National Statistics (ONS) currently publishes a range of consumer price indices. There are two main indices - the Retail Price Index (or RPI) and the Consumer Price Index (CPI) - and a number of variations of each. The two main indices can give very different results due mainly to two factors: the RPI includes a measure of housing costs whereas the CPI does not; and the 'formula effect' arising from the different formulae used in calculating the two indices. The formula effect alone means that RPI inflation will normally be higher than CPI inflation (recently by 1.0% based on data to October 2014).
In June 2010 the coalition government announced a move from RPI to CPI for increasing public sector and private sector pensions and most benefits. Some costs such as rail fares are still linked to RPI.
When originally compiled the RPI was intended primarily to measure the cost of living as a tool for adjusting wages, military pensions, etc. for the effects of inflation. However, the widespread adoption of inflation targeting over the past quarter century has meant that governments are mainly interested in using consumer price indices as macroeconomic variables. There are significant differences between how to measure the cost of living and how to measure macroeconomic performance – at present, there is no measure published by the ONS solely to measure the cost of living across the population.
Among the report’s key recommendations are:
- In future CPIH should be the main measure of inflation, replacing CPI.
- Government and regulators should work towards ending the use of the RPI as soon as practicable. RPI should not be used for new contracts. Taxes, benefits and regulated prices should not be linked to the RPI.
- Government should move away from selling debt linked to the RPI.
- RPIJ should be discontinued.
The UKSA will need to decide whether to adopt his recommendations.
'True' measure of inflation?
For a number of years there has been a confusion amongst the public over which of the many different statistics published is the ‘true’ measure of inflation in the country. Mr Johnson’s analysis is clear and direct – if these recommendations are adopted in full, this will send a clear message as to which statistics our experts consider reliable and much of this confusion will be resolved.
However, this review focuses more on clarifying the best index to be used for macroeconomic purposes and commercial contracts. It is disappointing that Mr Johnson has not recommended that a new index is developed to measure changes in living standards over time.
This popular confusion also affects defined benefit pension schemes – there have been some significant recent legal cases brought by trustees and employers to challenge which index they should be using (considering RPI, CPI, CPIH and RPIJ), including a case relating to pensions operated by the Arcadia Group (owners of Burton and Topshop). The report as it stands contributes usefully to this debate and, if these recommendations can be accepted, over the longer term we might see schemes which felt confident that they should be providing CPI-linked benefits rather than RPI-linked – reducing pension liabilities for UK companies.
The UK government has over £350bn in RPI-linked debt. The Debt Management Office considered as recently as 2011 whether to move to issuing CPI-linked debt but decided not to do so – fearing that there would not be enough demand from investors. It seems surprising that this situation might have changed in a space of only four years, but CPI-linked debt would be strongly welcomed by pension schemes, who could use them to more efficiently protect themselves against inflation risk.