From the TUC

ONS pronounce on the future of inflation measurement, with RPI “not a good measure”

09 Mar 2016, by in Economics

Today the National Statistician John Pullinger published advice on the ‘future of consumer inflation statistics in the UK’ (here). The trade union movement has taken a close interest in these issues and has contributed to the debate at various points. The purpose of this blog is to outline the current position, set out background and offer advice to those involved in wage negotiations.

The key points:

I am inclined to consider that CPIH should become the ONS preferred measure of consumer inflation and the focal point of ONS commentary in due course.

… users have sought clarification on the future of the Retail Prices Index (RPI). Put simply, I believe that the RPI is not a good measure of inflation and does not realistically have the potential to become one. I strongly discourage the use of RPI for as a measure of inflation as there are far superior alternatives. Nonetheless, RPI is still used for a number of legacy purposes and its production is mandated by legislation. My intention is that from the start of 2017, ONS would publish the minimum of RPI-related data necessary to ensure the critical and essential needs of existing users are met. This would consist of the aggregates RPI and RPIX. Component indices of RPI (for example, the RPI indices for food, clothing etc) would also continue.


In addition the ONS will begin to develop a household inflation index (see below).

This latest statement follows advice from newly convened advisory panels: a stakeholder panel and technical panel. These follow in the tradition of drawing on wider views from the user community when it comes to inflation measurement (normally definitions etc. are up to the statisticians), given the many ways in which inflation affects all of our lives (originally there was an RPI advisory committee and then a consumer prices advisory committee). I represented the TUC on the stakeholder panel. The ONS also published today the advice Pullinger received from these groups, the results of intensive deliberations over the past couple of months (here).

The detailed advice is clear that a number of members (* see end) had reservations about the way forward ONS were originally promoting, in general terms that one size cannot fit all. A single index cannot be suitable for both macroeconomic uses (e.g. inflation targeting, cross country comparisons) and capturing the cost of living as experienced by households (and hence for the uprating of pensions and benefits etc).

(There were in addition specific problems with the chosen CPIH measure, first that it is still failing to meet the standard as a National Statistic (given problems with its measurement of housing), and second conceptual problems with the way in which owner-occupier housing is measured by the so-called rental equivalence. See Andrew Sentance piece from the Sunday Telegraph – here.)

The ONS have moved some way to meet these concerns by agreeing with the Royal Statistical Society’s proposal for a ‘household inflation index’. This should permit some better understanding of how inflation experienced by households might be different to the headline measure, though detailed proposals are yet to be announced. It should help to capture better the cost of living idea.

The TUC’s main  reservation remains around the position on the RPI, and in particular the so-called formula effect. As close followers of the debate will know, this technical issue (on how individual prices are averaged together) has meant that RPI inflation is about one percentage point higher than CPI inflation. The ONS view is that RPI is wrong and CPI is right. The TUC find work by Dr Mark Courtney (ex of the government economics service) persuasive that the case is not so clear cut. His arguments have been published in a refereed journal (** – reference at end), and my position remains that his arguments have still not been given proper consideration. (I will try and blog separately on this at a later date; it’s not straightforward stuff.)

The advice of the advisory panel includes this statement from the ‘minority’ (the underline is my emphasis):

An examination into the choice of elementary aggregate formula used in CPIH and a resumption of the research into dealing with clothing prices and other goods which have the largest contribution to the formula effect. These members felt this work was needed to counteract a perception and (in the view of these members) a very possible reality that CPI and CPIH underestimate inflation in areas such as clothing and to improve the accuracy of all indices.

The technical panel however were unconvinced by Courtney’s arguments and Pullinger’s advice does not pick up on this minority proposal.

Understandably the ONS are desperate to clarify the position, and the panels, chairs and ONS secretariat worked intensively and very hard to come to a solution that commanded majority support and took into account minority opinion. It is for the ONS through its future work to continue to make the case and convince.

In the meantime the TUC remain concerned that CPI/CPIH understate inflationary pressures facing households. But even leaving aside any potential inaccuracies, there is colossal inconsistency in practice. The RPI will continue to be used for interest payments on index-linked gilts, for increases in energy and utility bills, and even by the government when it operates to its advantage (e.g. increasing duties and for interest rates on student loans). For as long as this remains the case of course the RPI should be used for pay bargaining.



* The ONS identify us: Jill Leyland of the Royal Statistical Society (RSS), John Astin – nominated by the RSS to sit on the technical panel and in the lead at Eurostat for the original development of harmonised measures, Andrew Sentance senior economic adviser at PricewaterhouseCoopers and previously a member of the Monetary Policy Committee and myself, involved with CPI issues in HMT Treasury over 2011–2014 and responsible for the development and launch of the UK end of the harmonised index of consumer prices programme – NB HICP was renamed CPI when it became the target measure for inflation.)

** Courtney, Mark (2013) ‘Consumer price indices and the identification problem’, Statistical Journal of the International Association for Official Statistics, Vol. 29, No. 1 (2013): pp. 41-51.