The future of the RPI

Can the Retail Prices Index be killed off? Should it be killed off and, if so, for what reason? Or is reform needed? A meeting is coming up (at the RSS in London, on 13 June, book here) to discuss the future of the RPI and the changes needed to all consumer price measures to keep them fit for purpose. Why not come and hear the views of John Pullinger, the UK’s National Statistician, and other experts?

The RPI is a measure that has been with us broadly in its current form since the 1950s, it is one of the most loved and widely used statistics and the only one specifically mentioned in legislation (the Statistics and Registration Service Act 2007, section 21) that must be produced. It is written into contracts and everyone is familiar with it. But in 2010, the ONS changed the way they collected clothing prices and it affected the index in such a way that the gap between it and CPI, the other main measure, widened. The CPI was already gaining popularity among economists not least because it was created on economic principles and is a harmonised definition common across the EU.

Seeing a chance to save money, various organisations have since then shifted contracts that were uprating by RPI to uprating by the lower CPI. The most notable example is the Treasury with regard to benefits and civil service pensions, successfully reducing public spending but also cutting the income to recipients. Following an UKSA assessment (no 243, March 2013) and its National Statistics de-designation, it is fashionable in Treasury/economist circles to rubbish the RPI and say that it is not to be used.

The UKSA-commissioned Johnson Review, published in 2015 suggested that the RPI and its derivative indices should be “discontinued” where possible and that “Government and regulators should work towards ending the use of the RPI as soon as practicable”. It continued: “Over the long term the Authority should look to phase out production of the RPI in consultation with users, amending the law as necessary.”

That said, as gilts (government debt) are pegged to RPI and the last one will mature in 2066, we have at least another 50 years of using it. Given the length of time, it really can’t credibly be left to wither without being updated and changed to keep it fit for purpose, even as a legacy measure. Some pension schemes, another area where RPI is deep in contracts, are shifting from RPI to CPI but most cannot due to the wording of the scheme rules. Some proposed switches (notably Thales and BT) have been challenged in the courts and been rejected, with judgments saying that RPI remains valid.

More generally a DWP White Paper on pensions (March 2018, paras 212-219) turned against the idea of allowing an uprating over-ride of RPI, and concluded that it was: ” ….. not persuaded by the view that employers or trustees should be able to override scheme rules on grounds of rationality and fairness, given the lack of consensus on what constitutes fairness in this circumstance.”

The legal cases and the DWP policy, have led some leading commentators to shift their view that if RPI cannot be disposed of as easily as thought, it needs to be reviewed. Leaving a system that in effect offers an up only ratchet in the inflation wedge between RPI and CPI is not acceptable. Given not all is perfect with the CPI, and the HCI is being developed, it is time to review all the measures on consumer price inflation.

The meeting will be a rare opportunity to hear the arguments of both sides.