The RPI measure of inflation is “without merit”. The retail price index has been an official measure of inflation for close to 50 years in the UK, but it requires a “deliberate and carefully timed” withdrawal. That’s according to Mark Carney, the governor of the Bank of England who made the case in a recent speech.
The seemingly arcane issue might, at first glance, seem like a comment on an obscure statistic. But the RPI is a measure that affects the day-to-day – and longer term – out of pocket expenses of millions of people. It is used as a basis, among other things, for calculating the interest rates on student loans, annual increases on train fares and, most relevantly for the Bank of England governor, the interest calculations on UK government gilts.
The RPI is not only used by the government, but by companies in setting terms for annual price increases, too. For example, phone company Vodafone uses it to set its plans. Given that the governor of the central bank feels the measure is without merit, one might question why the measure is so widely used?
The simple answer is that the rate of inflation reported by the RPI is likely to be higher than that reported by other measures, such as the more common measure of inflation the Consumer Prices Index (CPI). This is good news for those that collect money using RPI (such as the Student Loans Company, rail companies or your mobile service provider). But it is bad news for those whose incomes only increase in line with CPI.
Out of date
The history of inflation measurement in the UK is a complex one, which represents the intersection of economics, politics, statistics and data science. The RPI was officially introduced in 1956 and remained the official rate of inflation until it was replaced by the CPI in 2003, a measure which has existed since 1996 in response to an EU directive to standardise the way in which inflation was measured in the eurozone.
In the process of this standardisation, the key point was that a formula used by the RPI was effectively dismissed as the accepted way to measure inflation. This difference helped ensure that generally inflation as measured by the CPI would be lower than the comparative measure from the RPI.
Having spotted that the CPI rate of inflation was lower than the RPI rate, subsequent governments have identified a handy tool by which to reduce their fiscal deficits. When paying money out, for example setting a triple-lock for pensions, use the lower CPI inflation rate. And when getting money in, for example collecting student loans, use the higher RPI rate.
This confusion expands beyond the public sector. For example, employers are likely to want to negotiate wage increases based on the CPI, while employees and their unions would rather use the RPI rate. This has led to the current state where an inflation measure, which might no longer be considered appropriate, is still used in a wide range of important areas.
This brings us back to Mark Carney’s comments about withdrawing the RPI. He’s not alone in thinking this. The UK Statistics Authority, the country’s statistics regulator, no longer even classes RPI as a national statistic. Paul Johnson, head of the important Institute for Fiscal Studies think tank, urged that it be scrapped in his wide ranging review of such statistics. And expert in the area Erwin Diewert broadly recommended the same thing in a 2011 consultation carried out by the Office for National Statistics.
So it might seem a no-brainer to let the RPI die. But a couple of things prevent this from taking place so easily. First, the strategic use of the RPI by governments has been advantageous in giving fiscal decisions the veneer of statistical propriety.
And second, there remains one potentially damaging area where an immediate withdrawal of the statistic might cause problems: UK government gilts. These are effectively debt, which the government issues to fund its borrowing. It pays back interest based on the RPI and the contracts for these instruments say that any material change to this measure will allow gilt holders to cash in their debt immediately. That would equate to £30 billion.
As simply switching out RPI for CPI could therefore represent a significant challenge to government finances, there is a need to exercise deliberate and careful timing (as cautioned by Carney) in removing this statistical artefact from the landscape of UK economic statistics. There are no such reasons why use of the RPI in other areas, such as train fares and student loan interest, should not be curtailed in the short term though.