MNI INTERVIEW: Seasonally-Adjusted Inflation Data To Help BOE

The UK should be ready to publish seasonally-adjusted inflation data for the first time in 2026, simplifying communications for the Bank of England, which already produces its own in-house adjusted figures, an advisor on the joint project between the Office for National Statistics and the National Institute of Economic and Social Research told MNI.

A technical progress report, under NIESR’s name, will come out in April, and it is likely that at the end of the development process the new series will be published by the Office for National Statistics as experimental data, Professor Huw Dixon said in an interview.

While the UK is playing catch-up, with a dozen national statistics authorities and the European Central Bank already publishing seasonally-adjusted data, the change will pose a communications challenge for the ONS, given the use of inflation figures for setting a range of economic variables, from returns on linker bonds to pay deals.

"Seasonal adjustment as a process naturally involves revisions," noted Dixon, adding that the effects will show up clearly only in monthly data.

“The seasonally-adjusted figure will be based on what's going to happen in the next two or three years to inflation, because they use a centred moving average," he said. "For the annual inflation rate, it doesn't make any difference. For the level, it doesn't. But for monthly inflation, the revisions can be quite significant, up to about 10% of the figure.”

Adjusting monthly data helps separate signal from noise in inflation trends.

"Normally, prices drop in January. So you have negative inflation in January ... And obviously that doesn't mean you're going to have negative inflation going forward, disinflation. So in that sense the seasonally-adjusted figure for January will nearly always be positive.... It does give you a guide ... to trend inflation," he said. 

Unadjusted data will remain the norm for headline inflation and setting the BOE’s inflation target, Dixon said.

"I think the seasonally-adjusted data will remain experimental data ...  if they use it as a headline in their CPI bulletin ... then they may want to make it an official statistic but so far as I'm aware, it will be an experimental statistic.” (See MNI INTERVIEW: UK's OBR To Cut Productivity Only Slightly-Bean)

BOE ON BOARD

The BOE already makes its own seasonally-adjusted inflation calculations, which it factors into its analysis, but it keeps the data to itself.

"The Bank of England does do it, but they don't publish it … I think they would find it quite useful to have an official statistic that they could report on, even if they disagreed with it slightly," Dixon said. (See MNI POLICY: BOE Looks Set To Edge Up Neutral Rate Estimate)

Technical questions still to be addressed include how the seasonal adjustment should be made. This can be done directly, by separately adjusting the headline CPI number and its components, the so-called Classification of Individual Consumption According to Purpose, or COICOPs, or else indirectly, by adjusting the components and then aggregating them to get the headline number.

"What we recommended was just do it direct, not indirect. But, of course, the ONS's preferred method is indirect. For GDP, they do it they do it using indirect,” Dixon said.

Another challenge is posed by questions over the reliability of some lower-level numbers, such as specific food prices, due to a shortage of readings, though this should improve from next month as the ONS brings in scanner data.

"The great thing about scanner data is you've got price and quantity data, whereas the CPI is currently weighted on price data, and then you have to use only the old expenditure data ... which is two years old ... for national accounts, for the COICOP levels," Dixon said.