By David Milliken and Andy Bruce
LONDON, Sept 11 (Reuters) – The Bank of England reported a sharp drop in the public’s expectations for future inflation in August after it changed the company that conducts surveys for it.
Public inflation expectations are a key factor for many BoE policymakers in assessing the risk of persistent inflation, but the central bank said it was hard to draw firm conclusions about the trend due to its change in survey providers.
“Comparisons of changes (in) inflation expectations between the headline May results – Ipsos – and the headline August results – Savanta – should … be treated with caution, as in part they reflect changes in the provider as well as like-for-like changes in expectations,” the BoE said.
Savanta’s August survey showed year-ahead inflation expectations of 3.2%, down from the 4.0% Ipsos had reported for May. Two-year expectations fell to 2.9% from 3.5%, and five-year expectations dropped to 3.2% from 3.9%.
But most of that decline likely reflected the change of provider rather than any real shift in sentiment.
The BoE said it had also commissioned Savanta to poll expectations in May, and those results came in around half a percentage point below Ipsos’s across all three horizons, at 3.6%, 3.1% and 3.3%. Both firms found an identical reading of 5.0% for the public’s perception of current inflation.
Pantheon Macroeconomics said the switch had flattered the numbers and faulted the BoE for not publishing a consistent data series.
“In our view, the BoE should have published a break-adjusted consistent time series to allow everyone to report the same, consistent, official figures rather than reporting an apples-to-oranges comparison,” said Rob Wood, Pantheon’s chief UK economist.
Adjusting for the break, Pantheon said expectations had eased only slightly, in line with lower motor fuel and food price inflation, with five-year expectations fractionally below the record high reached in the second quarter.
The BoE said it awarded Savanta the contract after a competitive retendering process.
The central bank did not immediately respond to a request for comment on the reason for the difference between Ipsos’s and Savanta’s results.
(Reporting by David Milliken and Andy Bruce; Editing by Susan Fenton)
