I am making contact due to the extraordinary nature of the average earnings figures which were released by the Office for National Statistics on Tuesday morning.
This saga starts in the following fashion.
“Growth in average total pay (including bonuses) among employees for the three months October to December 2020 increased to 4.7%, and growth in regular pay (excluding bonuses) increased to 4.1%.”
When you consider that we are in a pandemic situation with the economy having contracted by around 10% that number immediately fails any basic sense test. If we stay within the same labour market release then we see that the hours worked figures have fallen by around 7% since the pre Covid-19 pandemic period.
Sadly things then get even worse for the analysis as shown below.
“In real terms, average pay has rebounded from the sharp falls during early summer 2020, in December total and regular pay are at a record high.”
So after more than a decade of weak real wage growth that has been so poor we have failed to recover the previous high from 2008 we see that according to these figures all we needed was an economic depression. If we project that forwards then if 2021 is as bad as 2020 then we will see extraordinary real wage growth! Checking back through the series we do not seem to have had annual wage growth as fast as we have now since April 2001.
Looking at your Code of Practice these numbers fail at least two of your pillars.
The statistics must be the best available estimate of what they aim to measure, and should not mislead. To achieve this the data must be relevant, the methods must be sound and the assurance around the outputs must be clear.
Trustworthy processes to create high-quality data may not be useful to the public if the statistics are not accessible, do not address key questions, are inefficiently produced, and do not add value or provide insight.
I do not wish to get too bogged down in detail in my initial supposition, but more can be found at the link below.
The problems I have identified are in fact features of what have been long standing issues. The statistician Simon Briscoe has investigated the matter previously. A link to his work is below
I would suggest an immediate investigation with a view to providing the evidence for a potential suspension of the National Statistics status of the Average Earnings data set.
Response from Ed Humpherson to Shaun Richards: ONS Average earnings figures