There are lies, damned lies, and statistics. Nowhere is this truer than in the UK’s retail sales numbers reported each month by the ONS.
December’s numbers, which were released earlier today, are a case in point. The headlines they have generated is that retail sales slumped in December and that retailers had a terrible festive period.
The headline is important, not least because it affects markets and it partly influences the confidence the public has in the economy — in this case in a negative way.
The trouble with the reporting is that it is, at best, a partial truth and is, at worst, highly misleading.
It is also confusing, especially given that it contradicts numbers out earlier this month from the British Retail Consortium which showed steady sales growth, and the plethora of solid trading updates that retailers have released.
So what really happened to retail sales in December?
Well, they didn’t fall. Indeed, an examination of the ONS own detailed figures shows that they increased.
The difference between what occurred and what is being reported arises because the ONS headline figure is one that looks at the change in the amount of stuff we buy compared to the previous month.
That is not a measure of the sales retailers made, it is a measure of the quantity of things they sold – an interesting, but entirely different metric.
The trouble with measuring quantity and looking at it in isolation is that it provides a very poor view of the true state of retail health.
If a retailer sold 10 jumpers at £10 one month and then sold nine at £20 the next month, would it be logical to say that their sales had fallen and that their performance had worsened? Of course not. The same thing holds true for the wider retail economy.
A further complication creeps in because of the ONS’s habit of reporting changes on a seasonally adjusted basis.
This is an illogical and totally unnecessary calculation that does nothing but distort reality. How many retailers, or indeed other businesses, release their own sales on some kind of seasonally adjusted basis?
The answer is none. Broadly, they report what they actually sold; they align with reality.
Admittedly, dealing with reality can be messy. Occasions like Easter move about. Some holiday dates shift.
Weather patterns change. And so on and so forth. But that’s life; that’s what retailers must deal with, and do deal with, on a daily basis. And that’s precisely what the fluctuations and movements in their sales reflect.
It is better to accept this messy reality, and to discuss and explain it, than to produce some artificially sanitised version of the numbers.
The more you move away from actual figures, the more difficult explanation and rationalisation becomes. You end up analysing something that didn’t happen.
When it comes to national statistics, gathering lots of numbers is important, but not half as important as interpreting and reporting them correctly. Alas, as the former US secretary of state Colin Powell once said: experts often possess more data than judgement.