Retail sales fell back last month after strong Christmas trading, leaving high street retailers and economists struggling to predict the likely strength of the consumer this year.

The Office for National Statistics said retail sales volumes fell by 1.3 per cent on a seasonally adjusted basis between December and January, reversing a steady climb in the index since the autumn. This was much weaker than the market expectation of a 0.2 per cent fall.

But sales between November and January, the three months spanning Christmas, were still pretty healthy. Volumes were 1.3 per cent up on the previous three months and 2.6 per cent higher than the same period a year earlier.

The annual growth in sales volumes was spread widely across most retailers, although household goods shops have suffered from rapidly declining prices, leading to a more than 2 per cent fall in the value of sales in the latest three months compared with the same period a year ago.

Investors interpreted the data as pessimistic for economic growth, pushing down sterling and market expectations of interest rates. The pound fell 0.25 per cent to $1.7352 and in the short sterling market, where investors trade on the likely future level of base rates, the implied interest rate at the end of the 2006 dropped by 0.05 percentage points.

When official figures have more than one plausible interpretation, City economists tend to favour the view that is in line with their economic forecasts.

True to form, economists give two contrasting accounts of the strength of the consumer. Vicky Redwood of Capital Economics said the “Christmas spending revival looks temporary”; while Nick Verdi of Barclays Capital said the trend in retail spending “has clearly recovered from the weakness seen in the first half of 2005”.

The argument among the economists of differing camps hinged on the importance of the January figures in the volatile trading period over Christmas.

Malcolm Barr of JP Morgan was of the view that there was “no reason why one should not take the message from the data at face value”. He concluded: “The upswing in spending in the fourth quarter appears to have evaporated in January.”

But Mervyn King, the governor of the Bank of England, has always chided those, like Mr Barr, who have taken a firm view of the Christmas trading period early in the first few months of the new year. Last January he said: “We should recognise that the true meaning of the Christmas story will not be revealed until Easter – or possibly much later.”

Mr King’s point was simple. ONS statisticians have to make huge adjustments to the raw spending data to arrive at their estimates for the underlying strength of retail sales in December and January. Even a small change in the pattern of spending from year to year can create spurious monthly movements in the figures.

The difficulties facing the ONS become apparent from a comparison of the past two years’ raw sales figures. In January 2006, retailers sold 32.1 per cent less than they did in December; a year earlier, the drop in January sales was 29.5 per cent.

First the statisticians must adjust these numbers for changes in prices to get the volume of sales in each month. Then, by looking at the historical pattern of falls in sales volumes between December and January over many years, they come to their assessment of the monthly change in underlying sales volumes. This year they said volumes fell by 1.3 per cent, last year they said volumes rose by 1.8 per cent.

The statisticians’ adjustments would work perfectly if sales patterns between December and January were similar every year. But, unfortunately they are not.

Some years, consumers rush to the shops before Christmas to avoid being disappointed, leaving little more to spend in January. Other years, particularly if retailers are reporting difficult Christmas trading, consumers hold back, waiting for bargains in January.

There was plenty of anecdotal evidence that consumers held back in December 2004 and waited until the sales of January 2005. This year, it appears the trends were reversed.

If true, the 1.3 per cent drop in retail sales volumes in January was simply a statistical fluke and nothing to be concerned about.

So far, the weight of evidence is with the optimists like Mr Verdi that the festive trading season was reasonable and only the three- month comparisons are of any real use.

But before optimists claim outright victory, there is a catch, pointed out yesterday by Danny Gabay of Fathom Consulting, among others. He calculated that unless there is a sharp improvement in February’s retail sales volumes, the three-month picture would start to look weak very quickly as the November and December figures dropped out of the calculations. For the first quarter of the year, “it will require growth of at least 0.9 per cent in February and then another 0.4 per cent in March, to get the three-month rate into positive territory,” Mr Gabay said.

Such growth rates are possible, especially if January’s level of retail sales has been depressed artificially by the ONS’s seasonal adjustment, but by no means in the bag.

Analysts will have at least another month of biting their fingernails before it becomes clear whether the high street is revealed to be growing strongly as the Bank of England expects, or returns to the doldrums as many in the City predict.

Get alerts on Central banks when a new story is published

Copyright The Financial Times Limited 2018. All rights reserved.

Comments have not been enabled for this article.

Follow the topics in this article