Mauled in the mall

‘Retail sales unexpectedly fall’, was the headline in Reuters yesterday afternoon, announcing the US Commerce Department’s latest report on the shopping exploits of the American public last month. Other agencies and media had more or less the same wording for one of the more important items of economic data on the monthly cycle. 

But the economic importance of the data, significant though it may be, is dwarfed by the sociological impact. First, what economists call retail sales is actually SHOPPING, the primary religious rite conducted by vast numbers of Americans in huge temples devoted to this purposes, on at least one or both days of their weekend and on or before all major holidays.

Economists recognise the centrality of retail sales, because they know that private consumption comprises over 70% of total US economic activity, as measured by GDP. But they do not accord retail sales any emotional – let alone spiritual – value, although funnily enough, many of them do have strong emotions about employment data.

Second, although there are retail sales data every month, Christmas cometh but once a year – and that’s in December, in case you didn’t know. Hence retail sales data for December is not just another month. Even for economists, it’s special – because the ‘holiday season’, which actually stretches from Thanksgiving in late November through the end of December – is critical for the retail sector, generating as much as half of total annual sales for some product groups.

In light of the foregoing, the fact that retail sales fell in December is very bad news for economists and sad news for religious shoppers. Obviously, it is further evidence that the economy is weak and that any talk of recovery is more wishful thinking than substance. But the fact that retail sales unexpectedly fell is a story in its own right. It is sad news for economists and very bad news for religious shoppers, with implications that stretch far beyond one month or even one year.

Why was this fall ‘unexpected’? For starters, what does that term mean? It means that Reuters, or Bloomberg, or whomever, did a survey of several dozen economists and analysts who work for financial firms such as banks, brokerages, fund managers, etc, and actually asked them: “what do you expect to be the change in retail sales in December?’ The answers ranged from a small drop to a gain of as much as 1.1%, with the average of all these answers presented as the ‘consensus expectation’. Needless to say, the bloke who predicted a drop is feeling very good today, whilst the guy at the other end of the scale is not the happiest or most popular chap in his firm right now. But never mind, there are more data next week to ‘predict’ and even retail sales will come round again next month.

Why, though, was this fall unexpected by most of the people who are paid handsomely to figure out what to expect? There are numerous answers, among them the usual suspects of laziness, charlatanism, etc. But there are also more interesting features. For instance, in recent days and as part of the run-up to the publication of this item of data, there have been many reports from specific companies about their sales in December – and most of these reports have been positive compared to last year. Were these reports wrong?

Not at all. There was indeed an increase in sales over last year – but that was not unexpected, that was a near-certainty, given the intensity of the atmosphere of gloom and fear that held sway in late 2008. The fall that was reported, versus the rise that had been expected, were with regard to November – in other words, a month-on-month (m-o-m), rather than year-on-year (y-o-y) change. M-o-m is much harder to predict, or even guess, than y-o-y.

But there is another pitfall, this one much more subtle. All these stories about the increase in sales over last year – what was their source? It was the companies whose stores made the sales – whether Sears, Best Buy, GAP or whomever. All the big chains, basically. Correction: all the big chains still around in December 2009. Circuit City, for instance, didn’t report its sales – because, having gone bust earlier in the year, it didn’t have any. This phenomenon is called ‘survivorship bias’ and it’s incredibly important, especially in recessions or market slumps. If you survive, you pick up market share from those who didn’t make it. This is true even within chains: all the big chains closed stores this year, so the dead stores didn’t report December sales – thereby effectively ensuring that the survivors would have improved sales over last year. Dead men tell no tales, and dead stores make no sales…

There is a still more fundamental reason why the m-o-m drop should not have been unexpected – i.e. why the analysts should actually have expected a fall. The data for consumer credit show a steady and intensifying trend – downwards. The quantity of credit-card debt held by the public is shrinking, which is almost unprecedented; it is, in fact, shrinking rapidly – which is totally unprecedented. Shopping without credit is meaningless, and in many cases simply impossible.

In other words, the religion of shopping is withering, because former believers are abandoning it in droves. Many others still visit the temples, but don’t actually take part in the essential ritual of buying things. Indeed, many temples are in danger of closing. For tens of millions, Haiti is just a television story. Falling retails sales, aka less shopping, is not mere data and statistics, it is the end of the world as they know it.