The seasonal adjustment otherwise known as Christmas


To mitigate the adverse economic effects of Christmas – which are large – one could think of a phased abolition of Christmas, raising annual GDP by as much as 1%. Or encourage a wider observance of Hanukkah, Ramadan and other festivals. (It would, however, be necessary to emphasise that only one observance per household could be tolerated.)

Among the many leaks recently from the Treasury, the following briefing paper recently came into my possession.

“You asked for a report on the economic consequences of Christmas. You will be aware that Christmas and New Year fall on a Sunday this year, and so Monday 26, Tuesday 27 and Monday 2 January are holidays. Much industrial production will cease on these days and will be reduced on Wednesday, Thursday and Friday. Similar effects are anticipated in business services and public administration, including education. A Treasury team will of course be available to you throughout although we understand you will be taking a short break with your family on Christmas morning itself.”

What ordinary people call the Christmas rush is known as seasonal adjustment at the Office for National Statistics. Seasonal adjustment is essentially a mechanical procedure, undertaken by computers which not only do not celebrate Christmas or late holidays, but are not informed of them: they deduce from recurrent patterns in December and August that something unusual is going on, and compensate accordingly. Moving festivals, like Easter, are especially problematic.

Seasonal adjustment means that the differences between raw data and published economic statistics which grab headlines may be large. Retail sales of non food items are thirty to forty per cent higher in December than the average for the year as a whole, and food and drink sales ten to fifteen per cent higher.

In Britain, industrial production varies by as much as 15% over the months of the year. December is not particularly low: the boost to demand seems to offset the interruptions to supply. The months with highest output levels tend to be those with 31 days and few holidays – March and October always perform strongly – and July and August are normally worst.

“You asked about effects on productivity. Overall, maximum total factor productivity (TFP) is generally achieved when inputs of labour and capital services remain stable over time. Significant variability of labour inputs due to many people absenting themselves from work simultaneously reduces aggregate TFP. Surges in retail sales during the period are productivity reducing: shops attempt to compensate for inadequate capital resources with additional labour, which distorts the optimal capital labour ratio. The labour itself is of lower effectiveness. Members of the Treasury staff report anecdotal evidence of a seasonal dip in economy wide productivity.”

In plain English: most industrial processes run best when inputs and outputs are relatively stable: shops in December are to busy and employ incompetent part-time staff: people are hungover after Christmas parties, they go for drinks at lunchtime and are less effective in the afternoon.

“Such productivity reducing factors are aggravated by the phenomenon known among economists as the Waldfogel effect. Shopping on behalf of other consumers is more time intensive than shopping on one’s own behalf and less accurately reflective of individual choices. Professor Waldfogel has prepared estimates of the economy wide losses from this phenomenon.”

The Waldfogel effect is exemplified by the tie your mother-in-law bought you, which is worth less to you than it cost her to buy. This puzzles economists, since rational economic man would not undertake such transactions. But your mother-in-law is not rational economic man and nor, if you hope for a merry Christmas, are you.

“You asked for specific policy suggestions to mitigate the adverse economic effects, which are large. A phased abolition of Christmas might raise GDP by as much as 1%. Alternatively the cost to the economy could be mitigated if reductions in labour supply were spread more evenly over the year. Consistent with the government’s support for multicultural policies, wider observance of Hanukkah, Ramadan and other festivals might be encouraged. It would, however, be necessary to emphasise that only one observance per household could be tolerated. In addition, a substantial reduction in the threshold at which gifts are taxed might discourage wasteful exchange and reduce the Waldfogel effect. Let me know if you require further briefing on these suggestions.”

Scrawled on the document by the messenger was the following. ‘These guys regularly talk about “effects on the economy” as if “the economy” was an end in itself rather than a means to human welfare. Get a life, or at least a drink.’

Print Friendly, PDF & Email