Nigel Hawkes: Love in a cold climate: Why baby boom is not down to big freeze

If you believe what some newspapers report, last year's winter storms are about to be followed by a blizzard of September births.

The evidence comes from Portsmouth, where the city's midwives expect heaving maternity wards after an unusually large number of 12-week pregnancy scans were carried out in March and April. Six hundred babies are expected to be born there in September – 20 per cent more than usual.

One city doesn't make a national trend, so it would be easy just to dismiss this story. But the way births are distributed across the year is a fascinating subject, and there is indeed evidence that the Christmas holiday has an influence. What is much less plausible is the suggestion that a cold Christmas has us all cuddling up for lack of anything else to do, appealing as it sounds.

In Europe, births peak in early spring and summer, and again in September. In the US, by contrast, births fall in spring, but there is a similar September peak. The pattern is persistent and hard to explain, since the populations share a common genetic inheritance – and Canada follows the European pattern, not the American one.

The differences can be quite large: in England and Wales, about 4 per cent more births are recorded in June and in September/October than in an average month, and 7 per cent fewer in February (a short month, of course).

Parents have a preference for summer births, so these effects are partly driven by conscious choices. But long-run studies have shown that seasonality pre-dated effective contraception. "Through wars, through social upheaval, through government-supported programmes of fertility control, through tremendous changes in marriage and divorce patterns, through both strengthened and weakened control over abortion, birth seasonality patterns in most parts of the world have remained constant," wrote Gabriele Doblhammer and Joseph Lee Rodgers, the authors of a study of seasonal birth trends in Austria in the 19th and 20th centuries.

The September peak, common to both the US and the European pattern, is generally attributed to conceptions occurring during the Christmas break. Given time to spend at leisure, and not much else to do, peoples' thoughts will naturally turn to private amusements. Two American economists, David Lam and Jeffrey Miron, found September peaks in births in every Christian country they studied, from both the northern and the southern hemisphere. This result argues strongly that Christmas has an effect, but that temperatures at Christmas do not.

Other bits of evidence also point to a rise in sexual activity over Christmas. Sales of condoms peak before Christmas, and slump the week after, Kaye Wellings and colleagues from the London School of Hygiene and Tropical Medicine reported in 1999. Sexually transmitted infections peak in the first quarter of the year, and so do terminations.

So if the Christmas effect is real, what about the possibility that a cold Christmas will enhance it? Lam and Miron suggest not, but they don't rule out temperature effects. Baking summers in the southern US are, they suggest, responsible for the spring trough in US births. My, it's just too hot...

As a simple check I compared average December temperatures for central England (a record exists from the 17th century) with births in England and Wales the following September.

Taking eight cold Decembers since 1960, and eight warm ones, I could find no link between temperature and the September peak. The exceptionally cold Decembers of the early Sixties were followed by no September peak at all, yet births did peak in that month after the warm Decembers we experienced in 1985, 1994 and 2002.

This is far from conclusive. A comparison with Christmas temperatures, rather than averages for December, might produce a different result. The figures I compared do hint at a more consistent September peak in recent years than in earlier ones, for which the obvious explanation would be the increasing length of the Christmas holiday. But that's a guess.

This month's deliveries will tell us if there really has been a surge in births, beyond the usual September peak, and if it extends beyond Portsmouth. It would be extraordinary if there were.

Nigel Hawkes (born in September 1943) is director of Straight Statistics (straightstatistics.org)