Many animals display a seasonality to migration, mating, and birth, likely coinciding with weather, food availability, offspring survival, or predator abundance. Unsurprisingly, in humans, seasonality of birth is associated with multiple metrics of birth outcome, weight at birth, myopia, risk of schizophrenia, autism, and even mortality/lifespan.
Remarkably, this sounds a lot like astrology or horoscopes, which are generally considered to be pseudoscience. Of course, astrology contains plausible and implausible facets (e.g. see commentary by Ball 2008). Given the above traits that are shown to be related to month of birth, perhaps a broader connection to personality, behaviour, development, or health is not unreasonable, while vague daily fortunes in the newspaper or fortune tellers may have no basis.
An obvious test for this would be whether the effect is mirrored across the equator, i.e. that those behaving like a capricorn in the southern hemisphere should instead be born in June-July, rather than December-January. This mirroring effect, for instance, is observed in the above refs McGrath 1995, comparing UK to Australia, or Doblhammer 2001, comparing Austria to Australia. More compelling evidence would be whether areas in the southern hemisphere had their own astrology or mythology that captured many of the same personality dimensions, but offset by 6 months. A difficulty here is the limited area in the southern hemisphere of temperate climates, mostly in South America (southern parts of Chile and Argentina). Large areas of both South Africa and Australia are subtropical deserts, historically unsupportive of large populations, so there may not be a detailed history of mythology or astrology from people there. Thus, the effect would then be whether migrants from the northern hemisphere show a mirrored pattern upon moving to South Africa, Australia, New Zealand, Chile or Argentina.
Plotted below are monthly birth rates, for various countries, from 1967 to 2020. Not all years are available for all countries. The traces are normalized by the monthly average for that year for that country, and then adjusted for number of days in that month. The average month would be 30.4 days, so Feb has around 10 percent fewer than those with 31 days.
Click here for a pdf of all countries with data, and excluding a few that were completely noise, typically due to very low population/birth rate (less than 100 or so).
Climate and temperature seem to have a strong effect:
- Several countries have a very even distribution, including Ireland and the UK, which have temperate, oceanic climates. Other large temperate-climate countries, like USA or Japan, show a mostly even distribution, with a slight, but consistent, peak in autumn.
- Many countries in Europe (and elsewhere) show a pattern with a July-September double peak, including Italy, Germany, Austria, Switzerland, Poland, and Slovakia, among others.
- The earlier parts of the timeseries show that most Nordic countries (Norway, Sweden, Denmark, Finland) formerly had an April peak, but that this is switching to the "European" pattern in modern times. This plausibly could be cultural, but also could be the result of warming or climate factors.
- Tropical countries, like Cuba or Jamaica, have a trough in springtime (9 months offset from hot summer) and a peak in autumn. One theory relates this to high temperatures in the summer time, hypothesized to affect motility of sperm citing a single study of only 56 patients. That is, people aren't having less sex in the summer, just fewer pregnancies.
In Austria, many studies examined effects of month of birth, including:
- Ward 1987 examined weight at birth, across seasons and other factors, for appx 10k births from 1865 to 1930, finding increased weight for babies born in autumn.
- Weber 1998 examined heights of 507k men, finding maximum heights for those born in April and minimum for October. They proposed that this was due to higher sun exposure following birth.
- Doblhammer 2001 included a cohort from Austria, of 681k people, arguing that those born in October-December live slightly longer than those born in springtime. They also note that those with higher education tend to be born in spring, rather than autumn.
For both Austria and Germany, births are high for summer, showing a peak in July and a peak in September. Nine months offset from these would be conceptions in October and December.
As the data only go back to the 70s for both countries, it does not give much of a historical view, that is, effectively modern culture with electric lights, hospitals, etc. One could imagine that there is no substantial advantage in modern times, and other social factors are responsible for any planning of conception (the above references may contradict this). If these data were similar in past centuries, this may coincide with conception around the equinox or a harvest festival. Compared to winter, food may be relatively abundant at this time, and could plausibly be related to the nutritional state of the mother during the first trimester. As these festivals also may involve socializing, heavy drinking and intoxicated romance, this could also contribute to the July spike.
Notably, the modern pattern differs from the results of Knodel 1981, who had examined births during the 18th and 19th centuries in villages in Germany, showing a peak in winter and trough in the summer. This difference was most pronounced in the earliest cohort, 1750-1774, and had mostly diminished by 1875-1899. Of course, Oktoberfest was only created in 1810, so any putative summer births due to Oktoberfest would not be observed until some time after.
- Note: the Jan low and the Dec high in Germany are from 2002, for unclear reasons. This could be an error.
Again, there is a history of study on effects of month of birth, for example:
- Samuelsson 1999, examining only 1248 patients (~2:1 M:F), found highest incidence of diabetes in those born in July/August, and lowest for October. The peak of births was in April.
- Doblhammer 2001 included a cohort from Denmark, of 1.371M people, found slightly longer lifespan for those born in autumn or winter.
Overall, the birth trend is different from Germany and Austria, with a sharp spike in the springtime, and then decreasing as the year goes on. The trend is similar for all 3 countries. As time goes on, they have been changing to the common pattern of European countries, with a broader peak during the late summer and fall. Possibly cultural factors play some role in this change, particularly as the time frame of the data spans births from 3 generations (children of baby-boomers, gen X-ers, and now millennials), and could potentially include major changes in city-living, job or vacation schedules, or education levels.
Cuba is a good example of a tropical pattern country, showing a very low springtime birth rate, and high autumn birth rate, reflecting low conceptions in summertime. Other countries or regions with this pattern include other Caribbean countries like the Bahamas, Barbados, Jamaica, Martinique, Suriname, or Trinidad, other tropical areas like Hong Kong and Macau, and even some Mediterranean countries like Cyprus or Israel.
Like the Western Zodiac, the Chinese Zodiac has 12 symbols, each representing an animal. These are combined with 5 elements (wood, fire, earth, metal, and water), each spanning 2 years, to end up with a 60 year cycle of the 5 elements and 12 animals.
There is a year effect on birth numbers that is evident in a few places in Asia, related to the good luck in the year of the dragon (龍), and corresponding bad luck in the year of the tiger (虎). Data are available for Hong Kong, Macau, and Singapore, all show a relative increase in births during the year of the dragon (eg. 1976, 1988, 2000), and a decrease during the year of the tiger (eg. 1986, 1998, 2010). Data for South Korea and Japan do not show this effect, and data were not available for mainland China.
Next year, 2024, year of the wood dragon, would then be expected to show this effect again.
