Trends in sex ratios at birth

The chart for England and Wales since the start of civil registration shows marked multi-year trends in sex ratio at birth (SRB) expressed as live male births per 1,000 female births. Ignoring the short-term year-to-year variation, from a 5% male excess at the start of the period there's a decline to 3.5% excess, an increase to 6%, then a drop to steadying off in recent years at about where it started.
According to the article Gender Ratio, there is no difference in the number of males and females conceived. "For births to be consistently male-biased, there must be gender differences in the probability of miscarriage through pregnancy."

"..there is a higher probability that an embryo with chromosomal abnormalities is male – in the first week of pregnancy, excess male mortality, therefore, means pregnancy is female-biased;
in the next 10-15 weeks of pregnancy female mortality is higher, which increases the ratio in favour of males;
male and female mortality is approximately equal around week 20;
between weeks 28-35 of pregnancy, there is higher male mortality.

Overall, a male-biased sex ratio at birth is the result."

A 2003 article Secular trends in sex ratios at birth in North America and Europe over the second half of the 20th century concluded that "we still cannot put forward any reasonable explanation for the observed trends, which may well be attributable to several factors and not just one."

One data point that stands out is for 1919 where the SRB was 1060, or 6% male excess. It has been speculated that it might be related to the end of the Great War. However, the peak in births was in 1920. Perhaps it's related to the 1918 influenza pandemic which resulted in more challenging conditions for the development of the female fetus.