Why is there a sudden drop in German births?

Could traditional vaules hold the key?

I really didn’t think this would be the topic of my first Substack post, but here goes… Other substackers such as Igor Chudov and Josh Guetzkow have already notably reported on this phenomenon and rightly pointed out it is not restricted to Germany but is also showing up in the data from many countries around the world.

For a number of reasons I am going to zoom in and focus on the births data for Germany. Firstly, Germany has become my home and I have family and friends who are potentially affected, so I want to know what is going on. Secondly, I speak the language, so perhaps I can offer some helpful additional information to the anglo-centric substacker reporting. And finally, the publicly available data(EDIT: births data - other areas are a datastrophe!) here is pretty good so I believed we should be able to get some clarity.

Okay, let’s get into the data. Births (live births) recorded for Q1 2022 in Germany show a significant drop in both year-on-year comparisons and with the pre-pandemic average. It is important to note comparisons with last year are to be made with caution as 2021 was a bumper baby year. Due to this fact and any longer-term trends I have primarily looked at the recent period of the last 5 years. As you can see births are quite cyclical following a quite consistent patter over the past 5 years. Until the first quarter of 2022 when it suddenly plummets: (note: non-zero y-axis, showing monthly live-birth totals):

Depending on how it is calculated (I excluded 2021 and used the Q1 average for 2017-2020 giving 185,105 annual births) we are currently seeing a 11% drop in Q1 for 2022 or ~20,500 births in single quarter. Predictably the German Federal Office for Statistics (Destatis) has reacted to the online commentariat and cautioned that the 2022 figures are still provisional, and attempted to attribute the sudden drop to parents having fulfilled their family plans in the bumper year of 2021. Destatis even went so far as to explicitly rule out any connection with Covid-19 vaccinations. Having used the WayBackMachine to see how updated reporting has previously affected prior monthly figures I can say that this far out from Q1 we are very unlikely to see the provisional figures change substantially and the percentage drop observed will not change by more than a single percentage point. Claims that the 2.9% annual arise of 6,875 births in 2021 is the reason for the year-on-year quarterly drop of 27,366 are laughably inadequate. (again non-zero y-axis below)

My approach is to examine the relationship-pregnancy timeline to try and figure out what might be responsible for such a sudden drop in Q1 2022 births. For a start, we can go back nine months to Q2 2021 to examine reasons for reduced conceptions, we can additionally go further back to examine earlier conditions from 2019 or 2020 which may have led to a reduction in couples trying to conceive in Q2 2021, or we can consider the latter two quarters of 2021 for causes of an increase in miscarriages, and finally we can look at stillbirths in Q1 2022.

The typical theories I have heard (mostly in the twitterverse) to explain away the anomaly are as follows:

1. Economic uncertainty affected family planning decisions

2. Lockdowns and other pandemic restrictions reduced the number of new couples/marriages

3. Coronavirus infections disrupted menstrual cycles or caused miscarriages

4. A particular high-profile public health campaign in 2021

Economic uncertainty is not any more so attributable to Q2 2021 than prior periods of the pandemic. If anything, winter lockdowns were ending, positive case numbers and hospitalisations were falling. Every week was bringing further loosening of the restrictions. The vaccines campaign was in full swing and promised to end the pandemic. All told, many were expecting a return to the old normal which should have been a cue for couples to kickstart any baby plans they might have put on hold during the uncertainty of the previous waves. And considering that births actually increased in 2021 following the first pandemic year, the economic uncertainty theory for Spring ‘21 is a non-starter for me.

Lockdowns: Yes, we all know Germany had some of the strictest, far-reaching lockdowns and other pandemic restrictions (I hope to write more about my personal experiences in another post). However, these measures in Spring ’20 and Winter ‘20/’21 apparently did absolutely nothing to dampen births 9 months later in 2021. As a matter of fact, 2021 was a mini baby boom year in Germany with annual births increasing 2% over the 2018-2020 average and reaching their highest total since 1997. Here you can see a breakdown of the monthly percentage changes in births for 2021 compared with the 2018-2020 average (red) and the 1st pandemic year 2020 (blue). It is clear that in 2021, Feb-Mar-April experienced a boost in births 9 months after the previous spring lockdown and Aug-Sep-Oct-Nov experienced a boost 9 months after the winter lockdown. Simple conclusion: lockdowns didn’t limit couples making babies, if anything it promoted it!

Dating culture: Moving further back the relationship/pregnancy timeline, what if the spring wave and winter wave ‘20/’21 lockdowns limited the number of new couples forming which then would have gone on to have children together in Q1 2022? The majority of babies in Germany are born into established relationships (even if one in three births are officially out of wedlock). For example, in 2020 the average age of a first-time mother in Germany was 30.2 years with only 8% of all first births (~28,500) born to mothers younger than 23 years old. Simply put, the vast majority of German mothers like to wait and plan their family.

Remember, Q1 2022 babies would have been conceived in Q2 ’21 and if we assume an established relationship is at least 12 months long (seems kinda short to me) then we are talking about couples who entered a relationship in Q1 2020 at the latest, i.e. before the pandemic. So, while it may have had some effect, I don’t see a restricted dating scene accounting for anywhere near the size of the drop in births we are witnessing.

Weddings: Moving along the relationship/pregnancy trajectory we come to marriage. Traditionally, people got married first and then make babies, so reduced numbers marrying should have affected births, right? Consistently over the past years, 2/3 of German births are within marriages. Interestingly, while 2021 saw the highest birth figures in Germany in decades, it simultaneously saw the lowest number of weddings. According to the Federal Office of Statistics (Destatis) marriages reached a historic post-war low in 2021 with 4.2% less marriages than the first pandemic year and 2020 producing 10.3% less marriages than the prior non-pandemic year!

These are significant drops from the pre-pandemic average (2015-2019) of 407,374 marriages (male-female) per year. These numbers need to be examined more closely as they are certainly of the same order as the drop seen in births. Curiously, although marriages in 2020 plummeted by about 10%, the next year saw the highest number of births since 1997.

(EDIT: I have rewritten the following two paragraphs to highligh the size of reductions from the pre-pandemic average for 2015-2019 of 407,374)

Consider now that new marriages predominantly lead to first-time births whereas second and successive children are predominantly born within existing longer-term marriages or relationships. The data for 2021 shows an annual reduction of weddings (between a man and woman) of round 14,000 (68,299 reduction from pre-pandemic average) and the drop in the first pandemic year was by ~10% almost 40,000 (43,009 from pre-pandemic average), which when combined represenent a considerable (~111,000 “missing pandemic marriages”) reduction from the pre-pandemic average.

The proportion of first-time marriages is consistently around 69% so I’ve estimated an approximate 47,000 reduction in first-time marriages for 2021 and 29,500 reduction in first-time marriages in 2020 alone. Could these “missing” first-time marriages” in 2020-2021 explain the fall in births in Q1 2022? And if the apparent number of cancelled/postponed wedding plans is at least large enough to explain the drop in births, the question remains: Why so suddenly and atypically starting in January?

Counterintuitively, first-time births actually rose by 1,2% in 2021, the year following the crash in marriages. Indeed Q4 ’21 births for October, November, December 2021 directly preceding the drop in births in the first quarter of 2022 were still all up significantly above the five year average. So how can reduced marriage figures during the pandemic years result in such a sudden sharp drop noticeable in January 2022? Here’s my suggestion: if couples typically wait about a year after marriage to try starting a family (i.e. the first baby comes 21-24 months after marriage) then the chart below comparing marriage totals by quarter could actually explain the sudden drop. (no zero y-axis again, please read the values)

Notice that the sizable drop in marriages in 2020 is overwhelmingly concentrated in Q2 and Q3 2020 presumably due to wedding plans cancelled or postponed because of the pandemic. I don’t have a breakdown by month of male-female marriages but it looks reasonably like a 25% drop of approx. 30,000 marriages occurred in Q2 2020 and about 12,000 in Q3 2020. These ~42,000 fewer weddings concentrated in Q2 and Q3 are a very strong effect. Indeed the following Spring Q2 2021 again witnessed significantly fewer marriages.

Data released for 2020 shows almost half of all births were first-time births and 60% of those were within existing marriages, in other words, 60% of Germans wait until after marriage to start having children. For example, looking at 2020 we see 214,000 first-time births were within marriages and accounted for 28% of all births that year. The magnitude of these figures seems to provide a plausible basis for the scale of reduction in births seen so far in 2022. The working hypohesis: did a major disruption to marriages in Q2 and Q3 2020 cause a delayed effect resulting in a major disruption to births in Q1 2022

Make of it what you will, I unfortunately still haven’t found data on the average length of time between marriage and births nor have I seen a breakdown of the births for Q1 2022 by 1st, 2nd child etc. If the proportion of first-time mothers has fallen much more than 2nd and successive births I think it would support this theory.

Conceptions

Returning to the pregnancy timeline let us consider whether SARS-Cov2 infections could have affected menstrual cycles and the ability to conceive. Yes, of course, just as any number of health-stress factors. But is there any indication that there was a sudden increase in corona infections within the child-bearing population of women in Q2 2021 considerably more so than during the previous winter wave? Recall Q2 2021 represents the falling of case numbers and the easing of restrictions. The following graph we can see the 20-39 cohorts saw no noticeable change. This does not suggest infections as a cause for such a sudden change in conceptions in Q2 ’21.

Moving forward along the relationship-pregnancy timeline we can ask, could coronavirus infections have led to more miscarriages in Q3 and Q4? That for me is a big question. Let’s look at the age profile of corona cases in Q4 2020 compared to Q4 2021 to see if more women of child bearing age were being affected. Yes, there were more positive cases of younger women and also hospitalisations just as there was across all age cohorts in Q4 ’21 peaking in late November (before skyrocketing again in Q1 2022 with Omicron cases). Could infections have caused such an increase in miscarriages?

Some anecdata from a midwife friend: she told me she had informally heard of an increase in miscarriages late last year from her colleagues working in hospital maternity departments (my friend offers home birthing).

Finally we come to vaccinations. In the 3 quarters prior to 2022 what effect could vaccinations have had on conceptions or miscarriages. When I asked my midwife friend if vaccination rates were perhaps lower amongst her clients opting for home-birthing than those opting for hospital deliveries. She mentioned two interesting points, firstly, first-time mothers are more likely to opt for hospital deliveries. Secondly, once the German body responsible for reviewing vaccinations (STIKO) formally recommended vaccination for women planning pregnancy, those pregnant (starting in 2nd trimester), and those breastfeeding that she saw a big uptake amongst her clients. When I checked to see when the German Vaccine Commision made this recommendation it turns out that was in September 2021. This made me think the effects of the vaccinations on this cohort should be more pronounced in the last quarter of 2022.

Note: the graph below from the Robert Koch Institute estimates that less than 40% of 18-59 year-olds had received 2 doses by the end of Q2 2021. It is safe to assume the percentage was much lower for 20-40 year-olds and perhaps even lower amongst those of them women since STIKO had not yet recommended the vaccines for those pregnant (2nd & 3rd trimester), breast feeding, or planning pregnancy. Unfortunately, I haven’t yet found German data for vaccination by sex, with smaller age-bands, and by dose.

Last but not least I’ve looked at the stillbirths data linked at the Jackanapes substack and while there may be a signal there, the totals are nowhere near explaining the drop in live births observed in Q1 2022. As a matter of fact, only one of the provisional monthly figures for Q1 2022 even shows an increase over 2021. Furthermore, stillbirths have been rising over the past 10 years, so am really not sure how significant the increases in 2021 are. The first graph below shows annual total numbers of stillborns. The second shows monthly totals for 2021-22 compared with the average from 2019-2020. What does stand out to me is the large deviation from average in Dec (boosters?/winter wave?) and April (1st dose?/spring wave?). What is not noticeable is any September/October deviation I might have expected (once STIKO recommended vax for this cohort).

Summary: I have tried to give a fair treatment of the main theories being offered to explain the SUDDEN drop in live births in the first quarter of 2022 in Germany. I was surprised to find a correspondingly sudden drop in old-fashioned weddings in Q2 2020 that may possibly explain the anomaly, although I haven’t explored any marriage data from other countries to see if a similar explanation is plausible. If anyone reads this, maybe they could chip in with their countries’ marriage data during the pandemic.

Remember I am not trying to debunk anyone else’s theories, I just wanted to look at the data more closely and examine the alternate explanations out there. I am fully prepared for there to be some Covid-19 or Vaccine effects hidden in the data. For example, I critically wasn’t able to find more detailed information about miscarriages since the rollout of the vaccines and stillborn data doesn’t capture 1st and 2nd trimester miscarriages.

For me the simplest explanation in the German instance remains the drop in marriages during the pandemic as primary driver of drop in births, although I still find the concentrated effect suddenly appearing in Q1 2022 surprising. If my theory is correct, we should critically see the numbers of first-borns dropped by a greater proportion than those of second/third births, etc. Also, birth figures could remain depressed for many months to come reflecting the long-tail effect of the significantly reduced number of marriages during the pandemic.

Any feedback very welcome.

==

UPDATE in new post:

The Wedding Crasher