As someone who came to grad school with basically zero programming and statistical software knowledge, I’m always looking for ways to improve my R skills. I thought I’d share the most recent fruits of this labor here. Huge thanks to my friend and fellow student Monica, who figured out how to rearrange a matrix of period data into a matrix of cohort data (and has also helped me with for-loops more than once). This data is from the Center for Demographic Research at the New Economic School in Moscow, a great, freely available resource.

My first impression of the Russian demographic crisis, when I was living in Russia and heard it discussed, was that it was a crisis of fertility – not enough babies being born. This is how it tends to be thought about, especially in official discourse. But it isn’t actually a fertility crisis. Russia’s birth rate is below replacement, but not desperately so; it’s substantially higher than in much of Europe, and close to the U.S. (Contrast this with the collapse in life expectancy in the 1990s, which was completely unprecedented, comparable only to sub-Saharan Africa at the height of the AIDS crisis.) For this R exercise, I thought I’d play around with fertility data to get a better picture of what the last ~25 years have looked like for Russian women of childbearing age.


Sorry this GIF is a bit like an old movie (and super fast) – my GIF-making skills aren’t up to par yet.

The curve here shows yearly age-specific fertility rates, from which we get the yearly total fertility rate (TFR) – the sum of fertility across all ages, and a very common demographic measure. In 1989, the curve’s peak was early and very high, meaning a large proportion of births were to women in their early 20s. (In fact, the peak rate of .18 in 1989 means that almost 1 in 5 21-year-olds had a baby that year!)

As the crisis years of the 1990s progress, we see a drastic collapse of the peak and a general fall in the level of the curve at all ages. The peak, though smaller, remains at age 20-21 until 2001, when it begins a rightward shift. Fertility begins to rebound, and the curve rises as it shifts rightward, with the slope down from peak fertility toward older ages growing much less steep than it was in 1989. This means that although the graph still doesn’t look as tall in 2014 as it did in 1989, the TFR has actually made a substantial rebound, as births are distributed more widely over the reproductive lifecourse:

Year TFR
1989 2.01
1990 1.89
1991 1.73
1992 1.55
1993 1.37
1994 1.39
1995 1.33
1996 1.27
1997 1.22
1998 1.23
1999 1.15
2000 1.19
2001 1.22
2002 1.28
2003 1.32
2004 1.32
2005 1.27
2006 1.28
2007 1.39
2008 1.47
2009 1.54
2010 1.57
2011 1.58
2012 1.69
2013 1.71
2014 1.75

We are seeing two phenomena combined here – one, the unique social and economic crisis of the 1990s and its attendant drop in birth rates, and two, a shift toward childbearing at older ages, which is a common phenomenon in developed countries. The latter has made (and may still be making) Russian fertility look lower than it really is, as shifting norms have left holes in the fertility schedule: older cohorts are finished with their peak childbearing, but younger cohorts have decided to wait a bit longer to start theirs – so even though women in both cohorts might intend to ultimately have the same number of children, in a given year or set of years, no one is having many.

A remedy to the distortions of period fertility measurement is cohort fertility measurement, or looking at the fertility experience of one birth cohort through time.


Although the line is incomplete (the last cohort shown here, born in 1986, is turning 30 this year and has half their reproductive lives left to live), this graph shows how different the fertility experience has been for different birth cohorts, and especially shows how childbearing at every age after 23 has grown with each cohort.  Amidst all its other transitions over the past 25 years, Russia has also transitioned rapidly to a fertility pattern like that of most of the developed world. It would be fascinating to project what the rest of these cohorts’ experience will look like.

It’s also interesting to think of what implications these patterns – both the old and the new – carry for Russian women’s lives and family life:

  • what was it like to grow up (or parent/not parent/make childbearing decisions) in a cohort where almost all of the cohort parents are the same age?
  • in recent times, grandmothers have often played a large role in Russian parenting – how much does that change with lengthening generations?
  • I’m also curious how much of the shift toward later childbearing has been influenced by high divorce rates; anecdotally, I know several women who married young, had a child, divorced, remarried in their 30s, and had another child.