Sergey Zhuravlev is a Russian economist who runs a wonky but eminently readable and very useful, interesting blog and writes for Expert (author profile), which I may add is an excellent publication. You have met him previously on my blog as the inventor of a clever – if, in my opinion, flawed – argument that the 2011 Duma elections were marred by 5%-6% fraud, but were clean in Moscow; and if you read the Russia blogs, you may also have come across Mark Adomanis’ translation of one his articles about Russian regional inequality. Now I am presenting a translation of his Feb 13 article on what I called as the end of Russia’s demographic crisis: The Reversal of the Russian Cross. In my opinion, it has a few weaknesses; in particular, he is too cavalier about dismissing the “alcohol hypothesis” about post-Soviet Russia’s “supermortality”. But overall it is a brilliant and deeply informative survey of the origins of the Russian Cross – the crossover of the births and deaths graphs in 1992 – as well as of its recent reversal, to the extent that natural population decline is now almost stabilized and the overall population is able to grow due to net migrants.
If we are to go by Rosstat’s figures, in the past year Russia’s population – for the first time in virtually the entire twenty years of Russia’s existence as a sovereign state – increased, exceeding 143 million people. The maximum population size was reached in 1992, at 148.56 millions, and has since decreased at a practically monotone rate. That said, it should be added that small population growth was previously observed in 1994 and 2009, and that the population fall in 2010 was, most likely, explained by cumulative errors over the period since the 2002 Census, and by the abnormal mortality during that summer’s heatwave [AK: There were c.56,000 excess deaths during the anomalous 2010 heatwave, which is basically equivalent to population decline of 48,300. Furthermore, the 2010 Census showed there to be 143.9 million Russians, which was one million higher than projections based on the 2002 Census; this implied that during the period, net immigration was underestimated by more than 100,000 per year. So its likely that even despite the heatwave, Russia's population still eked out an increase in 2010].
Caucasian Mountains only bested by Urals Mountains
The aggregate growth in our country’s permanent population was 165,000 for the past year [AK: This was a preliminary estimate that seems to have discounted December's migration stats; the final figure is population growth of 189,000]. Although overall positive growth is only enabled by migrants – net immigration is estimated at 296,000 for this year – the rate of natural population decrease continued to decline at a fast pace. Whereas in 2005 there were 828,000 more deaths than there were births, this past year it declined to 131,000.
Russia’s population is substantially affected by the effects of migration from the former Soviet Union. In the 22 years after 1990 – the year when ethnic problems in the former USSR exploded – some 7 million people have moved to Russia for permanent residency. This figure is in net terms, accounting for reverse flows from Russia, and discounting temporary labor migrants. Although net population outflow from Russia into countries of the Far Abroad constituted 80,000 annually throughout the 1990′s – in total, 1,050,000 Russians have officially moved into countries of the Far Abroad for permanent residency since 1990 – it has practically ceased from 2006 [AK: The Far Abroad is the world outside the former USSR, minus the Baltics and (recently) Georgia. Note also that Russia's "brain drain" came to a dripping halt at precisely the time when hacks in the Western media began to propagandize it].
Russia hosts the world’s second largest migrant population, after the US; it slightly exceeds Germany in this respect, and doubly so the next five largest migrant centers: Saudi Arabia, Canada, Great Britain, Spain, and France. A third of Russia’s migrant inflow from 1990 to 2010 from the former Soviet bloc accrued to Kazakhstan. But in the noughties Kazakhstan ceded leadership as a source of migrants to Uzbekistan, and after the Orange Revolution Ukraine caught up with them, and Kyrgyzstan after the Tulip Revolution [AK: Zhuravlev has a separate blog post noting that emigration waves typically accompany revolutions in the former Soviet space. I guess its something to look forwards to if the White Ribbon crowd seizes power.]
The only former Soviet republic with which Russia has had a negative migration balance these past 21 years – in which more people left than came in – is Belarus. That said, it should be noted that starting from 2005 the migration balance with Belarus too has turned positive, albeit it remains modest (net immigration from Belarus constitutes less than 8,000 people over the past six years). It is unclear why more people left for Belarus before this date; perhaps because the Russian provinces neighboring Belarus, such as Belarus, aren’t exactly the richest ones. Maybe it was tied to family reunification – parents returning to their children, or Belorussians returning to their homeland, for instance from Komsomol construction projects. Perhaps for this same reason Russia had a net outflow of migrants into Ukraine in the very early 1990′s.
As regards internal migration, the statistics do not reveal any special revelations that could refute or even complement intuition. There are three main destinations for internal migrants: The city of Moscow and Moscow oblast (in the past year the entire agglomeration absorbed 125,000 people, or three quarters of Russia’s population growth), and St.-Petersburg (33,000 migrants in the past year). There is also substantial migration into the Southern Federal District (in significant part from the neighboring North Caucasus) and into the Urals Federal District.
An important caveat is that in the two latter cases, population growth carries an exclusively point-like character. In the Urals Federal District, it is almost entirely concentrated around Tyumen oblast, the richest province in Russia today. Due to the high levels of social support in Tyumen oblast, fertility is also high: Young families get generous housing benefits, there are special programs for families with children. On its part the situation is similar for the Southern Federal District, which grows entirely thanks to Krasnodar krai, which is also understandable: Sochi.
It is clear that Russia’s demographic situation has improved in substantial part on account of the Northern Caucasus, where a strengthening baby boom started from about 2005. The other more or less demographically balanced Russian region, experiencing positive natural population growth, is the Urals Federal District thanks in turn to Tyumen. But contrary, perhaps, to popular belief, the Northern Caucasus isn’t the main source of migrants to the Central Federal District. In 2010, the most recent year for which internal migration data is available, only 16,000 people from the North Caucasus got permanent residence in the Center. This is but a drop in the ocean to the 19 million population of the Moscow region.
The biggest “donors” to the Moscow agglomeration are the Center itself and the Volga Federal District. These two regions, which constitute the primordial Russia as it developed in the 16th-17th centuries, experience not only the maximum natural population decrease in Russia but also the maximum mechanical loss of population, which in its turn is getting fairly intensively replaced by migrants from Central Asia (and in Siberia, apparently, from China [AK: Here I disagree with Zhuravlev. While there are significant numbers of Chinese labor migrants and shuttle traders in the Far East, very few of them choose to stay. This is not the case for Central Asians.])
Wartime Losses in Peacetime
Russia’s natural population decrease has declined as a result of a significant improvement in mortality, as well as a modest increase in fertility. The fall in mortality, just as its rise earlier in the 1990′s and early 2000′s, for the most part affected men, and substantially affected their expected life expectancy. From a remarkably low level for a civilized country of 58.9 years six years ago (the minimum was 57.4 years in 1994) it has now improved to 63.6 years. This is still far from a result to write home about, but at least it is now almost equal to the best Soviet-era indicators in the early 1960′s and late 1980′s. As for mortality among under 40′s, which has always been the scourge of Russian men, the current curves are even better than the Soviet ones (granted, the share of men living to 35-40 years is now higher mostly thanks to significantly lower infant and child mortality rates).
The phenomenon of “supermortality” from 1991 to 2009 – the 6.24 million excess deaths in the past 19 years, of which 3.2 million accrue to the 1990′s, that would not have occurred had age-specific mortality rates remained fixed at 1990 levels – has yet, in my opinion, to be endowed with a rational explanation [AK: This is the weakest point of Zhuravlev's essay. Yes, there is a rational and very convincing reason: Alcohol. There is a very close correlation between alcohol consumption and mortality since the late Soviet period, when an anti-alcohol campaign reduced consumption and improved life expectancy, to local peaks in consumption - and mortality - around 1994 and the early 2000's, to the past few years, when mortality reductions have occurred in lockstep with less boozing. There are similar correlations between alcohol consumption and mortality by geography, sex, and socio-economic sex; see the evidence here.]
Despite the hugeness of the number itself. It is equal to or even exceeds the “supermortality” caused by collectivization, is almost an order of magnitude greater than the number of victims of the Great Terror, and has the same order of magnitude as the rear losses of the USSR during the Great Patriotic War.
Falling living standards? This fit the maximum in 1994, but not the second local maximum in 2003, when normality was returning. And on the whole, while living standards fell during the transition period and reattained Soviet levels only in 2003-2005, the depth of the fall was nowhere near deep enough to explain this “supermortality” as during the war years with reasons such as malnutrition, poor sanitation, and the unbearable conditions of mobilized labor. The “supermortality” of the past twenty years carried some war front characteristics: Excess mortality among males from 25 to 44 years of age in percentage terms relative to Soviet norms was maximal, at 57%. As if Russia had a war.
To this day another very popular explanation is the “alcohol hypothesis.” Booze became more accessible, people got more free time on their hands, and parasitism was no longer a jailing offense. It is probable that more accessible spirits, and especially drugs – which were little known in the USSR – played their role. However, during the period, people didn’t start to buy fewer spirits; it remained at a constant 9-10 liters of ethanol per capita annually (the contribution of homemade moonshine is purely evaluative, often they add on about 10 liters of ethanol per capital, but who’s doing the counting?).
Be that as it may, the reduction in external (“non-natural”) causes of death in the past few years was very significant and was visibly faster than the reduction in mortality from all other causes. For instance, if aggregate mortality declined by 2.9% in 2011, for non-natural causes – homicides, suicides, alcohol poisoning – it fell by 9%-17%. Albeit, mortality from traffic accidents did increase by 1.3%.
The causes for this reduction in “non-natural” mortality should probably not be sought for beyond rising living standards. Especially revealing in this context is a comparison between large megapolises, especially Moscow, with the rest of Russia. In the capital, the numbers of murders and suicides, not to even mention alcohol poisonings by counterfeit vodka, are many times lower – by up to five to ten times lower – than in the rest of the country.
In aggregate drunkenness, banditry, the increasing number of auto accidents, and the war in Chechnya explain much less than 100,000 of the annual number of abnormal deaths, which in some years have reached up to 600,000 in the past decade. Furthermore the rise in mortality also affected women, albeit to a lesser extent, for whom the chances of meeting one of the deaths described above are much less characteristic.
The melancholy arising from a career loss is surely an important factor, especially when it comes to people near the end of it. But then its unclear why mortality increases afflicted 25 year old youths; there are cases of suicide even among party and Komsomol activists of this age, even though they fit perfectly into the new capitalist economy.
The mere fact of the demise of the state of “Kuzmich” could hardly have caused such an overpowering depression, as to invoke the desire to end it lethally [AK: Кузьмичи refers to a person who grew up on Soviet kitsch and later became disillusioned by it, but was forced to continue living the lie to retain his power. This cynicism and obscurantism described the Soviet nomenklatura by the 1970's-80's.] To be honest, it was sooner the other way round: They had annoyed everyone by then. One final consideration: We may be dealing with a statistical artifact from Soviet times. It’s well known that to a Soviet economic statistics were just rubbish to a significant extent. Is it possible that similar techniques were applies to mortality statistics, even though its more difficult? [AK: I very much doubt it, not only because falsifying demographic stats is more difficult but because the picture they reveal is damning nonetheless: Stagnant life expectancy (an overall decline for men) and an infant mortality rate that actually, unique among industrialized countries in peacetime, that actually increased under the late Brezhnev period.]
Girls, Ask your Girl Friends
The shifts taking place in fertility were no less interesting. In the 1990′s, the quantity of children per woman younger than 25 years nearly halved. This decrease barely affected older women; however, because it was specifically “youth fertility” that was high in the USSR, the aggregate result was dramatic. The total fertility rate (TFR) – the number of children a woman can expect to have in her lifetime – fell from 1.89 children in 1990 to 1.16 (!) in 1999, which is, of course, very far from level required to assure population replacement. Although the noughties observed an increasing TFR on account of more births among older women – in 2009, the TFR reached 1.54 children – the total “shortfall” of births from the reduction in “youth fertility” during the 1991 to 2009 period consisted of 11.292 people.
Up until 2007, the influence of these changes on the crude birth rate – the numbers of births per 1000 people – was slightly offset by the increase in the numbers of women in their childbearing age.
In the graph below, it is clear that in this indicator, adjusted for changes in age-specific mortality, was actually growing in the 1990′s and the first half of the 2000′s. This is not surprising, as fertility was mostly formulated on account of women born in 1975 or younger, when we had a repeat demographic spurt (an echo of the baby boom of the 1950′s). After 2007, the crude birth rate is starting to be affected by the echo of its own collapse in the 1990′s and by population aging. That is why the birth rate has remained almost flat since that year, despite the number of children per woman markedly increasing. This “echo effect” is going to influence fertility in the coming decade, since women from the small 1990′s cohort will be reaching child-bearing age.
It is difficult to say with certainty what caused this fertility shift towards women of greater age. In the Soviet period, a significant contributory factor to early childbearing was that it was figured as a condition for registration for the provision of housing. Apparently, postponed childbearing was enabled by growing income inequality (as a result of which, women began to take more care in choosing a mate, with economic factors playing a significant role in the process), new opportunities for international migration, or something else.
It’s clear that under the Soviet Union, the presence of kindergartens, schools, the Constitution’s guarantee – which was more or less followed in practice – of free housing constituted significant social supports, which enabled high fertility rates. One can also add that many Soviet cities – maybe, all of them – were developed like a “company town”, with social and housing infrastructures closely tied to the town-forming enterprise. When markets were introduced, and it became clear that nobody wanted so many tractors or so many tanks and the revenues of these enterprises dried up, all this infrastructure were left hanging in thin air. There was nothing left to finance the kindergartens and nurseries, no funds to build housing. And the destruction and uncertainty, of course, also influenced decisions on having children.
The economic stabilization of the 2000′s, and especially the new social support measures introduced in 2006-2007 – maternity capital, credit programs, etc. – launched a “delayed fertility” effect, a shift of births towards older women. In general fertility has matured, albeit one shouldn’t exclude the possibility that further concerted efforts to provide social support for families and children will return TFR back to Soviet levels. In any case, more than half of the movement back is already behind us.