Paul Goble

Vienna, September 22 – In preparing for what was to have been the 2010 census, the Russian State Statistical Committee (Rosstat) concluded that Russia’s population may now be below 140 million, two million fewer than it has been reporting and a difference that highlights both why a census is so important and why Moscow officials may have postponed this one.
According to a report on the Slon.ru portal, “it is possible that now live in Russia not 142 million people as has been considered all this year but 139.98 million,” a figure Rosstat fixed on but does not yet report “in the process of the preparation for the census,” which will now take place in 2013 (www.slon.ru/articles/137535/).
In an article entitled “The Miracles of Statistics,” Konstantin Gaaze, Ekaterina Chekmareva, and Boris Grozovsky say that the decision to delay the census means that no one in Russia will be in a position to definitively resolve this and other differences until at least 2014 when the results of the 2013 enumeration will become available. 
Russian officials have explained the delay almost entirely in terms of cost, citing budgetary shortfalls because of the current economic crisis. But the three analysts argue that cost may be far from the most important reasons for the delay: after all, they say, the census would cost less than one-tenth of one percent of the government’s annual budget even now.
The two groups most disappointed by the Russian government’s decision to delay the census by three years are regional officials who now face a more difficult time in arguing for greater funds for what they claim are larger populations and the leadership and staff of Rosstat itself.
“We are experiencing the putting off of the census as a personal grief,” one Rosstat employee said. “We had prepared for it, awaited it, and wanted it.” And some at that agency are convinced that the delay will sooner or later cost Vladimir Sokolin, the head of Rosstat who argued passionately for the census, his job. 
But as both Rosstat and the Russian government have pointed out, the preparatory work for the now-delayed census will not be wasted, although the Slon.ru article implies that some of the results of that work may have played a role in the decision of the government to ensure that there will be no census before the next presidential vote.
First, the census officials, working with regional authorities, clarified the borders among various administrative-territorial and municipal bodies. Second, they compiled a list of houses where census takers were to be sent, although this process was anything but accurate Rosstat officials say, with some enumerators put off by barking dogs!
And third, the census takers compared actual residents with officially registered ones in a number of places, a comparison critical because in the 2002 census, officials used the latter rather than the former as the basis for completing the census after the Putin government claimed it had no money to pay for a full-blown enumeration, often finding 10 percent variance.
.Summing up its findings, Rosstat calculated that there are 137.8 million people registered and resident and approximately 2.3 million more registered but not resident (the figure drawn from the 2002 count) for a total of 139.98 million – far less than the 141.9 million that Rosstat put out as the country’s population on July 1.
Rosstat employees warned the Slon.ru journalists that its data are preliminary and suggested that no one should take them too seriously. But another said that “the difference between the data of the census and those registered should not be [as] large” as Rosstat had found in this case.
The enumerations suffer from other problems as well. Valery Elizarov, a demographer at Moscow State University, told Slon.ru that the only “reliable” data concerns the number o births and deaths, but those are far from the only developments that affect the total. The counting of immigrants has improved, but it is still not good. 
The number of immigrants is almost certainly higher than many Rosstat reports say, the three journalists report, but the number of homeless people it assumes is equally likely too high, something that overstates the total population just as undercounts of immigrants understate it. And if the 2002 census is any guide, the numbers from the North Caucasus are very problematic.
On the one hand, there is a tendency to double count people there who are working elsewhere. And on the other, local officials have boosted the numbers of residents they repot as a way of getting more aid from Moscow, something they are probably going to do again, especially in the absence of a more reliable all-Russian count. 
Indeed, the various problems with the 2002 census are so great that Rosstat and other Russian agencies are still trying to cope. In the words of one demographer, “the 2002 census was conducted in a defective way,” raising the question of whether the now delayed 2013 count will represent a genuine improvement.