Rosstat and Levada are Russophobia’s Bane

The evil Russian Bear. But not a substitute for stats.

The evil Russian Bear. But not a substitute for stats.

Still no economic collapse. Still no anti-Putin bunt. Still no demographic apocalypse. As the years pass by, Russophobe canard after Russophobe trope is relegated to the dust-heap of history, only to rise back out of its grave, zombie-like, whenever Boris Nemtsov pens a brilliant indictment hysterical screed on the failures of Putinism or when the militsiya roughs up a few hundred (unsanctioned) protesters in a Russian city of millions. “Surely,” the Western commentariat says, “the system is rotten, the people hate their Chekist oppressors, and guys like Kasparov and Latynina will soon lead the people’s revolt back to pro-Western democracy?”

Unfortunately for their purveyors, these Manichean narratives mostly rely on anecdote, hearsay and the fluff and snake oil that is more commonly known as “political science”. When one looks at the objective evidence – things like economic and demographic statistics and Russian opinion polls – a rather disquieting picture emerges, for Russian limousine liberals and Western Commissars of Transitionology alike. This picture shows that Russians do more or less like “Putinism”, that liberals are despised when they are not ignored, and that most socio-economic indicators really are improving. True, it would be ridiculous to claim that they constitute a full vindication of the regime. Russia still has many serious problems and Russians frequently complain about the system’s corruption and social injustice. But the hard data from Levada Center (Russia’s Gallup) and Rosstat (state statistics service) does tend to invalidate around 90% of what is written about Russia in the Western press and political science*. The onus is on them to present serious evidence that these two organizations manipulate their figures to serve the Kremlin’s interests. And if they can’t, they’ll continue to rant and rave in big media while I spitefully snipe at them from my little blog and accomplish nothi… anyway, let’s not go there.

It is not my intention in this post to demonstrate the full range of ways in which the Russophobe narrative falls face down faced with the evidence from Rosstat and Levada. Though I’ll give just one or two examples, it is easy to extend them near indefinitely.

Let’s first take a look at Rosstat. Now one of the most prevalent narratives about the failure of Putinism is that Russia’s population is in “free-fall”, a “death spiral” (insert your own appropriately apocalyptic-sounding term)… The government couldn’t care less about the soaring murder rate or the plight of Russia’s children and HIV sufferers… Russian women are voting on their country’s future with their wombs and life expectancy has sunk to unimaginable lows… etc in a similar vein. There’s really no need to cite any examples here – anyone familiar with the Western commentary on Russia (or knows how to Google) can easily find many, many articles with these premises in “respectable” publications.

Yet according to the statistics, this narrative is increasingly obsolete, and sustained only by ever more brazen manipulations and misinterpretations of the data. Just to throw out some figures, from 2000 to 2009: the fertility rate rose from 1.20 children per woman to 1.56; life expectancy rose from 65 years to 69 years; infant mortality fell from 15.3/1000 to 8.1/1000. The rates of death from alcohol poisoning, murder, suicide and accidents have all fallen by around half relative to the early 2000’s. Now this is NOT to say that Russia’s demography is all nice and prim nowadays, nor that all the improvements can be chalked up to Putin’s policies. Death rates amongst middle-aged men remain stratospheric relative to the developed world. And it is not clear to what extent recent falls in mortality were due to better anti-alcohol or healthcare policies, and what share was accounted for by Russians simply beginning to drink less hard booze**. Nonetheless – and unless Rosstat is lying through its teeth – the improvements are real enough and denying them will not make them go away nor cause the “bloody Putin regime” to collapse any time soon.

Ironic that an institution once infamous for its statistical manipulations for the USSR now serves as a weapon against them... in the Western press.

Ironic that an institution once infamous for its statistical manipulations for the USSR now serves as a weapon against them… in the Western press.

The main argument remaining to the Russia pessimists is that Rosstat is simply lying. It is, after all, descended from Goskomstat (its web address,, underlines this), an institution which used to cover up the Soviet figures on infant mortality when they increased in the 1970’s and whose bogus accounting of Soviet economic growth implied that the USSR should have been several times wealthier than America by the time of its collapse. Michael McFaul, in his response to a blog post debunking many of his supposedly “factual” assertions in The Myth of the Authoritarian Modelclaims that “the real experts on this stuff (which I am not) have become very suspicious of goskomstat’s work of late”. Funnily, as if in anticipation, Rosstat makes sure to proclaim the exact opposite on its front page: “International expert examinations confirm that the data of the Federal State Statistics Service are reliable.” I guess everyone is susceptible to appeal to (unsourced) authority when their integrity is at question! 😉 So who’s right?

To be absolutely honest, there is no real way to find out (unless official stats are grossly out of sync with perceived reality as in the late USSR, but that cannot be said for today’s Russia). Let me try to explain. In general, only national statistics services have the manpower and regulatory resources to compile comprehensive demographic (economic, etc) statistics on their own countries. The stats you see from international institutions like the World Health Organization or the World Bank are mostly drawn and aggregated from national statistics services. We just have to take them at their word. The only exceptions are when the countries they operate in are so chaotic (Somalia) or closed (North Korea) that their stats cannot be relied upon, in which case multinational organizations try to come up with their own guesstimates (with the emphasis on the “guess” part). Russia is not one of these exceptions. International institutions do use Rosstat’s figures. Heck, guys like McFaul and Nemtsov use them, even though they cherry-pick them wildly to make their ideological points.

Furthermore, it is not entirely clear who will benefit from expending massive stats to subvert Rosstat. Cui bono? Certainly no private interests I can think of. While Putin or his circle may wish to “pad” some bad stats, this would be a very risky endevour. It explode in their faces (analyses from outside expert observers, revelations from whistle-blowers, etc) – and even if they can keep up the deception in the long run, the cessation of reliable information on the country will severely hurt the strategic vision of the leadership as happened in the late USSR. So given all the arguments for Kremlin non-interference, and in the in the absence of convincing evidence to the contrary***, we must assume Rosstat reliable.

Now let’s go over to Levada Center and a couple more examples. Though I know they have their limitations, I am a big fan of opinion polls. Why listen to the inane ramblings of self-important political scientists from their comfortable armchairs, when one can listed to the voice of the people directly? The Western chauvinists have one compelling reason to stick to the former, of course. What Russians say is deeply discomfiting to their worldview, in which Western values are held to be some kind of universal religion. For what Russians say goes far beyond expressing stratospheric approval ratings for Putin (at least that can be “explained” by the pro-Kremlin “propaganda” on state TV or Russians’ “traditional” preference for a strongman at the helm). But “explaining” the following is much harder for them:

1) The Internet is no more censored in Russia than in the West (which is to say very little), and the latest figures show penetration in Russia steadily creeping up to encompass more than a third of the population, which implies near universal access amongst groups like educated, urbane Muscovites. So one would presumably expect most Putinistas to be old, sour-mouthed “sovoks”, right? (As per classist, Russophobe thinking). Wrong. Support for the Kremlin – and disillusionment with the West – runs highest amongst young, university-educated Muscovite men, the very segment of the Russian population that is most exposed to the West through the the Internet and foreign travel! (Hell hath no fury like a Westerner scorned…)

Though the dinosaurs in the MVD may temporarily confiscate Nemtsov’s scribblings on how Putin is really, really bad, they could be freely accessed in cyberspace throughout the whole affair. Apparently, his works simply do not make much of an impact on their own (de)merits! All said, it is hard to see the merits of the Western chauvinist argument that Russians would reject Putinism if only they could discern the beacons of freedom beyond their borders… No. Said beacons already caused a Russian housefire in the 1990’s, and they have no desire to repeat the experiment.

2) Another cornerstone of the Russophobe narrative is that under Putin, elections have become so fraudulent that they have completely decoupled from reality. The corollary is that the regime no longer has democratic legitimacy. Now I’m certainly not one to deny that the Kremlin doesn’t make ample use of its “administrative resources” to slant election results to its liking, both formally (e.g. stricter registration requirements, unequal TV access) and informally (e.g. state employer pressure to vote for the party of power). I am also not denying that in a few regions, like Chechnya, elections really are risible and entirely meaningless. Yet is there really this huge black hole between public sentiment and the ballot count?

Well, we could actually take the unimaginably revolutionary and incomprehensibly convoluted extremely obvious and logical step of actually asking Russians whom they intend to vote for and whom they actually voted for, and compare it with the election results. In fact that is what Levada did for the 2008 Presidential elections:

Medvedev Zyuganov Zhirinovsky Bogdanov
Voting Intentions 80 11 9 <1
Voting Reminiscences 71 20 7 1
Election Results 71 18 10 1

[Medvedev is Putin’s anointed successor and of United Russia, the pro-Kremlin party of power; Zyuganov is of the Communist Party of the RF; Zhirinovsky is of the (fake) nationalist Liberal Democratic Party; Bogdanov was the token “liberal”].

Based on the above, it is fair to say that Russians got whom they wanted in the Presidency. The March 2 election results match both the February voter intentions and voter reminiscences some two weeks later. While one can certainly question the amount of real choice Russians got to exercise in what was a managed succession, it was hardly foisted on them by the jackboot.

3) Last but not least, most Russians themselves think they live in a free country and a democracy. Political scientists may disparage them for it, claiming that Russians don’t understand what democracy is all about. This misses the point. Democracy is more than just free, fair elections and some civil rights. Above all, it needs popular support for its long-term survival. Without that, the political scientists can (and do) go to hell.

Quite an indictment of most Russia commentary in the press today, wouldn’t you say? The Russophobes have two responses to this. First, as with Rosstat, they claim that “Levada’s institute is no longer fully reliable”**** (remember that getting results that can be construed as being pro-Kremlin disqualifies you from being “reliable” almost by definition). This is really laughable. I mean the director of Levada Center, Lev Gudkov, writes things like this:

… Putinism – is a system of decentralized use of the institutional instruments of coercion, preserved in the power ministries as relics of the totalitarian regime, and hijacked by the powers that be for the fulfillment of their private, clan-group interests. The regime is unstable, with questionable chances of long-term survival or peaceful transferal of power.

Yes, Gudkov sure sounds like a raging Russophile maniac skeptical sociologist with no particular love for the Kremlin!*****

The second critique is downright loony, and is never made by even halfway serious Russia watchers. They say that Russians are too afraid to answer opinion pollsters truthfully or reveal their real feelings towards Putin. There’s really no way to argue with such people. To them, if Russians say things are bad in Russia then they are bad, and if they say things are good in Russia then they are either paid shills or trembling slaves of the Kremlin. It’s a closed loop, unfalsifiable, fallacy.

There are three main conclusions to be made. First, the “moderates” in the Russia debate (Stephen Cohen, Ben Aris, “The Mutual Admiration Society“, etc) can rest assured that they’re on the right track. Second, the (extreme) Russophiles and Sovietophiles shouldn’t rejoice. The polls indicate continued low trust in most institutions, unsatisfactory access to healthcare and education and a very corrupt bureaucracy. Likewise, despite recent improvements, Russia’s demographic situation remains very unsatisfactory: middle-aged Russian men still have the life expectancy of their late Tsarist forefathers! Third, the (extreme) Russophobes would be wise to reconsider most of their positions in a fundamental way, because as it stands they are wrong on almost everything. Unless they are really, really good at digging up dirt on national statistics agencies and opinion pollsters, in which case they should get to work on “exposing” Rosstat and Levada!

* For a standard statement of the “Russophobe” position by which the Western mainstream media perceives Russia, see McFaul and Stoner-Weiss on The Myth of the Authoritarian Model. Their mendacity and cherry-picking is exposed by Fedia Kriukov here.

** Russia’s life expectancy is tightly coupled with per capita alcohol consumption. For more info see here.

*** The one serious criticism of Rosstat’s reliability that I’ve encountered in my Russia-watching career was made by Russian economist Gregory Khanin in “Economic growth and the mobilization model” in Ellman’s Russia’s Oil and Natural Gas: Bonanza or Curse? He argues that Rosstat’s methodology caused GDP growth to be overstated by 3% points from 1999-2003, most egregiously during the first two years. His alternate figures do not appear very rigorous. They are derived from analyzing concurrent growth rates of physical proxies like freight transport and fuel consumption, constructing three (widely differing) alternate GDP series based on said proxies, and averaging them to arrive at one alternate GDP series. There is little in the way of explanation why this is the logical and correct course to follow.

Incidentally, Khanin’s own methodology appears to be very similar to several (mostly) American attempts over the years to “prove” that China is *not* growing at 10% per annum by pointing to (occasional) dips in its electricity consumption… I would also like to add that not even the US is immune from suspicions that it is fiddling its numbers:

… Since the time of Reagan the definition of inflation used by the government was being continuously reworked to make the figures appear better than they otherwise would have been, using substitutions and hedonics to spruce up the figures (i.e. adjusting for consumers switching to other products when similar products become expensive, and trying to put values on quality improvements). If the BEA… continued using its old standards, then a) the economy would have been in stagnation during the 1990′s and recession in the 2000′s, b) inflation would have been steadily increasing to a peak of nearly 14% in 2007 and c) median incomes would have been in steep decline.

So there you go. I haven’t studied the issue in detail, and I don’t know whether it is the establishment statistics services or their contrarian critics who are on the ball. As usual, I suspect the truth is somewhere in between.

**** The political scientist who made this claim also recommended The Forensics of Election Fraud by Ordeshook, Myagkov, and Shakin. Anyone know if it has anything convincing or of interest?

***** Actually the Levada-Center is an independent offshoot of VTSIOM, which was brought under Kremlin influence around 2003. This caused most of VTSIOM’s sociologists to migrate with Yuri Levada and Lev Gudkov to the new outfit. There may be grounds to consider VTSIOM’s results suspect, but again there is no hard evidence to support this. For instance, its conclusions that most post-Soviet countries actually quite like Russia are the same as those produced by Gallup. And I certainly hope no-one will now try claiming that Gallup is controlled by the Kremlin! 😉

If you like the words I write, and want me to write more of them, consider donating or supporting me on Patreon.