“Subject to caution”, “to be taken with caution”, “probably partial”… Whether it is the level of Russian support for the war or the “brain drain”, it is an understatement to say that in the West, Russian statistics are appreciated with the greatest skepticism. And for good reason: if there is little doubt that a lot of data is well manipulated, there is little or nothing to penetrate the confines of the Russian bureaucracy and its calculation methods. Thanks to the war in Ukraine, many experts have encountered the difficulty of accurately estimating the evolution of the internal situation in Russia, judging the official figures sometimes inflated, sometimes below reality.
In addition to the vagueness that it maintains on the real situation of the aggressor country, the mystery which surrounds the production of Russian statistics is insidious in that it carries its share of fantasies. Should we see in these manipulations the sole arm of an authoritarian and ideological power, giving orders to falsify any compromising data, at the risk – if the deception is discovered – of increasing the distrust of a part of public opinion, and even make decisions that are counterproductive for your country? “The unity of federal power and its control over the rest of the state are largely overestimated, says Yoshiko Herrera, professor of political science at the University of Wisconsin Madison. Sometimes, Putin’s government can actually act at the source to demand that figures do not come out, or that they are favorable to him. But it is not a question of a well-oiled machine which would simply give the general order to manipulate the figures in its favor.”
Electoral pressure
It is 2020. The Covid-19 pandemic has forced the government to suspend the national referendum – initially planned for April – aimed at authorizing new presidential mandates for Vladimir Putin. A thorn in the side of the manager, who risks suffering from excess mortality due to Covid. In other words, it would be good for the statistics not to appear too catastrophic, in order to reassure the population and thus resume the course of political life… Result: numerous research studies have demonstrated that the mortality figures due to Covid have were indeed falsified in Russia (where the referendum was finally rescheduled for June 2020).
Researchers have, however, shown that despite appearances, “the falsification of Covid death figures” cannot be explained by the authoritarianism of the regime alone. In their study entitled “Encouraged to Cheat? Federal Incentives and Career Concerns at the Sub-national Level as Determinants of Under-Reporting of COVID-19 Mortality in Russia“, published in the British Journal of Political Sciencethe researchers point to a “complex interaction between bureaucratic and political actors to understand the propagation of the logic of manipulation”.
Indeed, in regions where elections were to be held within two years (between 2020 and 2022), reported mortality due to Covid-19 was minimized by 61% on average. When elections were held a year earlier, it was reduced by only 23%. Clearly, the under-reporting of mortality due to Covid-19 was correlated with potential political risks.
And for good reason: on paper, regional governors hold their position for a period of five years, and can run again thereafter. But in practice, the federal government largely determines whether a governor will stay in office or not. Therefore, governors have every interest in showing their loyalty to federal power towards the end of their mandate if they wish to remain in office.
Performance
The hand of a governor loyal to the central power, however, is not enough to influence the quality of the data produced in the various bureaucracies of his region. “The manipulations come from people whose activity and performance are evaluated using statistical indicators,” summarizes Andrei Yakovlev, former professor at the Higher School of Economics (HSE) and who now works as a researcher at the University of Harvard. But in Russia, this criterion applies to many people. Starting with the bureaucrats who participate in the statistics production chain, but also the senior civil servants who, if they hope to keep their position or benefit from a promotion, have every interest in producing data satisfactory to the Kremlin. “Central power is faced with a paradox: the government needs reliable data, but the fear of inspections or punishments still pushes the regions and the bureaucracy to distort the data,” continues the professor.
Falsified death certificates
“During the Covid-19 pandemic, bureaucrats and doctors alike changed death certificates, replacing “Covid” with “heart disease” or “diabetes”,” explains demographer Alexey Raksha. For daring to express his doubts on Facebook and in the international media about the reliability of official data concerning Covid, he was fired from his job within Rosstat, the public statistics agency in Russia.
According to Alexey Raksha’s calculations, even population census data for many regions are overestimated, for some of them to a large extent. “The regions obtain a large part of their budgets from the federal state on the basis of per capita funding. By inflating their figures, they obtain better ones.” According to the expert, these manipulations, when combined, would lead to an overall overestimate of the Russian population of… “5 or 6 million”.
Police officers, too, are evaluated based on key performance indicators (KPIs), such as the number of crimes recorded and resolved. The more cases a police officer sends to court and the more successful prosecutions are initiated, the more positive his or her record is. “Police officers are caught between the prospect of promotion if they arrest enough people, and the anticipated risk of being punished if they don’t do enough. This pushes them to seek to increase their numbers. One of the most famous ways to improve the police officer’s record is to place on a person the exact amount of herbal drug that could send him to prison,” explains Arnold Khachaturov, head of the data department of the Novaya Gazeta Europein charge of the “To be Precise” project, which collects statistics on social problems in Russia and monitors the openness of state data.
May decrees
Should we see in this “statistical dictatorship” the hallmark of Putin’s methods, knowing that Russia has a heavy tradition in terms of data manipulation? Under the USSR, the government sought to reduce infant mortality. To do this, the Soviet authorities had changed their methodology: babies less than seven months gestation, weighing less than one kilo and measuring less than thirty-five centimeters, and who died in the first week of their birth, were not not considered “live births” and were therefore not included in infant mortality statistics. “Under Putin, it’s the same system. They don’t completely lie. But they find ways to make people believe that the situation is not so bad,” analyzes Yoshiko Herrera.
But these methods gained new impetus under the Putin era, after the promulgation of the “May decrees” in 2012, the objective of which was to provide Russians by 2018 with targets for improving the standard of living on diverse topics. This involved, for example, reducing the mortality rate for tuberculosis, cardiovascular and oncological diseases, or even increasing the salaries of teachers. “Most of these goals were at best untenable and at worst incomprehensible,” recalls social scientist Mariia Vasilevskaia, former head of the data analysis team at the Center for Advanced Governance, and now founder of the Hannah Arendt Research Center. This was already somewhat the case before, but from these decrees, the bureaucrats began to use statistics as an instrument to measure the achievement of these objectives. So we had to find a way to show results to the central power.
One example among many: “In order to comply with the decrees of May 2012, many schools reduced their staff and increased the workload of the remaining employees,” writes the Expert Center of the Russian Confederation of Labor in one of his reports. Consequence: the salaries of the remaining teachers were higher, while in reality they worked more.
“Twenty years late”
A final factor is likely to modify the quality of the data. “Bureaucratic practices in data science (data collection, storage, analysis) are twenty years behind the times in Russia,” explains Mariia Vasilevskaia. “Bureaucrats are poorly paid and poorly trained, so those who falsify the data do not act always on purpose.”
It’s hard to imagine, however, clumsiness or lack of training when data is simply removed from public view, such as the number of Russian soldiers who died on the Ukrainian front… For the researcher, “most bureaucrats adhere to a sort of culture valuing secrecy. The less they publish, the better, because it means fewer questions, therefore fewer potential problems. This culture of secrecy is such that even ministries are not necessarily aware that the figures are misleading or data is missing.
Is this really what matters to Vladimir Putin? “If it were a normal government, we could say that the lack of reliable information would be really problematic. But in the case of Russia, judges Yoshiko Herrera, the power does not care. Just as it does It doesn’t matter how many soldiers die on the Ukrainian front.