As you may be aware, Russia yesterday held its presidential “election” and, as was widely expected, Vladimir Putin was “elected” president for a new six-year term. The share of “voting” won by his United Russia party matches their pre-election goal almost exactly: 70% of the vote on 70% turnout. In addition to the unlikelihood of such an accurate prediction, it takes only the most cursory glance at voting data to recognize the Russian “elections” as the sham they are.
Russia openly states – proudly states – that it is a “managed democracy.” As far as I can tell, this means that the government makes some choices for the people with respect to elections (specifically, closely controlling who can run), but that within these limits, the voting is intended to reflect the actual will of the people. Of course, this is already far from a free election; no legitimate opponent of Putin’s is permitted to run. Some, including the popular former Deputy President Boris Nemtsov, are assassinated for attempting to. Those permitted to run, like the Communist fossils dusted off and placed in front of the electorate, are carefully curated to offer the veneer of a choice, without in any way challenging Putin’s authority or policies.
Today, I’m going to ignore this management portion of the Russian “elections” and focus on the democracy part. I was following reporting closely over the weekend, finding RUElection Observer to be an especially helpful at pointing out strange anomalies. However, when reading the inestimable G. Elliott Morris I was referred to an analysis by Sergey Shpilkin, who was able to show in just a few charts that even the vote counting in Russia is absurd. The analysis is in Russian, but Google translate did an admirable job on it.
The analysis by Shpilkin involves categorizing precincts based on turnout. Normally, we would expect turnout by precinct in most elections to follow a normal-ish distribution, perhaps with some right skew. If you bucketed precincts based on turnout on the x-axis, with the number of votes on the y-axis, the chart would probably look something like this:
Nothing in there looks at all weird, right? (Ignore the low overall turnout for now). Just like many types of statistical samples, from the heights of American females to the points scored in NBA games, most of the values in fall close to the average, with fewer values the further you get away. These numbers could come from a falsified election, but there is no evidence of that here. If you didn’t look carefully, you might be surprised to know that this analysis is from yesterday’s Russian “election” – but with the vote for Putin’s party, United Russia, ignored (hence the low overall turnout of around 25%).
When we add back in United Russia, we get a very different picture:
This distribution is called bi-model, meaning that it has two peaks. Now there are some types of data that have two peaks; for example, if we charted the number of people in a restaurant by hour, we would see a noon peak for lunch and a 7pm peak for dinner. But I can think of so reason why many election precincts would have turnout around 35%, few around 70%, and then many again above 90%. At least, not in a legitimately-conducted election. Note that United Russia’s vote share (green line) is much greater in the precincts where turnout was 80% or higher. Also, note the consistent spikes for turnout at the multiples of 5% towards the higher end – including a lot of votes cast with an improbable 100% turnout. These are massive anomalies, and not one that is consistent with United Russia fairly winning such a share of the vote. If you can give me any legitimate explanation, I’d love to hear it.
If you look in the chart, you’ll see an area shaded in pink. This is Shpilkin’s estimates of the extra votes “won” by United Russia, were the distribution of turnout to be legitimate. Totaled, they represent around 12,000,000 votes – enough to downgrade Putin’s electoral romp to a simple plurality of around 40% of the vote.
Putin didn’t start this way. While his first election, in 2000, was based on stoking cultural divides and generating an existential fear of “others” residing in Russia, the actual vote counting appears to have been mostly legitimate. It was only as he took control of the local government of Russia’s states (called federal subjects) that he was able to move from tactics of fear and voter suppression to outright manipulation. When you hear of those in our own country who would delegitimize our voting process, using bogus claims of voter fraud to implement partisan election procedures, modern Russia serves as a cautionary tale. For those of us who want our elected leaders to be chosen based on the will of the people, the time to fight against Putin-style authoritarianism is before a would-be autocrat gains sufficient power. We have a Republic only if we can keep it.