#6304
#13
Use of Influence, Manipulation in Pskov Oblast Assembly
Elections
Obshchaya Gazeta
30 May 2002
[translation for personal use only]
Article by Boris Vishnevskiy: "Mathematics Is the
Enemy of the Electoral Commission"
The theory of probability pinpoints "administrative resources."
People often suggest that the results of one election or another were misrepresented, and with good reason. After all, there is ample cause for suspicion, for instance, when an incumbent governor or mayor is losing according to public opinion polls, but he ultimately wins with a margin of 2-3 percent--and in the precise precincts where the opposition had no observers. We have seen this happen in many elections, which are increasingly similar, in this era of the "vertical chain of command," to a game with a preprogrammed outcome.
The losers shrug their shoulders and say that "administrative resources" are unbeatable: It is almost pointless to fight them using legal methods. The "resources" in question are not just the use of government-controlled media, the recruitment of groundskeepers to get rid of opponents' campaign materials, and the issuance of instructions to the personnel of schools, health care facilities, and housing offices. There is a much more outrageous (but also more effective) way of using one's official status: direct intervention in the casting of ballots and the tallying of votes. There are criminal penalties for using this method, but the cases rarely get to court because everything is done surreptitiously and there are no witnesses.
Nevertheless, it is possible to find witnesses (or, to be more precise, testimony). This is the 21st century, after all, and we have mathematical statistics. Despite the seeming imperceptibility of "administrative resources," statistics can grab the tail of a "black cat in a dark room"....
Assistant Professor Yevgeniy Mironov from the St. Petersburg Polytechnic Institute (SPbGTU), an expert in this field, analyzed the results of the recent elections to the Pskov Oblast Assembly. Mironov feels he was able to find the "smoking gun" of two characteristic methods of using the notorious "administrative resources."
The first method reduces the number of ballots cast for an "undesirable" candidate. The ballots voters cast for this candidate are deliberately spoiled: Commission members simply add another checkmark to any line "for" another candidate or to the line "against all candidates."
The second method increases the number cast for the "desirable" candidate. It is used when representatives of the electoral commission take the ballot boxes to small populated communities with no permanent election precincts of their own. When they get there, they either fill out the ballots themselves or "humbly" suggest the right choice to the voters (usually elderly). Meanwhile, for some reason, there is never any room for observers in the car when they drive the ballot boxes out to all of the rural communities, and the process is completely under the control of the electoral commission.
It would be extremely difficult to prove all of this in a court of law: It would require the commission members responsible for the fraud to testify against themselves, and the Constitution says they are not obligated to do this. This is the point at which the theory of probability can be of assistance....
"Human behavior, including behavior in elections, is subject to statistical patterns," Mironov explained, "and if the patterns in one district are perceptibly different from patterns in the rest, this usually means that there has been some deliberate intervention in the electoral process in this district. The theory enables us to determine the probability of random deviations. If this probability is negligible, we know we are dealing with the kind of 'random situation' requiring thorough administrative preparations....
"We can begin with the deliberate spoilage of ballots. We know that a certain percentage of the ballots will always have to be invalidated as a result of random circumstances. The percentage, however, has remained remarkably constant in all of the elections of recent years and rarely exceeds 2-3 percent. The percentage of spoiled ballots in the State Duma elections of 1995 and 1996, for example, was virtually the same in both cases (1.9 percent)."
It was easy to judge the probability of this method of using "administrative resources" in the Pskov elections: All of the data are on the Internet. After analyzing them, Mironov found that the proportional number of invalidated ballots in the oblast was 380 per district, which was odd in itself. Further analysis revealed a much more intriguing pattern in one of the districts, the 10th: 845 spoiled ballots, or 6.1 percent of the total. Of course, we could presume that particularly stupid voters live in that district, but before we make that unflattering assumption, we should check the statistics. They tell us that the chances of this kind of event are 1 in 16,000, or less than a hundredth of a percentage point. In other words, the probability that ballots were deliberately spoiled in the 10th district is equivalent to 99.994 percent.
Mironov continued his analysis to learn who benefited from these "miracles." He checked the data on the out-of-precinct ballots in that district. His calculations indicated that percentage of votes for one of the candidates rose at an incredible rate as the number of ballots in mobile ballot boxes increased! In other words, he had much stronger voter support "outside the precincts." In all, this candidate garnered 32.2 percent of the vote. In precincts where the ballots in the mobile ballot boxes represented less than a third of the total, however, he was supported by only 20 percent of the voters. In the places where "mobile" ballots represented more than 60 percent of the total (which seems suspicious in itself), on the other hand, 53 percent of the votes were cast for the "miraculous candidate." What other explanation could there be than the use of "administrative resources"?
The name of the candidate who "beat the numbers" is Vladimir Yanikov. We wanted to know more, of course, about the "people's choice." We learned that the election results were not completely coincidental. Independent candidate Aleksey Bolshakov, who was also running in the 10th district and had been in the lead since the campaign started, had an obvious advantage, which was confirmed by several sociological polls, but Yanikov ultimately defeated him by a margin of about 250 votes (less than 2 percent). Furthermore, Bolshakov's problems had begun at the very start of the campaign (Obshchaya Gazeta reported them in March). First he was taken into custody on suspicion of murder, but they had to release him after a week because the charge was so patently absurd. This, however, was not the end of his "accidental" misfortunes. As soon as he was released from custody, the electoral commission revoked his candidate status. A court reversed this decision five days later. On the last day of the campaign, the electoral commission suddenly disqualified him again, but the court reinstated Bolshakov again on the Saturday before election day. Then Yanikov won the election....
The winner, we learned, was the deputy head of the Pskovskiy Rayon Administration. He was supported in every way possible by that administration--to the point at which the head of the administration took a leave of absence without pay to campaign for his protege personally. He probably obliged his colleague with the mobile ballot box maneuvers. Incidentally, the other tried-and-true method--spoiling ballots--was not overlooked either. What does this tell us about miracles? And about mathematical statistics..., which provided such clear evidence of everything that the analysis of political facts later confirmed?
Yevgeniy Mironov has come up with a seditious plan: Election laws should be amended to accommodate statistical audits for the verification of election results. Any candidate suspecting fraud could request an audit. An audit revealing deviations from the norm would be grounds for a thorough inspection of the ballots. This inspection might reveal that all of the invalidated ballots have checkmarks opposite the names of the "undesirable" candidate and some other candidate, but the ink does not match. This would be grounds for a criminal investigation of suspected fraud.
Current election practices in Russia arouse feelings of futility because of the seeming impossibility of countering "administrative resources," as in the example described here. Is it possible that the existence of a method of detecting violations, even in the absence of witnesses to the crime, will discourage people from breaking the rules? So far, our courts have not used the statistical method, so there is no precedent for this. We all know, however, that the first step is the hardest....
Back to the Top Next Article
- Back to the Top -