FOR IMMEDIATE RELEASE
Policy Forum Armenia (PFA) just concluded its preliminary analysis of the official results of Armenia’s 2013 Presidential Election based on the data provided by the Central Election Commission. Consistent with eyewitness accounts and a large body of local media reports, our analysis—which we summarize below—confirms the following:
- Presence of a significant and systemic effort to artificially inflate the voter turnout (Figure 1, Panel 1). The deviations from the expected pattern (i.e., the normal or Gaussian distribution) appear to be larger than those found in our analysis of the 2008 Presidential Election (see Figure 1 in PFA’s Report on “Armenia’s 2008 Presidential Election: Select Issues and Analysis”).
- This turnout-enhancing fraud was conducted on behalf of the incumbent, Serge Sargsyan, and against the main challenger, Raffi Hovannisian (Figure 1, Panels 2 and 3). The deviations from the expected pattern are massive and exceed the range found during the 2008 election (see Figure 2 in PFA’s Report on “Armenia’s 2008 Presidential Election: Select Issues and Analysis”).
- In addition to artificially inflating the voter turnout, votes cast in favor of Raffi Hovannisian and other opposition candidates were fraudulently assigned to Serge Sargsyan: for every 100 ballots cast in polling stations with artificially high voter turnout, Serge Sargsyan received 105 votes (as measured by the slope of the straight line on Figure 2, Panel 1).
- Voter turnout was inflated by ballot stuffing, among other means. As shown on Figure 2, Panel 3, the number of invalid ballots declined asthe voter turnout increased, a clear sign of ballot stuffing. This result contrasts with the findings of the 2012 Parliamentary Election, where ballot stuffing was not found to be a major factor (See PFA’s report on “Armenia’s 2012 Parliamentary Election”).
- There are indications that invalid ballots were counted in favor of the incumbent (Figure 2, Panel 4). The share of votes assigned to Serge Sargsyan is larger in polling stations where the ratio of ballots treated as “invalid” is lower.
- There is strong evidence of vote-counting fraud in polling stations in rural areas. The digit tests (Figure 3) reveal signs of human interference in voter counting (e.g., reassigning of ballots from one candidate to another, etc.) as measured by a significant deviation from the uniform distribution (of votes cast in favor of two front-runners) in polling stations outside of Yerevan and Gyumri.
All in all, we conclude that in polling stations not affected by falsifications, the main opposition candidate, Raffi Hovannisian, shows a lead of at least 4 percent of the vote. This, however, is likely to be the lower bound of the difference between the two front-runners, because of the extent of violations in favor of Serge Sargsyan, as mentioned earlier.
This analysis challenges—in the strongest possible way—the official version of the events, the statement of the OSCE-ODIHR observer mission, and the reports in Western media outlets about the extent of violations that took place on February 18, 2013.
Our results strongly indicate that the final outcome of February 2013 election was subject to massive manipulations and interference and did not reflect the free will of the Armenian citizens.
They also call into question the role of foreign election observers in the Armenian context, specifically their impartiality and ability to detect election fraud that is becoming ever more sophisticated.
We strongly encourage Raffi Hovannisian and his supporters to forcefully pursue either a system-wide recount of ballots (which would undoubtedly reduce the share of the incumbent’s votes and lead to the second round election) or a complete annulment of the election outcome, as the only ways forward for the country and its people.