Who Won the Disputed Iran Election?

The June 12th anniversary of the disputed (or rigged?) election in Iran is looming, and the debate over the legitimacy of the government of the Islamic Republic hasn’t quieted a bit. Recently, there was a heated debate on the Huffington Post between Omid Memarian, a long time ally to those covering the Green Movement, and Flynt and Hillary Leverett, long time deniers of the “rigged election” theory and downplayers of the Green Movement. The articles were about the strength and relevancy of the opposition movement in Iran, but the side debate was also focused on the opposing claims about the election results.  Both sides believe that the truth is misrepresented in the major media, and the blogosphere, and on Twitter, and by the politicians.

So this is the first in a series of articles reviewing the last year in post-election Iran. This article is designed to be a comprehensive coverage of the evidence and arguments about the validity of the election results.

In the weeks leading up to the election, Ahmadinejad seemed to have high poll rankings, but his support seemed to be rapidly eroding as election day loomed.

Between May 11th and 20th, a Terror Free Tomorrow poll showed a 2-1 margin of support for Ahmadinejad. In the poll, 34%  of respondents said they would vote for Ahmadinejad, 14% favored Mousavi, 2% favored Karroubi, 1% favored Rezaee and 27% were undecided.  However, of the “1,731 people contacted, well over half either refused to participate (42.2%) or did not indicate a preferred candidate (15.6%).”¹ Clearly, despite a 2-1 lead, Ahmadinejad had less than a majority needed to win outright. Because election cycles in the world outside the U.S. are short, a few weeks (by law, election campaigns can only start 30 days before the election in Iran),  large shifts are possible in shorter periods of time. With three weeks left, the campaign just starting up, and the vast majority of those surveyed still in play., it was clear that the race was going to be an interesting one.

The Terror Free Tomorrow poll  is often cited as the most scientific pre-election poll, as it was conducted by an independent organization outside Iran. As a result, it is often used by pundits who believe that the results of June’s election were valid. But that raises as many problems as it solves. First of all, the same poll, only a month earlier, showed that a runoff election was likely. Also, many of the opinions of this poll did not match up with the platform of Ahmadinejad or the hardliners.  First of all, 90% of the respondents ranked the economy as the most important priority. This stunning number was the result of an economy in complete disrepair, an economy that Ahmadinejad was elected to fix only four years earlier. The poll also showed that 4 out of 5 Iranians wanted to choose the Supreme Leader directly.  Iranians in the poll ranked “free elections and a free press as their most important priorities for their government (behind the economy),”  and over 70% wanted to cooperate with weapons inspectors and stop development of nuclear materials, and over 77% wanted to normalize relations with the United States.  Ahmadinejad, on the other hand, an avowed holocaust denier, has vowed that Israel, the “Zionist regime,” would be “wiped off the map,” he has proclaimed that the ”‘satanic power’ of the United States faced destruction,” and he has repeatedly blocked weapons inspections while defending Iran’s “peaceful” nuclear program. He is also, and obviously, far from a freedom of the press advocate, and he has defended the governmental structure of the Islamic Republic. In other words, the poll that definitively shows Ahmadinejad as in the lead, according to defenders of the election, also shows that more than 70% of Iranians disagree with major segments, perhaps even the core, of the Ahmadinejad platform.

In fact, as the poll itself states:

‘ A close examination of our survey results reveals that the race may actually be closer than a first look at the numbers would indicate. More than 60 percent of those who state they don’t know who they will vote for in the Presidential elections reflect individuals who favor political reform and change in the current system.’²

According to this leading indicator, if Ahmadinejad received ALL of the other undecided votes, he still wouldn’t get to 50%, the minimum needed to avoid a run-off.

One week into the campaign, with more than three weeks left of the campaign, and with 50% of the voters in the poll still not backing a candidate, is it really that hard to believe that Ahmadinejad could lose? Mousavi didn’t think so.³

Mousavi’s and Karroubi’s campaign posters in Tehran claimed that a high turnout would reduce Ahmadinejad’s chance of winning the election.^[46] Karroubi’s campaign manager, Gholamhossein Karbaschi, claimed that the chance of Ahmadinejad losing the election would be over 65 percent if over 32 million people voted, but less than 35 percent if less than 27 million people voted.^[46]

Within the next three weeks, many inside the regime were afraid that the hardliners were losing their grip on the populace. Polls that initially showed President Ahmadinejad leading by a landslide had shifted dramatically.  In fact, outside analysis of Iranian polls, combined with statistical analysis of past election results, was enough to convince Five-Thirty-Eight’s Renard Sexton that Mousavi was a serious contender, but probably not in the first round, as a run-off was almost guaranteed.

It seems that the government of Iran also knew this. On June 6th, a secret government poll commissioned by the IRGC (Iranian Revolutionary Guard Corps) was leaked to NEWSWEEK .  It showed that Ahmadinejad’s main opponent, Mir Hossein Mousavi, had the support of 16 million to 18 million Iranian voters, compared to 6-8 million who were planning on voting for the hardline incumbent. This report even showed large numbers of the IRGC itself were planning on voting for change, as frustration with Ahmadinejad’s “erratic” foreign policy and domestic leadership was damaging Iran’s economy and security abroad.

This certainly was not the only poll that showed Ahmadinejad was in serious trouble. Polls in Iran are generally considered to be a mess, and a leader of major polling organization was arrested and the organization disbanded in 2002. Most polls in Iran use small sample sizes, and often rely on slanted audiences to draw data from (small geographical areas, students, government employees, ect.) However, many polls showed Mousavi ahead, and according to the Tehran Bureau (Frontline PBS)  as many as 40 independent polls showed Mousavi winning by a 2-1 margin in the week before the election. Obviously, there were many other polls that showed Ahmadinejad winning, some by slim and some by huge margins as well. So what can we tell from all of this?

Clearly, the only thing clear about these polls is how unclear they all are.  However, the only thing that is clear is that it was certainly well within the realm of possibility that Ahmadinejad could lose.

In the lead-up to the election, the allegations of fraud were already starting up. Mousavi “accused influential Ahmadinejad supporters of handing out cash bonuses and food, increasing wages, printing millions of extra ballots and other acts in the run-up to the vote.” Hossein Shariatmadari, adviser to Ayatollah Ali Khamenei, responded by accusing Mousavi of treason. Ahead of the election, websites (including Facebook) were blocked.  There were also alleged leaks from inside the government that said the regime had been planning fraud for “weeks.”

Then the election occurred, a process that was far filled with controversy.

“Initially there were sporadic reports that opposition observers were barred from entering some voting stations. Officials from the Mousavi campaign also alleged that a number of stations in the northwest and south had run out of ballots. There were further complaints that many voting stations did not comply with the order issued by the Interior Ministry to extend voting hours.“

The next sign that things weren’t going as expected was that the election results were announced on the same day as the polls closed. Within one hour of the polls closing, Fars News (funded by the IRGC) announced Ahmadinejad was the winner. They even knew by what percentage. Despite the fact that the 2009 election had the largest turnout in the Islamic Republic of Iran’s history, in previous elections the results took days to tally.  In fact, election rules in Iran state that the election results should not be made official by the Election Commission for three days, allowing for the examination of irregularities. Instead, the results were announced within hours, and Supreme Leader Khamenei certified the results.⁴ Also, in at least 50 cities (the government number, and perhaps in as many as 170 cities, though it is unclear what a “city” is considered in these statements) there was more than 100% voter turnout.  Some have suggested that this is because Iranians can vote at any polling station, and the voter turnout at each precinct is determined by where you live (the argument is that if you voted near where you worked, not where you lived, it could swing that number).  However, Friday is not a work day in the Muslim world, and it would take a lot of migration (3 million votes) to create so many areas with more than 100% turnout.  The government of Iran even admitted that there may have been 3 million votes in question, but they claimed that this is not enough to effect the overall results. Sure it isn’t.⁵

Suspicious, but what about the actual numbers? Besides flaws in the process, the handling of the post election turmoil, the debate around the polls, and the widespread allegations of fraud (including allegedly leaked statements, documents, and other evidence from inside the government), the main evidence that the election may have been fixed stems from analysis of the vote count via two methods: historical, and statistical.

In 2005, Iran elected Ahmadinejad, an event widely considered a referendum to George Bush’s “Axis of Evil” speech.  Previous to this election, Iran had a reformist president, Khatami, so it is hardly beyond the scope of possibilities that a reformist could return to the Presidency, especially after Bush was replaced by Obama as the U.S. president. In fact, as soon as 2006 the pendulum appeared to be swinging away from Ahmadinejad and towards more moderate reformists.

Many other historical trends were bucked by the official 2009 election results. During the 2005 campaign, Mehdi Karroubi managed to win 5 million votes. Somehow, in 2009, despite all of the press, exposure, and buzz surrounding the rise of reform candidates, the official numbers only give Karroubi about 300,000. He only manages to win 3% of the votes in his hometown, Aligudarz, where Ahmadinejad wins 15 times more (despite Karroubi having beaten him in 2005 by a margin of 6-1). In fact, as the chart below shows, despite being highly competitive in the first round of elections in 2005, only falling 2% short of moving on to the second round, and despite historical trends that point towards Karroubi picking up additional votes, he was decisively crushed.⁶

This evidence, combined with statistical analysis by one of the world’s most accomplished pollsters that shows Ahmadinejad’s margin of victory far outside of the predicted deviation (given the turnout), point towards some suspicious election results in 2009.

Highly suspicious. In fact, no matter where the results came in from, Ahmadinejad seems to hold a commanding and evenly distributed lead, regardless of time or location. In past elections, results trickled in over a matter of days and were reported by regional districts. In 2009, the results came in waves of total vote counts, and each wave kept the exact same proportional lead for Ahmadinejad.

In the chart below, compiled based on the data released by the Ministry and announced by Iran’s national television, a perfect linear relation between the votes received by the President and Mir Hossein Mousavi has been maintained, and the President’s vote is always half of the President’s. The vertical axis (y) shows Mr. Mousavi’s votes, and the horizontal (x) the President’s. R^2 shows the correlation coefficient: the closer it is to 1.0, the more perfect is the fit, and it is 0.9995, as close to 1.0 as possible for any type of data.

Perhaps it wasn’t necessary to report  from each district because this year, unlike the wildly distributed areas of support shown in every other election in Iran, all areas voted almost exactly the same.  As Dr. Juan Cole, from the University of Michigan, puts it:

It is claimed that Ahmadinejad won the city of Tabriz with 57%. His main opponent, Mir Hossein Mousavi, is an Azeri from Azerbaijan province, of which Tabriz is the capital. Mousavi, according to such polls as exist in Iran and widespread anecdotal evidence, did better in cities and is popular in Azerbaijan. Certainly, his rallies there were very well attended. So for an Azeri urban center to go so heavily for Ahmadinejad just makes no sense. In past elections, Azeris voted disproportionately for even minor presidential candidates who hailed from that province.

So, according to this logic Mahmoud Ahmadinejad, despite a dismal economy (featuring inflation as high as 25%, unemployment over 11%, and 25% of bank loans defaulting), despite international sanction, despite religious and ethnic tension, despite the loss of political momentum in 2006, despite further political losses in 2008, and despite all historical precedent and voting trends, managed to rally all corners of his nation to vote for him as he crushed his opponents  by a nearly 2-1 margin. He’s an impressive guy.

History isn’t the only way to judge election results. Luckily, we have two other forces at work; statistics, and the habits of liars.

People are VERY bad at lying about numbers. The logic goes like this:

If one were to just look at the last two digits of a number, in a random event (like an unrigged election), the frequency of the amount of times “1″ appears should be about equal to the frequency that “2″ appears, and so on. If a specific number keeps showing up in unusual frequencies (for instance, if 50% of the last two digits was a “5″) then this would be a sure sign that something was drastically wrong.  By analyzing mock, unrigged elections, to develop a formula, and testing this formula on non-contested election results, one can easily determine the likelihood of a particular outcome being rigged.

Apparently, when rigging elections it becomes very difficult to generate totally random numbers. Also, for psychological reasons, humans have a hard time generating non-adjacent numbers (for instance, numbers like 54 would come up more often then 64). Knowing these two different issues, a study was done on the official election results.  The findings of this approach (published in the Washington Post) are very compelling:

Each of these two tests provides strong evidence that the numbers released by Iran’s Ministry of the Interior were manipulated. But taken together, they leave very little room for reasonable doubt. The probability that a fair election would produce both too few non-adjacent digits and the suspicious deviations in last-digit frequencies described earlier is less than .005. In other words, a bet that the numbers are clean is a one in two-hundred long shot.

Recap

So, the polls tell us with little certainty who was in the lead, but they tell us with a great deal of certainty that the people of Iran were ready for a tight election. History shows us that there were many abnormalities with the results, and statistics show us that there is only a .5% chance that the results weren’t rigged.  At this point, I could show you a bunch of unverified documents (possible forgeries, but supposedly leaked by a young worker from the Ministry of Interior) from ministry officials talking about fraud⁷ or I could show you official ballots, all voting for Ahmadinejad, and all written in the same handwriting⁸ but, instead, I think a few things are already clear from the evidence I have provided above.  Anyone who tells you that they are positive the election results from June 12, 2009, are the will of the Iranian people is passing off, with certainty, something which is incredibly uncertain. There were obvious flaws and deviations in this election, there were many historical trends that were not followed, and there were a great deal of “fraud alert” alarms that have gone off through the analysis of the results.

Dissenting Viewpoints

While careful analysis can never definitively prove  whether or not the vote on June 12th was rigged, careful analysis also fails to produce a single piece of evidence that would even strongly suggest that the  election was fraud free. And yet, only three days after the election, when the smoke had not even begun to settle, long time pro-regime apologists, Hillary and Flynt Leverett wrote a piece for Politico entitled “Ahmadinejad Won, Get Over It,” which begins with a humble assertion:

Without any evidence, many U.S. politicians and “Iran experts” have dismissed Iranian President Mahmoud Ahmadinejad’s reelection Friday, with 62.6 percent of the vote, as fraud.

The Leveretts then proceed to list 1300 words that assert, without a single piece of evidence but with plenty of spirited condemnation of “Iran experts,” that the election results were completely legitimate. Such certainty, in the face of so many questions, is highly suspect.

Propaganda.

So imagine how happy the Leveretts must have been when the ultimate vindication of the validity of the election results, the World Public Opinion Poll (run by the University of Maryland) was released in September 2009. The poll asked Iranians their opinions on America, Obama, their own government, and the June election. According to the poll, 8 in 10 Iranians consider Ahmadinejad to be the legitimate leader in Iran. Now the defenders of the election results had a real piece of evidence to back up their half-cocked argument.

While writing this evidence review, I thought that this poll would be the hardest piece of evidence to refute (in part because all of the other evidence seemed so easy to analyze). 8 in 10… that’s a hard number. Many scholars have argued that after 2-3 months of heavily repressive tactics, arrests, and executions, the idea that Americans could cold-call Iran and get people to speak out against the “legitimacy” of their own government was a preposterously flawed concept. I agree.  Looking deeper into the actual survey, however, I was surprised at two findings: A) the poll did not say what the Leverettts and their ilk believed it said, and B) the contradictions between this poll and the other reliable polling data, as well as internal contradictions inside the poll itself, show that the poll is significantly less reliable than many make it out to be.

First of all, I turn to a paragraph in the poll itself:

Naturally a question that arises is whether respondents are freely speaking their minds in sucha poll, especially when the Iranian government has been recently cracking down on dissent. As discussed below, the fact that one in four respondents refused to answer the question about who they voted for in the presidential election suggests that some people may have felt uncomfortable answering and thus the findings need to be viewed with caution and not as a clear indication of how people voted. Some questions for which we have trendline data also show a bit less readiness to take controversial positions in the current poll.

Indeed. The next paragraph was a gently argued defense of the legitimacy of their own data, but the pollsters themselves attempt to distance themselves from the “smoking gun” association that some have tried to force upon the survey.

In fact, 52% of the people called refused to respond to an American polling organization.⁹ It seems reasonable that some of these people refused to answer for fear of retribution, and if that is the case their opinions would likely be anti-establishment. It also seems reasonable, then, that if there is a swing in the numbers of those who did respond in order to placate government censors these numbers would hurt, not help, Ahmadinejad. 1-in4 refusing to answer the question of whom the voted for: a further problem for the “Ahmadinejad Won” camp.

From here, it gets really interesting. Only 55% of those responding claim that they voted for Ahmadinejad, a far cry from the 62% that the official results claim. Also, when asked whom they would vote for if the election were to occur again, Ahmadinejad’s numbers drop to 49%. Again, with a quarter of those who did respond refusing to answer these two questions (a dramatically larger percentage than many of the other questions) it is clear that this poll does not verify the Ahmadinejad victory.

Also, if the poll is at all reliable,   then one might look at other indicators buried within the poll. For instance, how satisfied were Iranians with the progress that their nation, their economy, or their own specific circumstances were making? Positive values on these questions would tend to favor an incumbent, while negative or stagnant values historically favor change. Well, Iranians have very unfavorable views towards their own economy. While only 27% thought their personal economic situation had gotten better over four years, 31% thought it had gotten worse and 42% thought it remained the same. When asked about the countries economic situation, 29% said it had gotten better, 45% said it had gotten worse, and 14% said it remained the same. NOT good news for an incumbent.

Beyond that, there are contradictory opinions about Iran’s relationship with the rest of the world, and opinions about the way the Supreme Leader is selected that differ greatly from the Terror Free Tomorrow Poll. In fact, very few opinions line up one poll to the other. Also, Mousavi’s vote tally in the World Public Opinion Poll is dramatically different from many of the other pre-election polls, and much lower than the actual election results. It is also interesting that distrust of the United States runs so high in the poll conducted by Americans, and yet those who also argue that the U.S. is pursuing an imperialist agenda in Iran (Leveretts) maintain that the responses to the poll are truthful and accurate. It seems that the only definitive answers, on the legitimacy of the election and the government of Iran, are answers where dissent could get you hanged outside Evin prison.

The fact is that the World Public Opinion Poll had too high a refusal number, too many discrepancies, and doesn’t align with any of the other data points. And the premise of the poll (Americans calling in and getting accurate responses after months of repression) is highly suspect. Other than that it’s a great poll, and we should believe it.

Conclusion

The overall problem can be rephrased like this: while huge amounts of doubt hang over the election results, while millions of Iranians themselves took to the streets to dispute the results, and while all data points (from polls to the official results) are inconsistent and unreliable, the only writers who have a clear position are those arguing that the election was not rigged. For those of us who rely more on evidence, there are still a few things we can take away from this analysis:

A) Leading indicators, including the economic situation in Iran and the Iranians’ opinions on the economy, pointed to a close election where the incumbent would be in jeapordy.

B) Mousavi’s standing, three weeks out, was well within striking distance of Ahmadinejad.

C) The likelihood of Ahmadinejad winning a majority of votes in the first round of the elections was very slim.

D) Historical trends tell us that the election results were highly anomalous.

E) Statistical analysis of the results show that the election is 200 times more likely to be rigged than to be legitimate.

F) All evidence pointing towards an Ahmadinejad victory is inconclusive.

G) All those arguing that the election results are clearly legitimate, clearly, are not themselves legitimate.

But here is the bottom line that people like the defenders of the election results don’t want you to hear: all Iran elections are rigged. The government of Iran chooses the candidates, controls the media, and suppresses dissenting voices.  Within even a few days of the election, the Green Movement had become bigger than the June election; it had become a referendum on the legitimacy of the Islamic Republic itself, and the place in the world occupied by the educated Iranian youth. It had become about human rights, freedom of speech, the rights of women, and establishing life, liberty, and the pursuit of happiness for a 21st century Iran.

And no rigging of an election can stop that.

1. (It is important to note that 42.2% non-responses is higher than the roughly 30% we see in American surveys. A minimun of 50% is required by most social-scientific publications. [↩]
2. Terror Free Tomorrow [↩]
3. http://en.wikipedia.org/wiki/Iranian_presidential_election,_2009#cite_note-AP_turnout_role-45 [↩]
4. Juan Cole: Stealing the Iranian Election [↩]
5. five-thirty-eight.com [↩]
6. compare Ahmadinejad’s vs. Karroubi’s results in 2005 here [↩]
7. Can anyone use handwriting analysis to verify Sadegh Mahsouli’s signature?

   Salaam Aleikum. Following your concerns regarding the results of the presidential election and per your given discretion to have Dr. Mahmoud Ahmadinejad remain as president during this sensitive juncture. Therefore, everything has been planned in a way that the public announcement will be made in accordance with the interests of the regime and the revolution. All necessary precautions have been taken to deal with any unexpected events of election aftermath and the intense monitoring of all the parties’ leaders as well as the election candidates.
   However, some believe that the real votes counted are as follows:
   Total number of votes: 43,026,078
   Mir Hossein Mousavi: 19,075,623
   Mehdi Karoubi: 13,387,104
   Mahmoud Ahmadinejad: 5,698,417
   Mohsen Rezaee: 3,754,218
   Void: 38,716

   Minister of Interior
   Sadegh Mahsouli

   

   [↩]

8. 
   [↩]
9. again, 30% being the average refusal rate in the U.S.A., and a refusal rate of over 50% is usually enough to throw out a survey [↩]