- Research Programs
- Regions & Topics
- All Publications
A nation must think before it acts.
In April 2014, shortly after the first round of the Afghan Presidential election, I wrote an FPRI E-Note publicizing the findings of two pre-election polls that my company Glevum Associates had undertaken in Afghanistan, funded by a grant provided by the Department of State. I assessed that Abdullah and Ghani would be the clear winners of the first round of the Presidential election but that neither of them would be able to secure an absolute majority. As a result a second ballot would be necessary. As predicted, that ballot will be held this coming Saturday, June 14th 2014 with Abdullah and Ghani contesting the outcome.
Based on these two face-to-face polls, conducted in December 2013 and January 2014, I had estimated that Abdullah and Ghani would each secure between 39 and 44 percent of the vote. I expected Ghani to secure slightly more votes than Abdullah in the first ballot and that the second ballot would be Ghani’s to lose. Not least, because I expected him to secure the majority of the Pashtun vote (the Pashtun are a plurality of Afghans overall) and with Dostum as his running mate, the majority of the Uzbek vote as well. I expected Abdullah to secure the vast majority of the Tajik vote and other minorities but that this would not be enough for him to win an outright majority.
When I wrote my E-Note, the election had just been held on April 5th and there were already early indications of some election fraud, although it did not appear to be on the same scale as was the case during and after the 2009 Presidential election. My concern therefore was that the supporters of both Abdullah and Ghani might feel compelled to influence the outcome of the second ballot by fraudulent means, given that the stakes would be even greater.
When the results of the first ballot were finally published, the results certainly validated my assessment that Ghani and Abdullah would be the clear winners and that neither would secure sufficient support to avoid a second ballot. The results were however still very surprising. Abdullah was the clear winner with effectively 45 percent of the vote while Ghani had only 31.5 percent. This outcome aroused my suspicion, as I expected Ghani to have a slight lead going into the second ballot and it further increased by concern about election fraud. I could draw only three conclusions. My polls and assessment were wrong, which is of course, possible. Abdullah had significantly out performed Ghani during the campaign, which I do not think the evidence supports, as both candidates appeared to have run effective campaigns. Or there had been a significant level of fraud, which had potentially undermined Ghani’s vote and reinforced Abdullah’s.
Sadly, the latest reports from Afghanistan have reinforced my original concern regarding the extent of election fraud. Last week, Afghanistan’s election commission fired more than 3,000 election staff, accusing them of malfeasance at polling stations and during the subsequent counting process. Independent election monitors have also indicated that many complaints were actually ignored by Afghanistan’s supposedly independent election commission, reportedly in an effort to meet deadlines. These monitors also stated that the entire decision making process lacked transparency. This would suggest that the extent of the fraud is even greater than the 3,000 firings suggest. According to media reports, more than 900 complaints were eventually classified in the most serious category. That is more than in the previous election, when more than a million votes were thrown out and it is well known that President Karzai’s supporters extensively manipulated the vote and the subsequent counting process. Around 300,000 votes have reportedly been excluded this time around.
This is deeply worrying not only for Afghanistan and its people but also the International Community. Another highly questionable election will almost inevitably lead to conflict within Afghanistan, probably along ethnic lines, and potentially the subsequent withdrawal of donor support, which would be disastrous for Afghanistan. It is highly unlikely for example that the Afghan National Army would survive for very long without international funds and American advisors. Nicholas Haysom, deputy head of the U.N. mission in Afghanistan, told journalists recently that “the worst-case scenario would be if the election is both polarizing and the results are not accepted by one of the candidates and that has the potential to lead to conflict.” The only winners from a discredited election and subsequent conflict would be the Taliban, who remain a potent force, and some of Afghanistan’s neighbors, mostly notably Pakistan and Iran.
Certainly both candidates have accused the supporters of the other of perpetrating significant election fraud. Ghani claims that close to a million fraudulent votes were included in the ballot, which should have been dismissed. And his supporters suggest that the inclusion of these votes largely benefited Abdullah. Abdullah has been less strident in his accusations but he has acknowledged that fraud did occur while strenuously denying that his supporters were responsible. He stated to the media that “The whole process would have been questioned had we gone further…exposing the faults of the process,” and “[s]ince there was a chance for the second round and for a correction of the system…we did not want to be responsible for that.” This is hardly an unequivocal condemnation of such fraud and seems more like deflection, which certainly does raise suspicions. Given that Ghani’s actual share of the vote was well below my expectations, while Abdullah performed above the absolute maximum (44 percent) that I expected for him, my suspicion now rests more with Abdullah and less with Ghani. Either way, all parties concerned do agree on one thing. There was significant fraud in the first ballot, reported and unreported.
These revelations and admissions encouraged me to do a comparison of the results of the first ballot with the two polls that Glevum undertook in December 2013 and January 2013 and with the recently released poll funded by Afghan Center for Socio-Economic and Opinion Research (ACSOR), which was undertaken in March 2014. Incidentally, ACSOR was one of the two other organizations given a grant by the Department of State to undertake three pre-election polls each, along with Glevum. However, under pressure from Afghan government officials and others, the U.S Embassy in Kabul canceled that grant after the three organizations had conducted just one poll each. ACSOR is to be commended for funding another poll with its own money. They have done the Afghan people a great service. It is a terrible shame that we bowed to pressure from President Karzai, which I can only conclude was intended to prevent the nine polls from being used to both deter fraud and to demonstrate after the fact that it had occurred.
My analysis of the stated intentions of likely voters in each of Afghanistan’s 34 provinces suggests that Abdullah secured a significant increase in his actual share of the vote in seven key provinces – Balkh, Daykundi, Ghazni, Heart, Kandahar, Logar, and Nooristan. These increases ranged from 15 percent to 25 percent, which is far greater than one might expect in a closely contested election. While Ghani also gained a sizable boast in a number of provinces, his vote surprisingly collapsed in others by between 10 percent and 20 percent. This is all the more surprising because in the Glevum polls, Ghani’s support appeared to be the most firm. The provinces where Ghani suffered a surprising decline again included Daykundi, Ghazni and Kandahar. Press reports suggest that some of the worst and largest scale examples of election fraud occurred in the seven provinces mentioned above. While this comparison is not clear evidence that election fraud has benefitted Abdullah more than Ghani, it is at least coincidental and worthy of further investigation. Unfortunately with the election less than a week away no such investigation will take place. And Abdullah at least is encouraging Afghans and monitors alike to look to the second ballot for a “correction.” Sadly that is unlikely to happen, even if 3,000 election workers have been fired.
An evaluation of the ACSOR survey, which was conducted in March, after campaigning was well underway, suggests that the differences are not as great as in the earlier Glevum polls and that Ghani’s support had indeed declined from January. ACSOR indicated that likely voters favored Abdullah by 46 percent to Ghani 35 percent. This is more in line with the official results of the first ballot. The other most interesting finding of the ACSOR survey was that likely voters, who did not support Abdullah or Ghani in the first ballot, favored Ghani by 57-34 percent in a second ballot. This is a significant increase from the Glevum polls, which did have Ghani as the second choice of more likely voters than Abdullah but only by a few percentage points. ACSOR also found that in a second ballot against Abdullah, Ghani’s support amongst the Pashtun would likely increase from around 49 percent to 75 percent. Given that a significant reason why Abdullah may have legitimately increased his support above my expectations, was the higher than expected support from some Pashtun voters, this finding could become decisive in the second ballot.
I believe it is clear that there was extensive fraud during and after the first ballot, exactly as I warned. The Afghan election commission has acknowledged it and so have both candidates. Ghani at least, along with independent monitors and media reporting, claim that the extent of this fraud was far greater than has been reported by the election commission, which is also accused of lacking in transparency. Sadly I think that there is an overwhelming likelihood of significant and decisive fraud taking place during and after the second ballot as well.
It seems likely to me that President Karzai has thrown his weight behind the candidacy of Abdullah and he has notoriously been involved in election fraud before. And despite International Community wishful thinking, he will remain a powerful force in Afghan politics after he leaves office, and therefore has a vested interest in seeing someone he can work with be elected President. I still believe that in a fair election, demographics alone should ensure a Ghani victory. It therefore seems likely that if anyone will benefit from such fraud and has the means to perpetrate it, it is far more likely to be Abdullah, whom I believe is the favored candidate of President Karzai. This would fly in the face of the Afghan peoples wishes, as they have overwhelmingly indicated in the ACSOR poll that they want Afghanistan taken in a new direction.
Unless we want to see another election tainted by fraud, with the Pashtun in particular potentially seriously aggravated and disenfranchised, it is vital that the International Community uses all the means at its disposal to encourage all parties concerned to actively resist the temptation to cheat. This must include much greater oversight, closer independent monitoring, external auditing of the votes counted and the application of significant political pressure on all parties. There is not much time left to beef up this oversight effort but the price of doing nothing more then is already complacently planned, will be far greater. While President Karzai and some candidates may complain of foreign interference, I think Afghans themselves would welcome such extensive oversight. There will not be another opportunity for a “correction”. And if this election is seriously tainted, there may not be another chance for democracy in Afghanistan, which is on the brink of either taking a huge stride toward a better future or a terrible step back to the 1990s.