Afghanistan Analysts Network – English

Political Landscape

Elections 2014 (37): Preliminary results before agreement on auditing

Kate Clark 7 min

The Independent Elections Commission (IEC) has announced preliminary results for the presidential run-off, today, 7 July 2014. The ‘raw’ votes, (largely before fraud has been taken into account) give a turnout of almost 8 million, with Ashraf Ghani getting 56 per cent of the vote and Abdullah Abdullah 43 per cent. For the last few days, the two teams, under the chairmanship of UNAMA, have been trying to thrash out an agreement that would see a deeper audit of the disputed second round votes and of Dr Abdullah re-entering the electoral process. A partial agreement appears to have come early this morning, but not a full one, which means Afghanistan could still be in troubled waters. Kate Clark (with input from Obaid Ali and Qayoom Suroush) reports.

Almost iftar: IEC announces the long-awaited and still disputed preliminary result of the Afghan presidential election 2014, Picture c/o Pajhwok Afghan NewsAlmost iftar: IEC announces the long-awaited and still disputed preliminary result of the Afghan presidential election 2014, Picture c/o Pajhwok Afghan News


The preliminary results, as announced by the IEC:

Total votes cast: 8,109,493

Male voters: 5,057,613  or 62.3 per cent

Female voters: 3,051,880 or 37.6 per cent


Dr Ashraf Ghani: 4,485,888 votes (56.44 per cent)

Dr Abdullah Abdullah: 3,461,639 votes (43.56 per cent)


Turnout figures includes valid votes: 7,972,727; and invalid votes: 136,766


Open polling centres 6172 (22,828 polling stations)

Total polling centres planned: 6,365

Total polling stations planned: 23,136

Contingency polling stations: 570

Closed polling centres: 193, with 579 polling stations

There were no ballots cast in an additional 299 polling stations.


Audits already carried out:

First audit of 299 polling stations (minimal criteria):

21 polling stations were affected

250 votes for Ghani were invalidated

684 votes for Abdullah were invalidated


Second audit of 1930 polling stations with 599 or more votes:

114 polling stations affected

6427 votes for Ghani were invalidated

4428 votes for Abdullah were invalidated


Journalists had been waiting for five hours for IEC chairman Yusef Nuristani to announce the results. Finally, just before iftar, in the early evening, he took the stage. That Ghani is in the lead is no surprise; turnout was higher, however, than anyone had expected. It should be noted, however that these are preliminary results and will change, depending on the work of the IEC and the Independent Elections Complaints Commission (IECC). However, for analysts and others, virtually starved of official data on the second round, they do mean, we can start scrutinising the preliminary results. (1)

The results appear to have been delayed while final efforts were being made to get an agreement between the two campaign teams on a deeper audit of the disputed second round vote. Abdullah withdrew his observers from the IEC and the IECC on 17 June 2014, in practice disengaging from the electoral process, after alleging massive and systematic fraud. He has released several wire-tapped telephone conversations allegedly between senior IEC staff and provincial government officials (see transcript of conversations in this AAN dispatch) to back up his claims and supporters took to the streets in peaceful demonstrations. The fact that the IEC was ready and keen to release preliminary results, but was waiting for the two teams, under the auspices of UNAMA, to try to complete last minute negotiations, suggests that the announcement of the preliminary results, the return of Abdullah’s observers to the IEC and IECC and a deeper audit would all have been part of any deal.

A partial agreement?

AAN was told by different sources that in the early hours of this morning, the two teams had come to an agreement on four audit triggers. Polling stations would be audited if: they contained 595 or more ballots (out of a maximum of 600), if there had been high female turnout relative to men, where there were ‘round number’ results (eg 300, 450, 500) and, also, of all those stations that been established for the runoff. This would have affected about 7000 polling stations with an unknown but substantial number of votes – possibly 2, 3 or more million.

However, there appeared to be/have been other demands on the table from Abdullah’s side which were not accepted –- an audit of stations based on a lower number of cast votes for one of the candidates per polling station, ie where 93 per cent or more of the vote had gone to one candidate and of female polling stations staffed by men; he also wanted the audit carried out by the IEC last week of stations with 599 or more ballots to be re-done. At around 2am, the talks broke down. It seems a full agreement was again tried for today, but could not be reached. This seems to have kept the whole deal, and the issue of whether the wider audit will take place, hanging in the air.

Nuristani appeared to refer to all this in his announcement today, while also deflecting responsibility away from his commission:

We know that there were negotiations between the two teams, and I know that they reached an agreement on four out of ten points. (2) And we have been told that Dr Abdullah’s team might dismiss another four points and that there are [ongoing] negotiations on the last two points.

Abdullah re-engages or fights on?

Support for a deeper audit of the second round vote has come not just from Abdullah, but also those supporting or observing the election, largely in private, but also publically by the European Union Election Assessment Team. It was also accepted by Ghani’s team “for the sake of transparency” and, apparently on the grounds, that he is confident in his votes and believes fraud was carried out on behalf of both candidates. Ghani, however, has also been insisting that the (already slightly delayed) electoral timeline must be kept to. That meant he wanted no delays in the release of the preliminary results and, in particular, is adamant that the new president must be inaugurated, as planned, on 2 August. There has been some inference from his team that calls for delays by Abdullah were just gamesmanship.

Before the results were announced, ​Abdullah had said he would only accept the outcome when the “clean votes are separated from unclean votes” and after all fraud allegations had been resolved. It was assumed he wanted to delay the announcement of preliminary results because they could lend  Ghani the aura of a leader in waiting. What Abdullah has is a partial, but not yet official agreement on a deeper audit and the release of the preliminary results despite his demand for them to be delayed. So far, there has been no substantial official reaction from either team, so we do not know whether Abdullah may return his observers to the commissions or not, or stay protesting from outside the electoral tent, or wait – if negotiations are still on-going.

Abdullah’s re-engagement in the election process is fundamental to any hope of an outcome to this election which is acceptable to all parties. However, the presence of his observers is also, in a very practical way, crucial to getting an audit that actually scrutinises the ballots. However good any new triggers might theoretically be in identifying ballot boxes which could have been stuffed, how the audit is carried out is critical. With Abdullah’s observers back in the room, a relatively shallow technical exercise aimed at being seen to get the audit done would be less likely. Instead, we might see a real attempt to separate genuine from fraudulent votes.

However, it is not clear if that deeper audit will now take place. “We have done our job,” said Nuristani in his news conference, “and now it is the IECC’s responsibility to investigate the fraud.” While appearing to suggest tackling fraud was now the IECC’s responsibility, he also said Ghani had agreed to an audit of more than 7,000 polling stations (which would clearly be within the IEC’s duties). However, he also said: “It is not the IECC’s job to audit. But we ask the IECC to stick to the electoral timeline. And we will fully cooperate with them.” It was really not clear what Nuristani meant by these statements. The IEC spokesman’s telephone was off, along with members of Abdullah and Ghani’s teams, so we have been unable to clarify. What one can say is that it seems that, although the IEC could opt for a deeper audit unilaterally, Nuristani appeared to rule that out and it would need both teams to need to make a joint request for the IEC to act further.

If the ‘deep audit’ was to go ahead, Abdullah could theoretically be back in the race. 7000 polling stations, containing an unknown, but substantial number of votes, would create enough room for the result of the election to change radically. On the other hand, we could still be facing more protests by Abdullah from outside the electoral tent and no end, as yet, to the deadlock.


(1) Until now, the only official data has been the list of the 1930 polling stations where 599 or more votes were cast. Some scrutiny has been made of these figures by Ian Schuler on the Development Seed website. The locations of the polling stations, he says, “currently largely track to areas with high levels of ballot box stuffing in the 2009 Presidential Election” and are fairly evenly split between those which returned a majority for Abdullah and those for Ghani in April (although, as the vote nationally was not evenly split, this shows some bias towards Ghani-supporting areas). Looking particularly at the stations among these 1930 which have been audited where there was a substantial jump in voting numbers between the first round and second rounds, Schuler finds most voted almost entirely for Ghani in the first round and those which jumped from fewer than 100 votes in April to the maximum in June are largely in Khost, Paktika and Wardak. His is a first look at trying to understand the numbers, but does not pick up everything. The first round ballot was scrutinised so minimally by the IEC and IECC that polling stations that were stuffed in the first and second rounds would not come up in this analysis.

(2) Proposals made by Abdullah in letters to the IEC in late June (one dated 25 June, the second undated, but sent soon thereafter) can be read, as translated by AAN here. His demands appear to have been streamlined into the following ten (obtained by AAN through two sources and apparently referred to by Nuristani today).

1 Cancellation of all those real and contingency polling stations which were added in the second round of the election.

2 Audit and checking of all polling stations where the number of voters was reported as less than 5 per cent in the first round of the election, but increased to 50 per cent in the second round.

3 Audit and checking of female polling stations where male employees were used instead of female employees.

4 Audit and checking of those polling stations where 93 per cent or more of the votes went to one candidate only.

5 Audit and checking of those polling stations where the percentage of void votes was reported as zero or more than 5 per cent.

6 Audit and checking of votes in those provinces that the number of voters was reported as more than number of eligible people for voting. (Appears to refer to national census statistics, see AAN dispatch on the shortcomings in Afghanistan’s statistics here.)

7 Audit and checking of those polling stations where the numbers of votes as were reported as 595 or more.

8 Audit and checking of those female polling stations in which the number of women voters was either equal to or more than the number of male voters in polling stations in the same polling centre.

9 Audit and checking of those stations which show round number results, eg 50, 100 or 200.

10 Audit and checking of those polling stations where the results were a series of numbers, eg 101, 102, 103, 104.


2014 Elections Ghani IEC Nuristani Result