On the eve of the election, AAN’s Martine van Bijlert takes a look at the final list of polling centres that the Independent Election Commission (IEC) has considered secure enough to open for voting tomorrow (5 April 2014). She looks at where the polling stations are – or rather: where they have they been removed – and wonders what this could mean for the election result in these areas. The location and spread of the polling sites, the number of ballots that they are provided with, and the likelihood that sites could be hijacked, she says, are a all crucial indicators of which way this election may go in the different localities.
The polling centre list is important because it determines how many votes can be cast in a certain area; every polling station is allocated 600 ballots. If there are too few polling stations in the areas where people are keen to vote, or if the stations are difficult to reach, voters may be disenfranchised. On the other hand, if too many stations are planned in areas that are difficult to monitor, the IEC may lose control over what happens on the ground. The more ballots are sent to areas that are out of reach, the greater the chances are of receiving back large numbers of fraudulent votes. And of course, not all areas are equal, particularly not in terms of which candidates may be drawing the greatest number of votes.
The total number of polling centres has fluctuated, as the Ministry of Interior (MoI) and the Independent Election Commission (IEC) have added and subtracted sites, based on population estimates and security assessments. A ‘final’ list was drawn up on 18 February 2014, detailing that 6775 centres out of a total of 7168 would be able to open. A further round of security assessments resulted in the removal of another 352 centres on 29 March 2014. The IEC has provided observer organisations and electoral staff with the details of these last removals, but they have not been released to the public, although the old list on the website has now been replaced by an updated one (so far, in Dari only).
More centres than in 2009
The latest round of removals means that the IEC is now planning for the opening of 6423 polling centres out of a total of 7168; almost 90 per cent. For a comparison: in the 2010 parliamentary elections, 938 of the 6835 originally planned sites were secured (slightly lower, at 86.3 per cent). The number, and the fact that it is higher than during previous elections, has been touted as a success. However, what really matters is not how long the list is, but whether the sites have indeed been well chosen (for details on how a lack of rigour can contribute to a loss of control, see here) and whether the distribution of the sites has been fair.
The province that was hardest hit in absolute terms in the latest round of removals (the 352 polling sites) was Nangarhar. The province, that had been left untouched in earlier rounds, lost 115 polling centres. The removal slashed almost 25 per cent of all polling sites – and potential votes – in the province (a loss of 166,800 potential votes in 278 polling stations).
The runner up was Ghazni, with 60 polling centres removed (175 polling stations representing 105,000 potential votes). Ghazni had already lost at least ten sites in an earlier round of removals, bringing the total to 70, which means that 17 per cent of the total number of polling centres in the province have been slashed. (1)
The removal of so many polling centres from selected districts obviously has an impact on the provincial council elections, particularly in ethnically mixed provinces like Ghazni, but also in other provinces where candidates may just have lost a large part of their support base (or the deals they made). But it also impacts the presidential election, as these are potentially very large numbers of votes, that may or may not have been consistently cast (or stuffed) for some of the presidential candidates.
An uneven distribution of removals
When looking at the total number of removals – figures are given below (2) – there are a few things that need to be watched. First, while the number of removals may represent only ten per cent of the polling centres all over the country, locally the picture can be quite different, particularly in the smaller provinces. There are, for instance, ten provinces that lost over 20 per cent of their sites. The provinces that were hardest hit, proportionally, are: Laghman (26 per cent), Farah (27 per cent), Badghis (32 per cent), Helmand (32 per cent) and Nuristan (where it was even 51 per cent).
These are high numbers, but they roughly correspond with what the situation might look on the ground, particularly if we compare these figures with what happened in 2009 and 2010. In 2010, in Nuristan over 70 per cent of the planned polling stations either didn’t open or were later disqualified. In Badghis this was almost 50 per cent, Farah 29 per cent, and Laghman almost 20 per cent. Only in Helmand it was rather low, with just 5 per cent of the planned polling stations either not opening or being later disqualified. (For more detail see here).
However, if we look, in general, at which areas in the country are insecure and which areas experienced have high percentages of closures and disqualifications in the past, it becomes clear that there has been a stark uneven distribution of removals – or rather of non-removals.
The provinces that have been most affected – either in absolute or proportional terms – by the full range of closures are Nangarhar, Ghazni, Nuristan, Zabul, Badghis, Helmand, Farah and Laghman. There are no real surprises there. A quick assessment of the affected districts indicates that, in most cases, these are areas that would indeed have been difficult to secure. (3) What is curious, however, is that areas in other provinces that are also highly vulnerable to insecurity, intimidation and the hijacking of whole polling sites, have been largely left alone.
In Nimruz, Baghlan, Khost, Paktika, Paktia, and Wardak – which are all provinces with precarious situations and known for serious and widespread irregularities in the past – not a single polling site was declared at risk. Kandahar, famous for mass ballot stuffing in 2009, lost only ten polling centres out of 244.
If we compare that with past closures and disqualifications, we see that in 2010 Baghlan, Paktia, Khost and Wardak lost around 40 per cent of their polling stations; Kandahar lost around 30 per cent; and Paktika even over 50 per cent. Only Nimruz was lower with 13 per cent, but anecdotal reporting at the time suggested that the fraud was considerably higher than was dealt with and there are reports of renewed preparations now.
It is not clear whether the difference in treatment is based on different styles of provincial reporting by electoral or security officials, or on decisions at the central level in either the MoI or the IEC, but recent conversations suggests that in several provinces local officials feel that not enough polling centres have been declared too insecure to open.
It has become difficult to predict which areas and which strongmen will support which candidates, with all the hedging going on. This means that it is not necessarily a given that the uneven distribution of removals is also a blatant and deliberate attempt to favour certain candidates – although it may well be. But what is clear, is that the large disparity provides grounds for speculation and suspicion. And that there are several areas that should be watched closely.
(1) The other provinces that lost polling centres in this round were: Zabul: 32, Herat: 31, Badghis: 19, Nuristan: 16, Daikondi: 14, Faryab: 13, Laghman: 12, Kandahar: 10, Farah: 8, Jowzjan: 7, Logar: 6, Ghor and Kunduz: both 3, Kunar: 2 and Badakhshan: 1.
(2) The original list, based on what had been used in previous elections, contained 6845 polling centres. After an initial assessment by the Ministry of Interior on which centres could be secured, the IEC announced on 11 January 2014 that it had taken 412 sites off the list, leaving a total of 6432. In the month after, the IEC submitted a list of 323 additional polling centres to the Ministry of Interior, apparently based on a 2012 assessment, while the Ministry of Interior separately announced that security operations had increased the number of sites that were safe enough to open. On 18 February 2014, the IEC finalised its list, which it announced on 22 February, settling on a maximum of 6775 polling centres containing 21,663 polling stations: 12,705 for men and 8,958 for women (for the difference between polling centres and polling stations see here).
The total number of polling centres that have been taken off the list (with the percentage of the total number of the provinces’ sites):
- Badakhshan: 0
- Badghis: 64 (32 per cent)
- Baghlan: 0
- Balkh: 0
- Bamyan: 0
- Daikondi: 14 (7 per cent)
- Farah: 61 (27 per cent)
- Faryab: 13 (6 per cent)
- Ghazni: 70 (18 per cent)
- Ghor: 53 (20 per cent)
- Helmand: 72 (33 per cent)
- Herat: 48 (10 per cent)
- Jowzjan: 7 (6 per cent)
- Kabul: 0
- Kandahar: 10 (4 per cent)
- Kapisa: 26 (24 per cent)
- Khost: 0
- Kunduz: 3 (1 per cent)
- Kunar: 25 (21 per cent)
- Laghman: 35 (26 per cent)
- Logar: 6 (7 per cent)
- Nangarhar: 115 (23 per cent)
- Nimruz: 0
- Nuristan: 36 (51 per cent)
- Panjshir: 0
- Paktia: 0
- Paktika: 0
- Parwan: 18 (12 per cent)
- Samangan: 0
- Sar-e Pol: 32 (22 per cent)
- Takhar: 0
- Uruzgan: 1 (2 per cent)
- Wardak: 0
- Zabul: 34 (42 per cent)
The provinces which saw no polling centres removed are Bamyan, Kabul, Panjshir, Balkh, Samangan and Takhar – which is fairly unremarkable – but also: Nimruz, Baghlan, Khost, Paktika, Paktia, and Wardak.
(3) The only notable exception is Daikondi, where, in northern Gizab, several sites in areas where the vote could easily be secured, were removed from the list. The polling centres are formally under the responsibility of the Daikondi IEC and are located close to Daikondi’s capital Nili, but were for some reason to be supplied by the IEC in Uruzgan, which claimed to have difficulty delivering the materials. The awkward arrangement appears to be part of wider pattern of local conflict and marginalisation.
This article was last updated on 9 Mar 2020