Afghanistan Analysts Network – English

Political Landscape

The Afghanistan Election Conundrum (7): A deficient polling centre assessment

Ali Yawar Adili 24 min

A fundamental question ahead of any Afghan election is where polling centres are located. In the past, they have been located in ways that deliberately disenfranchised some voters. New regulations designed to prevent this are aimed at making sure locations are ‘balanced’, ie there are now strict criteria which should ensure centres are near voters, their number is proportional to the local population and they are not sited in insecure areas. Given how vulnerable this aspect of Afghan elections is to manipulation, the Independent Electoral Commission’s (IEC) assessment of polling centres is crucial to the credibility of the forthcoming provincial and district elections (now scheduled for 20 October 2018). Here, AAN’s Ali Yawar Adili takes a close look at the IEC’s assessment and examines the doubts which have surrounded its findings.

Entrance to the main polling centre for men in Gardez city during the 2009 presidential elections. Photo: Thomas Ruttig 2009.Entrance to the main polling centre for men in Gardez city during the 2009 presidential elections. Photo: Thomas Ruttig 2009.

This is part seven of a series of dispatches about preparations for the parliamentary elections. Part one dealt with political challenges; part two with an initial set of technical problems, including the date, the budget and the use of biometric technology; part three with electoral constituencies; part four with controversies around the appointment of one new IEC member, after its former chief was sacked by President Ashraf Ghani; part five with a demand by political parties to change the electoral system and; part six with the date of the polls and with voter registration.

Why do polling centres matter?

Polling centres are the actual places where voters cast their ballots. They are often located in schools, mosques, government offices, large private homes or even in tents when no sufficiently large building is available. For example, many polling stations for Kuchis have been in tents during previous elections (see here).

Each polling centre contains at least two polling stations, one for men and one for women. (1) Identifying how many polling centres are needed and where they should be located is fundamental to preparing for any election in Afghanistan, a country marred by conflict and with insufficient – and by now, largely unfavourable – experience with elections.

Prior to the 2014 elections, AAN released its own analysis of what to watch out for in elections and their preparation. Polling centres featured prominently in this analysis. On the one hand, AAN argued, polling centres need to be known and accessible, so that voters keen to vote are not disenfranchised. On the other hand, the number of polling centres needs to be manageable, as each additional centre increases the likelihood of fraud, whether this is petty, spontaneous and localised, or large scale and well-prepared. This is especially the case in insecure areas where monitoring is limited. The more ballots sent to areas that are out of reach of observers, the greater the likelihood of receiving back large numbers of fraudulent votes.

The establishment of the polling centre list is, in itself, an important process. In advance of previous elections, the number of polling centres was altered several times. This led to complaints that the changes were intended either to disenfranchise or favour voters in certain constituencies (see this AAN paper) (2).

What does the new legal framework say?

The 2016 Electoral Law (see AAN’s previous reports here and here) goes further than its predecessors. It does not merely define what polling centres and polling stations are, but also stipulates that polling centres should be established in a balanced manner. In this context, ‘balanced’ means taking into consideration the number and geographical location of voters, so that the proportion of centres to voters and the proximity of centres to voters is the same across the country. This extended provision was included, based on recommendations by the Special Electoral Reform Commission (SERC), which was established by the government in 2015 to come up with proposals for electoral reform. The SERC specifically proposed that polling centres be revised in order to ensure “easy access” by voters to the centres. Article seven of the Electoral Law states that:

  • The Commission is obliged to establish polling centers taking into consideration the number of voters and their geographical locations in a balanced manner;
  • The Commission makes available all materials necessary for election at the polling centers prior to the election day, and provide the possible facilities to the voters and candidates to participate in elections and exercise their right to vote.

Based on the Electoral Law, the IEC developed and approved a regulation for a polling centre assessment. This was reported by the IEC on 9 May 2017 in a press release, which said that it had also approved the procedure and operational plan and that the budget for the assessment had been submitted to the government for it to provide funds. The regulation (available in Dari here) specified ten criteria for the allocation of polling centres, as follows:

  • Polling centres should be established in schools, or other public facilities if there are no schools;
  • A polling centre should be established for each residential area with a minimum of 200 voters, and the distance between a voter’s residence and the polling centre should be a maximum of three kilometres in urban areas and five kilometres in rural areas;
  • The minimum distance between two polling centres in densely populated urban areas should be one kilometre;
  • Each polling centre should consist of at least two polling stations;
  • The IEC determines the number of voters in each polling station and the number of polling stations in each polling centre for each election through a separate document and based on the voters list;
  • The existing polling centre list and Central Statistics Office’s data about the population will be used as the basis for allocating new polling centres;
  • The location of polling centres should be suitable for providing safe logistical services;
  • The location should also be suitable in terms of the safety and security of election workers, voters, agents, observers and of elections materials;
  • The polling stations within a polling centre should be large enough to be used for polling and counting activities;
  • Telecommunication coverage is an advantage in the allocation of a polling centre, but is not a determinant criterion.

The regulation also said that, in cases where these criteria do not exist when selecting a polling centre, the head of the IEC provincial office, in consultation with local authorities could recommend another suitable alternative location, which should be approved by the IEC central office. (3) Officials from the United Nations who are dealing with the Afghan elections, in conversation with AAN, said it was the first time the IEC has had such a clear regulation and called it “a big achievement.”

The IEC’s 2017 ‘countrywide’ polling centre assessment marred by doubts

As mentioned in the regulation, the point of departure for the polling centre assessment was the list of existing polling centres from the 2014 election. The assessment was, in practice, aimed at analysing whether the existing centres fulfilled the new criteria, or if adjustments had to be made. According to the IEC, it had a master list of 7,180 polling centres from 2014. However, not all of those centres were activated during the 2014 election due to security reasons. At that time, a list published on the IEC website contained 6,775 polling centres with 21,663 polling stations, 12,705 for men, and 8,958 for women (see AAN’s previous report on how the list changed constantly here). It went through even further revisions as per security reviews, and the number of polling centres that actually opened in the first round of 2014 presidential elections was 5,928. (4)

Kuchi Polling Centre with staff near Gardez with no voters (all day) during the presidential election in 2009. Photo: Thomas Ruttig - 2009.

Kuchi Polling Centre with staff near Gardez with no voters (all day) during the presidential election in 2009. Photo: Thomas Ruttig – 2009.

The IEC’s current assessment began on 7 August 2017and was conducted by 1,000 temporary staff (who, according to the IEC, had been hired and trained for this purpose), first in the provincial capitals and then in the districts. According to an IEC update, from 30 August 2017, the staff used tablet computers with the capability to obtain GPS and other information. The tablets, it said, had been designed to work “online and offline,” so that they could automatically transmit data from an area without internet coverage to headquarters when they came into an area with internet connection.

The IEC’s 30 August 2017 update also provided a breakdown of territorial coverage of the assessment (press release in Dari here). It said it had been able to access 385 (92 per cent) of the country’s 418 administrative units, but did not specify how many polling centres it had been able to visit in those units. (5) The IEC’s press release added that, in the remaining “33 administrative units (8 per cent) with high security threats, the polling centre assessment has not been done, but [the IEC continues to] work to pave the way for [their] assessment.”  Again, it did not specify how many polling centres were located in those areas. On 28 October 2017, Wasima Badghisi, the IEC deputy chair for operations, told the local website Khabarnama that, if the IEC could not assess these areas before election day, it would approve sites used in previous elections.

The IEC, in its first open session to decide the polling centres, on 7 November 2017, indeed made the decision that polling centres that had not been assessed due to security threats would be retained in their previous location, so that eligible voters were not disenfranchised should the security of those areas on election day be ensured. This indicates that the IEC had already come to terms with fears that it would not be able to assess these areas in advance of election day. In early January 2018, AAN received a list from the IEC that showed that 32 districts in 14 provinces had not been assessed. (6) Afghan media outlets also published a slightly shorter list. The lists largely overlap and include well-known areas of intense Taleban activity. On 18 February 2018, the head of the IEC field operation department, Zmaria Qalamyar, told AAN that the list of 32 districts was correct and that a 33rd district, named in an IEC press release was either a mistake, or that one ‘administrative unit’ had been added later to the list of those units that had been assessed.

Some election observers doubted the IEC’s claim of having assessed 92 per cent of the country’s districts. Sughra Sa’adat of the Transparent Election Foundation of Afghanistan (TEFA) told AAN on12 September 2017 that TEFA’s observations showed that “three districts in Balkh province (Chemtal, plus two others) haven’t been assessed at all. Out of Ghor’s ten districts, only three and in the provinces of Sar-e Pul, Faryab and Jawzjan, only the provincial capitals’ districts have been assessed.” She told AAN that TEFA’s assessment was based on observations collected from all 34 of the country’s provinces collected by 43 partner organisations which work with TEFA, either under a project contract, or as volunteers.

Polling Centre for women in Gardez city during the presidential election in 2009. Photo: Thomas Ruttig - 2009.

Polling Centre for women in Gardez city during the presidential election in 2009. Photo: Thomas Ruttig – 2009.

Sa’adat also raised concerns about the assessment in general. She said that “in the last elections, we had a lot of ghost polling centres, others inaccessible to women and [or] located in the houses of powerful men.” (This is confirmed by AAN research; see here and here.) She indicated that this might be the same this time.

Similarly, on 18 February 2018, Muhammad Yusuf Rashid, the executive director of the Free and Fair Election Foundation of Afghanistan (FEFA), another major observer group, wondered how the IEC – which had announced that the aim of the polling centre assessment was to balance polling centres – could do that, if it did not know the exact number of voters. A UN election specialist in conversation with AAN said that it was a ‘chicken and egg’ problem. This meant that it would also be hard to tie voters to specific polling centres if the IEC does not have a list of polling centres (according to article six of the Electoral Law, a voter should be tied to a specific polling centre, ie they should vote at the same poling centre where they registered to vote). Rashid also said that FEFA’s follow-up research showed the IEC employees had not visited many polling centres. Instead, they had either met elders or talked to people on the phone, asking if they were happy with the polling centres used in 2014.

The UN also seems to have its doubts. On 25 September 2017, the UN Special Representative to Afghanistan, Tadamichi Yamamoto,reported to the UN Security Council that the IEChad “completed its assessment of polling stations, reportedly [our emphasis] reaching more than 90% of districts.” Together with voter registration as the next step, he said this “would lay the foundation for what will be the most important electoral reforms since 2001: the establishment of polling station-specific voter lists and the completion of a viable voter registry.” (see here.)

Shahla Haque, the acting head of the IEC secretariat, (7) in response to the allegations, told AAN on 20 March 2018 that the IEC, in addition to the fact that it could not access 32 districts at all, also could not assess some polling centres in the districts that it accessed. She said that 32 districts included only around 360 polling centres, while the IEC could not assess an aggregate number of 1,744 polling centres and that this was what the IEC had reported publicly.

Shahla, in response to Rashid, told AAN that it was important for the IEC that employees brought the coordinates and other physical specifications of the polling centres they had assessed. This, she said, would not be possible unless they had visited those centres in person.

The outcome of the assessment

On 6 December 2017, during a consultation forum with political parties and civil society organisations (also attended by the author), the IEC gave these detailed statistics:

Out of 7,180 former polling centres (the IEC’s master list of polling centres from 2014), 5,436 (76%) were assessed and 1,744 (24 per cent) were not assessed due to insecurity.

1,015 polling centres (14.2 per cent) were moved from their 2014 sites to an alternative location, 617 (61 per cent) to schools and 398 (39 per cent) to mosques.

1,341 new centres were proposed. This would have increased the total number of centres to 8,521, an increase of 18 per cent. However, out of these, 1,176 (14 per cent) were found to be contrary to IEC criteria, while 586 others were merged.

In the end, 7,355 centres were approved. This number includes the 1,744 centres (23.7 per cent) which could not be assessed because of insecurity. Crucially, it is not known whether fulfil the new criteria or not.

A provincial breakdown of the centres can be seen in footnote 8.

Criticism from within the IEC

Apart from doubts about the actual territorial scope of the assessment, there were also questions – even within the IEC itself – as to whether or not the assessment achieved its objectives. For instance, the media received a copy of a letter (see here in which IEC Field Operations Chief Qalamyar wrote critically to the head of the IEC secretariat on 24 November 2017 that, despite the fact that the assessment was considered a huge achievement, the GIS (Geographical Information System) information showed that it had not met one important goal, to balance the polling centre assessment. He also called into question the budget and the efforts invested in the project.In a conversation with AAN on 7 January 2018, Qalamyar gave more detail:

I will not tell you the names of the provinces, but in some provinces, for each three villages within a radius of ten kilometres and with 1000 voters, one polling centre has been allocated, while in some other provinces, for each nine villages within a radius of six kilometres and with four thousand voters, one polling centre has been allocated. [This] will lead to an imbalanced distribution of polling centres.

If this is correct, it will affect the accessibility of polling centres to voters in certain areas. This, in turn, might lead to their disenfranchisement, as well as violating the Electoral Law’s provision that calls for balanced allocation of the polling centres.

Pajhwok News Agency reported on 10 December 2017 that, as a result of the new polling centre assessment, the number of polling centres had decreased in Nangrahar, despite its having seen a high rate of returnees and so has more voters. The number of polling centres for Nangrahar approved by the IEC (on 29 November 2017) is 457, a decrease of 41 polling centres from 498 in 2014. (see here.)

On 27 December 2017, the IEC’s Shahla Haque rejected Qalamyar’s criticism that polling centres had not been balanced. She told AAN that his criticism wasbased on additional criteria that he had wanted to introduce (such as the territorial size of each province). She further pointed to the on-going complaints procedure that could lead to adjustments to the polling centre list.

Allegations of corruption

In the process of carrying out the polling centre assessment, the IEC also found itself dealing with two major allegations of corruption. These allegations were associated with the procurement or the assessment itself. First, on 14 September 2017, the IEC announced that it had suspended the head of its field operations, Zmarai Qalamyar. This came after the IEC obtained an audio tape in which, as the (former) head of the IEC secretariat, Imam Muhammad Warimach told Tolonews, Qalamyar allegedly asked the head of a provincial IEC for 500,000 Afghanis. Warimach said Qalamyar couldbe heard demanding, “just cash three hundred thousand of it and keep the other two hundred thousand and pay the salaries.” (see here). It is unclear how this audio was obtained.

Polling Centre (provisional tent) near Gardez with no voters during the presidential election in 2009. Photo: Thomas Ruttig - 2009.

Polling Centre (provisional tent) near Gardez with no voters during the presidential election in 2009. Photo: Thomas Ruttig – 2009.

An IEC member who has listened to the tape and preferred not to be named told AAN that actually a larger sum was at stake in this case and that Qalamyar had spoken to the head of the provincial election office in Kandahar, asking him to return 1.5 million Afghanis allocated for the polling centre assessment in that province. The IEC has not published any report about budget allocation and expenditures or the corruption allegations. It was not clear, therefore, whether the 1.5 million afghanis was part of, or the full amount of the money spent on the assessment in Kandahar. The IEC member told AAN, “Ghost people were sent to carry out the polling centre assessment.”

This IEC member claimed to have told IEC officials to send the clip to the judicial agencies for prosecution and that, as a result of this pressure, Qalamyar had resigned on 6 September 2017. IEC spokesman (now chair) Gula Jan Abdul Badi Sayyad (9) told AAN on 23 October 2017 that Qalamyar’s resignation had been approved and that the case was now with the Attorney General’s Office. Later, on 6 December 2017, Sayyad backtracked and claimed that the Attorney General’s Office had sent a letter to the IEC saying Qalamyar could continue to work with the IEC and that he was already back at work.

On 7 January 2018, Qalamyar himself provided a diametrically conflicting story about the corruption allegation against him in a conversation with AAN. He alleged that the Kandahar provincial election officer holds two tazkeras (ID cards), one from Kandahar and one from Zabul, but had concealed his Kandahar tazkera and showed only his Zabul tazkera when he was hired as the provincial election officer in Kandahar. Qalamyar said that, after he received the provincial election officer’s Kandahar tazkera in 2017, he notified the Human Resources Department that the Kandahar provincial election officer had two tazkeras and asked whether they should continue to work with him or not. Qalamyar said that, according to the law, provincial election officers could not work in their own province. Qalamyar also alleged that, since then, the Kandahar provincial election officer has held a grudge against him and had forged the audiotape. He said he had resigned after the allegation came to light, but withdrew it after the Attorney General’s Office cleared him.

In the second corruption allegation against the IEC, an IEC member told AAN on 7 September 2017 that there had also been a misuse of resources during the purchase of the tablets used for the assessment. This allegation was about over-pricing and violation of procurement requirements. The IEC purchased them for 46,000 afghanis (around 700 USD) each, while, both according to IEC sources and AAN enquiries, the market price for such tablets is between 200 and 500 USD. The tablets were also purchased using a single-source procurement procedure. Another IEC employee confirmed both these facts to AAN. According to rule 22 of the general procurement rules of procedure, such a single-source procurement may only be used where: 1) a particular potential bidder has exclusive rights in respect of the provision of goods, works and services or the procurement can only be conducted from single source; 2) there is an emergency need for the goods, works or services, involving an imminent threat to public health, welfare or safety or an imminent threat of damage to property, and engaging in open tendering proceedings or other procurement methods would be impracticable or; 3) where the estimated value of the procurement does not exceed 5000 afghanis. (See the Dari version here).

Again, Qalamyar was involved in this issue. A source from the UN, which is privy to the IEC’s internal discussions, told AAN that Qalamyar very particularly pushed for single-source procurement. Qalamyar rejected this in a conversation with AAN. He even asserted that he was against this method. He said the IEC had not faced an emergency and there was more than one dealer readily available in the market.

However, then IEC spokesman (now chair), Sayyad, played down this allegation of corruption in two conversations with AAN, on 23 October 2017 and again on 6 December 2017. He claimed there had been insufficient time available for open-source or limited tendering methods, but the IEC would still deal with the issue seriously.

AAN is not in a position to judge who is right in either of these two cases. However, the allegations have left a stain on the IEC’s credibility. Some Afghan election observer organisations have expressed similar concerns. For example, Jandad Spinghar, the director of Election & Transparency Watch Organization of Afghanistan (ETWA), was quoted as saying that, although corruption existed everywhere, within the IEC it could “create distrust among the people about elections.” (see here.)

On-going assessments

There are two other assessments of polling centres.

One: Security assessment

After its approval, the IEC submitted the list of the 7,355 polling centres to the security agencies for a security review. Coordinating and approving the final list of polling centres that will open on election day with the security institutions responsible for the security of each of the centres has been the procedure ahead of every previous election.

On 18 February 2018, the head of the IEC field operation department, Zmaria Qalamyar, told AAN that, according to their procedure, the security agencies should provide their list to the IEC four months before election day. This would be 20 June 2018 for the new election date set on 20 October 2018 (it was scheduled after it became clear that the previously planned date, 7 July 2018, was not feasible). (10) After 20 June, the security agencies will no longer be able to clear any additional polling centre on the 7,355 list, but could still cancel centres which are under threat. According to Qalamyar, the IEC does not have any say in the security assessment of polling centres. If security agencies, for example, say they cannot ensure the security of a certain polling centre, then the IEC cannot open it.

On 25 March 2018, Senior Deputy Minister of the Interior Murad Ali Murad provided the security agencies’ assessment to the IEC. He said that out of the total 7,355 centres, 1,122 remain under “medium threat,” 1,120 face “high-level” threats and another 948 are not under the control of government forces. This would mean that out of the 7,355 polling centres proposed by the IEC, 3,190 or 43 per cent are either under a medium or high threat, or are in areas not under government control.

On 5 April 2018, Qalamyar told AAN that this meant the security agencies needed to take extra measures for those facing threats (both medium and high) and to regain the control of areas in which the 948 centres were located. This would be necessary for the delivery of election material, voter registration – another key step that the IEC plans to carry out between 14 April and 12 June 2018 (see also AAN analysis and on election day itself.)

Two: the ongoing complaints process

Simultaneously with the security assessment, a complaints process has been underway with regard to the IEC assessment’s results. According to the Electoral Law (article 28.1), the Electoral Complaints Commission (ECC) is responsible for addressing challenges and complaints arising from negligence, violation and identification of crimes related to elections. Pursuant to this provision, the ECC said, on 29 November 2017, it was ready to receive complaints regarding the balancing of polling centres, either at ECC headquarters in Kabul, or the IEC provincial offices. It said it had dispatched working teams to the zones between 2 and 11 December 2017.

The ECC has five members. Three are appointed for five years and two for three years. In addition to the central ECC, Provincial Electoral Complaints Commissions, comprising three members, have to be established one month before the candidates’ registration starts. Two members of the provincial bodies are appointed by the central ECC and take into consideration gender (there should be one woman and one man), while the third member is appointed by the Independent Human Rights Commission of Afghanistan upon the recommendation of the central ECC and approval of the president. According to article 29.6 of the Electoral Law, the government may also appoint two international election experts as non-voting members of the ECC in consultation with the United Nations for the purpose of ensuring further transparency in addressing challenges and complaints arising from the electoral negligence and violation. On 30 March 2018, ECC spokesman Ali Reza Ruhani told AAN that the provision of the appointment of non-voting international members was intended to boost the confidence of the people, as well as the international community if there are disputes, but it was not binding. He said the need for such an appointment has not come about yet. UN sources told AAN that some political party officials had flagged this; without being more specific, they wished the UN to play a greater role in elections.

On 16 December 2017, the ECC extended the complaints period for ten days until 21 December 2017, but without mentioning the reason (see here) (11). On 1 February 2018, it announced it had received 373 complaints from the provinces regarding the IEC’s performance in carrying out the polling centre assessment. It said it had sent the list of complaints to “the relevant authorities” to obtain the required information and would make decisions thereafter. The EEC indicated that the IEC had asked for one week in which to respond. After a preliminary review and analysis of the complaints, it published a list of them. According to Ruhani, this included an analysis of who complained and what was the complaint about (see it in Dari here).It showed various types of complaints. Some demanded more polling centres be allocated or reallocated to a certain area in various provinces (due to a number of reasons, including long distances from the polling centre, concentration of population, removal of polling centres and lack of polling centres).

The ECC held an open meeting on 28 March 2018 with representatives of the election observer groups to address the 373 complaints. (Ruhani had confirmed already to AAN on 20 March 2018 that the ECC had received the IEC’s response in three categories: invalid complaints; valid complaints which should be acted upon and; those that required the IEC to send its teams to the areas to reassess polling centres.) The ECC made the following decisions for these 373 complaints:

  • 19 complaints were plausible and accorded with the specified criteria such that these polling centres should be established;
  • three cases had not been surveyed due to security problems;
  • five complaints were valid and new polling stations should be established on election day;
  • 13 complaints were against the removal of polling centres, but it was established that the centres had not been removed;
  • the locations of 30 cases had not been given to the EEC by the IEC;
  • 44 complaints were duplicate cases;
  • 71 cases needed further review by the IEC’s technical team;
  • 184 complaints were considered implausible and rejected by the IEC, but needed more discussion and review;
  • There were complaints about four polling centres that did not exist.

On 5 April 2018, the IEC’s Qalamyar told AAN that it had added 11 centres based on the complaints procedure. This increased the total number of centres to 7,366. As the complaint process is ongoing, there could be further changes to the initial 7,355 polling centre list approved by the IEC.

Conclusion: the ‘balance’ of polling centres still in doubt

The IEC worked on the existing master list of 7,180 polling centres from the 2014 elections (not all actually used) to assess whether or not they met the new criteria established for the coming elections. The IEC claimed it could not assess 32 districts and 1,744 of the centres –almost one quarter of the total – due to security reasons.Although it stated it would work to ensure those districts and polling centres would be assessed before the election, the practical follow-up so far shows it has put no urgency into this task. Indeed, these districts and polling centres may not be assessed at all. If the IEC is not able to open polling centres for voter registration and voting in areas that were open during the 2014 election, this would be a serious matter, the effective disenfranchisement of the population there.

An outdoor Polling Centre near Gardez during the presidential elections in 2009. Photo: Thomas Ruttig - 2009.

An outdoor Polling Centre near Gardez during the presidential elections in 2009. Photo: Thomas Ruttig – 2009.

As a result of this exercise, the IEC has, so far, approved 7,366 polling centres for the next elections. Thishas been one of the key steps that the commission has taken in preparation for the next parliamentary and district council elections. As a complaints procedure and a separate security assessment of the centres are both ongoing, the final numbers may still change.

The IEC’s polling centres assessment was designed to find out also whether the centres’ distribution across the country’s population was balanced, or needed adjusting, and whether or not the centres were accessible to the voters. Although the IEC has, for the first time, put in place a clear regulation that specified the criteria for measuring this, doubts remain as to whether they were used in practice, as claimed by the commission. Some election observer groups, in particular, have cast doubts on the number of polling centres that the IEC claimed to have assessed, saying the number of districts and polling centres that remain un-assessed must be larger than those admitted to by the IEC. They have also accused the IEC of not carrying out part of the assessment on the ground, but by using indirect, remote methods. Doubts about the validity of the assessment have been compounded by allegations of corruption, which have not been refuted convincingly by the IEC.

As yet, the IEC’s polling centre assessment exercise remains deficient. If matters are not clarified, this means the integrity of the forthcoming elections is already in doubt.

Edited by Sari Kouvo, Thomas Ruttig and Kate Clark


(1) The size of a polling centre can vary greatly. Large polling centres can have up to 12 polling stations. Every polling centre has a code indicating the province, district and centre. Within the centre, each polling station is also numbered. (see AAN analysis here) A polling station is provided with one ballot box per election and a maximum of 600 ballot papers (see also this AAN’s paper here).

(2) For instance, as AAN wrote previously, “in June 2009, a leaked letter from the Ministry of Interior suggesting to decrease the number of planned polling stations in the North by 1000, due to shortage of security staff, met with protests in the media and led to allegations that certain actors were seeking to disenfranchise the non-Pashtun constituencies. In the south of Afghanistan, several interlocutors commented in the run-up to the elections on the potential willingness of local government officials to sacrifice security in certain districts where opposition candidates may make a strong showing. Due to insecurity, mainly in the Pashtun areas, several hundred planned polling stations may need to be relocated to more secure areas (in many cases next to existing polling stations or into the houses of tribal elders), while others may not open at all”

(3) The IEC in its polling centre assessment regulation also said it would determine the number of necessary mobile polling stations prior to each election. These are the polling stations to which voters are not specifically linked. It further said that these mobile polling stations would be set up in; 1) hospitals, prisons and military bases, one polling station for each 100 or more eligible voters; 2) impassable areas and villages that do not meet the abovementioned criteria. These would be identified during the polling centre assessment and reported to the IEC. The IEC may then decide to allocate mobile polling stations to those areas, based on their circumstances.

(4) Both Shahla Haque, the acting head of the IEC secretariat, and Zmarai Qalamyar, the head of the IEC field operations, told AAN that the 2014 polling centre list was based, in turn, on an earlier IEC polling centre assessment conducted in 2011 which, according to a concept note developed by IEC field operations, dated 24 June 2011 (which AAN received from Qalamyar) aimed to establish a “robust polling centre database (with polling centre risk factors, staffing needs and logistical support requirement).” According to this concept note, the assessment included two months of data collection and data compilation each with later updates.

It provided the following sample from the then-current list of a polling centre, which the concept note said, was not sufficient and that “many other facts about a given polling location must be managed.”

Province District Polling Center Name and Location PC Code Est. Voters Total PS/


Male Female Kuchi
Herat PROVINCIAL CENTER Sultan High school, PD 1 2401001 4500 8 5 3 0


Both IEC sources provided the following list of the data that it proposed be collected for the assessment of polling centres. It pointed out this was still not exhaustive.

  1. GPS coordinates
  2. Village/nahia (urban neighbourhood)
  3. Photos of building
  4. Type of building: mosque/school/madrassa/private house/open field/clinic
  5. Chairs and tables available (yes/no)
  6. Contact who manages the site on a day to day basis
  7. Number of rooms available
  8. Communications infrastructure for voice and data services
    • Mobile network (AWCC, Roshan, Etisalat)
  9. Village catchment by the polling center (to be tested against G[eographical] I[nformation] S[ystem] data)
  10. Access mode: road/track
  11. Smallest vehicle access: truck/4X4/donkey/camel/helicopter
  12. Tents needed (yes/no)
  13. Male and female stations co- located (yes/no)
  14. Kuchi stations (yes/no). If yes, which season are they present? Where are they during the off-season? (This will need considerable planning and accurate information in order not to disenfranchise them.)

(5)According to a list of the Independent Directorate of Local Governance (IDLG) that AAN received from its spokeswoman Munera Yusufzada on 18 December 2017, there are currently 364 official districts and 21 ‘temporary’ (unofficial) districts in Afghanistan. On 27 December 2017, Shahla Haque, the acting head of the IEC secretariat, told AAN that the IEC also used these figures (364 plus 21) and added the 34 provincial capitals, but not the urban districts, called nahias or nawahi (which, however, would give a total figure of 419, not 418 ‘administrative units’). (More on the relevance of this in a future dispatch about the district council elections.)

The IDLG in an 11 June 2017 official letter to the IEC (AAN has obtained a copy of it), introduced 21 temporary districts to the IEC to include in the electoral processes. According to the letter, ‘temporary districts’ refer to those districts that have been approved, after entry into force of the 2004 Constitution by the president of the country due to some security considerations.

However, they have yet to be approved by the parliament. However, according to another list that AAN was given by Obaid Ekhlas, executive manager to the deputy head of IDLG in mid-February 2018, there are 384 official and 23 temporary districts.

(6) This is the district list that AAN has received:

1) Maidan Wardak: Jaghatu; 2) Nangarhar: Hisarak; 3) Baghlan: Dahana-ye Ghori; 4) Ghazni: Zanakhan (under Taleban control), 5) Giro (only district centre with the government) 6) Ajristan (only district centre with the government) 7) Nawa;; 8) Paktika: Neka, 9) Gyan and 10) Dela; 11) Badakhshan: Warduj and 12) Yamgan (both are completely under the Taleban (see this AAN’s report); 13) Kunduz: Qulbad and 14) Gul Tapa of (both under the Taleban control); 15) Urozgan: Chora, 16) Shahid-e Hassas and 17) Chinarto; 18) Kandahar: Miyaneshin, 19) Shorabak and 20) Reg; 21) Faryab: Kohistan; 22) Helmand: Nawzad, 23) Sangin, 24) Musa Qala, 25) Reg (Khanneshin), 26) Baghran and 27) Disho; 28) Badghis: Ghormach; 29) Herat: Farsi, 30) Zer Koh and31) Pusht Koh; 32) Farah: Bakwa.

(7) Shahla Haque was appointed as the acting head of the IEC secretariat (also known as CEO) on 16 January 2018. This came after the former head, Imam Muhammad Warimach, was dismissed by President Ashraf Ghani on 21 October 2017 (see AAN’s previous reports here and here). Before this, Shahla was the head of the IEC training department. According to the electoral law, the IEC should introduce three candidates to the president who appoints one as the CEO. The IEC advertised the position with a six-day (20-26 February 2018) deadline for application. According to an IEC press release, 40 potential candidates had collected application forms, but only 29 people actually applied. The IEC started shortlisting on 12 March 2018. Shahla Haque herself toldAAN that she did not apply for the position as she lacked some legal requirements – the applicant should have degrees in certain fields that did not include her medical field. A UN election specialist expressed concern that, if Shahla were replaced, things could unravel, especially if a new CEO with no experience was appointed because, according to this source, things had been moving well with the election preparations since she became the CEO.

(8) The IEC approved the following polling centres in four sessions:

On 7 November 2017, for the following seven provinces: Samangan (115, increased from 101), Jawzjan (129, increased from 126), Kunduz (220, increased from 219), Nimruz (67, increased from 62), Sar-e Pul (149, increased from 143), Logar (82, decreased from 84) and Farah (224, decreased from 227). (see here.)

On 13 November 2017, for the following nine provinces: Baghlan (284, increased from 281), Badakhshan (322, increased from 315 in 2014), Ghor (289, increased from 240), Faryab (238, increased from 216), Kunar (137, increased from 119), Uruzgan (65, increased from 60), Zabul (99, increased from 81), Helmand (246, increased from 219) and Parwan (170, increased from 164). (see here.)

On 22 November 2018, for the following seven provinces: Paktia (197, decreased from 212), Panjshir (97, decreased from 108), Khost (192, increased from 186), Badghis (210, increased from 197), Kapisa (93, decreased from 107), Herat (462, decreased from 480) and Maidan Wardak (157, decreased from 160). (see here.)

On 29 November 2017, for the following 11 provinces: Kabul (553, decreased from 557), Balkh (314, decreased from 319), Bamyan (220, increased from 188), Paktika (237, decreased from 266), Daikundi (256, increased from 184), Kandahar (225, decreased from 245), Ghazni (406, increased from 397), Laghman (123, increased from 136), Nangarhar (457, decreased from 498), Takhar (241, increased from 215) and Nuristan (79, increased from 77). (see here) All the above links above include tables for each province.

(9) These allegations of corruption (as well as divisions within the IEC) also provided an opportunity for political groups to question the credibility of the IEC and demand the replacement of all of the IEC – and ECC – members with new ones. (See AAN’s previous report here) As a result, President Ghani sacked the IEC chairman on 15 November 2017 and a new IEC member was appointed on 13 January 2018. (See AAN’s previous reports here and here).

On 31 January 2018, the IEC elected its former spokesman Sayyad as its chairman for two and half years and Wasima Badghisi as deputy for operation and Mazallah Dawlati as deputy for finance and administration both for a period of one year. However, it has yet to elect a spokesman who, according to the Electoral Law, also serves as its secretary. According to paragraph two of article 11 of the Electoral Law, the chairperson is elected for a period of two years and six months, and the deputies and secretary (spokesperson) for a period of one year.

(10)For instance, before the 2010 parliamentary elections, the security institutions decided that 938 of the 6,835 originally planned polling centres (ie 13.7 per cent) could not be secured. The IEC accepted this in its final list of polling centres, which then consisted of 5,897 polling centres, representing a total of 18,731 planned polling stations. (See this AAN’s paper).

(11)According to the IEC regulation for complaints arising from the polling centre assessment, the IEC has to issue its final decision about the complaints received and send it to the IEC within 30 days for implementation (a Dari version of this regulation is available here).


election Elections Independent Election Commission parliamentary elections


Ali Yawar Adili

More from this author