«« Ïðåä. | ÎÃËÀÂËÅÍÈÅ | Ñëåä. »»
10. Counting the Votes and Reporting the Results
Generally speaking, the law provides for an appropriate mechanism for the immediate counting of votes at the precinct level and an orderly frame work for the rapid reporting of results. As provided for in Article 52 of the Law on Election of President, ballots are counted at the polling station by the Polling Station Electoral Commission immediately after the close of the polls at 10:00 p.m. Under the law the votes are to be counted directly by members of the Polling Station Electoral Commission with the right of deciding vote. Members with deliberative vote are excluded from any role in the counting process. In spite of the clarity of the guidance provided by the law and in the CEC's instructions there continued to be sporadic instances where international observers noted that deliberative voting members assisted in the count. On a few occasions observers of candidates were also seen participating. Although noted as a lapse in standard legal procedure, in none of the cases cited by international observers did there seem to be concern that their participation contributed to disruption of the process or contrived manipulation of the results.
The law provides a sound but superficial procedural outline of the counting and reporting process. In particular, provisions attempt to cover a logical sequence of steps to be undertaken in the counting of the votes and incorporate a laundry list of specific kinds of information which must be accounted for and reported on the protocol of election returns. The law also offers general guidance as to the rules which are to be applied in determining whether a voted ballot is valid or invalid affecting whether or not it may be included in the counting of votes. In addition, Article 52 dictates that the counting must be accomplished without a break until the results are established.
Before the Counting of Votes Begins
According to Article 52 of the Law on the Election of President, the polling station election committee is required to count and cancel (render void) the remaining unused ballots in the presence of all authorized observers before the ballot boxes are opened and counting begins. The number of the unused ballots is to be declared and entered into the protocol which will ultimately be completed to report the results at the polling station. The committee is then supposed to inspect the seals on the ballot boxes allowing those authorized to be present to verify that they have not been damaged. Only then is the committee ready to begin the actual counting of votes.
Article 52 also dictates the order in which the ballot boxes are to be opened. According to the law, «ballot boxes shall be opened by turns -- first mobile ballot boxes and then stationary ballot boxes.» With regard to the ballots contained in the mobile ballot boxes, the Article goes on to say that «their number should not exceed the number of written applications for the conduct of the voting outside the voting premises.» Under the law, if the number of ballots contained in the mobile ballot boxes exceeds the number of applications, they are all declared null and void by a decision of the Polling Station Electoral Commission. An acknowledgment of the decision must be made in writing, and must include the names of the members of the commission responsible for assisting voters voting outside the polling station. The statement must be attached to the protocol of results reported for the polling station. This is an important addition to the legal foundation because it attempts to add safeguards against abuses of the mobile voting opportunities.
Only after the preliminary steps have been completed are officials allowed to open the stationary ballot boxes and begin counting the votes.
For Consideration
Article 52 requires no other advanced preparations other than the cancellation of the unused ballots and the «counting» of the ballots from the mobile ballot boxes. However, there are other steps that should be considered for completion before the actual counting of the votes begins. In an earlier section of this report related to advance preparations and opening of the polls it was suggested that certain entries should be made on the protocol before the voting begins. Specifically, it was recommended that before the first voter votes, the following entries be made on the protocol to initiate the ballot accountability record:
number of voters on the voter list;
the number of voters who applied for absentee certificates; and,
the total number of ballots received by the polling site election commission.
These figures are already known before voting begins and should not be subject to change. They should represent the base information against which election day activity should ultimately balance.
In its Unified Procedure for the Tabulation of Vote Returns and Compilation of Protocols, the CEC stipulated, «The protocol of Polling site election commission on Vote Returns shall be filled out after the verification of accurate counting of the votes.« (Emphasis added.) However, implementation of this proposed procedure for completing base line entries in advance of the voting would seem to be complementary with Article 8 of the Law on Basic Guarantees of Electoral Rights with specifies that «modification of voter registers after the commencement of the vote counting procedure shall be prohibited.»
12.1. At the close of the polls, and before any ballot boxes are opened, it is also recommended that the law require that additional entries be made on the formal protocol to complete the base line information against which the ballot activity and the vote counting should ultimately be balanced. These entries which should be enumerated separately on the protocol include:
- the number of unused ballots that have been canceled;
- the number of ballots that were spoiled by voters and canceled;
- the number of voters who have been added to the list throughout the polling hours; and,
- the number of voters who have signed the voter list to acknowledge that they have received a ballot (the number of ballots given to voters at the polling station on election day.)
12.2. It is also recommended that as these figures are entered, the Chairman of the Polling Station Election Commission should announce them to the committee members, observers and candidate representatives who are present. This information will reassure them and help them understand how ballots are being accounted for, and how the results will be balanced to actual voters.
Only after these initial preparations are completed and the basis for the audit trail has been established should the first ballot boxes be opened. (See Completion of the Protocol later in this Chapter beginning on page 104.)
Counting the Ballots Voted Outside the Polling Station
The disposition of the ballots in the mobile ballot boxes serves as an example of the kinds of technical deficiencies that exist in the current law. Procedural details become particularly important when considering the vulnerability of this component of the election system to abuses, whether real or merely perceived. Any time ballots leave the polling station and are used outside the view of the majority of the officials and the observers, special attention to detail must be a priority. However, except for dictating that these ballot boxes be opened first and that the number of ballots may not exceed the number of applications, the law is silent on the procedural details. In fact, the provisions of Article 52 leave a lot of questions unanswered.
- When the law says that the ballots from the mobile ballot boxes are to be «counted» first, what does «counting» actually mean? Does it also mean counting to determine the votes cast, or just to determine the quantity of ballots involved?
- In the event that a polling station has more than one mobile ballot box, are the number of ballots counted separately for each box against the applications of voters using that box? Or, are the ballots from all the mobile ballot boxes commingled to be counted in one total?
- How are the results of the votes on these ballots integrated into the reporting of results for the precinct as a whole?
- Are the ballots from the mobile ballot boxes maintained separately, or are they commingled with the other ballots cast at the polling station?
The answers to these questions speak loudly to the ultimate accountability and auditability of the counting process.
The failure of the law to provide procedural guidance left it to the Central Election Commission to devise the mechanism as to how these critically important ballots would eventually be integrated into the overall results. In its Flip Chart for Polling Station Election Commissions, the CEC instructed officials to count the ballots for each mobile ballot box separately. This approach has merit because it ensures that in the event the ballots in one mobile box exceeds the number of applications related to that box, only they need to be excluded from the count. The ballots from the other boxes are not tainted by the error made in the distribution of ballots contained in one box.
However, the CEC's instructions for the presidential election did little to overcome other deficiencies in the law. One example relates to the unclear intention of lawmakers in requiring that the ballots from the mobile ballot box be «counted» before the stationary ballot boxes are opened. As written, it is open to subjective interpretation as to whether they are to be counted for quantity or if they are to be counted to determine the votes cast upon them. The instructional hand book also neglected to clarify whether or not the votes cast on these ballots were also to be counted before the stationary ballot box is opened. Both the law and the CEC's instructions neglected to address how these ballots were to be packaged or otherwise segregated. Nor did the Flip Chart make any mention or provide any clarification as to how the votes cast on these ballots would be accounted for or reported in the overall results.
In spite of having no specific legal basis or written instruction to stipulate their handling, there appeared to be a general assumption among officials that these ballots were initially to be counted without regard to the actual votes cast. Once their quantity was determined and officials were sure their number did not exceed the number of applications, the stationary ballot boxes were opened and the ballots inside were simply dumped on the table and commingled with the ballots from the mobile ballot boxes.
There are advantages and disadvantages to this approach. On the upside, commingling of ballots voted outside the polling station with those voted on the premises ensures that no conclusion can be drawn as to whether or not voters voting at home voted differently than those voting in person. However, in the vast majority of cases the mobile ballot box option was used by a sufficient number of voters to ensure the secrecy of any individual's vote as is required in law. The disadvantage is that when these ballots are commingled with those cast at the polling station, the audit trail for ballots in the mobile ballot box is obliterated. Any challenges which might subsequently arise regarding alleged improprieties or misuse of the mobile ballot box affect all the ballots from the polling station rather than only the ballots directly involved. These issues will deserve continued consideration as legal reform ensues.
For Consideration
12.3. The provisions of Article 52 related to the counting of the ballots cast outside the polling station should provide procedural details designed to allow for adequate accountability and to leave an audit trail sufficient to reasonably withstand the scrutiny incumbent in a legal challenge. At the very least the law should be embellished to provide answers to key procedural questions such as those identified above, or which are currently left open to interpretation.
12.4. It is suggested that the law specifically require that in comparing the number of applications to actual ballots those related to each mobile ballot box be accounted for separately. Documentation should record the information identifying each ballot box identified by a number, the officials responsible for the box and the quantity of ballots and applications related to the box.
12.5. It is recommended that the law require that votes cast on the ballots from the mobile ballot boxes be counted separately and not commingled with the ballots from the stationary ballots. Once the initial check has been completed to ensure that the number of applications and ballots are equal, the ballots from all the mobile ballot boxes found to be in compliance should be commingled for the purposes of actually counting the votes. Commingling all ballots from the various mobile ballot boxes together would help to ensure the secrecy of the voters' choices.
12.6. The actual votes counted for each candidate on ballots cast outside the polling station should be recorded separately on the protocol and the ballots should be packaged separately. By keeping these ballots separate from those voted in person at the polling station, any challenges to the mobile balloting would not taint the results from the in person voting.
12.7. Because this procedure would take a little longer resulting in further delay of the opening of the stationary ballot boxes consideration may be given to changing the order in which the ballot boxes are opened. It would be advisable for the stationary ballot boxes to be counted first.
Invalid Ballots
As each ballot is reviewed during the counting procedure, the commission must make a determination as to whether the ballot can be counted or whether it must be declared invalid. The Law on Basic Guarantees of Electoral Rights provides only a general statement as to when voted ballots are to be declared invalid. Article 31 establishes two grounds on which the ballot must be rejected:
- when the voter's will cannot be identified; and,
- when the ballot form is of «non-standard manufacture.»
Curiously, however, an omission of the requirement cited in Article 30 that the ballot «shall be stamped with the stamp of the precinct electoral committee or be signed by at least two committee members,» is not included among the grounds subsequently identified in Article 31 on which a ballot must be rejected.
Article 52 of the Law on the Election of President clarifies the grounds for rejecting a ballot, in particular, stating circumstances which would cause the will of the voter to be remain in question. It also addresses the omission of the committee's «certification» on the ballot. Under its provisions a ballot must be found to be invalid if:
- it is not in the approved format (is not an official ballot);
- it does not contain the certification of the electoral commission;
- it has more than one choice is marked;
- it is left blank with no choice marked;
- the intent of the voter cannot be determined.
If there is a dispute with regard to whether a ballot should be accepted or rejected a decision is arrived at by a vote of commission members. When such a decision is adopted and the ballot is determined to be invalid, the grounds on which the decision is based is to be written on the back side of the ballot. The requirement that the notation must be signed by no less than 3 members of the commission is consistently applied in Article 31 of the Law on Basic Guarantees of Electoral Rights, and Article 52 of the Law on the Election of President.
Invalid ballots, including those from a mobile ballot box which were declared void because their quantity exceeded the number of applications to vote outside the polling station, are maintained separately.
One of the key successes that is worthy of note, is that there appeared to be relatively few ballots that had to be declared invalid. In the presidential elections invalid ballots made up less than 2% of the number of ballots cast. Of particular significance is the fact that of the invalid ballots only a very small number were attributable to voters improperly marking their choices by the traditional method whereby all candidates were crossed off leaving the preferred choice exposed. This accomplishment was directly attributable to the effective voter information campaign undertaken by the Central Election Commission to educate voters regarding the new voting procedures.
For Consideration
It is recommended that lawmakers and officials review the omissions and inconsistencies in the various laws and develop appropriate amendments which will provide uniform criteria upon which a ballot must be declared invalid. The following examples represent some of the issues which should be reviewed.
12.8. Article 30 of the Law on Basic Guarantees of Electoral Rights does not include failure of the ballot to contain the official certifying seals or signatures as grounds for invalidating a ballots.
12.9. Although Article 52 of the Law on Election of the President addresses the grounds on which a ballot must be declared invalid, not all disqualifications are found there in one provision of law. Unfortunately, officials must peruse various sections to extrapolate other grounds on which a ballot might be rejected. For example, members must recall Article 49 which briefly describes the voting premises. Among its sections is one calling for the secrecy cabins in which voters may vote in private. In describing that these cabins must be outfitted with tables and writing accessories, the provision dictates that «Use of pencils for these purposes is not allowed.» Technically, this imposes a requirement on officials in what they must provide, not necessarily on a voter who could potentially use their own pencil. The question is left open further by the fact that failure to mark the ballot in ink, is omitted from the criteria identified specifically in Article 52 on which a ballot must be declared invalid.
12.10. From the instructions prepared by the Central Election Commission it is clear that they interpreted Article 49 to impose a restriction on the voter because they included use of pencil in their description of the grounds causing a ballot to be invalid. However, the CEC's instructions also add that «corrections or erasures» will also constitute grounds for rejecting a ballot, although there does not seem to be a clear legal basis for this determination. It would be difficult to argue that because a «correction or erasure» appears on the face of the ballot that the intent of the voter is not clear. As legal reform ensues, it would be helpful to identify all grounds on which a ballot will be declared invalid in one provision, and to close those areas open to interpretation.
The Counting Procedure in Practice
While the laws and conforming regulations and instructions produced by the Central Election Commission provide a number of clarifications and important details regarding the counting process, they fall short in defining the actual manner in which the ballots themselves are to be handled and physically counted. The practical aspects of handling the ballots and organizing the counting is left to the discretion of each Polling Station Commission. In fact, in its 12 April 1996 Resolution of adoption of its Unified Procedure For Tabulation of Vote Returns and Compilation of Protocols, the Central Election Commission specifically dictated that:
«For the sake of satisfaction of the requirements of the Federal Law and this Unified Procedure concerning transparency and publicity of the vote counting procedure, the Polling site election commission shall be entitled to set independently how to process the filled out ballots including the procedure of announcement of the ballot contents.» (Emphasis added.)
Ultimately, it means that there is little uniformity from precinct to precinct or, from one election to another. This degree of flexibility and the ultimate inconsistencies that go with it does not seem to conform with the provisions of Article 15, Section 2 which dictates that the Central Election Commission exercise control over adherence to the legitimacy in the preparation and conduct of the election of the president, and «provide for uniform application» of the federal law.
Among the reports of various observer delegations present for the previous elections there appeared to be a common theme. A number of delegations reported that polling station officials seemed confused and disorganized during the counting of votes. There appeared to be no consistent method by which they organized themselves to carry out the actual counting procedures. Instead, each site seemed to be devising its own system for handling, sorting and counting the ballots on the spot. Observers frequently noted that officials seemed unclear as to the procedures, and attributed such shortcomings to inadequate training.
Indeed, for the presidential elections the Central Election Commission made significant advances in its training efforts. In particular, the Flip Chart for Members of the Polling Station Election Committee prepared by the CEC served was lauded by precinct committees as being the most helpful tool in preparing them to carry out their duties on election day. The guide book also identified the specific articles of law which provided the legal basis for the instruction being outlined.
Not only did this hand book reiterate the law in layman's language, it added some helpful clarifications, diagrams, illustrations and suggestions to guide officials in implementing the legal requirements where the law provided inadequate detail. For example, it elaborated procedures dealing with circumstance where the number of ballots in the mobile ballot box exceeds the number of applications from voters wishing to vote outside the polling place. In this instance, it offered a sample of a form which should be used to report the situation and document the reason why the ballots were excluded from the count. In other examples, the Flip Chart clarified the kinds of documents which would be considered acceptable for identifying voters at the polls, as well as procedures for handling of voters who cannot produce identification, or who present two or more passports and want to vote on behalf of friends or relatives. The instructions also encouraged officials to display a larger form of the protocol for disclosure of the counting data and the results to anyone present in the polling station.
However, just as the law makes no mention of the actual physical process to be undertaken in handling the ballots and counting the votes, the Flip Chart for Polling Station Election Commissions also falls short of offering any definitive instruction as to the mechanics involved.
For the presidential elections international observer delegations also made note of some apparent confusion and doubt as officials prepared to count the ballots although there was a general acknowledgment of dramatic improvement in this area over previous elections. At the OSCE debriefing after the first round of the presidential election, for example, one of the common themes reported by a number of delegations was that «counting procedures at the polling stations often appeared confused and disorganized.» At a number of sites some initial unrest was noted as officials disagreed as to how to proceed. IFES witnessed observers at some sites expressing complaints that officials were proceeding with tasks out of the appropriate sequence.
From its own observations in Moscow and rural precincts in the vicinity as well as in Stavropol Kray, IFES also noted inconsistencies in the procedures being followed during the counting process. For example, in Moscow polling site officials made use of a tally sheet during the counting of the ballots. Rather than sorting the ballots by candidate, stacks of commingled ballots were read by one individual while another official made strokes on a tally sheet recording the individual votes for the various candidates. In Stavropol Kray, however, no tally sheets were used. Instead, ballots were sorted by candidate and stacked with the number of ballots in each stack subsequently counted.
Neither did there seem to be uniform consistency within the same precincts between first round elections and the repeat voting. During the repeat voting IFES had the opportunity to return to observe the activities at some of the same precincts visited in Moscow during the first round. It was noted that use of the tally sheet had been abandoned, in spite of their apparent success, and the fact that officials had advised team members that the tally sheets had been particularly useful.
It is acknowledged that once initial confusion at polling sites was overcome and officials settled into the routine they decided upon, the counting generally went smoothly and efficiently. The lapses in procedure noted by observers were not perceived as deliberate attempts to alter the outcome. In fact, most observers acknowledged that the counting at the precinct level generally believed to be an accurate reflection of the popular will.
As election administrators would generally agree, the ideal objective is to try to develop strategies and mechanisms that minimize opportunities for participants to find grounds on which to bring challenges. When it comes to counting and reporting the results, it is particularly important to ensure that the process is consistent, orderly, transparent and accountable. Virtually all of these elements depend on clearly defined uniform practices which are applied in the same manner at all polling stations. Officials at all levels must have a thorough knowledge of how to carry out their responsibilities with confidence and efficiency. It is when there seems to be a lapse in transparency, when polling site officials seem unsure or disorganized during the counting of votes or when discrepancies or lapses in the procedures are noted that candidates, authorized observers, and media representatives begin to question the credibility of the results and raise doubts about the competence of election officials. For these reasons it is important that officials at polling sites throughout the Russian Federation are trained in the same formalized procedures rather than being allowed to determine their own.
For Consideration
12.11. It would be beneficial if the «processing of the filled out ballots»was more clearly defined in the law to provide uniformity and consistency usually considered necessary to a fully accountable system. At the very least the Article 15, Section 15 should augmented to clarify the authority and responsibility of the Central Election Commission not only to «establish a uniform procedure of processing the results of the voting» but also a uniform procedure for the counting itself. Polling sites should not be given authority to «determine their own procedures.» The procedures adopted should be mandatory and uniformly applied at all polling sites involved in the election.
12.12. During the presidential elections there were basically two types of counting systems employed: the tally method, and the stack and count method. Either system can be effective as long as the prescribed procedures include measures to ensure security, accuracy and transparency. However, it is recommended that the law require the use of tally sheets in the performance of the counting process. There are some advantages to the tallying method which are worthy of note. The chief advantage is that the recording of individual votes on a tally sheet provides physical documentation to validate and support the totals reported on the protocol. The tallying method also provides fuller transparency of the process as observers can actually hear and observe the distribution of votes among the candidates. The tally method is also efficient. During the first round of the presidential elections it was noted that completing the counting by use of tally sheets did not seem to take any longer than the counting at other precincts where the ballots were sorted, stacked and counted.
12.13. Regardless of the type of procedure that is ultimately adopted there are procedural mechanisms that can be introduced that maximize accountability while preserving transparency. What follows are examples of how either of the two methods can be formalized and organized to maximize to fulfill both objectives.
What follows are examples of how instructions might be written based on a Stack and Count approach, and a Tally Sheet approach. The step by step details described in the sample instructions attempt to establish a uniformity of procedure required by Article 15 (15) of the Law on the Election of the President, while creating internal mechanisms for verifying accuracy and providing better transparency.
Sample Instructions No. 1: The Tallying Method
The tallying method can be implemented in a variety of configurations. During the first round precincts in Moscow (where the simultaneous election of mayor was also being conducted) the tally sheet process involved initial sorting of the ballots to segregate mayoral ballots from presidential ballots. The ballots for the separate elections were handled by teams at separate tables. At each table, one member of the commission read out the ballots while another member made the marks on the tally sheet. Observers and remaining commission members sat by the sidelines to await the results. Some commission members absorbed themselves in other activities such as organizing the materials for packaging and transport, counting the number of voters added to the voters lists, etc. If use of a tally sheet were formalized in law, there are number of procedural enhancements of the method might be beneficial.
One such alternative that might be considered involves the polling site election commission being divided into smaller counting teams made up of four members. Each team working at a separate table would be given a portion of the ballots to count. Two team members would sit on each side of the table. The assignments of duties among the counting team members would include:
1. a pair of «readers» on one side of the table - one member who reads the ballot out loud while the partner confirms that the reader has announced the vote accurately; and,
2. two members on the other side of the table who record the marks as each vote is called. One member makes marks on an original tally sheet, while the partner makes the same marks on a duplicate tally sheet.
To improve the speed and efficiency of the process, certain preparatory steps would be helpful. Other team members could be assigned to count out stacks of ballots in groups of 25. By counting ballots in groups of 25, it would be easier for counters to verify the accuracy of their work along the way and to isolate errors. The readers set aside invalid ballots so that they can be segregated, reviewed further and enumerated later. As counting of each stack of 25 ballots is completed, the members making the marks on the two sets of tally sheets announce their result to ensure they have the same totals for each candidate. If there is a discrepancy the error will be found in the last groups of 25 ballots and the error can be corrected immediately. An additional tool that can be helpful in streamlining the process is using two different colored pens with a switch in color of pen made between each group of 25 ballots.
When the counting is completed the total votes for each candidate are entered onto the protocol. The original tally sheet could be attached as supporting documentation with the protocol submitted to the Territorial Election Commission. The duplicate could be retained by the Polling site election commission.
Sample Instructions No. 2: The Stack and Count Method
If it is ultimately decided that the stack and count method is to be retained as the standard counting procedure, there are still some suggestions which would serve to improve transparency and provide internal mechanisms for verify the accuracy of the count. As counting by this method was carried out during the presidential elections, there was little to suggest that the results were tainted or inaccurate, in spite of the fact there was really very little transparency and no standardized method or procedure for confirming the accuracy of the count within the procedures being utilized. Typically, when the ballot boxes were opened and the ballots tumbled onto the center of a large table, committee members immediately surrounded the table to assist in the sorting of ballots by candidate. Although not intentional, the view of observers was generally obstructed. Commission members simply took stacks of ballots from the center of the table and created their own candidate piles. Each member then individually inspected the ballots and placed them on the piles they had personally created for the various candidates. Once the various members had completed their sorting, their various stacks for the individual candidate were added together. An individual member counted the cumulative stack of ballots for each candidate. At some sites the ballots for a candidate were cross hatched into groups of 50. For each candidate, the chairman then counted the stacks of 50 adding the total number of any remaining ballots for that person.
This rather lax approach to the counting process invited participation by deliberative voting members and observers who are precluded from taking part by law. Nor were specific steps uniformly applied to verify that the ballots were accurately read and stacked in the correct pile. Nor did these practices give observers any opportunity to actually view the process in a way which would allow them to verify the count for themselves. Any errors in the sorting, for example, would not necessarily be discovered under this scenario. Rather, observers could only await the announcement of the results. To improve transparency and to ensure accuracy there are some measures that could be considered.
1. Officials could be instructed to prepare and place signs for each candidate appearing on the ballot. The signs could be spaced along a table in the same order as the candidates appear on the ballot. Additional signs could be made for the «Against All Candidates» category and for invalid ballots. During the sorting process stacks of ballots for each candidate could be created and piled in the space marked with the candidate's name.
2. A team of at least two commission members could be assigned to be responsible for the maintenance and counting of the stacks of ballots for individual candidates, for the invalid ballots and for the ballots marked «Against All Candidates.»
3. The Chairman, Vice Chairman and Secretary and remaining members could be assigned to actually accomplish the sorting by inspecting the ballots, and determining for whom the vote was cast, or if the ballot is invalid. These members would pass the ballots to the appropriate teams who would verify that the ballot is, indeed, marked for the candidate for whom they are responsible. If they note a mistake, they would pass the ballot to the team responsible for the correct candidate.
4. Once the ballots have been sorted, each team would count the ballots in their stacks, marking the total and their initials on a piece of paper. (Cross hatching the ballots into groups of 50 to assist in the count could still be utilized.)
5. As a safeguard to ensure correctness of the counting, teams could trade places so that the ballots are recounted by a second team. The results of the second team's count would be compared to that of the team originally assigned. If there is a discrepancy, the ballots could be recounted by the team responsible for them.
6. When the team is satisfied that their count is correct, they would report their results to the Chairperson so that they could be entered onto the protocol. Ultimately, when the ballots for each candidate are packaged, the team responsible for the counting of the ballots could sign their names to the package at the time it is sealed.
While there are many options that could be employed for either of these counting methods, the examples provided here demonstrate how mechanisms can be incorporated that promote accuracy, accountability and transparency. As candidates and their representatives, deliberative voting members, observers and representatives of the media become better acquainted with their roles and more knowledgeable about the system there will be less tolerance for perceived confusion or ineptness on the part of polling officials. The necessity for adherence to uniform practices on the part of polling station election committees will only become more important in closing the opportunities for complaints or challenges related to the counting process.
Completion of the Protocol
In spite of the fact that there was little formal guidance or uniformity as to how the physical process of the counting was to be accomplished and that sporadic lapses in procedure that were encountered, observers at the polling sites generally agreed that the voting results for candidates were accurately determined. Once procedural issues were resolved polling station commissions generally completed the task smoothly and efficiently.
The next step in the counting process is completion of the protocols on which the balloting activity and the election results are recorded. The protocols are prepared in triplicate in the presence of all authorized observers and signed by all members of the Polling Station Commission. Use of pencil for the completion of the protocols is not allowed. In addition, the protocols must be devoid of any corrections or erasures. Any member of the commission who does not agree with any or all of the information provided on the protocol is entitled to put his remarks in writing. A record is made of the dissenting opinion on the protocol and the written statement is attached to the first copy. Certified copies of complaints and related decisions of the Polling Station Commission are attached to the first and second copies of the protocol.
One copy of the protocol and relevant attachments are immediately forwarded to the Territorial Electoral Commission. The second copy along with the sealed voted ballots, lists of members of deciding votes, observers of the candidates, electoral associations, blocs, foreign observers and representatives of the mass media who were present for the counting is filed and kept by the secretary of the Polling site election commission. The third copy of the protocol is made available for scrutiny by attorneys of the candidates, observers members of the commission with right of deliberative vote, and representatives of the mass media.
Ultimately, voting documents, including the ballots are transferred to the Territorial Electoral Commissions no later than 10 days after the official announcement of the results of the election. For the most part, observer teams found that polling station election commissions delivered all the materials at the same time they submitted their protocols to the Territorial Commissions in the late hours after the close of the polls. The law does not clarify the circumstances under which a delay in the delivery of materials can be justified. Presumably, the allowance of 10 days for the delivery of materials accommodates those sites that are distant from their Territorial Commissions. Unfortunately, the law makes no provision for proper security for the materials which are not delivered immediately.
While the counting of votes at the precincts went smoothly, it was an almost universal observation that the completion of the protocol did not. One common theme among reports by various observer delegations was that it took longer for commission members to complete the protocol than it did to count the ballots. The confusion and frustration experienced by commission members may be attributable to three circumstances.
1. In describing the data which is to be enumerated on the protocol, Article 52 fails to identify a number of entries needed to accommodate various provisions of law which are critical to the proper accounting for the ballots issued to the polling station and their use in servicing voters on election day.
2. The protocol form designed to follow the law is not user friendly. The ballot accountability portion of the protocol offers no guideposts as to the calculations which must be made before many of the individual entries can be made.
3. Polling Station Election Commissions do not have a clear understanding of the ballot accountability portion of the protocol. Nor has the significance of this portion of the protocol taken hold as a test of the integrity of the process.
Article 52 of the Law on Election of President also dictates the reporting of precinct results on a precinct protocol which will ultimately be delivered to the Territorial Election Commission for inclusion in summarized returns. When the ballots have been counted, the precinct officials complete the protocol of the election results. In their Uniform Procedures Regarding Tabulation of Vote Returns and Compilation of Protocols, and in their Flip Chart for Members of Polling site election commissions, polling officials are advised to fill out the protocol after verifying the accuracy of the vote counting. The protocol includes the following information:
1. the number of voters on the voters lists including names that have been added to the lists;
2. the total number of ballots originally issued to the polling station;
3. the number of ballots issued to the voters at the polling station on election day
4. the number of ballots issued to voters outside the polling station;
5. the number of ballots declared void (canceled);
6. the number of ballots contained in the stationary ballot boxes, excluding ballots which are not in the official format;
7. the number of ballots contained in the mobile ballot boxes, excluding those which are not in the official format;
8. the total number of valid ballots;
9. the total number of ballots determined to be invalid with an additional entry separately identifying the number of invalid ballots that contained no markings at all;
10. the full names of the candidates, and in the event of their similarity, other data necessary to distinguish them;
11. the number of votes each candidate received;
12. the number of ballots cast «against all candidates.»
These line items cover the general parameters usually associated with ballot accountability. However, they are not enough to give the full picture based on other legal requirements and allowances that the law provides in various articles throughout its text. Unfortunately, it would be impossible for a full accounting of ballot usage at the polling site without additional information.
For Consideration
The following represent examples of the information which the law should more specifically identify if a valid accounting is to be transparent and complete.
12.14. As written, Item 1 accounts for the voters originally on the voter list plus those that have been added throughout the voting process. Taken literally without a clarifying adjustment identified by the CEC in its Uniform Procedure and in its Flip Chart instructions, it would not appropriately reflect a later provision which requires that the voters who received an absentee certificate to allow them to vote elsewhere are to be excluded from this number. Without this adjustment the number of voters on the list would be inflated and potentially result in voters being counted on two lists: one for the precinct from which they received their Absentee Certificate, and a second list at the polling station where they cast their votes and were added to the list on election day.
12.15. Item 3 is not articulated in a way that serves to provide an accurate figure against which to actually balance ballot usage. According to the wording of this item officials are to identify «the number of ballots issued to the voters at the polling station on election day.» Under this wording officials could calculate the number of ballots issued at the polling site by a mathematical calculation based on the total number of ballots issued to the station minus the number ballots voted outside the polling, the number remaining unused, and the number spoiled by voters. Although the mathematics might be accurate, it does nothing to justify or balance the number of ballots «issued» against anything. To provide a meaningful comparison, the wording of this item should be modified to require officials to count and enumerate:
«the number of signatures on the voter list acknowledging that the voter received a ballot at the polling station.»
12.16. The items for enumeration listed in Article 52 do not include guidance as to accounting for the ballots spoiled by voters and replaced. Voters who believe they have spoiled their ballots may request a replacement under the final provision in Article 51. Without addressing spoiled ballots officials would find it difficult to balance their returns.
12.17. Item 9 requires identification of the number of invalid ballots, and calls for a separate entry for invalid ballots that contain no markings. As worded it does not make it clear whether or not the initial entry is to represent the total invalid ballots including those without any markings, or only those that contain markings. It is not clear why the law specifically requires that the unmarked ballots are to be listed separately.
12.18. None of the items address the disposition or accounting for ballots from mobile ballot boxes declared void because their number exceeds the number of applications.
12.19. These same issues should also be reevaluated related to Article 31 of the Law on Basic Guarantees of Electoral Rights. In addition, it must be noted that some of these entries called for in the Law on Election of President vary slightly from those required in Article 31. For example, Article 31 also requires the protocol to include the number of ballots cast early. In presidential elections advance voting, although not allowed for the population at large, is allowed for remote site, on military and commercial vessels, polar stations, etc. Article 31 also refers to the identification of the number of ballots voted at «places of residence» while the related provision in the Law on the Election of President refers to ballots voted «outside the voting premises.» Clearly, the presidential election law offers broader options for voting at places other than one's residence. Article 31 does not require a separate indication of the number of ballots with no markings, or exclusion of ballots of non-standard form from the total ballots found in the ballot boxes. Nor does it require a separate entry for the number of ballots from the mobile ballot boxes.
Format of the Protocol
The Central Election Commission did a commendable job overcoming the deficiencies in the law as they relate to the accounting for election day activity and ballot usage at the polls in its publication of Uniform Procedure for the Tabulation of Vote Returns and Compilation of Protocols. Rational decisions were made to integrate information that the law failed to address. For example, the CEC determined that ballots spoiled by voters and replaced, would be included in the number of unused (canceled) ballots. Likewise, instructions reminded officials that ballots from mobile ballot boxes nullified because they exceeded the number of applications were to be included in the total number of invalid ballots.
Unfortunately, the protocol form itself only provided the categories as they were identified in the law. A number of calculations which are actually necessary are not identified on the protocol itself. That meant that officials had to continually refer to the regulations to figure out how to arrive at the correct entry for several of the categories. At the end of the counting on the nights of the first and second rounds of the presidential elections, it became clear that the process of completing the protocol was very confusing and difficult for most commissions. In particular, officials struggled with Items 1 - 10 of the protocol which pertain to ballot accountability.
For Consideration
12.20. Some of the problems could be alleviated by redesigning the protocol to provide space and detail which would allow officials to actually perform the calculations on the face of the protocol itself. Not only would such modifications make the protocol form easier for officials to use; there would also be an additional benefit in that actual display of the individual entries for each element of the mathematical equations involved would also retain the audit trail in a manner which is obliterated by the existing procedure. Instead of just entering lump sum of each equation, entries of the detail would allow observers, superior election commissions and other interested entities to see how totals were arrived at. This level of detail would also be helpful for review by the courts in the event of a challenges that might be filed based on the results.
What follows is an example of how this portion of the protocol could be modified to ease the burden on polling station election commissions while helping preserve a more meaningful audit trail. Not only could the form include all the specific line items needed to accommodate the various provisions of the law not already addressed, the form could remind officials how to perform the calculations.
Misunderstanding Ballot Accountability
Training of Polling Station Election Commission related to the completion of the ballot accountability portion of the protocol needs to be revisited. There is reason to question whether officials truly understand what ballot accountability is all about, and its significance in ensuring the integrity and reliability of the process. In addition, completion of the ballot accountability portion of the protocol is also the least understood by observers and the least transparent element of the counting process. There are several contributing factors.
First, as discussed earlier in this report, there does not yet seem to be a reliable audit trail created beginning with the printing and distribution of ballots. Ballots are not printed on watermarked paper, packaged in uniform quantities, printed with sequential numbering, or with counterfoils or stubs. There is no requirement that the number of ballots issued to the polling station be counted and confirmed in the presence of members of the commission and observers before the voting begins on election day. IFES encountered officials who acknowledged that precinct commissions were not always careful about counting for themselves the ballots that were delivered to them, often relying instead on the number they have been told they have received from superior commissions.
Aside from their Uniform Procedure for Tabulating Vote Returns and Compilation of Protocols of Precincts, Territorial Election Commissions and Election Commissions of the Subjects of the Russian Federation adopted by the CEC on April 12, 1996, the CEC also ordered production of the Guide Book of the Territorial Election Commissions Concerning the Election of the President of the Russian Federation. This instruction booklet included some procedural detail on a number of commission functions including management and distribution of election funds, status of observers, logistical assistance to polling site election commissions, organization and procedures for sessions of the commission, transfer and storage of electoral documents and adjudication of complaints. The hand book also provided samples of a full complement of forms that territorial commissions could use in the organization of their work. Included among the forms was Form No. 9, Act on Transfer of Ballots on the Election of President. This form offered a sample of a receipt that could be used in which the Territorial Election Commission could write in numbers and in words the quantity of ballots being transferred to the polling site. The sample form provided a place for the member of the territorial commission and the precinct commission member to sign their names.
It is important to note that in the preface to Guide Book , territorial commissions are advised that, «All forms of electoral documents offered in this Guide Book are not binding and may be amended and adjusted with due regard to the conditions set at each territorial election commission.»
Once again, the omission of formal guidelines regarding ballot accountability in the law is not overcome with a uniform and standardized procedure. The procedural system simply falls short in addressing the issue. The actual number of ballots issued to the polling site is as critical to a reliable audit trail as is the number of voters. Laxness with ballot security also feeds seeds of doubt and opportunities for the kinds of allegations that were raised during the presidential elections about pre-marked ballots being found at some precincts as was alleged in Rostov and in other regions of the Federation. There can be no true accountability unless there is a verifiable beginning ballot quantity against which to balance.
Accountability procedures may also be impacted by a sense of urgency among precinct commissions nurtured by the fact that under Article 52 also prohibits any corrections to appear on the protocol and the fact that the State Automated System is programmed to reject any entries that are not perfectly balanced to a preset formula. To assist officials in their work copies of the data control formula were provided to them. A description of the formula to which they were to conform in filling out the protocol follows.
The concern is that officials may have been more motivated in successfully satisfying the formula than on accomplishing the manual steps precisely in their efforts to complete the accountability portion of the protocol. IFES had the opportunity to observe a situation which may serve to illustrate a problem that may be more wide spread than the «accuracy» of the accountability portions of the protocols as entered into the SAS would indicate.
At one of the precincts IFES observed for the count it was noticed that the commission's working papers had lots of adjustments and corrections as they tried to achieve the required results among the various sets of calculations. The commission ultimately ran out of blank forms. Once they were satisfied that they had determined the correct figure to enter into each line item of the protocol the chairman left the polling station and went to the local administration to get more forms. In his absence the secretary read off what was purported to be the final correct figures as IFES made up its own handwritten copy of the results. Even on the work sheet with the «final» figures there appeared to be a number of adjustments and many of the figures had been written over many times. The secretary said that everything was balanced perfectly and indicated that IFES could check the calculations on their own record for themselves. It immediately became apparent to the secretary and to IFES observers that an error still existed in the balancing of used, unused and spoiled ballots to the total number of ballots received by the polling station. They had apparently misapplied the formula balancing the used, unused and spoiled ballots to the total number of voters on the voter list instead of the ballots received. As a result of their miscalculation, their protocol would have been ten ballots off in the accountability portion.
Upon the chairman's return he and other members of the commission were clearly confused as to how the problem should be resolved. Ultimately, it appeared that they «manufactured» a correction to achieve the required balance to the ballot accountability portion of the protocol. To «correct» the problem they merely added ten to the number of canceled unused ballots. This «correction» did not impact the voting results which had been recorded properly. When they produced the final protocol just minutes later, IFES asked how they had found the error and determined where the correction had to be made and was told that they had recounted the unused ballots. However, during the canceling of these ballots observers witnessed that they had been torn in half. In addition, though IFES observers remained present during the entire time after the error on the protocol was discovered, no additional counting activity took place. In fact, the packaged materials had already been removed from the room. The correction appeared to be instantaneous. It could only be concluded that the correction was simply an artificial adjustment made out of frustration and not knowing what how else to rectify the problem. It seemed clear that they only had rote understanding of the ballot accountability process. They relied on the mathematical formula, but did not really understand how the ballots, voter lists, applications and other materials actually related to one another. Therefore, when they were unable to balance to the formula, they did not know which steps had to be repeated to find where the error had been made.
IFES later learned that the chairman was a new official serving in this capacity for the first time. It is also acknowledged that although the ballot accountability portion of the protocol was probably not completed properly, the vote totals for the candidates did appear to have been reported accurately. In talking with a member of the Territorial Election Commission about this circumstance, he acknowledged his belief that such adjustments were probably more frequent than realized.
The fact that precinct commissions are given the formula for balancing the figures combined with the fact that the SAS system will not allow any entries that do not balance perfectly creates an urgency which potentially encourages officials to make artificial adjustments just to be able to close out their activities. This component of the counting process will need continued attention if there is to be consistent, accurate and reliable reporting and accounting.
Test of Ballot Scanning Devices at Selected Sites
IFES had the opportunity to observe the counting at a precinct which had been selected as a test site for newly installed ballot scanning devices. In Moscow where the scanner devices were utilized the election for Mayor was being conducted simultaneously with the election of the President. Instead of dropping the ballots into a regular sealed ballot box, each voter inserted his ballots into a scanner which read the markings and counted the votes automatically. The devises being tested were programmed to differentiate between the mayoral and presidential ballots. The computer was also able to identify invalid ballots based on criteria written into its program.
The holding box in the scanning machines remained sealed until after the polls closed. At the end of the polling, the boxes were opened and the manual counting was carried out in the usual manner. The devise would cut a notch into the edge of each ballot it read as invalid so that they could be segregated by commission members when the ballot boxes were eventually opened. This allowed the members to review the ballots that had been defined by the software as invalid so that they could make their own determinations. Although the results of the scanner count were not intended for general publication, IFES had the opportunity to review and compare the computer results with those of the manual count. In a few instances, there was a one vote difference between the scanned count and the manual count total attributed to a particular candidate. However, these discrepancies may be attributed to subjective decisions made by the commission regarding whether or not a questionable ballot could be counted.
Voters seemed to have a generally tolerant, if not enthusiastic, attitude toward inserting their voted ballots into the scanner rather than a traditional ballot box. It was interesting to note that even older voters were open to the new technology. There were a few voters who remained reluctant to insert their ballots into the scanner, however, the devices were equipped with a separate slot and holding box which precluded their ballot papers from being scanned. When this box was opened with the main holding box, the few ballots it contained were put through the scanner and included in the results recorded on a special tape.
For Consideration
12.21. IFES believes the test was very successful. It is difficult to imagine wide spread use of the scanning devises throughout the 93,000 polling sites established for a Federation-wide election simply because of the cost factor. However, the scanning devices show real promise. There may be opportunities to expand their use in specific Subjects with high density populations. Another application which is worthy of investigation, is their potential use at the Territorial Electoral Commissions to scan precinct protocols, with a direct link to the SAS.
12.22. If scanning devises ultimately find a permanent place in the conduct of elections, it will be important to not only build pre-election testing of the program into the system; it would also be advisable to incorporate a standard policy for random manual count testing to ensure the integrity of the program as part of normal activity after the close of the polls.
Summarization of Results
Results from precincts throughout the territory as a whole are summarized by the Territorial Electoral Commissions who report their calculations to the respective Subject Electoral Commissions. The SECs summarize Subject-wide results which are then reported to the Central Election Commission. Ultimately, the Central Election determines and reports the final results for the Federation as a whole.
The State Automated System (SAS)
The Central Election Commission established a system of computers linked together to transmit precinct results up through the administrative election hierarchy called the State Automated System (SAS.) The SAS was instigated by a Decree of the President in August of 1994 to increase the speed, accuracy and transparency in the reporting of election returns. The Law on the Election of President provides license to utilize an automated system for processing the results while posing some limitations and controls on its use. The entire coverage of the SAS in the Law in the Election of President is capsulated in Article 59. Under its provisions, there is not a requirement that an automated system be used. Rather, the article provides parameters, «in the event of the use of an automated information system in the process of elections...» Among those parameters is a requirement that each commission responsible for the use of the system form a control group made up of both members with deciding vote and with deliberative vote to «exercise control» over use of the system. In practice these Control Groups were responsible for overseeing and auditing the data entry process and verifying the accuracy of the results as they were reported.
The law also mandates that all members of the election commission have a right to familiarize themselves with any information input into the automated information system or produced as output from the system.
Article 59 also states that from the beginning of the voting until the final protocol is signed by the Central Election Commission the system is to be used «exclusively» for supervising the process and the results of voting «by means of transmitting data from lower election commissions to higher election commissions.» This provision is intended as a safeguard against adjustments to election returns being made after submission of the protocols from the levels where they originated and where transparency is most accessible. This article also articulates that the returns reported through the SAS are «preliminary and of no legal importance.»
Development of the State Automated System and the procedures for its use in processing the results of the election rested with the Central Election Commission and adjunct Center for Information Technologies.
The results from the manual counting of ballots at the Polling Stations reported on protocols are delivered to the Territorial Election Commission. Upon delivery, the data from the protocols is entered into the SAS computer network to be integrated into cumulative totals Federation-wide. Based on data received from the Territories, the Subject Election Commissions will follow similar procedures for summarizing Subject-wide results. They in turn, ultimately transmit their summarized data to the Central Election Commission. The data control room serving as the headquarters for the SAS was established at the Central Election Commission which was linked to the Information Center where media representatives, observers, electoral associations and participants had access to broadcast of ongoing preliminary electronic returns throughout election night.
In compliance with Article 59, the results reported through the SAS are considered preliminary and «unofficial.» The computerized reports of these «unofficial» results during the presidential elections were available beginning at 11:00 p.m. in Moscow after the polls in the farthest western area of Kaliningrad close. The earliest initial returns were those coming from parts of the Federation in the eastern time zones where polls close first.
The official returns for the election are the manually prepared protocols and summary tables prepared and signed by election commission members at each level of summarization. The computerized results serve only as preliminary advance returns to apprise the population of results as quickly as possible.
For the presidential elections, over 2800 computers had been distributed among 88 of the 89 Subjects of the Russian Federation. Chechnya was the only Subject for which computers were not scheduled for delivery. These numbers reflected delivery of computers to 2713 of the 2735 Territories established for the presidential elections. According to information provided to IFES by the CEC, as of 10 June 1996, 82 Subjects had reported that their computers were installed and operational in the territories in their jurisdiction. In six Subjects, installation and preparation were still underway in the week immediately preceding election day. In addition to computers provided to Territorial Commissions, 39 computers linked to the SAS were assigned to Embassies beyond the borders of the Russian Federation, 1 computer was placed in the Ministry of Foreign Affairs, and 14 computers were assigned to frontier military units serving outside the Federation.
As of 12 June 1996, efforts were still underway to ensure that appropriate expertise and trained personnel were on-line at all computerized sites. There was concern that there would be a few areas where the SAS would not be totally operational by election day, although it was anticipated that these sites should be few in number. A back up plan was devised so that in the event results could not be reported from a Territory or remote site due to malfunction or other unanticipated contingency, the reporting could be accomplished via telephone or fax where possible.
Throughout election day, there were also call-in reports of voter turnout from the precincts to the Territorial Commissions on a regular schedule. It was been pre-arranged that these reports would be scheduled at 10:00 a.m., Noon, 2:00 p.m. and 6:00 p.m. From these estimates entered into the system, the SAS could project overall turnout at intervals throughout the polling hours.
Although the SAS was most recognized for its processing and reporting of election returns, it was also intended to store associated election information. In particular, the system was to store the full text of election laws and regulations and information on the candidates. The addition of automation into election processing is relatively new and represents only initial efforts which will more than likely be expanded in the future. There is also a growing interest in developing a system where users of network communication systems would be able to have direct access to the SAS on a read only basis. In fact a bill which would have provided for such opportunities was narrowly defeated just prior to the presidential election.
The success of the SAS was generally applauded on all fronts as a significant achievement and lauded for its timely reporting of summarized results and overall efficiency. Observers generally agreed that the data entry and processing of results through the SAS was fast, efficient and professional.
For Consideration
12.23. As the evolution of the system continues to unfold, it is recommended that additional data become a standard part of election programming. Ideally, pre-set data should, for example, include the number of voters on the voter lists prior to the opening of the polls, and the number of ballots issued to each polling station. Under the current system this data is entered, not in advance, but based on the protocols prepared by precinct officials after the close of the polls. However, pre-entry of this information would provide additional opportunities for programming more meaningful internal checks and balances and establishing another layer to the audit trail. It would also allow the Central Election Commission in accumulating statistical information on the election, including projections of voter turnout throughout election day.
12.24. It is recommended that officials consider how the SAS data can be organized and consolidated into suitable formats for providing election related information to candidates, electoral associations and other nominating organizations, as well as the new media during the pre-election period. Advance information about the details of election organization, polling sites, voter lists, etc. can be extremely useful in helping election participants prepare their campaign strategies. It can also be helpful to the media in organizing their coverage of the process.
Processing Returns at the Territorial Commission Level
The CEC issued useful instructions to territorial commissions on processing the precinct protocols, accomplishing the data entry, and managing the summarization of the results. Territorial officials seemed to be well organized and fully knowledgeable about their responsibilities in performing the first level of summarization. During both the first round and the repeat voting IFES observers had the opportunity to witness the processing of precinct protocols at a number of Territorial Commissions in Moscow and rural territories in the vicinity, in Stavropol Kray, and in the Rostov area. All IFES teams were impressed by the speed and efficiency of the data entry and the organization of the summarization process at the territorial level. However, there appeared to be some significant regional differences in the procedures for acceptance of receipt of precinct documents, verification of protocols, and accommodation of observers.
Instructions from the Central Election Commission called for the Control Group at the Territorial Election Commissions to review each protocol submitted by the precinct commissions to manually verify the entries to ensure that they achieved the appropriate balance required by the established pre-set formula. This was found to be a standard practice at all the territorial offices observed by IFES teams during the two rounds of elections. Only after the manual verification demonstrated that the protocol met the requirements of the pre-set formula was the data entered into the SAS.
One of the major differences in procedure related to the involvement of polling station officials in acknowledging the accuracy of the data entry. In Moscow the manual verification as well as the data entry was accomplished in the presence of the polling station election commission member. In fact, once the data from the protocol was entered into the computer, it was not actually transmitted until it had been verified against the original protocol by a member of the Territorial Commission and the Chairman of the Polling Station Election Commission. For this purpose, a printout of the precinct results was generated and reviewed against the original protocol. Together the Territorial and the Precinct official verified the accuracy of the data entry. It was also observed that the printout of the precinct results was sometimes signed by senior members of the territorial commission and by the data entry technician while in other instances only the data entry technician signed the precinct printout. In most cases the precinct printout was given to the polling site official to retain, although at one site the printout was kept by the Territorial Election Commission. When the IFES observers asked for copies of the precinct printout they were generated and made available immediately. At some territorial commission offices IFES teams encountered domestic observers who were usually given copies of these protocols on request.
In Budyonovsk, as each precinct's results were entered and verified, the printout was signed by the computer operator and the polling official. The polling site official was then required to sign a log being maintained by the computer operator listing the number and location of the precinct, and the time of completion of the data entry. In this region the precinct printout was also given to the polling station official.
In Leninsky Territory polling station officials were required to sign a log in which the time of submission of their materials was recorded. However, IFES observers found that after the Territorial Commission manually verified the protocol, subsequent procedures differed from those implemented in Moscow and Budyonovsk. For example, polling site officials were not present in the room where the data entry was being accomplished. In addition, no printout of the data entry was made available for polling officials to verify that the data entry had been done accurately. Another variation was that the Territorial Commission maintained a parallel table on which the Chairman recorded all the figures from the individual precinct protocols.
When IFES observers asked for a copy of the precinct printout they were told that no such printouts were possible. Territorial officials indicated that since the computer would only accept figures that balanced, there was no need for such a step. IFES was informed that only when all precincts had been entered, would there be a copy of the summary table printout. When IFES observers were given a copy at the end of the process, it was not certified, signed or stamped; the commission chairman indicated that only the handwritten territorial protocol was considered «official.»
There were occasions when errors were found by the Control Group or when the SAS refused to accept the data on the basis of the test criteria incorporated into the software. This criteria was the same as that provided for polling station officials which required an absolute balancing of the ballot accountability data on the protocols. Reasonableness standards were also applied to the information provided concerning the election results. When errors were found by the control group, polling officials were required to return to their precincts where members remained on duty, to reconcile the numbers and prepare and certify new copies of the protocols. Once the polling officials returned to their sites, it is not clear how the correction process was approached. There is no way of knowing, for example, whether the sealed packages of ballots, absentee applications, voters lists, and other materials were actually opened so that they could be reviewed and recounted as necessary.
Paragraph 12 of Article 52 of the Law on Election of President provides that in the event mistakes in the protocol are revealed or there are doubts about the accuracy of the protocol, the higher election commission can adopt a decision calling for the «repeated calculation of votes.» Upon adoption of such a decision the repeated calculation is to be conducted by the polling station election commission. The Article goes on to say that the «repeated calculation of votes» is to be conducted in the presence of the member(s) of the higher election commission. Most often, however, the errors or discrepancies found did not relate to the votes, but only to the failure of the ballot accountability portion of the protocol to balance according to the pre-set formula. The law dictates no guidelines for seeking corrections when these kinds of «mistakes» are revealed or there are doubts about the accuracy of the accountability figures. The possibility exists that «artificial» adjustments could be made just to satisfy the requirements, with no real verification process being undertaken. Since this part of the process is also the least transparent , the cause of the errors as well as any adjustments made to force a balance could remain undiscovered.
The difficulty with the rigidity and inflexibility of a system that absolutely requires that all calculations balance perfectly is that it means that throughout the entire process there is no room for human error. Realistically, however, it is hard to imagine that among 93,000+ polling stations, the entire process and accountability could be accomplished without an inadvertent mistake. It would mean that during the peak rush hours when polling places were most crowded that no voter ever failed to sign the voter list or received two ballots because two ballot papers inadvertently stuck together. It would mean that official inadvertently missed making the proper notation on the voter list regarding a voter who applied for an Absentee Certificate. It would mean that no official ever made a mistake counting the number of ballots delivered to the polling station or the total number of voters on the voter list at the end of a long voting day. As the election process matures and as public confidence in the reliability of the system grows, there may room to consider new ways to address the reasonable discrepancies that are a normal part of election administration.
For Consideration
12.25. The law also dictates that the counting must be accomplished without a break until the results are established. The law fails to impose any similar requirement during the summarization of results. Concern was expressed by Yavlinski representatives that in Chita the summarization process at the subject level was interrupted for a six hour break. Reportedly, observers, and more importantly, territorial commission members with protocols were told to go to a local hotel and return in the morning. Observers who offered to stay overnight in the commission headquarters were refused. Obviously, any interruption can breed suspicion and allegations of manipulations. It serves the interests of officials, candidates, observers and the public alike to avoid such mine fields of opportunity for allegations of impropriety and subsequent challenges.
12.26. In reviewing the varying approaches, it is suggested that the presence of polling station commission members and their involvement in verifying the accuracy of the data entry contributes an important element of accountability in the process. The issuance of a precinct printout can be a valuable tool in the verification process as is the signature of the polling site official acknowledging that they have reviewed it and found it to be an accurate reflection of their protocol.
12.27. While training, procedural instructions and the personal commitment on the part of all officials should be dedicated to ensuring as perfect an election as possible, it is a human process. Protocols should be an accurate reflection of actual activity even if minor errors must be ultimately be acknowledged. In the long run, it would be preferable for precinct officials to accurately and precisely accomplish the individual steps in counting related to the individual categories of information. Where discrepancies are noted, officials should be required to recount as necessary. If a discrepancy cannot be reconciled, perhaps it would be preferable for the commission to document the error and describe the steps taken resolve the problem. This information could be made part of the permanent record delivered to the Territorial Commission with the precinct protocol. The Territorial Commission could still require a full recount of the votes as prescribed in the current law if they felt the circumstances warranted it. Without a real and accurate accounting there can be no reliable audit trail. A reliable audit trail is fundamental to the integrity of the election process.
Observer Presence During the Counting and Summarization of Votes
The electoral laws also provides that authorized observers are allowed to remain present for the count. The rights of observers to be present during this part of the process has been dramatically expanded, not only in the laws themselves, but most positively affirmed in regulations regarding the rights of observers adopted by the Central Election Commission. The degree of transparency afforded the counting and summarization process has the potential to contribute tremendously to the accountability and integrity of the system, although sporadic lapses in implementation during the presidential have tested overall fidelity to the fundamental principles intended.
The rights of observers to be present during the count is expressed in the Law on Basic Guarantees of Electoral Rights, and, expanded most definitively in the regulations promulgated by the CEC. Article 14 of the Law on Basic Guarantees states that observers sent by «public or electoral associations or by candidates and the foreign observers....have the right to be present at the polling stations on election day starting from the commencement of the polling station election commission's activities until the signing by the commission members of the election returns protocol.» Under Article 31 of the same law which covers the counting of votes, polling station commissions are obliged to announce the returns to all members of the commission, «the observers representing the candidates or the electoral associations, as well as the attending foreign observers.» A critical portion of the safeguarding of transparency through the meaningful participation of observers is stated in a later section of Article 31 that obliges the commission to «supply a certified copy of the protocol of the voting returns» to commission members and to any observers, upon demand.
The Law on Election of President expresses the rights of observers during the counting of votes in more general terms. The language of the presidential election law virtually duplicates that of the Law on Basic Guarantees regarding the eligibility of observers to be present at the polling place from the beginning of the activities of the polling station commissions until the completion of the documents recording the election results. However, Article 52 of the presidential election law falls far shorter in covering the rights of access of observers to documents at the end of the count. Under its provisions, counting is to continue without a break, and then observers and commission members are to be notified of the results. Then, it only requires that a third copy of the protocol be «given for familiarization» to representatives of the candidates, observers, members of the polling station election commission with the right of deliberative vote and representatives of mass media.
For Consideration
12.28. A critically important omission in the law is that neither the Law on Basic Guarantees nor the Law on the Election of the President offers clear enough guidance as to the eligibility of observers to be present at the Territorial and Subject levels during the summarization process. Unless they are also guaranteed access to observe the process and to receive certified copies of the protocols and summary tables at the higher levels, the loop is not closed and transparency is shortchanged. It is the comparison of the precinct protocols against the entries on the summaries that allows observers and representatives of the candidates and electoral associations to verify the accuracy of the results.
12.29. The presidential election law does not conform to the more effective language of the Law on Basic Guarantees that guarantees observers the right to receive certified copies of the results. It is through the dissemination of certified copies of the protocols that a meaningful tracking and confirmation of the results through the summarization process can take place. This is another example of a disparity between the two laws which should be overcome in favor of the language of the Law on Basic Guarantees as legal reform ensues.
Distribution of Certified Copies of the Protocols
Accessibility to certified protocols at all levels is one of the most significant features of the election process in the Russian Federation. Not only does it provide open transparency of the process, it ensures its integrity. The positive and progressive potential of this right was established on solid ground during the 1996 presidential elections. However, there are two issues which will continue to need attention if the full impact of this critically important component of the election system is to reach full fruition:
1. overcoming lingering misunderstandings among some election commissions as to their obligations in making certified copies of protocols available to authorized observers and representatives immediately upon request, and not subject to prior review by a superior election commission;
2. strengthening the cognizance of election commissions about the official nature of certified copies of protocols and their significance as evidence in court proceedings in the event the of grievances and challenges.
In spite of CEC regulations that support and extend the rights of observers to receive certified copies of protocols of precinct results as well as summarized results at the territorial and subject levels, some election commissions still seemed confused about their obligations in this regard. While most commissions complied cooperatively there were still enough complaints about reluctance of officials to release certified copies of the protocols to authorized representatives and observers that it is clear that the issue deserves additional attention. Certainly the Tartarstan case being heard in the Supreme Court involving allegations of falsification of Subject Election Commission summaries points to the significance of the issue. In the weeks following the election, complainants were still struggling to get certified copies of protocols with which they sought to support their case. Yabloko representatives were refused copies of the protocols in Tartarstan. Even the court itself was confronted with a regional bureaucracy which refused to cooperate.
This represents perhaps the most strident of cases, but it does not diminish the negative impact of similar cases that were reported at scattered locations throughout the Federation. Observers of the Communist Party of the Russian Federation reported difficulties in acquiring copies of protocols, as did representatives of Yavlinski. Complaints were also filed in Rostov where candidate observers reported that they had been refused copies of protocols. There was even some question as to whether IFES observers would receive a copy of the protocol from a Territorial Commission in Rostov. When a copy was provided it was not certified with the signatures or the seal of the Commission.
The importance of this transparency measure cannot be understated. With a legal basis founded in the Law on Basic Guarantees of Electoral Rights and supported by clear regulations and directives of the Central Election Commission, there should be no misunderstanding as to the rights of authorized observers and representatives to have access to this documentation.
For Consideration
12.30. Of particular importance in ensuring that officials do not disrupt this vitally important transparency mechanism is enforcement by higher authorities. In its Unified Procedure for the Tabulation of Vote Returns and Compilation of Protocols, the CEC evoked Article 40 (13) of the Code on Administrative Offenses in advising chairmen of election commissions that failure to comply with the requirement that the third copy of the protocol be made available for examination would result in the imposition of fines from 20 to 50 times the monthly wage. The regulating instructions fall short in applying the same penalties to officials who fail to provide personal certified copies upon request. Perhaps the CEC was reluctant to include this violation under the jeopardy of penalty because the Law on Election of President fails to specifically require providing certified copies as Article 31 of the Law on Basic Guarantees of Electoral Rights does. But clearly the effectiveness of this guaranteed right can only be measured by its enforcement. Hopefully, full attention will be focussed on overcoming the deficiency in the Law on the Election of President, and any misunderstandings officials may have over their obligation in this regard.
12.31. The second issue that must be addressed is further training of officials regarding the official standing of the certified copies of protocols. Officials at all levels need to have a solid understanding of the importance of absolute accuracy in the copies of protocols they provide. Based on the observations of IFES observers, it became apparent that often the observers themselves were making hand written copies of the data from the official protocol. This of certainly an efficient way of ensuring that copies can be provided since copy machines are not available at most sites. However, officials often affixed their signatures or the seal of the polling site without actually verifying the accuracy of the data written by the observers. They simply assumed that the information had been written correctly. More than likely it was correct, however, without reviewing the copies thoroughly officials leave themselves open to misrepresentation.
12.32. It would also be helpful if officials maintained a list of the observers to whom certified copies were provided. One of the concerns identified by officials was that in the event the Territorial Commission Control Group found an error on the protocol and it was ultimately corrected by the precinct commission, the observers who had requested copies would not have the corrected figures. This is a legitimate concern that points to the need for retention of documentation as to what corrections had to be made and when and by whom the adjustments were made. The original copy should be kept as part of the permanent record; the corrected copy should include a notation that it is a subsequent version.
12.33. The Unified Procedure requires commissions to inform the members of the commission with deliberative vote and observers attending the procedure of the completion of the original protocol about the decision made regarding the correction. However, there was no indication that this notification was provided as a standard practice during the presidential elections. It should also be noted that in this directive members of the commission with deliberative vote were not identified among those who are to be notified. An obligation on the part of the commissions to post or otherwise apprise individuals who had taken certified copies of the protocols of the nature of any corrections that were required should be imposed by law. Upon request the chairman should also be required to provide a certified copy of the updated protocol which includes a notation that it is a corrected copy.
12.34. Additional training should emphasize the official status of the certified copies of protocols and their significance as evidence in court proceedings in the event the results are challenged.
«« Ïðåä. | ÎÃËÀÂËÅÍÈÅ | Ñëåä. »»
| 
