Strasbourg, 16 January 2003

CDL-AD (2003) 1
Or. Engl.

Opinion no. 226/2002

















adopted by the Venice Commission

at its 53rd Plenary Meeting

(Venice, 13-14 December 2002)


on the basis of comments by


Mr Richard ROSE (Expert, United Kingdom)

Mr Kåre VOLLAN (Expert, Norway)













On 28 August 2002, the Venice Commission was asked by the Secretary General to analyse the Electoral Law of Moldova (CDL (2002) 141). Following this request, the Venice Commission appointed Messrs Richard Rose and Kåre Vollan as rapporteurs.


At its 53rd Plenary session (Venice, 13-14 December 2002), the Venice Commission endorsed the comments of these rapporteurs (CDL (2002) 156 et 157) and authorised the Secretariat to prepare a consolidated opinion based on these comments, in order to be submitted, following approval by the rapporteurs, to the Secretary General on 15 January 2003.


This opinion is based on:

- the Election Law of Moldova (translation) (CDL (2002) 141);

- the Constitution of Moldova;

- the Election observation reports from both CoE and OSCE;

- the Code of Good Practice in Electoral Matters by the Venice Commission, including the Guidelines on Elections (CDL-AD (2002) 23);

- the Language Law of Moldova;

- the opinion on the Draft law on political parties and socio-political organisations of Moldova (CDL-AD (2002) 28);

- comments on the Election Law of the Republic of Moldova, by Mr. Richard Rose (CDL (2002) 156);

- comments on the Election Law of the Republic of Moldova, by Mr. Kåre Vollan (CDL (2002) 157).

Preliminary remark


1. It should be mentioned that the English translation of the Election Law of Moldova leaves something to be desired, and occasionally is technically misleading: for example, some provisions refer to candidates, whereas the voter can vote for a (blocked) list of candidates (see infra ad Article 10).


The Parliamentary elections – Electoral system and minority representation


2. It should also be noted that the Parliament of Moldova has 101 members, all elected in a single constituency in a proportional list system. The system allows for independent candidates. There is a Representation Threshold of 6% for parties, 9% for blocks (pre-election coalitions) consisting of two parties, and 12% for blocks of three of more parties. For independent candidates the threshold is 3%.


3. In Part III and V there is a lot of repetition of the general articles from earlier parts, which is unnecessary and confusing (e.g. Article 74).


Representation thresholds


4. Substantively, the representation threshold is high by absolute and comparative standards. The system asks voters to endorse a single candidate rather than a party list, and then allocates seats in proportion to the electoral strength of parties (that is, blocs of candidates). It also makes it possible for an independent candidate to be elected, if he or she wins 3% of the vote.


5. In eight of the ten post-Communist countries now seeking admission to the European Union the representation threshold is 5%, which it also is in the Russian Federation. In Bulgaria the threshold is 4% and in Slovenia, 3%. A higher threshold for blocs or coalitions of parties is frequently found in the region - but not at the levels specified here, and it may not be a wise arrangement. For example, in Romania a two-party bloc has a threshold of 8%, with 1% added for each additional member of the bloc. In Slovakia, a two or three-party bloc has a threshold of 7%; and for four or more parties it is 10%. In the Czech Republic the threshold rises steeply from 5 percent for a single party to 10% for a two-party coalition, 15% for 3 parties and 20% for four parties. Minority ethnic groups are sometimes catered for by establishing different and lower thresholds or by reserving a small number of seats (Table 1).

Table 1.- Proportional representation thresholds




Threshold for seats (%)

PR Index (%)



3+ parties
































Czech R










































(Proportionality Index: 100% equals exact match share of votes and seats)


a. Two seats reserved for ethnic minorities.

b. No threshold for minority parties.

c. And 20% for four parties.

d. For a coalition of four or more parties, 10%.

e. Guaranteed representation for minority parties.


Source: Calculated from official electoral data to be reported in Richard Rose and Neil Munro, Elections and Parties in New European Democracies. Washington DC: Congressional Quarterly Press, forthcoming, spring, 2003.


6. In Moldova the high threshold to qualify for seats in Parliament has resulted in only three parties winning seats there in the election of 25 February 2001. No independent was able to clear the 3% barrier for election.


7. Collectively, three parties secured 72% of the vote; the remaining 28% was cast for other parties or independents. The result is a degree of disproportional representation virtually unheard of in established democracies.


8. Whilst the 1994 Moldovan election was not so disproportional (Index: 82%), it too was below the median country in Table 1.


9. The fact that no proportional representation election produces an exactly proportional result is not a justification for maintaining a grossly disproportional election law. If a straight 5% threshold had been applied in the most recent Moldovan election and all electors had behaved the same, the relative size of parties would have been maintained and the Communist Party would have retained an absolute majority of seats in Parliament (Table 2). However, there would have been five rather than three parties in Parliament, and the Opposition would have had 40 rather than 30 seats, thus enabling it to operate more effectively (Table 2). A 1% reduction in the threshold would have produced, ceteris paribus, an increase of 10 percentage points in the Index of Proportionality, raising it to 82, well within the range of countries negotiating European Union membership. Therefore, we would recommend a maximum threshold of 5% for the representation of single parties. Given the limited number of parties currently contesting elections in Moldova, there would appear no reason to raise the threshold higher for coalitions of parties. In summary, coalitions should be encouraged, rather than penalised.


Table 2.- Simulated effects of a 5% threshold in Moldova


Actual threshold: 6%

If threshold 5%:

2001 vote







Braghis Alliance




Christian Dem People’s




Rebirth & Conciliation




Democratic Party




All others





Source: “Election Results”, OSCE/ODIHR Final Report Moldova Election (Warsaw, 3 April 2001), p. 13.


10. Allowing votes to be cast for individual candidates to allow independents to win seats in a PR system is acceptable in situations of weak and unstable party formations. However, the threshold of 3% specified in Article 87(3) is unreasonably high. Even though the threshold for independent candidates has been reduced from 4 to 3%, it is still so high that the election of an independent candidate is highly unlikely. To take an extreme example, if 33 independents each qualified for a seat, not a single party would pass the 5% threshold in a 101 seat Parliament. A strictly proportional quota for independent candidates would set the threshold for being awarded a seat at 1%. If 10 independents won seats by taking at least 3% of the vote, then a party would need to win 7.1% of the vote for party blocs (that is 5/70ths) to pass the election threshold.


11. Hence, it is recommended that the threshold for the representation of independents be lowered to 1%, a point particularly relevant to ethnic minorities. One may even consider altogether removing the threshold for such candidates, considering that in many countries it is difficult (or impossible) for independent to be elected.


12. The existence of numerous ethnic minorities, e.g. Ukrainians, Russians, Gagauz, Bulgarians and Roma, raises additional questions of principle, which will even increase in importance when the predominantly non-Moldovan ethnic population in the Transnistria region are to be given full practical access to the Moldovan elections. The reduction in the electoral threshold to 5% would allow large minorities, e.g. Russians and Ukrainians, to be represented in Parliament - if a substantial portion of the ethnic group wished to be represented by ethnic parties. The Gagauz minority does not appear to be large enough to clear a 5% threshold, and in particular not with a multitude of parties.


13. Lowering the threshold for PR to 5% would enable larger minorities to win a significant bloc of seats in Parliament, if their nominal members voted along ethnic lines.


14. Lowering the threshold to 1% would allow smaller minorities to get at least one voice in Parliament, if there was a significant degree of cohesive voting along ethnic lines. However, introduction of constituencies as described below seems to be the only effective way of providing proper representation to the Gagauz.


15. The long-term effect of maintain the current representation thresholds will possibly be that voters will concentrate on the main movements, but it is doubtful if such a high threshold is helpful at the stage Moldovan democracy has reached. Even more extreme are the increasing thresholds for blocks. It is probably beneficial for the effectiveness of the Parliament that pre-election coalitions are being formed, and it provides good information to voters. They should therefore be encouraged. It is not easy to understand why the threshold for such blocks needs to be higher than for parties. In that way, the incentive for building larger groups will be reduced, and work contrary to the purpose.


16. In any case, these decisions should not be taken by legislation but by the electorate.


Constituencies and minority representation


17. Very few countries are electing the Parliament in one constituency only. In Europe the Netherlands represent a prominent exception. In most other countries there is a degree of geographical representation secured by elections held in a number of constituencies. If a country is rather uniform in terms of population or other relevant criteria, elections in one constituency may work well. It will then be up to the parties to secure the geographical representation when compiling their lists of candidates. However, when minorities are concentrated in certain areas, constituencies can be the most effective instruments for securing reasonable minority representation in the Parliament. The report from the CLRAE Delegation to the Moldovan Local and Regional elections of 23 May 1999 quotes the Venice Commission stating that it is


"necessary for States to take into account the presence of one or more minorities on their soil when dividing the territory into political or administrative subdivisions as well as into electoral constituencies" (Opinion on the interpretation of Article 11 of Recommendation 1201 (1993) of the Parliamentary Assembly of the Council of Europe, CDL-INF (96) 4).”


18. The issue discussed in this report concerns the boundaries of the districts of local government, where the Bulgarians of the district Taraklia have been diluted when being merged into Cahul. However, the argument is just as strong when it comes to the representation of the Gagauz in the Parliament, and it has obvious relevance to the representation in the Parliament from Transnistria (at the moment the conflict is being resolved) as well.


19. A minority like the Gagauz would under the current system have two ways of being represented: by gaining positions at prominent places at the lists of the countrywide parties or by forming minority based parties. The first alternative does not provide sufficient guarantee for representation from the minority. The second would force the Gagauz to give priority to the minority dimension over the regular political dimension, since there will hardly be room for more than one party with representation in the Parliament, and with the current high threshold even that seems unrealistic.


20. OSCE/ODIHR has therefore both in 1998 and in 2001 recommended introduction of a system based upon constituencies. There are many such systems available, some based upon majority votes, other with proportional representation. Moldova has used a proportional system so far, and it is most likely that one would want to keep this principle. One alternative would then be the German or Albanian system with a combination of single member constituencies and a countrywide (in Germany state-wide) party list election to compensate for a disproportional representation in the Parliament. In this case the voter would cast two votes.


21. A simpler system which requires only one ballot would be the one adopted in Bosnia and Herzegovina in 2000 and which is commonly used for example in Scandinavia and in Romania in a slightly different form. The system is based on multi-member constituencies. The largest share (e.g. 70-80%) of the members of the Parliament would be elected by a proportional system in the multi-member constituencies, and the rest would be distributed as compensatory seats according to the countrywide results. No threshold should be applied at local level, whereas a threshold could be applied for the compensatory seats.


22. With constituencies following the population characteristics, regular political parties would compete locally on regular political issues, but the candidates would be coming from the constituency and most of them would therefore most likely be of the minority.


23. The second alternative with minority based parties is - as already stated - not a good one. For the representation to be effective, most of the voters would have to vote along the ethnic dimension only. That would mean that the voter would be deprived of the possibility of both voting for a certain political direction and candidates who are members of the minority group.


24. In addition to that, the Law on political parties used to have as a requirement that the party could show more than 5000 members from at least half of the districts of the country. The new Draft law on political parties and socio-political organisations of Moldova has brought this requirement further. A party representing the Gagauz would therefore fail to register, since they are represented in a concentrated geographical area only. This is in itself a violation of minority rights.


25. If constituencies are introduced, the possibility of being elected as an independent candidate may increase as well, in such cases where the candidate has a strong local support, even though the share of the votes needed to be elected would be higher within the constituency.


26. The introduction of constituencies would not need a change of the Constitution.



Other important issues




27. Article 1. The definition of referendum is unclear when it is question of a suffrage with regard to “the most serious issues of the state and the society in general”.


28. Articles 142 ff. It would be necessary to say in a more precise manner what kind of questions may be submitted to referendum. In particular, it should be said whether they are general questions (which is not so much suitable) or precise constitutional or legal texts.


29. Articles 145(2) and 162(2). Hold More Referenda at the Same Time. Article 145(2) seems to prevent two referenda to happen on the same day, whereas Article 162(2) regulates the case when there are more than one simultaneously.


30. Article 152. Details on Referenda. Article 152 regulates in detail the internal life of a public initiative for initiating a referendum. This falls into a long tradition in post communist countries but seem unnecessary. On the other hand, a general requirement that the wording of the ballot paper should be neutral and not favour any alternatives may be helpful (e.g. in Article 162).


31. Article 154. There is no specific reason why only members of the initiative group should be allowed to collect signatures. Cp. Article 42(2) according to which “persons empowered” by the candidates etc. may collect signatures in case of elections. This norm could restrict strongly the right to collect signatures if the members of the initiative group had to assemble (Article 152(1) could be interpreted in this way). The same is valid for Article 182.


32. Article 186. Local referendum. Paragraphs 1, 2 b): is the decision not to hold a referendum discretionary? If not, the legal grounds should be indicated in the law.


33. Article 187(2). Why should special electoral districts be constituted for referendum? (Cp. Article 120(1)).


34. Article 197. Does this provision mean that the normal provisions on appeals do not apply? If this is the case, it does not seem to be justified.


Election commissions


35. Article 16(2) vests the choice of the composition of the Central Elections Commission in the President of the Republic, the Parliament and the High Council of Magistracy. If the election is decided by majority or plurality vote in the last two bodies, all posts could be in the hands of a single political party. To avoid this, it would be suitable to elect the Commission by a system of proportional representation. The election of the president of the Central Elections Commission seems to depend on the majority in Parliament. Article 17(1) vests the choice of the Vice President and Secretary in a majority of the Commission members. In order to ensure political pluralism, the three above-mentioned officers of the Commission, or at least the Vice President and Secretary should be elected by PR, for instance with the single transferable vote system.


36. Article 16(9). Central Elections Commission Appointment. It may be the translation, but the word ‘proportionally’ seems wrong in Article 16(9). It should probably be equal shares or similar. It is also not clear who nominates the Central Elections Commission president, but that probably means that the Parliament makes its own choice.


37. Article 19. The training of the election commission members should be better ensured (see point II.3.1.d of the Code of good practice in electoral matters). In the explanatory report of the Code of good practice in electoral matters (Paragraph 84), it is advised that “members of electoral commissions have to receive standardised training at all levels of the election administration. Such training should also be made available to the members of commissions appointed by political parties.”


38. Article 20(2). The possibility to dismiss members of the Central Elections Commission for deeds incompatible with their quality could lead to abuses. It would be preferable to provide with an exhaustive list in Article 19(2), including possibly a list of penal and administrative offences.


39. Article 21. What is the gradual reference of first and second rank for Central Elections Commission members?


40. Articles 22 and 26. There are repetitions of tasks in both articles. It would be better to simplify these provisions.


41. Article 25

25(1). The right to call a meeting should not require a majority of the Central Elections Commission. It should require no more than four and preferably three members to request a meeting.

25(2). Sessions should be announced earlier than 48 hours before taking place, insofar as possible.


42. Article 26 c) and Article 27(1). Constitute Electoral Districts. These articles make a difference between constituting electoral districts and constituting district election councils. However, the district election councils shall correspond to administrative units of second level. In that case, why would the Central Elections Commission have to create the administrative units and the district election councils in parallel?


43. Articles 27(1), 29(10). The Central Elections Commissions should also use a system of proportional representation for the appointment of the district electoral councils. This should also be the case for the composition of polling stations by the district electoral councils. The present system (see e.g. Articles 27(3-4), 29(11)) is not very clear.


44. Article 27(7). Decisions of the district electoral councils. This article prescribes that decisions are taken by a majority of its members, not by majority of those voting. This seems to be too strict and is not in line with decisions e.g. in the Central Election Commission. Is that a mistake? It appears to differ from the approach taken in Article 32(2), where decisions of the councils and electoral commission members, if they do require the presence of half the members, only require consent from a majority of the members at the meeting in order to adopt a decision. In summary, it is recommended to clarify the distinction between both Article 27(7) and Article 32(2).


45. Article 29(2). This paragraph explains the possibility of having from 30 to 3000 voters in each polling station. 30 is a very low number, except in very small villages and remote places; categories of localities with a very low number of voters must be specified in the Law. As to the larger polling stations, provisions of the law should guarantee that they are provided with adequate staff.


46. Article 29(12). The election commissions’ members should also be elected by proportional representation to prevent one party or clique from dominating it.


47. Article 32(2). There should be a possibility to vote by secret ballot.


48. Article 33(2). The possibility to dismiss members of electoral commissions at will cast doubts on their independence. It should be deleted.[1]


Electoral lists (Voters registers)


49. The accuracy of the voters’ lists has decreased over the last election. The number of voters entered into the additional lists increased from 6% in 1998 to 10% in 2001 according to OSCE/ODIHR. This problem can only be solved to some extent by legal actions. It depends on the accuracy of a civic register, if one could legally connect the voters’ lists to such register.


50. The minimum is, however, that the voters are given a fair chance to scrutinise the lists for their own as well as for other voters’ entry. In Article 39(1) and Article 40(1) it is said that the lists will be published at least ten days prior to the Election Day. This is too late for making a printed update, but the list can be amended by hand. Article 40(2) does not give a deadline for a challenge, which probably means all the way up to Election Day. It also does not mention challenging other entries than his own. There should be ways of challenging the voters’ lists before they are printed, including other people’s entry as well.


51. Article 39(7) says that the polling station shall issue a certificate to a voter who has moved, but it is not clear if it is the one from where the voter leaves from or where he has moved to.


52. Article 40(2). How can voters claim? Before a court? Before an election commission?


53. Article 53(2). The registration should not take place at the polling station on election day, but the voters should have access to an administrative procedure – subject to judicial control – or to judicial procedure allowing for their registration.[2]



Electoral campaign / Media


54. Article 45(1). The number of representatives of each electoral competitor should be the same.


55. In many new democracies the media have often been biased in favour of the incumbents. Therefore strict rules on election advertisements have been introduced. The effect has often been that the editors have been so scared of being biased that they have not covered the campaign in a critical manner at all, and left the media outlet’s coverage to the parties own commercials transmitted free of charge. Alternatively the editors have covered the incumbents in excess, claiming this is not campaign but covering regular Government business.


56. In Moldova we have mainly observed the first effect, even though there has been violations by giving airtime with campaigning content to incumbents as well. The result is not very informative to voters. This is a symptom of the lack of tradition for independent, critical and neutral editors of public electronic media.


57. The Code of Good Practice in Electoral Matters issued by the Venice Commission covers in its Section I.2.3 the possibility to treat the competitors, according to strict equality in some instances, but ‘proportionally’ in other:


Equality of opportunity should be ensured between parties and candidates and should prompt the state to be impartial towards them and to apply the same law uniformly to all. In particular, the neutrality requirement applies to the electoral campaign and coverage by the media, especially the publicly owned media, as well as to public funding of parties and campaigns. This means that there are two possible interpretations of equality: either “strict” equality or “proportional” equality. “Strict” equality means that the political parties are treated without regard to their present strength in parliament or among the electorate. It must apply to the use of public facilities for electioneering purposes (for example bill posting, postal services and similar, public demonstrations, public meeting rooms). “Proportional” equality implies that the treatment of political parties is in proportion to the number of votes. Equality of opportunity (strict and/or proportional) applies in particular to radio and television airtime, public funds and other forms of backing. Certain forms of backing may on the one hand be submitted to strict equality and on the other hand to proportional equality.”


58. Article 46(1) guarantees each competitor equal right in using mass media. However Article 47(3) uses the term equitable about the private broadcasting of debates etc, but even here it is explicitly stated that the time given to each competitor during a debate has to be equal. The way equality is implemented should be made clearer (concerning in particular the time allocated in mass media: does it mean that each competitor has the same time?).


59. Article 47(11) does give an indication that the press agencies of the Government, Parliament and President should not misuse their allocated broadcasting time for electoral propaganda. However it does not cover the editors’ obligation to be extra cautious in their coverage of the incumbents during elections times. This should be included in the law.


60. Article 47(1), (12). The reference to ethic rules is not precise enough and could lead to abuses. Such imprecise restrictions of the freedoms of expression, assembly and association could run counter Articles 10 and 11 ECHR.


61. Article 47(4) is understood as applying only to free of charge broadcasts, otherwise there would be a contradiction with Article 47(6).


62. However, the main problem in the election campaign in the media in Moldova, as well as in many other countries without a long tradition for a free press, is to provide informative, interesting and unbiased coverage of the campaign. The parties own advertisements are obviously not critical, and do not alone give the voters the information they need. Therefore the questions raised by critical and neutral journalists may offer more information. Article 47(4) goes far in preventing such journalism in the fear of giving an advantage to some competitors. One may argue that such a restrictive line is necessary to prevent abuse by incumbents, and a strict rule should be maintained.


63. On the other hand, a campaign is also about political initiatives, and editors are supposed to cover those. The tendency has been that initiatives by Government or the President have been referred by media without a critical approach or without requesting the opinion by the opposition, referring to it as coverage of normal Government business. This is an issue far beyond an election law, but with the detailed regulation of the current law, there should also be a general provision, which gives the editors a possibility to cover political issues brought up during the campaign, with a strictly neutral and unbiased approach. One possibility would be to issue a more detailed Code of Conduct for editors during election period.


Voting procedures


64. Article 48(1). Ballot. A clause should be added to the effect that the paper used should be opaque, so that the voter’s mark cannot be read or, if this is not done, then voters’ should be given an envelope in which to seal the ballot they present for dropping in the ballot box. The list of candidates of each party should be on the voting paper or at least on boards in the polling station.


65. Article 55(4-6). Mobile voting boxes. This practice from the Soviet era is not followed in established democracies where allegations of fraud or political pressure are far rarer than in the Commonwealth of Independent States. The objections are that it removes the act of voting from full scrutiny of the election commissions’ members; it opens up the opportunity for impersonation of the elector and/or intimidation; and it also takes officials away from their duty at the polling station. The reference at the end of Paragraph 5 must be to Paragraph 6.


Votes’ counting, aggregation and publication of results


66. Article 56 describes the process of the count in the polling station. After having reviewed a high number of protocols from the count, it is safe to say that the entries for number of voters having been issued ballot papers and the number of voters that took part in the elections were not fully understood. The following improvements are recommended:


Article 56(5) should move up to after (2), and be done before opening the ballot boxes, not only before counting the votes.


Article 58 c) and d) should be made clearer, because they have often been misunderstood to mean the same thing. This could be done by adding to c) ‘... according to the voters lists’, and adding to d) ‘... according to the total number of ballots in the ballot boxes. A new ‘j) Number of ballots issued by mobile teams’ should also be added.


There are no rules for reconciliation at the polling station. A rule should be introduced for the case that d) is a higher number than c) (the ballot stuffing situation), e.g. a recount and an entry in the protocol, and another rule for if d) is substantially lower than c) (e.g. recount if there is more than 5% missing, which is less serious).


One should also check that i) + g) + e) is close to h).


58(4). One protocol remaining in the polling station should be posted there immediately.


67. The verification process should, however, be made more clear. The Central Elections Commission will receive all the polling station results. In 1998 these were entered into a spreadsheet for checking of the district electoral councils results, and inaccuracies were found. This verification should be explicitly mentioned, as well as the obligation to make the tabulation of polling station results public. This publication does not have to happen within the five days of the official results, but should definitely happen within reasonable time, e.g. ten days. The purpose of this is for observers and the public to be able to check the polling station they observed was entered correctly in tabulation. It is a crucial element of transparency and can remove a lot of suspiciousness against the tabulation process.


68. A few points are unclear:

Article 57(3): who are the authorised persons?

Article 61(2): what is the end of the electoral period?


69. Partial results are being published and that is good. These would in most cases be accurate, at least when it comes to allocation of seats.


Invalid elections


70. Articles 91, 93(3) and 136 provide the turnout criteria for valid elections. It is a question whether there should be a turnout requirement at all for elections. For referenda such requirement is reasonable since a decision is not a necessity. However, for elections it often ends up in a stalemate, and the next attempts do often give even lower turnouts.


71. In Article 92 it seems like the whole election needs to be annulled, but a more practical approach would be to only annul the parts affected by a fault (at the level of the polling stations).[3] The same is valid for Article 137.


Family voting


72. Family voting is wide spread in Moldova. The conditions for giving assistance to voters as per Article 54(1) should be formalised, for example by a signature in the voters register.




73. The Articles 48(6) and 162(3) refer to the language law. The law seems to offer due protection of both Russian and Gagauz languages in public documents, and therefore also in election material including ballots.


Nomination and registration of candidates


74. Article 41(2). The nomination of candidates belongs first to registered parties and other social-political organisations. It is therefore fundamental for the legislation applying to such registration to make registration easy.


75. Article 43(3). Making the persons collecting signatures liable for the authenticity of the data included could lead to abuse and/or prevent many people from collecting signatures. Those people collecting signatures could be asked to get signatories to produce a valid means of identification. The possible sanctions should be defined in a precise manner and in conformity with the principle of proportionality.


76. Article 43(4). The text appears to be incomplete. It is supposed that the words “are invalid” should be added.


77. Articles 46(6), 80. The possibility to change the lists of candidates up to 5 days before the elections is confusing. In practice, it will be difficult to have the proper list of candidates on the ballot papers. This could also lead to undue pressure. It would be preferable to exclude every possibility to change the lists.


78. Article 78. It seems that all signatures are checked. This is in conformity with point I.1.3.iv of the Code of good practice in electoral matters.




79. Article 37. Many countries do make such cash payments to candidates who receive a pre-established percentage of the vote – but this is a non-returnable grant rather than a repayable loan, as appears to be required in clause 4. A non-returnable grant to candidates / parties with a specified amount of votes would appear to be justifiable state support for competitive elections.


Financial declaration of candidate


80. Article 44(1) e). Information should be required about other members of the candidate’s family too; the Russian Federation’s law could be used as a template.


81. Article 48(3). The order of candidates should be determined by lot rather than the time of registration.


Keeping the electoral documents


82. Article 61(3). There should be a minimum period for retaining electoral documents, such as one year from the date of the election, or a fixed number of months from the expiry of the legal right to mount a challenge - and there should certainly be a clause that all documents must be retained indefinitely as long as there is a legal challenge being heard by the appropriate authorities.


Opinion polls


83. Article 64(4). A high threshold in a proportional representation election makes it desirable for electors to have an idea which parties are likely to pass the threshold and which are not, in order to avoid wasting votes. Reputable public opinion polls conducted according to scientific standards provide such information and WAPOR (World Association of Public Opinion Research) is therefore justified in arguing against a ban. Moreover, a ban encourages rumours about “secret” polls, which may be non-existent, falsely reported or mis-reported.


84. If there is a concern with the reporting of fraudulent results, then the law could stipulate that no results of opinion polls may be published without a statement of the number of persons interviewed, the dates of interviewing, an estimate of the margin of sampling error and that details of the sample be lodged with the Central Elections Commission at the time of publication by the organisation responsible for the survey.


85. Further safeguards against “last minute” announcements of fraudulent results would include a ban on publication of any poll within 48 hours of the opening of election day. That would give critics of last minute polls an opportunity to inspect the sample details filed with the Central Elections Commission and challenge the reliability of results through the media before votes were cast.


Court procedures


86. Chapter 12, Section 1 (Articles 65-68) seems to deal only with appeals to courts. However, Article 65 provides also for appeals to hierarchic superior authorities. This should be made clearer. According to the Code of good practice in electoral matters, point II.3.3.c: “The appeal procedure and, in particular, the powers and responsibilities of the various bodies should be clearly regulated by law… Neither the appellants nor the authorities should be able to choose the appeal body”.


87. Article 89(2-3). The Constitutional Court has to validate the elections within a 10-days term after the documents have been received from the Central Elections Commission. This deadline is quite short and the decisions of ordinary courts will not be final. In order to avoid a denial of justice, the respective roles of the election commissions, the ordinary courts and the Constitutional court should be made clearer.


Liability for violation of electoral legislation


88. Articles 70-71. It is understood that the criminal and administrative offences mentioned in these texts are developed more in detail in the criminal and administrative offence codes.




89. Article 15, but also Section II, Articles 27 and 28, Article 31. Observers should be authorised to assist at electoral authorities’ meetings. Article 15(2): The grounds allowing refusal of the representatives of the electoral competitors should be specified in the law.


90. Article 63(5). Observers’ roles. When observers’ roles are mentioned the word ‘assist’ is being used, e.g. in Article 63(5). This can be misleading, and observe should rather be used.


Sanctions and proportionality, incompatibilities


91. The proportionality principle must be applied to severe restrictions of such a fundamental right.


92. Article 13(1) c) on deprivation of the right to vote can be understood as applying only to persons who have not yet (completely) served their sentence, but it would be suitable to write it explicitly.


93. Article 13(2) c) seems to indicate that anyone who has been recently convicted for any crime – even a small one – cannot stand for elections. This is unusually harsh, if the sentence does not explicitly remove this important civil right.


94. Concerning incompatibilities, there was a long dispute in 1998 on the issue of who would have to withdraw from office after having been nominated as candidate to the Parliament. It seems like this has now been resolved by Article 13(3). However, the present opinion does not analyse the legislation applying to such incompatibilities.




95. Article 88. Article 88 gives the rules for substitutes. It is unclear on what basis the Constitutional Court had objections to the term, but the actual arrangement seems reasonable.


Vote to one candidate
96. Article 10. The translation of this article says that a voter should vote for a single candidate, whereas the correct term must be a single competitor. See e.g. Article 54(4). The same question arises e.g. for Articles 15(2), 37(3), 42(4-5), 60(1) f).


Confirmation of mandates


97. Article 90. Article 90 states that the validation of at least 2/3 of the total number of deputies’ mandates shall be forwarded to the Parliament. There may be a good reason, but it is not obvious why only at least 2/3.


Special limitations of the right to vote


98. Article 123(2): it would be much simpler to provide for the vote at the place of residence.



99. The unification of the whole electoral legislation has to be welcome.
100. Thresholds should be lowered, in order to reduce the number of lost votes. This is especially important for national minorities. If the main threshold, for candidates from one single political party, is lowered from 6 to 5%, minorities would be better represented, and the Parliament would not be composed of only three parties, as since the last elections of 2001. Moreover, independent candidates would also stand a better chance to be elected.
101. By changing the single constituency into a system of local constituencies, geographically concentrated minorities such as the Gagauz would have a fair chance to be represented, even with representatives from a variety of parties/political directions. With such a change, a voter from a concentrated minority would not be forced to choose an ethnically based party to be represented by a person of his/her own ethnic group, but would be able to combine ethnicity and regular political directions in the vote. The individual voter could therefore determine the importance of such ethnic considerations for his/her vote.
102. Other important issues should be improved in the law. This includes:
103. Referendum issues; several provisions should be improved, such as the questions which could be submitted to referendum, or the collection of signatures.

104. Election commissions. It would be suitable that members be elected by proportional representation, instead of majority or plurality of votes; the same solution should be adopted for election of the election commissions’ members. The proportional representation system for taking decisions by members of the district electoral councils should also be changed; voting by secret ballot should be made possible. The training of election commission members should be better ensured; it is also recommended that the possibility to dismiss members be avoided. Distinction between constituting electoral districts and district electoral councils is unclear.


105. Electoral lists. The accuracy of the voters’ lists has to be increased. The law does not provide voters with the possibility to make a claim on electoral lists’ issues.


106. Electoral campaign / Media. It would be suitable to make the law more precise, in order to ensure equality of opportunity between parties and candidates, in particular through equal access to mass media.


107. Counting. The process for counting in polling stations should be improved. Moreover, the verification process should be made clearer.


108. It would be suitable to revise other points which are developed in the Section Other important issues. These notably include: the question of voting procedures, votes’ counting, invalid elections, family voting, nomination and registration of candidates, financial declaration of candidates, court procedures, liability for violation of electoral legislation, observers and sanctions and proportionality.

[1] See Code of good practice in electoral matters, point II.3.1.f.

[2] Code of good practice in electoral matters, point I.1.2.iv.

[3]Code of good practice in electoral matters, point II.3.3.e.