The votes, the referendum and the difference in the Zeros
The votes, the referendum and the difference in the Zeros

This analysis aims to focus our attention on two issues - both directly related to the election process in Bulgaria.

The first one is a conventional, well-known and has already become a cliche, which in turn is described with clichés like "favorite gum in the mouths of politicians", "threat to democracy", etc.

The second is unknown and is also unprecedented in our recent history – I will be very happy if someone manages to refute me but about that - after a moment.

The first problem lies in the burgeoning voter lists.


By law, for each election in Bulgaria, the electoral lists are drawn up by CRAS /Directorate General of Civil Registration and Administrative Services/ and they reflect the number of Bulgarian citizens who are entitled to vote. Before the publication of these lists, all municipal mayors are obliged to send to CRAS a report with the actual number of adult citizens in all municipalities, where these reports should reflect even the deceased citizens through their death certificates, and respectively they should be excluded from the lists. However, in practice, these deceased citizens are often not removed from the lists and this lead to an amazing number of centenarians in the country. On the other hand, a large majority of people appearing in the lists are living abroad but are still registered at a permanent address in Bulgaria - therefore they are entitled a vote.

So far, this explanation shows significant gaps, but still, the system of calculation follows certain logic: all Bulgarians aged 18 or over can vote in elections, young people just turning 18 are being added to the lists, and those citizens who have died are being removed. If we look closer however, at the absolute number of eligible to vote citizens over the years, we will find something illogical - the number varies with large differences, both downwards and upwards. For the purpose of the analysis, I have used only the national elections and have omitted the local ones with all their complications. The data is from the CEC /Central Election Commission/ website:

At one of the elections, the eligible citizens were 6,720,941 (Parliamentary Elections, 2005), the next year their number dropped to 6,445,260 (Presidential Elections, 2006) but at the elections for European Parliament in 2007 (just 7 months later) their number grew again. The largest margin was between the presidential elections in 2006 and the parliamentary elections in 2009. In absolute terms, we speak about 684,705 people, and remarkably, we talk about growth upwards, while at the same time there is a negative population growth rate and mass immigration (the whole graphics, for all national elections and the number of voters can be seen here).

Apparently, no one in Bulgaria knows exactly how many people are eligible to vote in elections and each time the calculations of CRAS are different. The constraint is that the electoral activity depends on the number of voters in the election.

In practice, this can be a problem for two types of elections. When electing a President of the Republic it is required to have a threshold of 50% of eligible voters so that the choice is legitimate in the first round (remember the 2006 elections). The second type of elections is the referendums, where it is necessary to have the votes of 20% of the citizens, in order the issue of the consultation to be sent back to the Parliament. The last referendum showed us vividly exactly that kind of picture, where the fight was to reach the sacred threshold of 20 per cent.

In these two types of activity choices, the main indicator for the legitimacy of the election is 1) if the voters are less than half of the legitimate citizens, we go to the second round, 2) if the voters at less than 20%, then the referendum is invalid.

Obviously, we have a situation of indiscriminately floating number, which however determines the legitimacy of the elections in the country. With such a problem, the hypotheses of backstage action, conspiracies, manipulations and "adjusting the results" is just a matter of asking ourselves a simple question. And the answers to follow depend on the personal imagination, not on simple logic.

What are the possible outcomes of this situation? According to some experts, it is necessary to introduce an active registration for elections in the country. However, in many cases, it is exactly the active registration, which is a prerequisite for discrimination, and besides, all this will probably lead to a drastic decline in those willing to vote, which certainly is not good for the still developing democracy in the country. Or, as my colleague Chavdar Naydenov says about the activity of voting abroad: "If I am at one end of Sweden, I have to go to the other end, where Stockholm is, just to vote, because there is only one section. In theory, nobody has forbidden me to vote, but in practice they have limited my right to do it". Similarly, pre-registration will restrict the majority of citizens. Not to mention the chaos that will occur if it is allowed new names to be added to the lists in the last moment (remember the chaos in front of the polls during the presidential elections, 2011, and multiply it by a random number). Moreover - if there is an active registration, how are we going to calculate the election activity? What will be the dividend or the numerator? For, it is the election activity itself that is the indicator for the legitimacy of the elections and referendums. So, we should fundamentally change the way we choose the president and the way we consult the sovereign.

Apparently, the active registration does not work. So, we should look for another model to update the electoral lists, which is fair and not discriminatory.

And now, to the second issue in this text, which I’ve stated to have unique character. At the referendum on nuclear power, which has just passed, there was no logic in the data of CEC. What do I mean? Basically, N number of voters vote at all elections. How does this happen? Citizens go to their Section Voting Commission /SVC/, they are being found in the list, then receive a voting paper and an envelope (with or without print), leave their ID, fill in the voting paper, place the envelope in the urn, sign in the election documents and get back their ID. That’s it. In other words, the signatures in the voting lists/documents should match exactly the votes, counted in the ballot box. It is another issue that one can express a "protest vote" or something else, dropping an empty envelope (up to a few years back in the past, some political parties were building their whole doctrine on this principle). There have always been invalid voting papers and empty envelopes. Discrepancies between signatures and voices - no.

So far, there has been no case when signatures in the list do not match the envelopes in the urn, and therefore, the counted votes (with one small exception - during the elections for members of the European Parliament in 2007 there was such inconsistency, with 15 units:

However, during the referendum that has just passed, CEC reported 1,415,183 signatures in the election lists and 1,405,463 votes in the polls. In other words, we have excess of 9720 signatures! Or about 10 000 Bulgarians have signed that they had voted, but somehow their voices went missing (data available here:!

What does this mean?

  1. My analysis shows that such disparities are a fact in 22 constituencies in the country.
  2. Only in four of them the difference is in the opposite direction - more votes than signatures, and their values are significantly lower.
  3. The difference is most significant in Haskovo - 2,648 votes or about 6 % of the voters, as well as in Plovdiv-city - 1266 or about 2% of the voters.

What follows from this?

  1. The actual electoral activity at the referendum should be a few tenths of the percent higher. This is very important for the official result, and the possibility that the referendum was considered illegitimate (+ the larger again electoral rolls, with nearly 80,000 more voters than at the presidential elections, 2011).
  2. The disappearance of nearly 10,000 votes is a direct violation of the rule of law and democracy.
  3. This fact speaks for possible manipulation of votes.
  4. Apparently, someone hasn’t done or did not want to do the work properly.
  5. At parliamentary or local elections such an asymmetry would clearly affect the results and the distribution of mandates.

At this point, I have not heard an explanation by CEC yet, of how it was possible for the above to happen. The fact is that this is unprecedented for at least the last 10 years (since results are available on the Internet).

All references to the inconsistencies are available here !

Everything described above sets questions about the fairness of the votes in the last years (do not forget that the 2011 elections were also accompanied by unprecedented scandals and suspicions of manipulation), and about the competence of the electoral administration - both at central and local level.


The next Parliamentary elections in Bulgaria will be held after six months. And the difference between 10,000 and 100,000 voice is obviously just a Zero.


Stefan Georgiev is a member of the team of ISI and of the General Assembly of ISI.


Sociologist by training and profession; works in "AFIS" agency.