«

»

RATING OF A. LUKASHENKO AND HIS RIVALS

Data in Table 1 shows that A. Lukashenko’s open rating dropped down right after the presidential election. At the same time, the number of respondents who found it difficult to answer “for whom they would vote if the presidential election is held tomorrow” increased step-wise. Thus, in March of 2003 the percentage of those who found it difficult to answer this question even exceeded A. Lukashenko’s rating.

Table 1. Dynamics of answer distribution to the question “If a presidential election is held tomorrow in Belarus, for whom would you vote?”, % (open question)

Answers

10’01

04’02

09’02

12’02

03’03

09’03

06’04

11’04

03’05

05’05

09’05

A. Lukashenko

46.0

30.9

27.0

30.5

26.2

31.7

34.2

47.7

46.4

41.7

47.3

Other candidate*

11.8

8.3

6.5

5.0

2.6

4.0

3.0

1.5

1.4

2.0

3.5

DA

24.1

15.7

25.5

18.7

31.9

25.0

23.1

16.5

17.6

22.3

18.1

NA

11.7

16.8

16.2

17.0

14.0

11.2

16.5

18.5

15.0

12.2

10.5

* Given is the highest indicator from all A. Lukashenko’s potential rivals for the forthcoming presidential election mentioned by respondents

After the referendum of 2004, which repealed the Constitution ban for a president in office to run for the third consecutive term, A. Lukashenko’s rating slightly increased and after a minor downfall in spring it boosted up quickly. This most likely happened because previously president’s supporters who believed in supremacy of Constitution couldn’t vote for him and after the referendum their choice simplified. As a result, the percentage of those who found it difficult to answer decreased mirror-like to the growth of president’s rating. Actually, growth of presidential rating became steady due to last summer strengthening of mass propaganda of Belarusian economic achievements supported by certain improvement of respondents’ financial state, A. Lukashenko’s trips around regions, obvious improvement of Belarus-Russia relations and unavailability of alternative information.

The number of respondents who didn’t answer the question of Table 1 right after the campaign of 2001 is about the same as the post-election rating of former candidate for democratic forces V. Goncharik. Later on, their number increased, reached its highest pick in November of 2004 and went down afterwards. (See Picture 1). One might argue on incentives of respondents who refused to answer a topical political question but the size of this group and the fact that it exceeds the highest indicator of A. Lukashenko’s hypothetical rivals in the pre-election period make it important for research. Who of presidential contenders can recruit votes for his/her support? How will members of this group behave on the voting day?


Picture 1. Dynamics of ratings for A. Lukashenko and his rivals, %

In anticipation of the presidential campaign, ratings of the politicians who could make a competition to the acting head of state seem especially interesting. Thus, in September of 2005, 25.5% of respondents stated their readiness in general to vote for a candidate for democratic opposition at the presidential election but the highest indicator of a particular A. Lukashenko’s rival reached only 3.5%. (See Table 1). Also, 10.3% of the polled declared their readiness to trust the decisions of the Congress for Democratic Forces and support the presidential contender it elects while 53.3% said “I’ll see from the circumstances.”

This nation opinion poll was conducted before A. Milinkevich was elected a sole candidate by the Congress and the very fact of his appearance changed disposition of forces in the Belarusian political field. Therefore, we can only state that hypothetical voters presently feel significant trust to this figure which the opposition forces are ready to present as their leader.

It is noteworthy that respondents were ready to support a politician even not knowing his name and they ensured support comparable to the after-election rating of former sole candidate V. Goncharik. We also understand those who want to “see from the circumstances” and therefore refused to buy a pig in a poke.

The advance 10% of support to the sole candidate is still to be won and even then it won’t be enough to win election. Current ratings of the Belarusian politicians very seldom exceed the margin of statistical accuracy. However, their summed rating in the open question shows that a quarter of respondents takes the side of rivals to the current president. This is why the goal of the sole candidate A. Milinkevich on consolidation of the democratic forces should go forth into even a more complicated goal which is consolidation of the electorate looking for changes.

A multiple-choice question simplifies the choice and therefore the number of those who found it difficult to answer is fewer here. (See Table 2). Yet, the ratings of both A. Lukashenko and his hypothetic democratic rival are almost the same as in the answers to an open question.

Table 2. Distribution of answers to the question “If A. Lukashenko runs for presidency for the third time and his opponent is a candidate for the democratic opposition, for whom will you vote?”, %

Variant of answer

%

For A. Lukashenko

47.5

For a candidate for democratic opposition

25.5

For none of them

12.8

DA

13.7

Thus, every seventh chose the “none of them” alternative. In the trampled political field of Belarus, A. Lukashenko’s rival will have to pay attention to these at first sight alien voters. They aren’t supporters of the current president and at the same time they don’t trust the democratic opposition. What urged them to make this very choice? Who are they disappointed in – A. Lukashenko or his rivals, or both? Answers to these questions as well as other questions pertaining to expectations of this group and, which is the most important, correct actions based on this information may help increase the rating of an opposition candidate.

As Table 3 shows, the number of those who believe in fairness and freedom of the forthcoming election is slightly more than the number of skeptics. It is obvious that the ratio of answers to this question to a certain degree reflects the ratio of supporters and opponents to the current political course. However, if the skeptics appear right, the election will not only be not free and unfair but it will be rigged. Are voters ready to stand up their choice? Data in Table 4 may answer this question.

Table 3. Distribution of answers to the question “In your opinion, will the next presidential election be free and fair?”

Variant of answer

%

Yes

44.5

No

39.1

Table 4. Distribution of answers to the question “Are you ready to participate in street actions of protest if the results of the presidential election are falsified?”

Variant of answer

%

Yes

15.5

No

76.0

Thus, only 15.5% of respondents are ready to uphold their choice in the street actions. Is this much or little? Since respondents involved are voters themselves and the number of voters in Belarus is around 7 million people, over 1 million people are ready to participate in the street actions. Such a scale of protest seems impossible in Belarus, although it did take place in early 90-ies. Those protests were economic and not political though.

How true is the intention aired by respondents? Its indirect assessment is given in the answers to the question on A. Lukashenko’s possible displacement through a “colored revolution” and readiness of respondents to participate in such a revolution. It appears that 14.5% of the polled believe in success of such a revolution and only 9.7% of them are ready to take part in it.

Consequently, the following can be concluded from above:

– for the three summer months, president’s rating has increased by 5.6 points;

– A. Lukashenko’s probable contenders at the presidential election and especially a sole democratic candidate have an advance of trust among voters standing for changes;

– respondents demonstrate high degree of readiness to stand up their choice in case the election is rigged but they scarcely believe in the possibility of a colored revolution in Belarus.