Thursday, January 15, 2015

Possible Distortions In Presidential Elections

By Laksiri Fernando 
The voter registration lists, Mahinda Rajapaksa’s majorities in ten districts and supposed high voter turnouts, all are unfortunately questionable without any prejudice to him or anyone.[i] If I am bias, I am bias for democracy and for a proper election system.

When Sri Lanka went for the controversial presidential elections on 8 January, there were 15,044,490 registered voters on voting lists. This is quite a high number for a population of 20.7 million people.

The Census population figure in 2012 was 20,359,439. The Department of Census gave an estimated mid-year population figure for 2013 as 20,483,000.[ii] This is also the figure the Central Bank gives.[iii] If I add a generous 1 percent increase, then the estimated mid-year population in 2014 would be 20,687,830.

Incredibility

If we take the registered voters as a percentage of the above population figure, it stands as 72.72 percent. This is quite a high figure by all standards.

Under the country’s election laws, only citizens who are 18 years and above can be registered as voters.

If I go through the same source and take the 2013 estimates, under 14 children are 5,431,000 or 26.3 of the 2014 population. The Department of Census doesn’t give an estimate for 15-17 group, but only for 15-19. When we take the necessary fraction, it estimates 994,800 as teenagers between 15 and 17. Therefore, the total ineligible under age group consists around 6,425,800 or 31.1 percent of the population. This also means the eligible population is 14,262,030 or 68.9 percent.

Therefore, how come that 15,044,490 people or 73 percent of the population are registered as voters?

On the basis of the above estimates, 782,460 or 5.5 percent are registered in excess of the eligible age group of population. This may be little less or little more. However, who are they? Who registered them? And for what purpose? Even these initial calculations give rise to serious suspicions when considering other factors. Another country which has this kind of an anomaly is Papua New Guinea.[iv]

Although it is a tall figure, let us assume eligible voters are 73 percent of Sri Lanka’s population. Even then how can you believe that 100 percent of them are (duly) registered? Even in a compulsory voting country, the actual registration does not exceed 75-80 percent. In Australia, it was revealed in a study last year that a quarter of the eligible voters are not registered.[v]

In recent times a high figure of registration (95%) of eligible voters was recorded in Scotland. But that was when a referendum was taken on the independence issue last September. But in the case of Sri Lanka, the figure or even the lists appear in excess of credibility which should be explained or inquired by the Department of Elections. It is true that there can be double entry errors. For example, the same name can appear in two polling or Grama Niladhari areas, after someone moving from one area to the other. There is also a possibility that a newly dead person’s name would still remain in the electoral list. However, these errors cannot even exceed 0.5 percent.

It is true that the Commissioner of Elections declared strict rules in identification and voting procedures. Even a finger of the voter is inked after handing over the ballot paper. However, when large numbers are flocked for voting, the election officers, might not be able to scrutinize the proper identification. I am not here raising the issue of possible stuffing of ballot boxes. My main concern is incredibility of the total number of registered voters.

So-Called Majority?

There are claims that Mahinda Rajapaksa obtained a ‘majority of the majority’ in ten districts that he won. These are Hambantota, Moneragala, Matara, Ratnapura, Galle, Anuradhapura, Kurunegala, Kalutara, Kegalle and Matale, as in Table 1, in the order of the voter percentages that he received.

The majority percentages are between 51.41 and 63.02. As the Table more clearly shows these are by and large the districts where a large percentage of voters happen to turnout. The turnout percentages are between 82.35 and 84.90. There is a clear or high correlation between the two, although it cannot be claimed 100%. I have not attempted a correlation coefficient calculation as my undergraduate statistics knowledge has gone rusty.

There are of course six other districts where voter turnouts were fairly high, but Mahinda Rajapaksa failed to win a majority because, in my opinion, of the tough competition from the opposition, irrespective of fraud/malpractices. These are Polonnaruwa, Colombo, Gampaha, Badulla, Kandy and Nuwaraeliya.

Argument

My present argument about the correlation derives from my main argument that registered voters are incredibly or suspiciously high. There is reason to believe that there were substantial numbers of ‘ghost voters.’ One may ask, if almost all the adults are ostensibly registered, then from where does the ‘ghost voters’ come? Although I don’t have a clear answer, the following are some possibilities. (1) Those who cast the postal vote, can easily impersonate for others, if the time or opportunity permits. (2) The underage also can impersonate like in PNG (if they look older or on the margin) and obviously Nil Balakaya comes to my mind. (3) If the finger ink can be washed, a crook can vote not once but even several times.

There is always a substantial percentage of voters not wanting to take the trouble of voting. I would reckon that this group to be around 20 to 25 percent in Sri Lanka based on the past figures (2005 and 2010). If one wants to impersonate this group, without creating suspicion, the best policy might be to inflate the registered voter numbers or ‘ghost voters.’ Given the procedures of voter registration, this is not a difficult task. But in the present case, it appears the effort has been gone overboard creating suspicion as I and others have raised. So much so it is worse than PNG. In PNG, some years back, around 4 million were registered out of a 6.1 million population although with a high young population like Sri Lanka.

In 2005, the voter turnout was 73.73 percent, largely due to LTTE induced boycott. In 2010, the voter turnout was 74.49. At the recent election, it could have reasonably gone up by around 3 percent due to higher turnout in the North and the East. However instead, the voter turnout has shot up to 81.52.

In my article titled “In Predicting Presidential Elections” (Colombo Telegraph, 6 January 2015) I estimated “2 percent being fraudulently casted or counted” for Mahinda Rajapaksa at the 2010 elections. This time my estimate is 5 percent. On the Election Day morning, responding to Uvindu Kurukulasuriya’s provocative article “Are We Too Late?” (Colombo Telegraph, 8 January 2015) I more concretely predicted saying “I would count the overall political swing to be around 10 percent and this means something like 52 percent for MS, 47 for MR and 1 percent for others.”

What I did not count without available election data at that time is the degree of possible systematic election fraud in 2015. Therefore, I wish to submit that if not for around 5 percent manipulation this time, Maithripala Sirisena would have won not only the 12 districts that he already has, but also the districts of Kurunegala, Anuradhapura, Kalutara, Kegalle and Matale. This is clear from the data in Table 1. Therefore, the claim that President Maithripala Sirisena has not won the ‘majority of the majority’ is not correct. There are some obvious distortions in the election results.

Conclusion

Much has been said about the coercion and violence; and the misuse of state resources, personnel and the state media for the former president’s election campaign. However, the long term subversion of the election processes through controlling of voter registration lists and schemes for impersonation and voter manipulation have not taken our enough attention irrespective of the good work carried out by election monitoring organizations.

Mahinda Rajapaksa has been in executive power since 2005, and the UPFA as a political organization since 1994. It is also my personal knowledge that the election processes were closely monitored, controlled and to a large extent manipulated through political and party apparatuses throughout these years. Therefore, there are major distortions within the electoral processes as a result. This is apart from the distortions in the electoral system in general.


To rebuild democracy in the country these matters have to be looked into in a careful and a systematic manner.

No comments:

Post a Comment