I have been asked many questions with regard to the 3.1 million potential non-resident voters which I first highlighted at a press conference together with members of the Bersih steering committee on April 23. I subsequently wrote about this issue in greater detail here.
In response to these questions, I would like to make the following points:
1. That not all 3.1 million of these potential non-resident voters are dubious voters.
2. There are many reasons why voters are not registered in the same constituency indicated by their IC address.
3. That one of the reasons is because of politicians registering their supporters in their respective constituencies even though these voters do not and have never lived in these constituencies.
4. A manifestation of this practice is the presence of addresses with many voters registered in them or voters registered without house numbers.
5. That the Election Commission (EC) has decided to move to a new registration system where a new/transferred voter could only be registered in the constituency indicated by his/her IC address.
6. That the EC still refuses to acknowledge the problem of non-resident voters as well as the problem of many non-resident voters being registered in the same address.
7. The example of the locality where Mimos (Malaysian Institute of Microelectronic Systems) ‘misinterpreted' the electoral data has a large number of newly-registered Kod 71 voters.
8. It is difficult for politicians and the EC to locate voters without house numbers, especially in the urban areas.
Not all are dubious voters but...
I have never said that all these 3.1 million potential non-resident voters are dubious voters.
What I wanted to highlight was that there were 3.1 million voters in Peninsular Malaysia whose IC address did not correspond with their voting constituency and that these voters were identified in a National Registration Department (JPN in BM) project called ‘Projek SPR'.
For those who doubt that I have access to this data, please see screenshots of the 11 CDs representing data for each state (right).
In response to these questions, I would like to make the following points:
1. That not all 3.1 million of these potential non-resident voters are dubious voters.
2. There are many reasons why voters are not registered in the same constituency indicated by their IC address.
3. That one of the reasons is because of politicians registering their supporters in their respective constituencies even though these voters do not and have never lived in these constituencies.
4. A manifestation of this practice is the presence of addresses with many voters registered in them or voters registered without house numbers.
5. That the Election Commission (EC) has decided to move to a new registration system where a new/transferred voter could only be registered in the constituency indicated by his/her IC address.
6. That the EC still refuses to acknowledge the problem of non-resident voters as well as the problem of many non-resident voters being registered in the same address.
7. The example of the locality where Mimos (Malaysian Institute of Microelectronic Systems) ‘misinterpreted' the electoral data has a large number of newly-registered Kod 71 voters.
8. It is difficult for politicians and the EC to locate voters without house numbers, especially in the urban areas.
Not all are dubious voters but...
I have never said that all these 3.1 million potential non-resident voters are dubious voters.
What I wanted to highlight was that there were 3.1 million voters in Peninsular Malaysia whose IC address did not correspond with their voting constituency and that these voters were identified in a National Registration Department (JPN in BM) project called ‘Projek SPR'.For those who doubt that I have access to this data, please see screenshots of the 11 CDs representing data for each state (right).
There are many reasons to explain these potential non-resident voters. Table 1 lists three possible reasons according to three categories.
This categorisation is not exhaustive since there are many other permutations and combinations which can be added. The key here is to identify the voters who can be classified as non-resident voters.
This categorisation is not exhaustive since there are many other permutations and combinations which can be added. The key here is to identify the voters who can be classified as non-resident voters.
Take a case under Category 1. A voter's hometown is in Ijok, Kuala Selangor. He then moves to Section 16 under the Petaling Jaya Selatan parliamentary constituency in Selangor. He registers as a voter here but his IC address still remains in Ijok. This voter is a resident voter.
This is the example which describes Selangor Menteri Besar Khalid Ibrahim whose IC address is in Ijok but whose voting constituency was in Petaling Jaya Selatan (now transferred to Lembah Pantai after the EC's boundary correction exercise).
Take a case under Category 2. A voter's hometown is in Kota Bharu, Kelantan. He registers as a voter in the Kota Bharu parliamentary seat. He then moves to Bangi, Selangor. He changes his IC address to Bangi. But he maintains his voting constituency in his hometown seat of Kota Bharu. This voter is a resident voter.
Voters in categories 1 and 2 are considered resident voters because they were residents in their respective constituencies at the time of their registration as voters. Even if they have subsequently moved away from these constituencies, they are still considered as resident voters as per Article 119 of the Federal Constitution.
Please note that voters who have moved from one place to another but still maintains their IC address and voting constituency in their hometowns are not included among the 3.1 million non-resident voters identified by JPN. For example, the ‘Project SPR' databases do not include the many Kelantanese who have moved to the Bangi area but whose IC address and voting constituencies are still in Kelantan.
Hence, it would be wrong, as this blogger has done, to say that we should not be concerned about this 3.1 million voters because there were still 1.1 million Malaysians who were still holding on to their old Malaysian ICs in 2005 or the fact that there were 3.6 million intra-state migrants according to the 2000 population.
The 1.1 million Malaysians who still had old ICs in 2005 would not have been included among the 3.1 million if their IC address was the same their voting constituency.
Just because one upgrades from the old IC to a new IC does not mean that the person's IC address would have changed. In addition, some of the 3.6 million intra-state migrants in 2000 would have maintained their voting constituency according to their IC address. Just because they have moved, does not mean that they would have changed their voting constituency or their IC address.
Inter- and intra-state migration certainly increases the complexity in defining and understanding the problem of non-resident and resident voters and I appreciate the efforts of this blogger for pointing it out.
However, it is voters in category 3 which I am most concerned about since these are non-resident voters who have never lived in the constituencies they were registered in 2002 when the Project SPR exercise was completed. Those in category 3 are those which I would define as dubious voters.
If it wasn't broke, why fix it?
It was widely acknowledge that political parties and politicians on both sides of the political divide were responsible for registering supporters in their own constituencies so as to boost their chances of retaining or winning these constituencies even though these voters did not live in these constituencies.
Standards of ‘proving' that a voter who wanted to register in a particular constituency actually lived in that constituency were rather lax. Some sort of bill (telephone, electricity) of a person's name and address was all that was required.
In many cases, even this requirement was waived. There are also many voters who were allowed to register without indicating their house number even though they were registered in urban areas.
The EC themselves tacitly acknowledged this problem when it decided to move to a new registration system where a voter could only be registered at constituency indicated by his or her IC address. After all, if the previous system was working fine, why change the registration criteria in 2002?
The results of ‘Projek SPR' which was carried out by JPN prior to the change of the registration system in 2002 validated these concerns since it revealed that as many as 37% (3.1 million out of 8.2 million) voters in Peninsular Malaysia at that time had IC addresses that was different from the voting constituencies.
Even if a third of these voters were dubious voters, i.e. those in Category 3 in Table 1 above, it would mean that one million voters were registered in constituencies which they have never lived in.
Without a further audit, we simply do not know the percentage of voters according to the different categories listed in Table 1. To my knowledge, the EC has never tried to find out, even after being given this data by the JPN back in 2002.
Many voters in one address
A clear manifestation of the problem of non-resident voters is the presence of house addresses with many registered voters since politicians - before 2002, when there was less detailed scrutiny on the electoral roll - would regularly register their supporters in a limited number of addresses in their respective constituencies.
This is the example which describes Selangor Menteri Besar Khalid Ibrahim whose IC address is in Ijok but whose voting constituency was in Petaling Jaya Selatan (now transferred to Lembah Pantai after the EC's boundary correction exercise).
Take a case under Category 2. A voter's hometown is in Kota Bharu, Kelantan. He registers as a voter in the Kota Bharu parliamentary seat. He then moves to Bangi, Selangor. He changes his IC address to Bangi. But he maintains his voting constituency in his hometown seat of Kota Bharu. This voter is a resident voter.
Voters in categories 1 and 2 are considered resident voters because they were residents in their respective constituencies at the time of their registration as voters. Even if they have subsequently moved away from these constituencies, they are still considered as resident voters as per Article 119 of the Federal Constitution.
Please note that voters who have moved from one place to another but still maintains their IC address and voting constituency in their hometowns are not included among the 3.1 million non-resident voters identified by JPN. For example, the ‘Project SPR' databases do not include the many Kelantanese who have moved to the Bangi area but whose IC address and voting constituencies are still in Kelantan.
Hence, it would be wrong, as this blogger has done, to say that we should not be concerned about this 3.1 million voters because there were still 1.1 million Malaysians who were still holding on to their old Malaysian ICs in 2005 or the fact that there were 3.6 million intra-state migrants according to the 2000 population.
The 1.1 million Malaysians who still had old ICs in 2005 would not have been included among the 3.1 million if their IC address was the same their voting constituency.Just because one upgrades from the old IC to a new IC does not mean that the person's IC address would have changed. In addition, some of the 3.6 million intra-state migrants in 2000 would have maintained their voting constituency according to their IC address. Just because they have moved, does not mean that they would have changed their voting constituency or their IC address.
Inter- and intra-state migration certainly increases the complexity in defining and understanding the problem of non-resident and resident voters and I appreciate the efforts of this blogger for pointing it out.
However, it is voters in category 3 which I am most concerned about since these are non-resident voters who have never lived in the constituencies they were registered in 2002 when the Project SPR exercise was completed. Those in category 3 are those which I would define as dubious voters.
If it wasn't broke, why fix it?
It was widely acknowledge that political parties and politicians on both sides of the political divide were responsible for registering supporters in their own constituencies so as to boost their chances of retaining or winning these constituencies even though these voters did not live in these constituencies.
Standards of ‘proving' that a voter who wanted to register in a particular constituency actually lived in that constituency were rather lax. Some sort of bill (telephone, electricity) of a person's name and address was all that was required.
In many cases, even this requirement was waived. There are also many voters who were allowed to register without indicating their house number even though they were registered in urban areas.The EC themselves tacitly acknowledged this problem when it decided to move to a new registration system where a voter could only be registered at constituency indicated by his or her IC address. After all, if the previous system was working fine, why change the registration criteria in 2002?
The results of ‘Projek SPR' which was carried out by JPN prior to the change of the registration system in 2002 validated these concerns since it revealed that as many as 37% (3.1 million out of 8.2 million) voters in Peninsular Malaysia at that time had IC addresses that was different from the voting constituencies.
Even if a third of these voters were dubious voters, i.e. those in Category 3 in Table 1 above, it would mean that one million voters were registered in constituencies which they have never lived in.
Without a further audit, we simply do not know the percentage of voters according to the different categories listed in Table 1. To my knowledge, the EC has never tried to find out, even after being given this data by the JPN back in 2002.
Many voters in one address
A clear manifestation of the problem of non-resident voters is the presence of house addresses with many registered voters since politicians - before 2002, when there was less detailed scrutiny on the electoral roll - would regularly register their supporters in a limited number of addresses in their respective constituencies.
As a result some of these houses have Malay, Chinese and Indian registered voters who in all likelihood do not even know of each other's existence. In some cases, some of these voters were not even allocated a house number but was registered using a street name.
What is both sad and troubling is the fact that the EC has never done anything, at least publicly, to find out the extent of this problem, much less do something about it. It is not rocket science.
Anyone armed with even a rudimentary command of Excel could go through an electoral roll and highlight houses with more than X number of voters registered in them.
Mimos was asked to identify possible problems in the electoral roll as part of the Parliamentary Select Committee on Electoral Reform. Their analysis revealed 324 addresses with more than 100 voters and another 938 addresses with between 51 to 100 addresses.
The PSC recommended that the EC publishes the names of all of the voters in these addresses so that these voters can step forward to verify their identity and for the EC to report back, in 45 days, to a specially set up parliamentary select committee which will oversee the cleaning up of the electoral roll.
What was the EC's response? In a booklet entitled 'The Truth Behind the Accusations and Lies towards the Election Commission' (henceforth referred to as ‘The Truth'), under Allegation 6, the EC responded by saying that Mimos "had misunderstood the large number of electors in a locality, so as to mean that all the electors in that locality are registered under one address".
The example given by the EC was the locality of Kampung Melayu Majidee in the parliament seat of Johor Bahru, which was assumed to have just one address with more than 50 voters. The EC said that "In reality, this locality has tens of complete addresses, as well as incomplete addresses that have all been combined to form a locality."
What is stated by the EC is technically correct. When examining in Quarter 3 (Q3) 2011 electoral roll, I found 96 names in the locality named "Kg Melayu Majidee" (Locality Code: 1604407014). 74 of these voters had complete addresses, 22 others did not.
In Table 2, I show 30 voters from this locality with complete house addresses, including house numbers.
What is both sad and troubling is the fact that the EC has never done anything, at least publicly, to find out the extent of this problem, much less do something about it. It is not rocket science.
Anyone armed with even a rudimentary command of Excel could go through an electoral roll and highlight houses with more than X number of voters registered in them.
Mimos was asked to identify possible problems in the electoral roll as part of the Parliamentary Select Committee on Electoral Reform. Their analysis revealed 324 addresses with more than 100 voters and another 938 addresses with between 51 to 100 addresses.
The PSC recommended that the EC publishes the names of all of the voters in these addresses so that these voters can step forward to verify their identity and for the EC to report back, in 45 days, to a specially set up parliamentary select committee which will oversee the cleaning up of the electoral roll.
What was the EC's response? In a booklet entitled 'The Truth Behind the Accusations and Lies towards the Election Commission' (henceforth referred to as ‘The Truth'), under Allegation 6, the EC responded by saying that Mimos "had misunderstood the large number of electors in a locality, so as to mean that all the electors in that locality are registered under one address".The example given by the EC was the locality of Kampung Melayu Majidee in the parliament seat of Johor Bahru, which was assumed to have just one address with more than 50 voters. The EC said that "In reality, this locality has tens of complete addresses, as well as incomplete addresses that have all been combined to form a locality."
What is stated by the EC is technically correct. When examining in Quarter 3 (Q3) 2011 electoral roll, I found 96 names in the locality named "Kg Melayu Majidee" (Locality Code: 1604407014). 74 of these voters had complete addresses, 22 others did not.
In Table 2, I show 30 voters from this locality with complete house addresses, including house numbers.
Foreigners in the house?
At the same time, I also found a list of 22 names in the same locality which did not have any house numbers in the same locality. (See Table 3 below)
A few points of concern immediately jumps out when comparing Tables 2 and 3.
Firstly, there are indeed many identifiable roads in this particular locality, according to Table 2. If this area is relatively developed and it should be since it is in the Johor Bahru constituency, then one wonders why is it that there are no roads as well as house numbers attached to the voters in Table 3.
Secondly, the voters without any house numbers in Table 3 were all registered in May and June 2011. One would have assumed that their ICs would give the full address of these voters given that this place is in the urbanised seat of Johor Bahru.
Thirdly, and more disconcerting, all of the voters in Table 3 have Kod 71 in their ICs indicating that they were not born in Malaysia. (Looking at their names, it is likely that they are of Indonesian ‘heritage' e.g. Nur'asiyah). By contrast, note that none of the voters in Table 2 have Kod 71 in their ICs.
Having aroused my suspicion, I then proceeded to examine newly-registered voters for this specific locality for Q4 2011 and Q1 2012. The results are show below in Table 4 and 5.
Secondly, the voters without any house numbers in Table 3 were all registered in May and June 2011. One would have assumed that their ICs would give the full address of these voters given that this place is in the urbanised seat of Johor Bahru.
Thirdly, and more disconcerting, all of the voters in Table 3 have Kod 71 in their ICs indicating that they were not born in Malaysia. (Looking at their names, it is likely that they are of Indonesian ‘heritage' e.g. Nur'asiyah). By contrast, note that none of the voters in Table 2 have Kod 71 in their ICs.
Having aroused my suspicion, I then proceeded to examine newly-registered voters for this specific locality for Q4 2011 and Q1 2012. The results are show below in Table 4 and 5.
According to Table 4 (above), there a total of 32 voters were in the Q4 2011 electoral roll update in this locality. Of these 32 voters, 30 were newly-registered voters and all of them do not have house addresses and all of them have Kod 71 in their ICs.
The only voter with a house address (No 24) was taken off this electoral roll because she has moved to another constituency. The only non-Kod 71 voter added to the electoral roll is not a newly-registered voter but who move to this voting constituency.
According to Table 5 (above), there were six newly-registered voters in this locality in the Q1 2012 electoral roll update. Two voters, both of whom are not Kod 71 voters, have house numbers. The other four voters, who are Kod 71 voters, all do not have house addresses.
To recap, what we have in this locality, which was highlighted in the EC's recent booklet to tell "the truth", is a situation, as for Q1 2012, where of the 134 registered voters in this locality, 74 voters have complete addresses (house numbers and street names), none of whom are Kod 71 voters, while the remaining 57 voters do not have house numbers or even street names and all except one are Kod 71 voters who have been newly registered in the past one year.
The EC, rather than giving reasons on why Mimos had misinterpreted the electoral roll data, should be extremely concerned that Mimos found more than 50 names in this locality without house addresses in an urban locality.
What is more, they should have done exactly what I did here, which was to examine the voters who did not have house numbers and would have found that all of them were recently registered and with Kod 71 in their ICs.
This should have prompted them to do a serious audit in this area to determine if the ICs of these voters have been legitimately issued and why so many of them are staying in one locality without any house number. Instead, the EC choose to look the other way by giving excuses.
This is just one locality. What about the other 1,261 localities with more than 50 registered voters? What would we find if the EC reveals what these localities are?
Locating voters without house numbers
The EC may be right in saying that there are voters who live in kampongs who do not have house addresses. But there are probably as many voters who live in semi-urban and urban areas where there are house numbers for each locality or street, but whose details do not show any house numbers.
What this means it is very difficult for a candidate who is running for office or an incumbent MP or Adun (state assemblyperson) to identify these voters to ascertain if they are indeed resident voters (or were resident voters).
For example, 22 voters from a locality in urban Petaling Jaya Utara (Locality Name: Jalan 1 Kg Bahru, Sg Way (Jalan 1-25), Locality Code: 1063510001) are shown in Table 6 below. All of them do not have house numbers and all of them were registered before 2002.
Malays, Chinese and Indians are all represented in this list of voters. Without the house numbers, how is one supposed to locate these voters?
If someone wants to object to the inclusion of some of these voters into the electoral roll during the quarterly updates (for newly-registered voters without house addresses), how will the EC send letters to these voters so that they can turn up to during the verification hearing?
These technical but important questions were not answered by the EC in their ‘truth' booklet.
What is more, they should have done exactly what I did here, which was to examine the voters who did not have house numbers and would have found that all of them were recently registered and with Kod 71 in their ICs.
This should have prompted them to do a serious audit in this area to determine if the ICs of these voters have been legitimately issued and why so many of them are staying in one locality without any house number. Instead, the EC choose to look the other way by giving excuses.
This is just one locality. What about the other 1,261 localities with more than 50 registered voters? What would we find if the EC reveals what these localities are?
Locating voters without house numbers
The EC may be right in saying that there are voters who live in kampongs who do not have house addresses. But there are probably as many voters who live in semi-urban and urban areas where there are house numbers for each locality or street, but whose details do not show any house numbers.
What this means it is very difficult for a candidate who is running for office or an incumbent MP or Adun (state assemblyperson) to identify these voters to ascertain if they are indeed resident voters (or were resident voters).
For example, 22 voters from a locality in urban Petaling Jaya Utara (Locality Name: Jalan 1 Kg Bahru, Sg Way (Jalan 1-25), Locality Code: 1063510001) are shown in Table 6 below. All of them do not have house numbers and all of them were registered before 2002.
Malays, Chinese and Indians are all represented in this list of voters. Without the house numbers, how is one supposed to locate these voters?
If someone wants to object to the inclusion of some of these voters into the electoral roll during the quarterly updates (for newly-registered voters without house addresses), how will the EC send letters to these voters so that they can turn up to during the verification hearing?
These technical but important questions were not answered by the EC in their ‘truth' booklet.
Tip of the iceberg
To recap, I started this article by pointing out that not all of the 3.1 million potential non-resident voters are dubious voters. Under the previous voter registration, it was relatively easy to register non-resident as voters and that even if one-third of these 3.1 million voters are non-resident, it would constitute one million voters.
One manifestation of non-resident voters can be found in houses with many registered voters (10 is usually a good cut-off point) or urban and semi-urban localities with many voters without house addresses.
The EC thought that this problem could be reduced by forcing everyone to register under their IC address. This change was introduced in 2002. But even after this change, the ‘legacy' issue of non-resident voters has not been resolved as seen by the voters without house addresses in one locality in Petaling Jaya Utara in Table 6.
More worryingly, in the locality which the EC used as an example of how Mimos ‘misinterpreted' the electoral roll data, 55 out of 57 newly registered voters since May 2011 did not have house addresses and 54 out of these 55 voters had the Kod 71 in their IC numbers, indicating that they were not born in Malaysia.
This may just be the tip of the iceberg. If the EC is willing to release the names of the other 1,261 localities with more than 50 registered voters, my team at the Malaysian Electoral Roll Analysis Project (Merap) are more than willing to analyse these localities and present the results for public consumption.
More worryingly, in the locality which the EC used as an example of how Mimos ‘misinterpreted' the electoral roll data, 55 out of 57 newly registered voters since May 2011 did not have house addresses and 54 out of these 55 voters had the Kod 71 in their IC numbers, indicating that they were not born in Malaysia.
This may just be the tip of the iceberg. If the EC is willing to release the names of the other 1,261 localities with more than 50 registered voters, my team at the Malaysian Electoral Roll Analysis Project (Merap) are more than willing to analyse these localities and present the results for public consumption.
ONG KIAN MING is the project director of the Malaysian Electoral Roll Analysis Project (Merap) and a lecturer and political analyst at UCSI University. He can be reached at im.ok.man(at)gmail.com.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.