Mr Speaker, Sir,
I beg to move, “That this House calls on the Government to review the process by which electoral boundaries are determined in order to increase the transparency and fairness of the electoral boundary review process for all political parties.”
The Progress Singapore Party (PSP) had earlier asked for Group Representation Constituencies (GRCs) to be abolished. Our proposals in this debate are based on accepting the existence of GRCs as reality and how we can make things better within that reality and is not an indication of a change of position on GRCs.
Sir, the process by which electoral boundaries are reviewed in Singapore has been long-standing. The Prime Minister appoints an Electoral Boundaries Review Committee (EBRC) and sets its terms of reference. Once the EBRC publishes its report with the new electoral boundaries, Parliament is dissolved and a General Election (GE) held shortly after.
However, the Prime Minister does not have to reveal when the EBRC will be formed or whether he has formed the EBRC. On previous occasions, the Prime Minister has only revealed to the public that the EBRC has been formed months after he has convened the EBRC.
In the interests of transparency and public accountability, I would first like to start by asking the Prime Minister whether he has formed the EBRC.
PSP has raised this Motion today because we believe that there are many areas in this process which can be reformed, so that electoral boundaries are reviewed in a manner that incorporates transparency, accountability, independence, fairness and respect for voters.
One, justify boundary changes for transparency and accountability. The EBRC needs to make public the reasons why the boundaries of certain constituencies are changed while others remain untouched.
Over the years, the reports of the EBRC have become shorter and shorter, providing less and less explanation of why the boundaries have changed. The reports of the 2011, 2015 and 2020 EBRCs were only four pages long and contained six paragraphs. For example, in the 2020 report, the EBRC merely stated, “The Committee reviewed the boundaries of the current electoral divisions taking into account the current configurations, changes in the number of electors due to population shifts and housing developments.”
This is in stark contrast to the early years of Independence. The report of the 1971 EBRC, for example, was 12 pages long and contained 20 paragraphs, and provided more details for its decisions than present-day EBRCs. For example, the 1971 EBRC explained clearly the reasons why it did not alter the boundaries of constituencies in the central areas of the city. It even explained the names for new constituencies.
Many constituencies now have strange boundaries that do not follow the boundaries of Housing and Development Board (HDB) towns or Urban Redevelopment Authority (URA) planning areas, or any relation to how people on the ground live their lives. In 1971, the EBRC clearly stated that “all changes and adjustments were made in keeping with proximity, identity of interest and similarity of association”. These do not appear to be considerations for the EBRCs of today. It is common to find residents of the same HDB town belonging to different constituencies.
I still remember when I first entered politics more than a decade ago, a veteran in an opposition party told me how hard he walked the ground in a Single Member Constituency (SMC) for four years, only to have that SMC disappear on him when boundaries were reviewed and the SMC was absorbed into a GRC. Is this an isolated incident or is there a pattern?
We examined the data on SMCs from GE1988 to GE2015, spanning seven GEs. Excluding opposition-held SMCs at the point of boundary review, we find that amongst the SMCs where opposition scored below 40% of the votes, on average, 58% were retained as SMCs. In comparison, where opposition scored between 40% and 50% of the votes, only 33% were retained on average. SMCs in GE2020 cannot be included here because the decision on whether they will be retained is yet to be made. We hope that these statistics will not be maintained or further aggravated by the next boundary review.
Without clear justifications for its decisions, it is difficult to tell whether the EBRC has changed the boundaries simply because of population shifts and housing developments, or for other reasons. Suspicions of gerrymandering naturally surface and erode public trust.
The number of voters in each constituency is often used as the justification for boundary changes. The average number of voters per Member of Parliament (MP) is not codified into law. In 2020, the EBRC worked with a ratio of one MP to about 29,000 voters. Since 1980, the EBRC has also allowed the average number of voters per MP to deviate by plus or minus 30%, up from 20% previously. This is very wide and creates the possibility that a constituency could be almost twice as large as another and still return the same number of MPs. For example, an SMC could have 20,000 to 38,000 voters while a five-member GRC could have 100,000 to 190,000 voters. This is unequal representation and unfair to voters.
Let me illustrate with some real examples in the last GE. In 2020, Sengkang GRC had 120,100 voters and elected four MPs, while East Coast GRC had 121,644 voters, only slightly more, but elected five MPs. Yuhua SMC had 21,351 voters, while Bukit Panjang SMC had 35,437 voters. This means that the votes of Sengkang and Bukit Panjang voters counted for less compared to East Coast and Yuhua voters.
In 1963, the EBRC adopted a rule allowing the numbers of voters in each electoral division to differ by no more than 20%. The permitted deviation was increased to 30% in 1980. This deviation should be narrower so that each electoral divisions has approximately equal numbers of voters. This is important as the electors’ votes need to carry the same weight. The electors’ votes should be respected and not be unduly diluted by having larger deviation in each electoral division.
The average number of voters per MP and the allowable deviations should be fixed into law. In the United Kingdom (UK), for example, under the Parliamentary Constituencies Act 1986, there must be 650 MPs and the average number of voters per MP can only deviate by plus or minus 5%. In Australia, the average number of voters per MP can only deviate by plus or minus 10% under the Commonwealth Electoral Act (No 2) 1973.
Putting aside the issue of unequal workload for each MP, we should adopt tighter limits on deviation to ensure that every Singaporean’s vote carries the same weightage and is represented fairly in this House.
In the interest of transparency, accountability and fairness to voters, PSP proposes the following changes.
One, upon formation of EBRC, announce the formation immediately and publish the current number of voters in each constituency based on boundaries in the last GE.
Further, identify the constituencies that require boundary changes based on fixed criteria.
First, constituencies with the number of voters falling outside the specified range can have its boundaries changed. The current range of 20,000 to 38,000 per MP is too wide, with the maximum number being 90% higher than, or nearly twice, that of the minimum. This means that some MPs are taking the load of nearly twice that of other MPs. This is not an ideal situation. PSP proposes that one MP should represent 30,000 voters, with a deviation of plus or minus 10%, such that each MP represents 27,000 to 33,000 voters. As such, the maximum is limited to 22% higher than the minimum. This is a fairer distribution of duties amongst MPs and fairer to voters in terms of their voting weightage.
Second, the boundaries of constituencies with number of voters within specified range cannot be changed unless justified on the basis of necessity due to an adjacent constituency having voters falling outside that specified range. The basis for such changes must be published.
Two, respect voters by reducing the potential for gerrymandering. Gerrymandering is an attempt to manipulate the outcome of elections. It is an act of disrespect for voters. Instead of working harder to make things better, serving the people better and being more responsive to the needs of the voters, gerrymandering seeks the easy way out of changing the rules. The genuine concerns of voters are not addressed and respect for voters is eroded. This is not acceptable and we should reduce the potential for gerrymandering in our system.
To achieve this, PSP makes the following proposal. Just like houses and buildings have structural walls that cannot be demolished in renovations and non-structural walls that can be demolished, we propose the introduction of major boundaries and minor boundaries.
Major boundaries are meant to remain unchanged for many elections. Each region bound by major boundaries could contain one or two GRCs and a few SMCs. These major boundaries should correspond closely to local ties and geographical considerations and could take reference from existing HDB town boundaries, URA planning area boundaries and postal district boundaries. We propose that these major boundaries cannot be changed unless agreed to by a bipartisan committee. I will elaborate on this later. The introduction of major boundaries that will not change frequently also make it easier for MPs to build community identities and bond.
Minor boundaries lie within the major boundaries and can be changed by EBRC based on the criteria mentioned earlier. Let me illustrate with an example. Suppose that a region enclosed by major boundaries is as follows: the number of voters was 200,000 in the previous GE and there was a five-member GRC with 140,000 voters plus two SMCs with 30,000 voters each.
Suppose that before the next GE, the number of voters changed as follows: (a) the number of voters grew to 220,000; (b) the number of voters in the GRC grew to 166,000 and the two SMCs dropped to 27,000 voters each. Under the proposed system, the boundaries of the two SMCs need not be changed as they are still within the range. But the boundaries for the GRC must change as it has gone beyond 33,000 per MP. The GRC must now be served by six MPs instead of five.
By altering the minor boundaries, EBRC can change the GRC to either a five-member GRC + one SMC; or a four-member GRC + two SMCs; or a three-member GRC + three SMCs; or two three-member GRCs. The justification for the option chosen should be provided by EBRC.
Three, ensure political fairness by imposing a minimum timeframe from finalising boundaries to GE. We propose a minimum period of three months between the publishing of EBRC report to the dissolution of Parliament. This will ensure a more level playing field by giving all political parties more time to prepare for the GE.
In response to a Parliamentary Question (PQ) earlier, Minister Chan Chun Sing had assured Parliament that “there will be sufficient time between the release of EBRC’s report to the time of election.” Does the Minister agree that a minimum of three months is a very reasonable time frame?
Four, enhancing independence in the composition, appointment and reporting structure of EBRC. In our system, the Prime Minister has full control over the electoral boundary review process. This starts with who gets to be appointed to EBRC. There is no law stating who gets to be appointed to EBRC, even though in recent years, EBRC has always consisted of the Cabinet Secretary, the chief executive officers (CEOs) of the Elections Department (ELD), HDB and the Singapore Land Authority (SLA), and the Chief Statistician. These are senior civil servants with deep domain knowledge that enable them to make decisions on how to redraw electoral boundaries in a practical way and we have no objection to them sitting on EBRC.
There are concerns about potential conflict of interest for EBRC to report to the Prime Minister. In the interest of enhancing the independence of EBRC, PSP proposes that the membership of EBRC should be codified into law to prevent any future abuses. In the UK, for example, the Parliamentary Constituencies Act 1986 specifies that the Boundary Commissions are chaired by the Speaker of the House of Commons and the deputy chair must be a High Court Judge. Some legal constraints are necessary to prevent the appointment of partisan members onto the committee. PSP proposes that EBRC be chaired by a High Court Judge with four other members being the CEOs of ELD, HDB and SLA, and the Chief Statistician.
If EBRC is proposing changes to the major boundaries, then these changes must be approved by a committee to be chaired by the Speaker of Parliament by right of his office, but he will not play any part in the review. The other members of the committee will be appointed by the President and comprise equal numbers of representatives from: one, the ruling party; and two, the other political parties that have won at least 5% of the votes at the previous GE. This committee could be bipartisan or multi-partisan. For the sake of simplicity, I shall refer to this committee as “the bipartisan committee” within this debate. If the EBRC’s proposed electoral boundary changes do not impact the major boundaries, then these changes do not need to be approved by a bipartisan committee.
At the end of the process, the changes will be compiled into a report by EBRC and submitted to Parliament. The report by the Boundary Commission for England contained a paragraph assuring that election results are not taken into consideration in the review of boundaries. We should do the same to enhance public trust. The changes will be approved and implemented without modification by Parliament to ensure that the process remains fair and transparent.
In summary, PSP’s proposal seeks to:
(a) increase transparency and accountability by: one, requiring the formation of EBRC to be publicly announced immediately; two, requiring EBRC to, upon its formation, publish the current number of voters in each constituency based on boundaries in the last GE; and three, requiring EBRC to publicly and sufficiently explain all changes;
(b) ensure fairness to voters by narrowing the range of number of voters per MP;
(c) respect voters and reduce the potential for gerrymandering by: one, implementing fixed rules through major and minor boundaries; and two, requiring changes of major boundaries to be approved by a bipartisan committee;
(d) increase political fairness by imposing a minimum timeframe from finalising boundaries to GE; and
(e) enhance the independence of EBRC.
With that, I urge all hon Members to support the Motion.