New Naratif is Fundraising!
New Naratif is a 100% ad-free, member funded, non-profit movement for Southeast Asian democracy. Membership fees fund high quality, independent journalism and create positive change in our communities. This year, we need to raise US$226,336 to continue our operations. Please consider becoming a member or donating today!
This article is part of New Naratif’s coverage of the Singapore General election 2020. For more, please visit our Singapore General Election 2020 portal.
Singapore has been governed by the People’s Action Party (PAP) for over 60 years. Beginning with its victory in the 1959 General Elections, the PAP has won 14 consecutive General Elections, many with a clean sweep of all seats. Former Prime Minister Lee Kuan Yew stated, “My policies have been endorsed by the electorate every four to five years by a clear majority, never below 60 per cent.” His son, current PM Lee Hsien Loong, has argued that these election victories legitimise the PAP’s policies, as they have won a “mandate” from the electorate.
In practice, many mechanisms unfairly disadvantage the opposition, so that, in effect, the PAP has not faced a competitive election since 1963.
However, such an argument is valid only if the elections are free and fair. In practice, many mechanisms unfairly disadvantage the opposition, so that, in effect, the PAP has not faced a competitive election since 1963. From the 1960s, these mechanisms included the detention and/or bankrupting of opposition leaders, activists, and critics; denial of permits to hold campaign rallies; shutting down printing facilities used by opposition parties; increasing the candidate registration fee, decreasing the nomination period, and other bureaucratic hurdles; holding snap elections and limiting the campaign period to the legal minimum of nine days; and the lack of an independent Elections Department. As a result, beginning in 1968, the PAP made a clean sweep of Parliament at every election for over a decade.
See: A Brief History of Elections in Singapore, by Thum Ping Tjin
This period of complete domination only ended in the 1980s. Following a series of disastrous policies and a major recession, Singaporeans responded by electing two opposition MPs. The 1981 and 1984 elections produced JB Jeyaretnam in Anson and Chiam See Tong in Potong Pasir. In response to this, from 1984, the electoral system itself was fundamentally altered in ways that maximised electoral outcomes for the PAP.
Firstly, the Constitution was changed so that the government could alter the number of Members of Parliament (MPs) at will.
Secondly, Group Representation Constituencies (GRCs) were created, with each GRC represented by three to six MPs, elected as a slate, at least one of whom must be a member of a minority race. The government unilaterally decides the number of GRCs and designates the minority race that must be represented in each. The GRC system was presented as a way of ensuring minority racial representation, but no convincing explanation was given as to why it had to involve a slate of candidates as opposed to other alternative systems, such as reserving individual seats in single-member constituencies for minorities. The main consequence of the GRCs has thus been to raise barriers for opposition parties who already struggle to find viable candidates in an environment shaped by fear of the detentions, lawsuits, and bankruptcy.
Thirdly, provision of municipal services was delegated to Town Councils, but headed by MPs instead of locally-elected councillors. This expanded the role of MPs from being just national legislators to further include municipal management, but crucially left funding for this function in the hands of the central government. This meant that MPs had responsibility for local services but no control over their funding, creating a mechanism through which the PAP could openly threaten to punish constituencies which elect opposition parties, by withholding public services until PAP constituencies had first been supplied, and warn the electorate against voting for the opposition. In 1985, after the PAP lost 2 seats in the 1984 election, Minister for National Development Teh Cheang Wan stated that the PAP would prioritise PAP constituencies in providing public services; future PM Goh Chok Tong stated that this would make voters “think a little more carefully before they cast their votes.” He reiterated this when in power, suggesting that opposition-held constituencies would be placed at the back of the “queue” when it came to upgrading works. More recently, Lee Kuan Yew warned that Aljunied’s residents would “pay a price, the hard way” if the PAP lost Aljunied GRC. “Voters [will] have 5 years to repent… You must expect the PAP to look after PAP constituencies first.” Aljunied’s Town Council has subsequently been largely denied upgrading funds by the government. In a fair electoral system, this selective use of national funds for party-political purposes may be defensible as an inverse form of pork-barrel politics; combined with the unfairness of the electoral system, however, it is morally dubious.
This article focuses on the redelineation (i.e. redrawing) of electoral boundaries as a mechanism that helps create unfair elections. Unlike many other countries, Singapore does not have an open, transparent system for this. The Parliamentary Elections Act in Singapore does not prescribe a specific procedure for delineating boundaries but gives the Minister the power to specify the names and boundaries of electoral divisions. The Electoral Boundary Review Committee (EBRC) is appointed by the Prime Minister and typically consists of senior civil servants from various Ministries and Statutory Boards. There is no requirement for public consultation during redelineation exercises and there is no minimum period between redelineation and the calling of elections. Prior to 1984, the average length of this period was five-and-a-half months. After 1984, the government no longer had to get Parliamentary approval of changes in the number of MPs, and the average period has fallen to one month. In 2001, elections were called one day after new boundaries were announced.
The consequences of these changes have been dramatic. Between 1991 and 2001, the PAP won three general elections in a row by default, because fewer than half of the seats were even contested. In 2006, only 47 out of 82 seats were contested.
There are well established and internationally accepted principles on which the redrawing of electoral boundaries should be based in order to create a fair system. By contrasting them with Singapore’s system, we can better understand how Singapore’s system of redelineation creates unfair elections.
How should constituencies be drawn?
The drawing of electoral boundaries can drastically affect electoral outcomes. In essence, constituencies can be drawn, using demographic data or past voting behaviour, to benefit one side or another. Delineation has to be fair and perceived to be fair to ensure that the outcome of the elections themselves are accepted as legitimate.
Delineation practices vary widely across different countries but the following principles are generally accepted as best practices:
Representativeness – An elected representative should represent a cohesive community, defined by such factors as administrative boundaries, geographic features, and communities of interest. Constituencies should thus be drawn taking into account such communities.
Transparency – The delineation process should be as transparent and accessible to the public as possible.
Impartiality – The boundary authority should ideally be an independent, non-partisan professional body. Where that is not possible, all political parties should be given an equitable opportunity to participate in the delineation process.
Equality of voting strength – The populations of constituencies should be as equal as possible so that voters have roughly equal voting strength (i.e, every person’s vote has about the same impact on the outcome).
How have these principles been observed—or not—in practice, in the case of Singapore’s constituency boundaries?
Many countries require that electoral boundaries be drawn taking “communities of interest” into account, so that legislators can more effectively represent their constituents. Even when Singapore was still a colony, this principle was expressed by the committee that drew the boundaries for the 1955 Legislative Assembly election: “The elector should see clearly why he is in one division rather than in another; lines of demarcation, therefore must be easy to follow … From the point of view of the candidate, the division must not be too large in number or diverse in interest for him to canvass his electors, or care for his constituents, once elected.”
At the next redelineation exercise in 1958, the All-Party Committee observed that:
“As the voters are not mere units in a mathematical calculation, but men and women sharing, and bound by, the common interests of the community of their neighbours, consideration must be given to identity of interest, either in the local community and its services as the focus of the pattern of their lives, or in the occupation followed by the majority, e.g. farming or fishing. Villages should not, therefore, if it can be avoided, be divided between two or more electoral divisions: and similarly villages without common interests should not be forced together.”
In contemporary Singapore, the concept of “communities of interest” has further practical implications, because Town Councils are formed based on constituency boundaries.
The HDB says that “towns are designed to be self-sufficient, with easy access to shops, schools, and social and recreational facilities”. Logically, Town Council boundaries should be aligned with those of Housing and Development Board (HDB) estates, but many HDB estates are divided into different Town Councils because they are split into multiple constituencies.
Logically, Town Council boundaries should be aligned with those of Housing and Development Board (HDB) estates, but many HDB estates are divided into different Town Councils because they are split into multiple constituencies.
For example, Sengkang is a relatively compact HDB Town with well-defined borders. The new town was established in the late 1990s, and has grown steadily since. By 2011 the number of flats in Sengkang was almost the same as or more than that in other estates such as Pasir Ris or even Ang Mo Kio (Table 1).
Despite that, during the 2011 and 2015 General Elections, Sengkang was split across four constituencies – Ang Mo Kio GRC, Pasir-Ris Punggol GRC, Sengkang West SMC and Punggol East SMC, as shown in Figure 1.
Figure 1 – During the 2011 and 2015 General Elections, Sengkang Town was split into four different constituencies – Ang Mo Kio GRC, Sengkang West SMC, Pasir Ris-Punggol SMC and Punggol East SMC. Heavy black lines represent borders of URA planning areas 
Figure 1A – Even after the creation of Sengkang GRC in 2020, Sengkang GRC still only constitutes half of the Sengkang planning area.
Only in 2020 has the EBRC created a new Sengkang GRC. Even so, this GRC only covers half of the Sengkang planning area. The former Sengkang West SMC has been split into half and divided between Ang Mo Kio GRC and Sengkang GRC. Residents in Fernvale and Jalan Kayu have thus moved from Ang Mo Kio GRC to Sengkang West SMC and then back to Ang Mo Kio GRC within ten years. To an objective observer, it would make more sense to merge the entire Sengkang West SMC into Sengkang GRC, so that the whole Sengkang HDB estate could be managed under one Town Council. One can only surmise that that was not done because the resulting four-member Sengkang GRC would have had more voters than East Coast GRC, which has five members after absorbing Fengshan SMC. Conceivably, the EBRC could have made Sengkang GRC a five-member GRC instead, but that may have been too big a step for an area which had never before been contested as a unified GRC.
Conversely, many Town Councils have to manage pieces of different HDB estates because GRC boundaries are not aligned with HDB estate boundaries. Aljunied GRC, for example, stretches from Serangoon Gardens to Bedok and Tampines and includes parts of Hougang, Bedok and Tampines HDB estates. (Figure 2).
Figure 2 – Aljunied GRC stretches from Serangoon Gardens to Bedok and Tampines
This is clearly inefficient and makes it more difficult for Town Councils to serve their residents. Furthermore, HDB buildings less than a hundred metres from each other receive different levels of maintenance and upgrading. This artificial division of neighbourhoods makes it more difficult to promote community bonding and spirit. Leader of the Opposition Pritam Singh made this criticism specifically in reference to the selective funding of Town Councils, arguing that it is a “divisive approach to politics… that will make Singapore a politically polarised society.”
Another example of a failure to observe the principle of representativeness can be found in Marine Parade GRC, which stretches from the sea at East Coast Park to the edge of Serangoon Gardens. It cuts across five URA Planning areas and includes parts of four HDB estates – Serangoon, Geylang, Bedok and Marine Parade itself! It does not contain a single unified URA planning area within its borders. Oddly-shaped constituencies like this recall the origins of the word “gerrymander”, first used in the early 1800s to lampoon Governor Elbridge Gerry of the State of Massachusetts for drawing electoral districts that resembled salamanders.
Figure 2 – Marine Parade GRC (outlined in red) cuts across five URA Planning Areas.
Unlike other countries such as Canada (ten years), UK (eight to 12 years), and Malaysia (at least eight years), there is no fixed or minimum period between redelineation exercises in Singapore. Some neighbourhoods have thus seen multiple changes within a short period of time.
Figure 4 – The Kaki Bukit neighbourhood was in four different GRCs over the course of six elections between 1988 and 2011.
For example, as shown in Figure 4, the Kaki Bukit neighbourhood in Bedok was moved from Eunos GRC (1988) to East Coast (1997), Marine Parade (2006), and then Aljunied GRC (2011). Frequent changes in electoral boundaries cannot be helpful in creating a strong relationship between residents and their MP, nor for long-term planning and development of neighbourhoods.
Until the 1980s, the number of MPs was fixed by law and Parliament had to approve any change in the number of MPs. In 1984, the Constitution was amended to give the Government the power to vary the number of MPs by simply changing the number of constituencies. Four years later, the constitution was amended again to allow for the creation of GRCs. There is no minimum notice period for the redrawing of boundaries. In 2001, elections were called the day after the EBRC report was released, with polling taking place just 16 days later.
During the colonial era, the 1954 and 1957 boundary delimitation committees included elected representatives from several political parties. In 1963, after Singapore joined Malaysia, the PAP government invited opposition parties to submit proposals on delineating constituencies for Malaysian Federal elections. In contrast, there is no public or opposition party input to the boundary delimitation process today.
Accusations of gerrymandering inevitably follow every redelineation exercise, but because of the lack of detail in present-day EBRC reports, it is very hard to assess whether changes in constituency boundaries are primarily due to shifts in population or are designed to favour one party. The EBRC reports carry scant or no explanation as to the criteria applied. The 2020 report, for example, is only 10 pages long, including annexes. It does not explain how it came to its conclusions despite the fact that it was given two additional tasks this time round, namely to reduce the average size of GRCs and to increase the number of SMCs, apart from the usual considerations of changes in the distribution of population.
There was a net increase of one SMC in the 2020 boundary revision, in accordance with the Prime Minister’s instructions to the EBRC to increase the number of SMCs. However, the three SMCs that were eliminated were also the three aside from Hougang (held by the WP) where the PAP had the lowest vote share .
Sengkang West SMC and Punggol East SMC were merged into a new Sengkang GRC. However, only half of Sengkang West SMC was included. As noted above, had the entire Sengkang West SMC been included, the total number of voters in (four-MP) Sengkang GRC would have exceeded that in (five-MP) East Coast GRC, a disparity that would have been even greater but forFengshan’s absorption by East Coast GRC. One can only speculate as to why the EBRC chose to expand East Coast GRC into a five-member GRC while capping Sengkang GRC as a four-member GRC. Another consequence of splitting the former Sengkang West SMC into two is that with 21,000 voters transferred to Ang Mo Kio GRC, Ang Mo Kio GRC has one-and-a-half times as many voters as East Coast GRC even though both are five-member GRCs.
Constituencies are divided into polling districts, and it is possible to determine how specific polling districts voted even though individual votes are secret. In 2011, seven polling districts in Aljunied GRC were transferred to Ang Mo Kio GRC, and the opposition Worker’s Party (WP) complained that this amounted to gerrymandering because they had significant support in those areas. Unfortunately,the Elections Department does not release detailed election results at polling district level, so outside observers are unable to assess whether the WP’s complaints are valid. One exception is the case of Joo Chiat SMC. Yee Jenn Jong of the WP came within 1% of winning Joo Chiat in 2011. In the 2015 General Elections, however, the constituency was absorbed into Marine Parade GRC, which had five times as many voters and had previously voted in favour of the PAP by a 13% margin. Expressing his disappointment, Yee lamented that: “[T]here is no clear justification for the changes. With the eraser and the pencil, the mighty committee has made the Joo Chiat SMC with such a rich and unique tradition disappear. It was a SMC from 1959-1988 and also for the last 3 GEs since 2001.”
Recognising the importance of a fair electoral process, most countries have independent commissions to depoliticise the redistricting process. In Canada, Malaysia, and the UK, for example, members of redistricting commissions are appointed by a non-partisan figure, such as the Speaker of the House or the Head of the Judiciary. The members tend to be diverse and from technical backgrounds, including academics (especially statisticians), lawyers, electoral officers, or senior judges. By contrast, in Singapore, the EBRC consists of five civil servants who report directly or indirectly to the Prime Minister. As voter registration and redistricting processes make extensive use of government geospatial and population databases, it is expedient for specialists from government agencies to be involved. However, the composition of the EBRC should be broadened to include private sector members and public hearings should be held as part of any redelineation exercise.
The opacity of the EBRC’s work is aggravated by the absolute discretion that the Government has on the timing of boundary revisions. In most countries with similar Westminster-derived systems, delineations are typically conducted after a decennial census or changes in the number of registered voters or administrative boundaries – typically, every eight to 12 years. In Singapore, however, there is no legally-mandated criteria for the interval between exercises.. In other words, most countries delineate boundaries only when they have to; Singapore does it whenever the Prime Minister wants to.
In Singapore, … there is no legally-mandated criteria for the interval between exercises.. In other words, most countries delineate boundaries only when they have to; Singapore does it whenever the Prime Minister wants to.
Moreover, while the legislature typically plays a role in approving or rejecting the proposal in other countries, the Executive in Singapore can unilaterally redraw boundaries and change the number of MPs, without any scrutiny from the judiciary or legislature.
Finally, in a fair election, sufficient lead time should be afforded to all candidates. However, the brevity of the period between the report and the calling of elections means that candidates do not have time to properly prepare for elections. Any groundwork they put into a particular constituency may be wasted if that constituency gets erased. Independents and smaller parties in particular find it more difficult to put together teams of candidates of the correct ethnicities to contest in GRCs. As election campaigns are frequently limited to nine days (the minimum allowed by law), voters also do not have sufficient time to get to know the candidates for their constituency.
The table below shows the interval between announcement of revised electoral boundaries and the calling of elections since independence. From an average of 22 weeks in the 1960s-1980s, the average interval has dropped to four weeks since the 1990s. In 2001, Parliament was dissolved, and an election called, one day after the EBRC report was issued.
This means that aspiring political candidates who have been ‘working the ground’ for years, meeting constituents and seeking to understand local issues, may sometimes find that very ‘ground’ being swept out from under them. Just months or even days before elections, a candidate’s ‘ground’ may be cut up into several parts and merged with other constituencies, or absorbed into a GRC.
The function of rules is to give certainty and stability to social, economic and political life. No one can plan ahead if the terms of electoral competition are constantly changed at the last moment.
By comparison: see our analysis of Malaysia’s last delimitation exercise, How Malaysia’s Election is Being Rigged by Ooi Kok Hin.
Equality is the principle whereby each voter should have a vote of equal strength. For historical and practical reasons, many democracies do not meet this ideal. Rural voters, for example, often have more “powerful” votes, in that it takes fewer votes to elect a representative in a rural district than in an urban one. In a system which delineates voters according to geography, it is virtually impossible to guarantee that every constituency will have the same number of voters.
Academic studies show that Singapore is characterised by a high level of disproportionality in its translation of opposition votes to seats. For the last 13 elections, for example, the average disproportionality based on two standard indices of electoral disproportionality are very high: 27.7 on the Loosemore-Hanby Index and 22.3 based on Gallagher Index. These systems both measure an electoral system’s disproportionality by calculating the difference between votes received and seats allotted in a legislature (lower is better). Looking at neighbouring countries with broadly similar plurality electoral systems, Singapore’s disproportionality is much higher than Malaysia (15.8, 1959-2004), Thailand (11.1, 2001-5) and the Philippines (10.4, 1992-8). In other words, the electoral system is highly unrepresentative of opposition votes, and heavily benefits the ruling party, to a far greater degree than in other electoral systems in the region.
Academic studies show that Singapore is characterised by a high level of disproportionality in its translation of opposition votes to seats…. In other words, the electoral system is highly unrepresentative of opposition votes, and heavily benefits the ruling party, to a far greater degree than in other electoral systems in the region.
In the absence of any law requiring some parity in the sizes of the various constituencies, there is nothing to prevent the system from being abused through malapportionment.
Singapore’s system of electoral redelineation is deeply unfair. Its constituencies are unrepresentative, arbitrary, and lack voter equality. The system for delineating them lacks transparency, is extremely arbitrary, and heavily disadvantages non-governing parties. The history and timing of the changes, and the way they have been utilised, suggests that the PAP amended the constitution and wrote the relevant legislation in order to design a system of elections that maximises the outcomes for the ruling party.
To have a fairer system of delineating boundaries, Singapore should, at minimum, adopt the following:
- An independent electoral boundaries delineation commission, appointed by a non-partisan person independent of the executive. Politicians should not be involved. The commission members should be diverse and include technical experts—for example, academics, lawyers, electoral officers, or senior judges.
- The criteria for drawing constituency boundaries should be clear and transparent, and seek to be representative, especially of communities of interest.
- The process should be transparent and the deliberations of the commission made public. There should be periods of public scrutiny and public feedback should be invited.
- The timing should be predictable. The process should be tied to the census, which is typically held every decade. Nomination Day should be a minimum of six months after the redelineation process is completed. This will guarantee sufficient lead time for all parties in the election.
These reforms would go a long way in allaying the people’s fears that the incumbent party is abusing its constitutional powers to maximise its electoral outcome through gerrymandering.
At the same time, these reforms are necessary but not sufficient for ensuring a fair electoral system. To do so, other mechanisms such as the Group Representation Constituencies, the timing of elections, the role of MPs in the Town Councils, and others, will need to be reconsidered as well. Public discussion should be fostered about creating an electoral system that is fair and ensures that all Singaporeans are adequately represented in Parliament.
Further Reading on Elections in Singapore
- Lee Morgenbesser (2017) The autocratic mandate: elections, legitimacy and regime stability in Singapore, The Pacific Review, 30:2, 205-231, DOI: 10.1080/09512748.2016.1201134
- Tan, Netina (2013) ‘Manipulating electoral laws in Singapore’, Electoral Studies 32(4): 632–43.
Change in Voting
- Tey, T. (2008) ‘Singapore’s electoral system: government by the people?’, Legal Studies 28(4): 610–28.
- Ong, E. (2015) ‘Complementary institutions in authoritarian regimes: the everyday politics of constituency service in Singapore’, Journal of East Asian Studies 15(3): 361–90.
- Rodan, G. (1996) ‘Elections without representation: the Singapore experience under the PAP’, in R. Taylor (ed) The Politics of Elections in Southeast Asia, New York: Woodrow Wilson Center Press, pp. 61–89.
- Rodan, G. (2009) ‘New modes of political participation and Singapore’s nominated members of parliament’, Government and Opposition 44(4): 438–62.
 Plate, T. (2010) Conversations with Lee Kuan Yew: Citizen Singapore: How to Build a Nation, Singapore: Marshall Cavendish, p. 182. He is not including elections from before independence, as the PAP only achieved 46.9% of the vote in 1963.
 https://www.straitstimes.com/politics/ge2015-election-results-will-bolster-confidence-in-singapore-says-pm-lee-hsien-loong; https://www.straitstimes.com/politics/ge2015-strong-mandate-means-mps-must-work-extra-hard-to-serve-says-pm-lee; https://www.straitstimes.com/politics/singapores-competitive-advantage-politics-and-policies-that-work
 Morgenbesser, L. The autocratic mandate: elections, legitimacy and regime stability in Singapore. The Pacific Review, Vol 30, Issue 2 (2017). https://doi.org/10.1080/09512748.2016.1201134 ; Tan, N. (2013) ‘Manipulating electoral laws in Singapore’, Electoral Studies 32(4): 632–43.
 Far Eastern Economic Review, 11 April 1985.
 Parliamentary Elections Act(Cap 218, 2011 Rev Ed), s 8(1)
 These were Goh Chok Tong’s only three elections as leader of the PAP. Thus, Goh never faced an election where the outcome was not known on nomination day.
 Electoral competition has since recovered, as all but one constituency was contested in 2011 and all constituencies were contested in 2015.
 https://www.ifes.org/sites/default/files/ifes_challenging_election_norms_and_standards_wp.pdf; https://www.idea.int/sites/default/files/publications/international-electoral-standards-guidelines-for-reviewing-the-legal-framework-of-elections.pdf; ACE Electoral Knowledge, ‘Boundary Delimitation —’ available at http://aceproject.org/ace-en/topics/bd/default
 If the voters in Sengkang West SMC who were transferred to Ang Mo Kio GRC were retained in Sengkang GRC, Sengkang GRC would have had 138,201 voters compared to 120,239 in East Coast GRC.
 Report of the Electoral Boundaries Review Committee, 1988, 1997, 2006, 2011
 “At present each time the number of elected constituency MPs in Parliament is to be increased, it is necessary to enact a law. It is considered that this is not really necessary because the procedure for the revision of electoral boundaries and the increase or reduction of the number of constituencies which, in turn, will determine the number of MPs is already spelt out in the Parliamentary Elections Act. The proposed amendment, therefore, provides that the number of elected MPs will be equal to the number of constituencies. This does not change the present position in substance. It, however, makes it unnecessary to enact a new law each time the number of MPs is increased.”, Singapore Parliamentary Debates, Official Report, 24 July 1984, vol 44 at col 1723, Lee Kuan Yew, Prime Minister.
 John Laycock (Progressive Party), A P Rajah and Dasaratha Raj (Labour Party)
 Tang Peng Yew (Labour Front), Darus Shariff (UMNO-MCA Alliance), Thio Chan Bee (Liberal Socialist Party), Toh Chin Chye (PAP)
 “Opposition asked to help draw up new federal polls boundaries”, Straits Times, page 11, 29 June 1963
 Sylvia Lim, “Worker’s Party Response To the Electoral Boundaries Review Committee Report”, 24 Feb 2011, archived at https://web.archive.org/web/20110302014323/http://wp.sg/2011/02/workers’-party’s-response-to-the-electoral-boundaries-review-committee-report/
 Xabryna Kek ‘Opposition Parties Studying Electoral Boundaries Review Committee’s Report’ Channel NewsAsia 24 Jul 2015, available at http://www.channelnewsasia.com/news/singapore/opposition-parties/2005766.html
 Netina Tan, “Pre-Electoral Malpractice, Gerrymandering, and its Effects on Singapore’s 2015 GE”, Change in Voting, Singapore: Ethos, 2015, p.182
 Netina Tan, “Pre-Electoral Malpractice”, p.183
 Burt L Monroe, ‘Disproportionality and Malapportionment: Measuring Electoral Inequity.’ (1994) 13(2) Electoral Studies 132–149, at p. 138.; David Samuels & Richard Snyder, ‘The Value of a Vote: Malapportionment in Comparative Perspective’ (2001) 31(4) British Journal of Political Science, p. 651–671. http://users.polisci.umn.edu/~dsamuels/BJPS2001.pdf
 Netina Tan, “Pre-Electoral Malpractice”, p. 185. See https://en.wikipedia.org/wiki/Loosemore%E2%80%93Hanby_index and https://en.wikipedia.org/wiki/Gallagher_index