General election 2015: Looking back, looking forward, part 1

Maruah held a post-election forum on 19 Sept 2015.

Maruah held a post-election forum on 19 Sept 2015.

Maruah’s post-election forum, 19 September 2015, was very well attended. They ran out of chairs and people had to sit on the floor.

I was one of the six speakers, but the best part for me was the way that comments made by other speakers and members of the audience generated further thoughts which will take three posts for me to cover. These thoughts can be grouped into seven themes. In this post (Part One of three), I will discuss the first two: The ‘apathetic’ voter, and the debate about whether parties need ideological platforms.

The ‘apathetic’ voter

I didn’t use this word. It came from someone in the audience during question time, though conceptually I alluded to something similar in a previous post. If I remember correctly, the question was about what we can do to cure apathy. I don’t think I even tried to answer this question because in my opinion the ‘apathetic’ voter will always be with us. We can’t expect everyone else to be interested in the same things as we are. Ask me about fashion or the culinary arts, and you’ll find me hopelessly apathetic on those scores.

Ah, but politics is different, you might say. It affects our lives.

Nope. Still doesn’t give people reason to be interested.


Consider this: it’s a well-known fact that many people who have put their savings into buying shares remain uninformed, apathetic shareholders. They only care whether today, tomorrow, the share price is going up or going down. They know next to nothing about the business, its strategy, operating environment, financial health or the quality of its management. Plenty of managements exploit the fact the their shareholders are passive to do whatever they want, and reward themselves handsomely.

Derek da Cunha, another panellist at Maruah’s forum, highlighted one fact (albeit in connection with some other point) which I thought was germane to the question of the ‘apathetic’ voter: voting is compulsory in Singapore. The table below, whose figures I obtained from here, shows the strong effect on voter turnout when voting is compulsory (Australia after 1924) compared to when it is not (United Kingdom).


The overall profile of voters who go to the booths in compulsory voting differs from those in non-compulsory voting.

Compulsory voting’s effects on turnout are more pronounced among certain segments of the electorate. . . . the young, the less knowledgeable, the poor, and those who are more detached from politics participate at roughly the same rate as their older, more knowledgeable, richer, and more engaged counterparts in countries where voting is compulsory and abstention is sanctioned.

Of course, by increasing participation among these typically dormant groups, compulsory voting produces voting populations that are more likely to include individuals who are apathetic or unknowledgeable about politics and government.

— Political Studies Association, UK. Beyond turnout: the consequences of compulsory voting. Link.

The ‘apathetic’ voter (whatever that actually means) is thus an ineluctable part of Singapore’s electoral landscape, and a significant factor too. We should however be careful not to read into this a deterministic view that they will always vote for the People’s Action Party (PAP). In 2011, some portion of them clearly didn’t. In Australia, despite compulsory voting, there have been regular changes of government.

Political parties need to have a strategy for winning their votes. What that strategy is, however, would call for another huge debate, so we won’t get into that here. What I think can be said though is that even the PAP may not have a clue about an effective strategy, seeing how they too were surprised by the landslide earlier this month. Oh my gosh, what did we do right? would have been their reaction.

Does party platform play any role?

This may not be an accurate summary of Derek da Cunha’s point, but is roughly what I sensed it to be: It is personal connection that secures votes for a party/candidate, ideology plays virtually no role. Certainly he disagreed with my point that parties need to make clear some kind of ideological position. In his view, manifestos play no role.

The Workers' Party's Low Thia Khiang (left)_ and Png Eng Huat (centre) greeting residents.

The Workers’ Party’s Low Thia Khiang (left) and Png Eng Huat (centre) greeting residents.

I think he is over-generalising from a tiny sample of two long-staying politicians: Chiam See Tong and Low Thia Khiang. While I don’t disagree that their personal connections with their constituents gave them staying power, and I have heard about how the other Workers’ Party members of parliament in Aljunied are likewise building up the same assets through regular interactions with their voters over the last few years, I think such a prescription is too reductive.

Nor is it sufficiently helpful advice for parties who have not yet secured a toe-hold in any constituency. How would they even begin to build personal connections?

In any case, we cannot pretend that all voters are the same. Different voters need to be satisfied in different ways. Some voters vote for you because they know and like you. Others vote because you’re saying what they want to hear. Yet other voters give you their support because you stand for something they strongly believe in even though they’ve never met you. A successful politician needs to be all these things to stitch together a majority.

Especially when considered in relation to another point (described below, and one that I have written about in previous Yawning Bread posts) to dispense with ideology is foolish.

Differentiation is key to brand-building

The point that I have made at the forum and previously, and wish to repeat here — and please bear with me here because it is layered — is this: Non-PAP parties handicap themselves by failing to distinguish themselves from each other. When voters mentally park them under the generic label of ‘opposition parties’, they run a substantial risk of doing worse electorally.


  • The sins of one party colours voters’ perception of all other parties (e.g. internal infighting, management prudence); the innocent party may be tarred by association through no fault of its own;
  • Makes it easy for the PAP to smear all parties with the failings of one;
  • Surge in support for any one party is seen (feared) as surge in support for all opposition parties, therefore ‘freak result’ is seen as more likely, therefore a tendency to reactively compensate for it and vote for the PAP.

To escape this associative trap, it is important that parties build unique brands and break out of the stifling ‘opposition’ label.

What better way than to infuse the party brand with ideology and related values? Imagine this scenario: Party A’s brand is strong and support surges in the lead-up to the polls. First of all, voters do not misconceive that Parties B to F are also surging. Secondly, even if Party A looks likely to eat into the PAP’s seat count in the legislature, voters don’t necessarily fear that A will join hands with B to F to form a non-PAP government. Voters will more or less know that ideologically A is so far apart from C to F that there is no real risk of such a non-PAP coalition. Even if Party A surges, the most likely post-election scenario may be a PAP+Party A coalition, thus moderating PAP policies rather than throwing the baby out with the bathwater.

There are two other reasons to put effort into building an ideologically-weighted brand.

Firstly, it helps to get voters to take notice of your party. To say the same things as other parties won’t make you stand out. After people have taken notice of you, then maybe you have the possibility of a conversation (directly or indirectly) with the voter to build the personal bonds, etc. Without it, you have no opening at all.

Secondly, it gives people a reason to vote for you, not just against the PAP. This second reason is painfully clear after the 2015 election. We now know that there was never any ‘New Normal’.  The 2011 results, as I have argued in an earlier essay, should be seen as a protest vote. People weren’t voting that year for non-PAP parties as much as they were voting against the PAP. How quickly the tide turned in 2015, when they were no longer angry!

Forget about ‘opposition unity’

At the talk, I presented a graph that showed how in this election some parties withstood the rout better than others. (The data for this graph can be found in the earlier article here.)


At Maruah’s forum I reiterated the point that the parties with relatively more solid branding were better able to withstand the turning tide. I then took the logic further: As these parties establish their own unique standing, it would make no sense to participate in “opposition unity” or horse-trading talks. Doing so would only reinforce the idea that all opposition parties are inter-substitutable, thereby undercutting their more valuable differentiation.

Derek da Cunha added the point that while some people have accused the Worker’s Party of arrogance in not fully participating in the constituency horse-trading prior to this campaign, the actual fact of the matter is that the accusation applied more to the lesser parties for even imagining they were in the same league as WP. I agree.

I would also add that as the last few years’ three- or four-cornered contests have borne out, the stronger opposition party can crush the weaker opposition party. Then why participate in horse-trading? To say that this is to avoid splitting the anti-PAP vote is to indulge in fantasy. Firstly, it won’t significantly split the vote — the voters will go to the stronger opposition party if they’re not inclined to support the PAP — and secondly, “anti-PAP” will not add up to enough votes to win. The 2015 election results show that anti-PAP only gives you about 20% of the votes. As I mentioned above, parties need to give voters a reason to vote for them.

Percent vote-share down, but absolute vote numbers?

pic_201509_80Since the total number of voters had grown, I was interested to see whether the decline in percentage vote-share experienced by opposition parties represented real declines in absolute numbers of votes or only relative declines. The next chart shows what I found in selected constituencies where a comparison could validly be made: i.e. no major boundary changes.

The data supporting this table can be found at right.


What this graph shows is really interesting. The Singapore Democratic Party (SDP) seemed to have nearly held their own in Holland-Bukit Timah and Bukit Panjang, performing a shade better than the Workers’ Party in their two core areas of Hougang and Aljunied. The SDP lost only 1,132 votes compared to the WP’s 4,062 votes in their respective core areas.

On the other hand, in Punggol East, even though the Workers’ Party’s Lee Li Lian lost the election, her absolute number of votes increased by 3,041.

Several lines of enquiry are suggested by the above table and chart:

  • Did the Aljunied Town Council saga affect WP’s level of support? It would seem to have had some effect looking at their loss of votes, but on the other hand, Punggol East was also within the town council’s scope, yet Lee Li Lian raised her haul.
  • That said, any analysis of Punggol East is complicated by the fact that there was a by-election in 2013, which Lee Li Lian won with 16,045 votes. If compared to that, then she lost 227 votes in the general election of 2015 which is quite small.
  • Did the “rehabilitation” of Chee Soon Juan’s image contribute to stemming the losses for SDP? But if so, why did a net 1,474 voters desert them in Yuhua?
  • Suppose you say that the Chee effect could only apply to Holland-Bukit Timah since that was where he was standing, then how to account for SDP doing even better in Bukit Panjang where a first-time candidate, Khung Wai Yeen, was standing?
  • One might argue that the personality (or credentials) of the candidate is key. Possible, except that without more information about how Khung Wai Yeen, Lee Li Lian and Jaslyn Go (SDP, Yuhua) were perceived by voters in those areas, I can’t say much.

The really dramatic numbers are SPP’s. They had a net loss of 16,097 votes in Bishan-Toa Payoh GRC and 2,510 in its former stronghold of Potong Pasir. I think we can guess why. Chiam See Tong was no longer standing for election. But this brings me back to the “personal connection versus ideology” debate. Inasmuch as personal connections with voters count for a great deal in the here and now, transferability to successors is low.

The same effect may perhaps be seen in Hougang, where there was a net loss of 1,823 votes for the Workers’ Party, five years after Low Thia Khiang moved to Aljunied.

However hard it is, however unrewarding it may be in the short term, a solid party brand is an essential step to progress.

In Part Two, I will share my thoughts on the ‘new citizens’ effect and how parties can better focus on swing voters.



6 Responses to “General election 2015: Looking back, looking forward, part 1”

  1. 1 Hawking Eye 21 September 2015 at 17:02

    Incisive analysis. Will wait for two other parts to come before expressing my point of views

  2. 2 yuenchungkwong 21 September 2015 at 17:12

    > last few years’ three- or four-cornered contests have borne out, the stronger opposition party can crush the weaker opposition party. Then why participate in horse-trading?

    the weaker parties might get just 1%, but if the race is close like this time in Punggol, 1% makes a difference; if the votes are 70/30, then of course it does not matter if you get 29 instead of 30

  3. 3 Jason 21 September 2015 at 21:30

    I think the last chart comparing absolute numbers is meaningless without looking at the total number of voters. What matters is the proportion of voters supporting opposition.

  4. 4 han 21 September 2015 at 21:49

    I am glad to see some analysis on absolute numbers here. Too often people says there is a swing of 10% from opposition to the PAP and then assume that this 10% are essentially the people who voted for opposition previously but now have swing over to vote for PAP. Once we look at the various data in terms of absolute numbers (e.g. increase in the number of new non-new-citizen voters, increase in new citizens who are eligible to be voters), then we can attribute part of that 10% to a part of these new voters. That should give us a more accurate picture of how many people switched from voting for opposition in 2011 to voting for PAP in 2015.

    • 5 yawningbread 22 September 2015 at 16:34

      At the same time, we need to be careful not to assume that those who voted for Party A in 2011 are the same individuals who voted for Party A in 2015. It is conceivable that some A-supporters from 2011 deserted the party in 2015, while others (who didn’t support A in 2011, or just moved into the constituency) gave their vote to A in 2015. All that the absolute numbers show is net gain, net loss. Despite this limitation, it is nonetheless helpful to know the numbers. Certainly net equivalence is a lot better than the whopping net loss that Reform Party and SPP suffered.

  5. 6 LC 22 September 2015 at 08:50

    For the WP held areas, there have been new voters moving into new residential developments. For example, there have been new developments that have received their TOP in Bedok Reservoir (under Low TKs division) and in Hougang since GE 2011. This may mean that new voters who may not necessarily voted WP or opposition in 2011 and have been unconvinced by WP’s arguments/ performance, could be part of the swing to the PAP.

    It is crucial for WP to reach out to these voters as well. Otherwise, they will see a gradual erosion of support, like on Potong Pasir from GE 2006 to GE 2011.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: