13 May 2016

Bukit Batok by the numbers

In our previous post, we applied the Dominant Party System model to the Bukit Batok by-elections and made some predictions of the vote, as well as initial critiques of the early campaigning by the People's Action Party and the Singapore Democratic Party. It's time we looked at the numbers, and measure the model against the competing narratives from the mainstream media's pundits as well as popular folk wisdom.

According to the Elections Department of Singapore:

Bukit Batok

Number of Electors: 25,727
Candidates Party Total Votes





We previously predicted a 67% baseline win for the PAP. Chee has outperformed our target, but how did we arrive at that number in the first place and was it a fair target?

We began with the GE2015 results: a 73% win for the PAP

Note it is not true that all by-elections will have a swing against the ruling party, even in a dominant party state. Just look at most of the by-elections before Anson, as well as the Marine Parade by-election. A "by-election effect" or more accurately, a significant swing away from the dominant party, is predicated on voter backlash (either from ruling party shenanigans or its fielding unpopular candidates). In the language of the Dominant Party State model, a by-election effect occurs when voters are dissuaded of the ruling party's hyper-effectiveness halo.

Having David Ong resign from his seat in a scandal qualifies for a voter backlash, whether or not it was capitalised as an issue during campaigning.

The literature on by-elections in normal electorates as well as dominant party states suggests a consistent average swing of 10% towards main opposition parties, and between 2 to 5% towards smaller parties.

The Workers Party, as we have mentioned, is the main opposition party in Singapore (despite having just the "merit" to win 5 seats in parliament). Whether it deserves it or not, WP has its own halo of hyper-efficiency. As Alex Au points out, "71 percent said the Workers’ Party would be their first choice; 24 percent said it would be the Singapore Democratic Party (SDP)." More importantly, Au's survey predicted a 63% total share for SDP, based on first and second choice votes.

We therefore predicted a 63% strength vote swing, which works out to our baseline figure of a 67% win.

Overperformed, underperformed, or failed?

Chee Soon Juan achieved a 12% vote swing against the PAP. He still lost badly at 38%. Is Chee therefore unelectable with insurmountable high negatives? Should Chee plan for SDP's leadership renewal, retire from politics, and play the role of an elder opposition statesman?

We insist on looking at hard numbers. Chee's 12% swing in Bukit Batok, 2016 can only be compared to Lee Li Lian's 11% swing in Punggol East, 2011. Going way back, the only larger swing of 37% was achieved by JB Jeyaratnam in 1981. Chiam See Tong won Potong Pasir in 1981 on the back of a 20% swing. Low Thia Khiang won Hougang in 1991 on a 12% swing.

Jeyaratnam and Chiam both achieved victories due to extreme voter backlash and anger against the PAP. Devan Nair had vacated his seat after Papalee appointed him president (voters punishing a close political relationship between a head of state and the PM is a recurring theme in Singapore). Papalee's pugnacious campaigning for the already off-putting and arrogant Mah Bow Tan, combined with his new harebrained policies (graduate mothers! PSLE streaming!) only served to undermine the PAP's hypereffectiveness halo.

Electorates are not equal or the same; demographically, Bukit Batok is not Punggol East and in general, the west of Singapore is far less fertile ground for the opposition than the east. This is why the number to look at, if we're credible pollsters and analysts, is the swing vote.

It is true that Chee is no Chiam or JBJ; but his swing is as good as Lee Li Lian's and even Low Thia Khiang's. Based on his ability to get a larger swing than predicted by the model, we say that Chee is a viable candidate. Bukit Batok was one of the safest seats in GE2015; turning it into the equivalent of East Coast GRC scarcely six months later is a feat in itself.

Why won't Chee please go away?

We have already refuted the idea that Chee is unelectable because he "only" got 40% of the vote in the by-election.

Let's look at the next most credible narrative: Chee is unelectable because the election proves he has high insurmountable negatives.

With due respect to pundits like Eugene Tan and Derek da Cunha, we suggest that their pronouncements fall far below the standards of political science. Their pronouncements on Chee have the value of folk narratives and coffeeshop chatter because their words are not backed by any quantifiable, empirical research.

A trained political scientist would point towards actual polling data to identify what percentage of voters would "never vote for Chee", which demographics Chee has key weaknesses in, etc. The silence of Tan and da Cunha on this matter speak for themselves, in the face of a 12% swing vote for Dr Chee.

We have noted previously that in a Dominant Party System, the electorate votes on the basis of party, not candidate. This has been proven by various Singapore surveys over the years. The election, despite the campaigns run by both parties, was never about the person of Chee Soon Juan or Murali Pillai. And given the clearwater strategy of the SDP, it was extremely unlikely for it to win the by-election on a clear fight.

Qualitative measures

We have insisted that the key to an opposition victory in the Dominant Party System lies in challenging the dominant party's halo of hyper-effectiveness. Could the SDP have done a better job?

The first half of the by-election campaign was won by the SDP, which capitalised on the social media's attack on Murali's recycled and stale upgrading carrot. It was the backpedalling by the PAP that proved the effectiveness of this strategy. By challenging the halo, by delegitimising what was heretofore an accepted and acceptable election tactic, the SDP made serious headway.

The second half of the campaign was won handily by the PAP, which sent what felt like the entire cabinet, past and present, to campaign for Mr Murali. Yes, it's blatantly unfair. But to voters, it's proof of the hyper-effectiveness of the dominant party. To complain bitterly about the unfairness of this arrangement is to further confirm to voters the hyper-effectiveness of the dominant party. Unlike in 2011, the SDP campaign failed to highlight the gaffes of the cabinet and senior ministers, parlay their attacks and paint them as out of control attacks by arrogant and out of touch politicians.

Byebye Dr Chee, see you next GE

If Minilee in his perestroika moment does get the reforms he wants to parliament, there will be a significant increase in single ward seats next GE. While the SDP may not have been a viable party under the GRC system in Singapore's dominant party system, it may be actually viable if there are sufficient single ward seats. We'll only know in time.

No comments: