GE13 Malaysia Forecast Using The Uniform Swing Theory

It is finally finished – my mathematical model for all 222 federal seats in Malaysia, and thus, the winner of the next and 13th General Election. Behold in all its glory.

Partisan Direction

Not all states vote follow the national trend. In fact, from my calculations, none do. Some lean to BN, others lean to PR. The challenge is figuring out which states lean where. So I benched to the previous general election in 2008.

Partisan Direction = State Vote – National Vote

E.g. BN lost Kuala Lumpur by -24.03%, but won Malaysia overall by 3.63%. Kuala Lumpur’s partisan direction is -27.66% (-24.03% – 3.63%). Kuala Lumpur is a safe PR state.

• Safe = Partisan Direction greater than 20%
• Strong = Partisan Direction between 10-20%
• Lean = Partisan Direction between 0-20%
• Conflicted = When there is a mismatch between state vote and seats. For example, BN lost Perak by -6.87%, but still won the majority of seats (13 out of 24)

The usual suspects, such as Kelantan, show a strong partisan direction to the opposition (PR), and PR holds sway in many urban centers. Rural seats tend to have a strong partisan direction to the government (BN).

Flipping Seats and States

We are confident that GE13 will see nett seats flip from the government to the opposition; this confidence is born of opinion polls, the proliferation of rallies, and increasing voter preference for a two-party democracy. However, forecasting exactly which seats flip requires maths. Therefore I propose a simple model I call the “Uniform Swing Theory”.

“Voters that change (flip) their votes from BN to PR will do so in equal proportions everywhere in the country. Hence, if 5% of Malaysians swing to the opposition, this means that 5% of Sabah swings to the opposition, 5% of Pahang swings to the opposition, 5% of Melaka swings to the opposition, and so on.”

All three of Perlis three seats went to BN in 2008, but they do not all have the same partisan direction. Kangar went to BN by 40.24%, a big margin, Padang Besar by 18.68%, and Arau by a very slim 0.92%. It is obvious that Arau is going to flip if the opposition improves at all. In fact, PR can lose GE13 by as much as -2.7% and still win Arau by the barest of leads, 0.01%.

Meanwhile, it would take a total landslide, 15.1%, to win the next seat, Padang Besar. Whereas Kangar (36.61%) is totally impossible until the map is radically altered.

The “Uniform Swing Theory” is incomplete, because it ignores local state developments and campaigns that target specific races, ages, and education levels. It is, however, a good starting point due to simplicity. This article will only consider events under the UST.

The Tipping Point

Barisan National currently holds 140 seats in parliment, because they won GE12 by 3.63% (I’m excluding all independent, spoilt, and third party votes). This drops to an exact tie if they lose GE13 by -7.5%.

The hung parliament is not pretty, because two states, Negeri Sembilan and Terenganu, will also hang. Firstly, neither party can declare victory. Secondly, many individual MPs will sell their vote to the highest bidder and jump parties like frogs. Thirdly, the country may descend into chaos. It reminds of the chinese curse, “May you live in interesting times.”

With that said, is it possible for -7.5% to occur? It would mean a majority larger than Obama’s victory over John McCain. Then again, Malaysian politics is not as polarized as America, because many people are still just waking up to the possibility of alternatives.

Popular Vote Tie, Modest PR Win, and Landslide

I simulated three other outcomes. Click to enlarge.

1. Popular Vote Tie – If an equal number of voters vote for both sides, the popular vote will be tied at 50%-50%. A BN victory, because they will capture 133 seats versus 89 for PR. (How does this happen? GERRYMANDERING.)
2. 5% PR win – If PR has 5% more voters than BN, they irrationally will still lose the election. (Isn’t democracy about getting more votes than the other guy? Nope, not in the cheat-prone and antiquated British FPTP democracy.) BN will get 118 seats versus 104 for PR.
3. 10% PR win – But if PR gets 10% more voters, then there will be a reversal. The opposition will capture Parliament for the first time in history. PR will now get 119 seats vs 103 for BN.

The UST versus Reality

The UST shows how national trends impact the election, but actual GE13 results can diverge from this.

1. Johor, Sarawak and Sabah were barely contested in GE12. BN’s thumping majorities there were not so much a product of popularity; it was the total absence of organized opposition. PR can overperform the model just by showing up.

Currently the UST estimates Johor, Sarawak, and Sabah to give 13 seats to the opposition in a landslide. However, this is far below what Pakatan can achieve. In a previous article, I pegged the likely outcome at 19 seats.

2. New voters. As DAP supremo Lim Kit Siang has highlighted, there will be 3 million voters, 60% young, added to the rolls in GE13. With such a huge number of young voters, it is likely that they favor the policies of the opposition. The young in Malaysia are also better informed and more liberal.

On the other hand, statistic after statistic show that the young fail to turn up at elections, meaning they talk a good game but fall asleep on the big day. The Obama campaigns of 2008 and 2012 prove that youth voters need to be bussed, cajoled, and given the convenience of several weeks of early voting, before they deign to participate in democracy. Yes, the 60 year old voter will make it to the polls rain or shine (or hurricane). But twenties are prone to “my vote will not make a difference anyway”.

So, let us not overestimate the new voters.

3. Lynas. It is a scandal that might single-handedly drive Pahang to opposition arms; getting 20,000 people to walk from Kuantan to KL is a BFD. The electorate in Pahang is energized.

In normal years Pahang has a BN partisan direction of 15.32%. The direction may be reversed by as much as 10 points. Scandals tend to have hyperlocal (rather than national) effects. Only robust local polling of Pahang can confirm this.

BN has other scandals such as Sabah corruption, AES, and Deepak. For PR, their main danger is PAS’ islamic state agenda.

MAILBAG