Malaysia election 2013

[jaichind]

One claim I made earlier which I think I misspoke on is the distribution of close elections.  PK leader Anwar Ibrahim claimed that there are a lot close elections (20-30) they will dispute.  I made the claim earlier that the ratio of winning close elections was 4 to 1 in favor of BN.  I looked at the data.  Out of the lowest winning vote share races, 37 was won by BN and 18 was won by PK.  It could be a matter of BN able to allocate resources to close races.  One example was Nurul Izzah Anwar, daughter of Anwar Ibrahim, that won by a reduced margin compared to 2008 due to BN concentrating recources on defeat her.  That BN can target her and actually make progress when she has t he superstar factor on her side is a demonstration of BN's mobilization skills.

[jaichind]

My tabluation of results


BN                    47.61%  (-2.66)        130    (-7)
   UNMO            29.32% (-0.01)         88    (+9)
   MCA                 7.98% (-2.37)           7      (-8)
   MIC                  2.64% (+0.43)           4     (+1)
   PBB                  2.10% (+0.48)         14     (-)   Sarawak regional tribal party
   Gerakan           1.54% (-0.73)            1    (-1)  Small liberal Chinese party
   SUPP                1.21% (-0.26)            1    (-5)  Sarawak Chinese party
   Others             2.82% (-0.20)         18    (-3)  Various Sabah and Sarawak regional parties

PK                   50.88%  (+4.48)          89   (+7)
   DAP              15.71%  (+1.94)          38 (+10)
   PAS              14.78%   (+0.73)         21    (-2)
   PKR              20.39%   (+1.81)         30    (-1)

The BN Chinese parties got killed and bled votes and seats to PK.  One thing that is intersteing is that both PAS and PKR also gained votes.  I suspect it is also part of the Chinese swing toward PK as PKR also ran Chinese candidates in some districts againist BN Chinese parties.  Also independent votes declined due to the closeness and polorization of the race.

What gave BN victory was the Sabah and Sarawak  vote.  The Chinese vote in Sarawak was lost to PK but the rest of state held solid for BN.  BN also contained losses in Johor.

[jaichind]

How likely was a PK victory in retrospect.  A net gain of 7 seats seems small given the 3.5% swing toward PK this election.  Structural factors were obviously in place.  There are 30 some seats that were close and could have gone PK way.  Another way to look at this question is how many seats actually could have voted PK.  How they voted in 2008 is a clue.  In 2013 BN captured 9 seats from PRK and 7 seats from PAS.   Of course it lost a bunch to PK as well.  But this tells us 89 seats voted for PK in 2013 plus another 16 seats PK won in 2008 that it did not win in 2013.  For sure that PK won in these 16 seats in 2008 means most likely it could have won in 2013 when it actually won more votes overall.  So that established a benchmark of 89+16 = 105 seats that PK at least could have won in 2013.

That PK beat BN by over 3% yet only potentally win 105 seats out 222 means the structural factors are large indeed.  Looking at the results it would take another uniform swing of another 2% for PK to get to 112.  In other words PK would have to beat BN by 7% to actually win a majority.  This might be doable in 2018 but that would take a significant swing within the Malay vote to do this.  That the Chinese vote went so much for PK might actually make this harder.

Anyhow.  One thing is for sure, there is a big risk for BN in using a system like this to keep its majority on the long run.  If and when PK actually wins a majority due to the contradictions built up within year after year of BN rule, PK will then change the election rules.  Even if it used a non-partisan method, unless PK totally fails in its first term, PK, if it manages to hold together, would most likely be in power several terms before BN comes back to power, if BN is still around at all.

[ag]

Another problem for BN is that it is increasingly becoming just UMNO, at least in the peninsular.  MCA was important in establishing Chinese acquiescence to the Malay nationalist policy. That it can no longer do. In much of the country BN is no longer a national coalition, but merely an ethnic Malay party. This puts their entire model of governance in question.

[jaichind]

Now that the election is over and PK won a clear majority of the votes beating BN by over 3%, the battle of how to read the results has begun.  Hardliners in UNMO talks about a "Chinese tsunami" that lead to this result while the PK talks about a "Malaysian Tsunami."  The hardline UNMO argument is that Najib was too moderate when it comes to interethic affairs and that being moderate did not work as Chinese vote went to PK in droves.  They look to oust Najib later this year for this election result and move to a approach of BN as defender of Malay interests and consolidate the Malay vote.  The opposition PK which wants to avoid this Malay consolidate explains this election result in terms of a multi-ethnic swing toward PK, especially in urban areas.

As to which side is right, I looked into this problem.  I divided the 222 seats into "Chinese" seats and "Non-Chinese" seats.  They way I define a "Chinese" seat as that the Opposition runs a Chinese candidate (either DAP or PKR) AND BN runs a candidate from a Chinese party (MCA, SUPP, Gerakan.)  I found 45 seats that fits this description and the other 177 seats that do not are labeled as "Non-Chinese."  Looking at the 2008 and 2013 election results.  I found that in 2008 PK won 30 out of 45 seats and 56.81% of the vote in "Chinese seats."  Likewise, in 2008 PK won 52 out of the 177 seats and 44.05% of the vote in "Non-Chinese" seats.  In 2013 PK won 40 out of 45 seats and 64.01% of the vote in "Chinese seats."  In 2013 PK won 49 out of 177 seats and 46.48% of the vote in "Non-Chinese" seats.  So PK achieved a swing of 7.2% in "Chinese" seats and only 2.43% in "Non-Chinese" seats.  PK also was plus 10 seats in "Chinese" seats and minus 3 seats in "Non-Chinese" seats (despite a positive swing of the total vote.)  All in all the data tends to support the UNMO hardliner narrative of an "Chinese tsunami" swing of "ungrateful Chinese" despite what they feel are reasonable policies toward inter-ethnic issues as there was a much larger pro-PK swing in "Chinese seats".  It also shows that BN did a good job of putting resources into swing "Non-Chinese" seats to maximize their seat count despite losing votes there too.  Also UNMO hardline allies like Ibrahim Ali and his Perkasa which is a Malay supremacy organization working in the 2013 election in Malay districts was also effective preventing the loss of more Malay votes to PK.  It is an effective setup where Perkasa whipped up ethnic resentment to the benifit of UNMO but gives UNMO plausible deniablity so it does not lose urban liberal moderate Malay votes.

Of course, even if they got their numbers correct, the hardline UNMO approach does take hold it could lead to much greater negative economic consequences for the Malaysia as a whole.  To project a Malay supremest government and away from a multi-ethnic front could lead to the Chinese to vote with their wallets.  The difference here and say Blacks in the Old South or Dalits in India, is that the Chinese population controls over 60% of the economy and has over 60% of the national income despite being 25% of the population.  The average Chinese income is 4 times larger than the averge Malay, a gap that did not get bigger the last 40 years due to pro-Malay affirmative action but a gap that did not get smaller as it reduces incentives for the Malay population to increase their education and economic capacity.  A policy that continues to use ethnic background for economic decision making will make the economy less efficient over time leading to wealthy Chinese to invest their money overseas.   

[jaichind]

Some final thoughts on this election.  I was thinking what sort of strategy would have given PR a chance for victory.  I broke down the 222 seats into 4 types
1) Chinese: urban/suburban districts dominated by Chinese including a few in Sarawak/Sabah
2) Tribal: suburban/rual districts in Sarawak/Sabah
3) Rural: Rural heartland of Malaysia dominated by Malays with UNMO vs PAS
4) Moderate: urban/suburban multiethnic districts with plurality of moderate liberal Malays

I then tracked results over these 4 types over 2008 and 2013
      
                                     2013        2013         2013         2008        2008      2008
                    Num      Avg Vote    PR wins     PR vote    Avg Vote   PR wins  PR Vote
--------------------------------------------------------------------------------------------------------------------
Chinese         47         61885         44           65.42%      45302        32         58.34%
Tribal             47         27076           1           29.67%      16394          0         22.96%
Rural              48         41725          7            43.40%      31200        12        43.40%
Moderate       80         61768         37           50.50%      44576        38         47.33%
Total            222         49793         89           50.88%      35663        82         47.32%

Some observations
1) The disparity in seat sizes seems to be one of disadvantage toward urban/suburban districts relative to rural and tribal areas and did not seem to single out Chinese districts for discrimination.
2) The PR swing is mostly in Chinese districts (where massive swings have an weak effect because PR already did very well there in 2008) and Tribal districts (where a large swing toward PR gains little for the opposite reason of PR being to weak to gain from the swing)
3) BN did a good job gaining seats in Rural and Moderate despite zero or negative swings.  This mostly have to do with large resources BN has access to to concentrate in swing districts.

BN in retrospect ran a brilliant campaign.  What it did was to
1) Correctly write off the Chinese districts as lost.  
2) Make sure promises of subsidies continues to flow to Sarawak/Sabah so the Tribal districts are still a lock
3) Run a campaign of warning rural Malays about the impending doom when the Chinese dominated PR takes over.  
4) Run a campaign of economic reform in Moderate districts which was effective in winning over enough moderate liberal Malays to go with the devil you know

PR ran mostly a campaign of anti-corruption across the board.  But PR also promising a bunch of subsidies dulled that message.  In rural districts this might have worked but BN's campaign of protection of Malay domination was able to counteract this.  In Chinese districts PR made sure they had a lock by talking about a multi-ethnic Malaysia but did not run such a campaign in Rural and even Moderate districts for fear of provoking the PAS Islamic hardliners.  In other words alliance harmony was given a high priority and hopes of making gains in Rural districts was part of its strategy.  

In retrospect PR made a mistake on this issue.  They should have confronted BN on the topic of Malaysia for Malays versus a multi-ethnic Malaysia across the board.  This could have lead to the defection of the PAS Islamic hardliners.   But PR was defeated across the board in rural districts winning only 7 of 48 seats anyway.  If PR ran a strong campaign confronting BN on the ethnic issue PR would have lost 5%-10% in rural districts.  But since 5 of the 7 seats won by PR in rural districts are actually PAS fiefdoms won by very large margins such a strategy would have lost 2 seats only.  On the other side of the ledger, such a strategy would have won for PR the 3 remaining Chinese seats that it did not win since those were lost by very small margins.  Also such a campaign could have turned enough moderate Malays over to PR.  I looked into this and found up to 23 Moderate seats where the election was close enough where PR could have won if they used this strategy.  In this alternative strategy PR could have won up to 89 -2 +3 +23 or 113 seats which is just enough for a majority.  Of course everything has to work for this to work out which is never the case but PR in retrospect should have gone with this more risky course given the results with their middle-of-the road approach.