Taipei City mayoral race reconsidered

The last time I wrote about the Taipei City mayoral race, I predicted that Sean Lien 連勝文 would eventually win the race. This prediction flew in the face of all survey evidence, which has for several months showed Ko Wen-je 柯文哲 with a consistent and sizeable lead. However, I had a very specific process in mind. We all know that Taipei City has historically voted solidly blue. There are a large number of voters who have always ended up voting blue but who this time were telling pollsters that they either were undecided or were planning to vote for Ko. I thought that as the election drew nearer, many of these voters would eventually have to face the conflict of their blue sympathies and Ko’s green nature. Once these people got into the voting booth, I didn’t think they would be able to go through with a vote for Ko. I was writing about seven weeks before the election, and I thought that the Lien’s comeback would have to start fairly soon. I expected that the race would be much tighter by the end of October.

By now, it is pretty clear that this hasn’t happened. The polling numbers have barely moved. If anything, they have tilted toward Ko, not Lien. It has slowly dawned on many of us that the KMT could actually lose this race. For me the “A-ha!” moment was this TVBS poll from October 21. It wasn’t so much the top level number (Ko 47, Lien 32), even though that 15 point margin was the biggest to date in TVBS polls. It was the other numbers that were more shocking. (Aside: You always have to be careful looking at subgroups since sample sizes can get very small. As a general rule, if the group has fewer than 200 respondents, you shouldn’t even bother thinking about it. Anything over 400 you can be fairly confident in. This survey had a fairly large sample N=1219, allowing us to look meaningfully at some of the subgroups. Mainlanders, for example, make up 17% of the sample (1219*.17=207), so there are just enough to make so cautious inferences about.) In this survey, Lien led among Mainlanders by 54-29 (Question 2-5). The exact numbers aren’t important (or meaningful). What is apparent is that Ko has a lot of support from Mainlanders. This is a group that the blue camp has historically won by overwhelming margins, usually on the order of 10 to 1. This survey says that Lien is only winning by 2 to 1. That should be terrifying for the KMT. The report also contrasts the Oct 21 results with April 22, which was Lien’s high point. In April, Lien led Ko 47-38. Since that point, Lien has lost support in all groups, but some are more dramatic than others. For example, Ko has improved a bit among both DPP and KMT identifiers, but the most dramatic results are in the non-identifiers and the identifiers of other parties (Q 2-6). Among voters who do not feel close to any political party, Lien went from a tie (37-37) to losing by more than 3 to 1 (51-16). In Taipei City, identifiers of third parties are mostly PFP and New Party supporters. This is a heavily blue-leaning group that Lien should win handily. In April, this group was tied; by October Ko had opened up a double-digit lead.

It looks to me like there are two processes at work. The very large group of non-identifiers is overwhelmingly swinging toward Ko. This group of voters will not automatically support one party or the other. Instead candidates have to convince them. Of course, many of these voters are more sympathetic to one of the party’s positions, so it is usually pretty easy to convince them to vote for that party’s candidate. However, Sean Lien might be that rare candidate who can alienate his own party’s habitual voters. It certainly seems as if these voters are collectively deciding that the more they get to know Lien, the less they like him. Rather than closing the deal with voters who were predisposed to vote for him, Lien has managed to alienate them and make them consider staying home or even supporting Ko.

The process among third-party identifiers is also interesting. I think that Ko’s campaign manager, Yao Li-ming 姚立明, has played a critical role in this campaign. Yao is a scholar, a Mainlander, a former New Party legislator, a creative thinker, and a veteran of the talk show circuit. No one doubts that he is a Chinese nationalist, and his presence in Ko’s inner core provides a lot of credibility for Ko’s campaign. Without Yao, I don’t think a lot of Mainlanders and PFP supporters would even consider supporting Ko. However, with Yao on board, many PFP candidates have openly campaigned with Ko, and this has allowed blue voters who are dissatisfied with the Ma government or who don’t like Lien to support Ko without feeling that they are making a radical decision to cross entrenched cleavage lines.

At any rate, this race looks like it is just about over. Lien is far behind in polls, the KMT coalition in Taipei City looks like it has cracked wide open, and there doesn’t seem to be any sign that things are reversing.


A month ago, Michael Turton wrote, “Surely the Lien Campaign has to be one of the worst run campaigns in the history of the island’s democracy.” At the time, I thought it was far too early in the campaign to make such a harsh judgment. No longer. If anything, he was too generous. At this point, Lien’s campaign looks like it will go down in Taiwan’s electoral history as THE worst major (mayor / magistrate / governor / president) campaign of the democratic era. Who else has managed to blow such an overwhelmingly advantageous starting position? In case you are wondering, here are some of the other nominees for this dubious honor:

Lien Chan 連戰, KMT, 2000 presidential election

Lien Chan 連戰, KMT, 2004 presidential election

Huang Ta-chou 黃大洲, KMT, 1994 Taipei mayoral election

Hsu Huei-you 許惠祐, KMT, 1997 Nantou County magistrate election

Lin Chih-cheng 林志成, KMT, 1997, Hsinchu City mayor election

That’s an impressive list of incompetency and clueless blunders, but Sean Lien’s effort beats them all. His braintrust has consistently failed to understand one of the most basic concepts in campaigning: framing. Every graduate student (and many undergraduates) in political science for the last quarter century has read News That Matters, a study in which scholars used fake news clips to show that people responded differently when different topics were presented. Heck, you don’t even have to read a book to understand this – most people know it instinctively. You will do better if voters are thinking about issues that they think you are better on than if they are thinking about issues that they think your opponent is better on. The Lien campaign has pigheadedly insisted on repeatedly forcing us to think about Lien’s privileged birth, his wealth, corruption, honesty, personal achievements, and character. The stupidest instance of this was a couple of weeks ago when the Pingtung County government was found to have leaked documents about a surprise inspection for a cooking oil factory. Up to that point, the DPP had completely owned the cooking oil issue. However, this was a perfect opportunity for the KMT to scream that the DPP was actually the one at fault and blur responsibility. Lien should have shut up and let this story dominate the media for a week. Instead, he changed the topic by telling reporters that the Buddha had come from a rich family. How could the media resist such an invitation? Ko responded with a sarcastic “a-mi-to-fo,” a monk came forward claiming that he had advised Lien but that Lien had completely missed his point, talk show hosts had another round of debates about whether a fortunate birth should disqualify someone from public service, and no one talked about cooking oil for a few days. I am simply bewildered by these sorts of decisions. In the TV debate, it was more of the same. Lien started on the wrong foot and set the tone for the entire debate. In the first minute of his opening statement, Lien told us that he didn’t need to run for mayor but that he was doing so out of a sense of duty. Oh, thank you so much for generously debasing yourself in the service of us mere commoners. He bookended the debate in his closing statement by assuring us that he was ready to devote himself to the public, using a term 奉獻 that implies personal sacrifice from a person of higher status.

Lien seems completely oblivious to how his actions will be perceived by the general public. Of the two top faces in his campaign, Alex Tsai 蔡正元 is incredibly the one with the better image. He managed to eventually get his corruption case dismissed by a higher court on appeal. The other face is Chang Shuo-wen 張碩文, who was convicted of vote buying in the 2008 legislative election and had his seat stripped. Again, these are the two people that Lien chose to act as his most visible representatives to the general public.

Sometimes the campaign seems to be run by people who think they are the smartest people in the room. For example, in the debate Lien repeatedly offered to have an accountants association scrutinize the books from both campaigns to see who was spending excessively. We weren’t supposed to realize that Luo Shu-lei 羅淑蕾, one of Lien’s main attack dogs, is the head of that association. Hmm. I wonder what they would find. Another example is the decision to coordinate the questions from the friendly social groups in the debate. They really messed up by thinking that they could ask extremely loaded questions and not have anyone notice. It’s as if they have never had to put themselves in someone else’s shoes and ask how other people might react. Maybe this is a result of things always working out for Sean Lien – because someone always had a strong incentive to make sure things went smoothly for Lien Chan’s son – and him believing that he was the one who, through his smart decisions, produced the favorable outcome. It reminds me of an interview I heard during the financial crisis with a Wall Street trader who insisted that he was rich because he was smart, oblivious to the fact that the game was rigged in Goldman Sachs’ favor or that traders made a lot of money by breaking the laws and undermining the free market.

Would it be tacky for me to suggest a theme song for the Lien campaign? Maybe, but I need an excuse to link to this fantastic performance by a trombone virtuoso. You also deserve a break from this screed.

I originally thought that this mayoral campaign was relentlessly disgusting and depressing. I’m happy to report that I have changed my mind. Once I came to the conclusion that Lien’s campaign was a lost cause, I started looking at his efforts as farce. Every day I wake up eager to see what sort of unbelievable and counter-productive statement the campaign will come up with. Stephen Colbert would be proud of Lien’s efforts at self-ridicule. This has been especially apparent in the past few days as the Lien campaign has started to publicly melt down. A few days ago, Chang Shuo-wen publicly told Alex Tsai to shut his face. More specifically, he demanded that Tsai stop posting things on Facebook. It seems the campaign insiders have started trying to avoid blame for the impending disaster. Most of Taiwan’s younger population has hoped Alex Tsai would shut his trap at least since the Sunflower Movement, and now they find their voice is … Chang Shuo-wen?. That’s an unexpected twist. This weekend was a complete disaster for the Lien campaign. One of the old geezers with an important title in the campaign appealed to voters not to allow Ko to become the mayor of the capital of “China.” That was merely the fourth worst gaffe of the weekend. Lien’s mother came off as an ungrateful jerk by insisting that Ko had nothing to do with saving her son’s life after his shooting in 2010. Sean Lien complained that he had been unfairly attacked, with the Ko campaign going so far as to insult his ancestors. Ok, but on that very day, Lien Chan criticized Ko’s father and grandfather for having the temerity to hold public office under the Japanese. Finally, just to remind us that the Lien campaign is performance art and not a serious political effort, Lien Chan crossed a line of public decorum. You simply cannot, as Lien Chan did yesterday, call your opponent a “bastard” 渾蛋” in public. (I heard the audio and Lien was joking in the previous sentence, but (a) this was not acceptable even in that context and (b) you have to be aware of how such language will sound in print.)


It would be hyperbole to say that the Lien campaign hasn’t done anything right. I previously suggested that they needed to play the identity card, though this had to be done deftly. They actually did this in a minor episode just recently. Someone in the Lien campaign dug up an old statement by Ko complaining about the names of roads, in which Ko ended by saying that the most disgusting road name was actually Chingkuo Rd 經國路. The Lien campaign then contrasted that radical, deep green version of Ko with the current not blue, not green version of Ko. Sean Lien ended by commenting, “He will change after the election.” This was beautifully done. It reminded light blue voters of CCK, their favorite figure from the authoritarian era. It painted Ko as a deep green figure cynically pretending to be tolerant of blue voters. It raised more general doubts about whether you can trust Ko on any of the wonderful things he is promising now. Do you really know who Ko Wen-je is? Perhaps most importantly, Lien didn’t overdo it. It was a small vignette, and Lien made a small point. What the Lien campaign really needed was about twenty-five of these stories, and they needed to start using them in July and August to define Ko’s image. By November of a losing campaign, it is too late. Voters’ bullshit meters are turned on by now, and with anything less restrained than this story, it simply looks like the Lien campaign is cynically pulling out the patriotism scare tactic.

The Lien campaign has completely missed other golden opportunities. Just last week, the Ko campaign handed Lien a sparkling invitation to attack him. Ko announced that his top campaign lieutenants would not be given spots in his administration. Instead, Ko would allow online voters to determine who should get positions in his mini-cabinet. I see two immediate ways to attack this. First, if Yao Li-ming is not going to be a core member of the governing team, will Ko’s team have voices from across the political spectrum? Or will the blue elements be cast aside the day after the election in favor of a team of deep green ideologues? Second, Lien could have a field day attacking Ko’s idea of leadership. A leader has to have ideas about how to govern, and it is up to him to talk to various people to find the ones who have a similar commitment and the talent to realize that vision. Ko will delegate the most important decisions to a group of people who have no detailed information on how to, for example, resolve traffic congestion at a five-way intersection or the difficulties involved in upgrading water pipes in old neighborhoods. How can Ko lead a team that he didn’t choose, who don’t necessarily feel loyal to him, who might not work together, and who might not know anything about the problems they are supposed to solve? Lien could thump his chest and proclaim that, as mayor, he would be in charge of all the important decisions, and he certainly wouldn’t try to avoid responsibility by delegating power to some unelected and self-selected group of uninformed and inexperienced faceless ideologues. He would lead!

Of course, Lien didn’t do any of that. Instead, his father called Ko a bastard.


While I don’t think that Ko has been a good candidate in any absolute sense, I think that he has been nearly the perfect foil for Lien in this specific race. He has just enough character flaws that Lien is duped into trying to attack his character. This backfires since it merely highlights Lien’s own character flaws. Ko is just wealthy enough that Lien can’t resist asking how Ko made his money on a doctor’s salary. Again, oops. Ko is somehow getting away with appealing to light blue voters even though he is so obviously deep green, and this infuriates the Lien campaign to the point of clouding their judgment. They go too far in pointing it out, and this makes them look like oblivious deep blue ideologues. If many light blue voters think, as I do, that Ko will probably not be a very good mayor, this may also work in Ko’s favor. If Lien were running against someone with presidential ambitions and potential, light blue voters might be hesitant to advance his career. Certainly, a more charismatic opponent would shift the spotlight away from Lien and onto himself. Ko has consistently been at his best when he shuts up and lets Sean Lien do the talking. In a sense, Lien has been running two negative campaigns, one against Ko and one against himself. Ko, like so many before him, claims to be running a new type of campaign to change election culture. This almost never works, but Ko has the advantage of running against Lien. If voters go to the polls asking whether they want Lien as mayor, Ko will win. If voters instead ask themselves if they want Ko as mayor, Lien has a shot. This is why Yao Li-ming is brilliant to try to run the anti-campaign. Ko doesn’t need lots of media ads. He can just let Lien do the talking. Watch a Lien campaign ad and see if it doesn’t convince you to oppose Lien. If there is really anything the Ko campaign needs to say, they can always take advantage of the plentiful free media coverage. No one in any other race, not even Eric Chu 朱立倫, can get as much free media as he wants whenever he wants it. Taipei City is privileged. Shutting off campaign contributions with nearly three weeks to go in the election was a stroke of genius for the Ko campaign. On the one hand, they don’t need the money because they aren’t running a big campaign. On the other hand, it highlights and delegitimizes the Lien campaign’s huge financial advantage. In a normal election, you’d be crazy to tone down campaign efforts in the final two weeks. In this bizarro race, it makes perfect sense.


How low can Sean Lien possibly go? We are entering uncharted waters, so it’s hard to guess. If the normal partisan patterns completely break down (at least on the blue side), we could see an absolutely jaw-dropping result. The two most stunning results in Taipei City of the past twenty years are probably Huang Ta-chou’s 25.9% in 1994 and Lien Chan’s 21.9% in 2000. However, both of those were running in a three-way race against a charismatic candidate from the unification wing of the spectrum. In a head to head race, Sean Lien will get many of those votes. I’m having a hard time writing any number lower than 47% because I am a slave to history. However, if the number isn’t lower than 45%, we shouldn’t be so sure that Lien is a sure loser, so maybe he could go as low as 42%? Really, I have no idea what to expect. Somehow, I think there may still be another outlandish episode or two left in this very peculiar race.



9 Responses to “Taipei City mayoral race reconsidered”

  1. taiwanlawblog Says:

    What a wonderful and thorough analysis, and I too “wake up eager to see what sort of unbelievable and counter-productive statement the campaign will come up with” every day.

  2. Kungwan Says:

    Taipei’s election is just pure comedy gold… shame we don’t have comedy stand-ups milking this for all its worth. The long running TV spoof show 全民大悶鍋 has long lost its street cred for being unwilling to lampoon the ruling KMT; in its place, PTT and facebook has spawned a new generation of satire/Kuso.

  3. taiwanben2000 Says:

    Another great analysis. I wrote this – – before I read your piece so it’s reassuring to read someone with a far more detailed understanding come to roughly similar conclusions.

  4. Pat Says:

    Bit of an aside, but I wouldn’t be surprised if Ko tips Huang Shanshan to be his deputy. She’s young, popular, and deep blue, and it could be a way up the ladder for her since, as you mentioned in you post on the district 2 race, her PFP association limits her prospects.

  5. ジェームス (@jmstwn) Says:

    Now the Lien camp’s even messing up the Ching-kuo Rd. story. Took out half-page newspaper ads asking why Ko didn’t criticize “Teng-hui Blvd.” The New Taipei City transportation bureau and LTH’s office responded there’s never been a road with that official name, and “Teng-hui Blvd.” is a nickname for Jindan Rd. that was originated by local residents.
    Tsai posted on his blog, why doesn’t Ko criticize Jinlong Bridge, named after LTH’s dad? A PTT user responded, “wasn’t that named after Hu Jin-long who made it to baseball’s major leagues in the US?”

    • frozengarlic Says:

      See! The Lien campaign really is brilliant performance art! The day after I mention that they handled one topic according to conventional campaign logic, they fix their “error” by going to grotesque extremes. It’s genius!

  6. The Real Essence of 2014 Taiwan Elections - Ketagalan Media Says:

    […] independent. It would be one thing if Ko Wen-je were effectively a DPP candidate in disguise, but if recent polls are to be believed, he is something else to many “light-blue” voters. Ko has campaigned with PFP candidates for […]

  7. Steven Lai Says:

    It’s interesting to note how the story ends, our brilliant author would hesitate on estimating Lien’s rate under 42%, here is the result: 40%.

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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: