campaign trail: DPP rally in Taipei

I’ve been wrong about Ko Wen-je more or less continuously for the past five years. In 2014, I was pretty sure KMT supporters would return home and vote for Sean Lien. It was a position I held to for months, until about six weeks before the election when it became clear that public opinion wasn’t shifting and the Lien campaign started saying some really crazy stuff almost as if they wanted to ensure that the shift would never happen. Even after Ko won (with 57%!), I refused to believe that he was really that popular, that he represented a fundamentally new force in Taiwanese politics, or that he was a reasonable bet as a future president. I believe Ko’s presence in Taiwanese politics is due far more to a historically inept opponent than most people admit.

This spring, when the DPP decided to cut ties with Ko and nominate its own candidate, I thought that Ko was pretty much doomed. KMT sympathizers would support Ting Shou-chung, who, though not exactly mesmerizingly charismatic, is a solid and inoffensive longtime party soldier. Ting is not Lien, so Ko wouldn’t be able to count on peeling off 10% of the electorate from the pan-blue side again. More importantly, the DPP would be able to rally its supporters to come home. Historically, the DPP has almost never seen its vote base abandon it for a rival candidate. Even though Yao was trailing Ko by a huge margin at the time, I thought that Yao would be able to pull back another 10% from Ko which would put them roughly in a tie. If Ko ever slipped into third place in the polls, the ensuing strategic voting would destroy him. I laid out these ideas in a blog post, and then Donovan Smith, Michael Turton, and I had a further discussion on Facebook, which Michael summarized on his blog. You can read the whole exchange, but to summarize, Donovan thought that I was wrong and that Ko would hold his support and win re-election. I fundamentally see Taiwanese politics through the lens of (the established) party politics, while Donovan thought that the old party lines were ripe for change, at least in Taipei. Donovan was right. Sometime in August or so, after yet another poll showing Ko expanding his lead over Ting and Yao continuing to languish around 10%, I sent Donovan a FB message telling him his crystal ball was better than mine.

So now that I’ve admitted I’m a stupid moron who doesn’t know anything, it seems that the race might be turning again to make me look even more stupid for abandoning my position too early. As the Kaohsiung race heated up, KMT partisans all over the island became more and more excited. When one of Ko’s cabinet members expressed support for the DPP in Kaohsiung, Ko supporters with KMT sympathies were incensed. In the last few polls published before the polling blackout, Ko’s previous double digit lead had disappeared. Instead, they showed that Ko and Ting were nearly tied, with Yao still languishing far behind.

What happened to Yao? For the first three months of 2018, he made a vigorous case against Ko, holding marches and rallies nearly every weekend. By May, his efforts had led the DPP to a fairly unanimous decision to cut ties with Ko. I thought that after Yao was nominated and any possibility of cooperation had been removed, the DPP attacks on Ko would kick into an even higher gear. Instead, Yao nearly disappeared. From May to September, Yao was nearly absent from view. DPP candidates in other cities and counties put together alliances, but they generally shunned Yao. The DPP never even got around to forcing Ko’s allies on the city council to take sides. Kao Chia-yu, for example, is still running for re-election on the DPP label, even though she is widely believed to be siding with Ko. It almost looked as if the DPP had given up on Yao and was tacitly signaling to its supporters that it wouldn’t mind if they voted for Ko. However, in the last week or two, as the campaign has heated up and Ko’s position in the polls has slipped, the DPP leaders have made more and more noise about being firmly behind Yao.

 

That brings us to Sunday. On the Golden Weekend, Yao had one last chance to demonstrate his viability and ask DPP supporters to stay with him. As I said in my previous post, most rallies have two goals, to pump up enthusiasm so that your base will mobilize their friends and families and to give them some talking points to use as ammunition. However, in this case, since Yao is the third candidates in a three way race, the goal was something else. Yao had to demonstrate that his base was still there and still large.

Three way races are unstable. People supporting the trailing candidate often vote for their second favorite candidate to prevent their most disliked candidate from winning. This makes information critical. If you have clear information about who is leading and who is trailing, it becomes obvious which voters should abandon which candidate (the one in third place). However, if the information is unclear, then voters don’t always know what to do. If the third place candidate can make a credible argument that he or she is not actually in third place, the results can shift wildly. Yao hoped to have an enormous turnout for his event on Sunday afternoon, preferably quite a bit larger than Ko’s event on Sunday evening, so that he could sow doubts about who was actually in second place. However, let’s be realistic. At best, this will have only a moderate effect. Yao has been consistently trailing by wide margins. It is highly unlikely that he is actually in second place, regardless of the size of his event.

A different way to think about the problem is one of credible commitment and common knowledge. Yao wanted a big turnout to demonstrate that he still has a base that supports him and is not planning to vote strategically. If they can credibly commit to supporting Yao, that sends a signal to Ko’s supporters that since Yao’s side will not defect to Ko, the only way to stop Ting is for Ko’s supporters to strategically vote for Yao. Moreover, with a huge turnout that gets widespread publicity, both Yao’s supporters and Ko’s supporters get this message. It probably won’t work, but it might convince wavering Yao supporters to stay with Yao rather than defect to Ko.

 

On Sunday, Mrs. Garlic and I went to see for ourselves how Yao’s event would turn out. Yao’s campaign organized a hold-hands-and-form-a-giant-ring event. In 2004, the Chen Shui-bian re-election campaign* organized a hold hands around Taiwan event on 2-28, and then a few weeks later Chen was re-elected. Many DPP people remember this event as a moving and galvanizing demonstration, so Yao wanted to remind his supporters of their history and galvanize to support him. This time, the ring encircled the Sun Yat-sen Memorial and the Taipei Dome project, ending at the steps of City Hall. The controversial Taipei Dome project is a centerpiece of Yao’s campaign, so it was a logical choice.

*Yes, I know it was nominally organized by some other group. Don’t be naïve.

We got to the event a little after 2:30 and walked around the entire ring to see what it would look like. Of course, people were still arriving when we went through the first few sections, so I assume they filled out a bit more after we left. We were supposed to hold hands at 3:33 (because Yao’s candidate registration number is 3), but the section I was on at that time never did. It seemed like the organization wasn’t great. Each side of the ring was designated as the responsibility of one of the six city council districts. All of the DPP city council candidates (except Ko ally Kao Chia-yu) had a presence, and most of them had mobilized a few hundred supporters in their section.

The instructions seemed to indicate that after 3:33, the city council candidates were supposed to go around high-fiving people (who were facing the road). Maybe that happened in other sections that I wasn’t on. Then, everyone was supposed to move to the steps of city hall for a few speeches, including an address by Yao. There were a couple of problems with that plan. For one thing, the stage in front of city hall was tiny, and they faced it in the wrong direction. My first thought was that there must be another, bigger stage somewhere else. Nope. The bigger problem was that it seemed like most of the people in the big ring never made it to the rally. After sitting around for two hours, I think a lot of people got tired and just took the MRT home.

Ok, but what was the turnout like? It’s almost impossible to estimate the crowd. They were spread out over a long set of sidewalks at various densities that changed over time. Since many people didn’t go to the stage area, we never saw the entire crowd all at once. If the idea was to show your supporters and the wider public how massive Yao’s base was, this event was a colossal mistake. We never saw that picture of an enormous, roiling crowd. Even the people at the event only saw their section of the sidewalk; they didn’t see how big the ring was or how the line kept stretching and stretching. If Yao’s campaign had wanted to minimize the impact of their crowd, they couldn’t have done much better. I guessed that the campaign would claim 50,000 (and it did) – implying an actual number around 25,000, but the number of people who attended some part of the event could have been anywhere from 15,000 to 60,000. I really don’t have any idea, and since I walked the entire ring, I had one of the best views.

Who showed up? It was the base. The people who showed up were older, and most of the signs and banners indicated that they were the Independence Fundamentalists. It wasn’t a particularly happy (or unhappy) or passionate crowd. They weren’t particularly angry or sullen, but they were more determined and dedicated than frenzied. I got the feeling that they were there because they felt it was important that Yao had a good turnout more than because they were excited to go out and cheer for Yao. I don’t know whether the message was transmitted to the rest of the general public, but I clearly got the message that the DPP base is going to stay loyal to the party. The fundamentalists are not going to strategically defect to Ko.

 

Strategically, this is bad news for Ko. I didn’t go to his event at the North Gate, but the TV reports I saw made it look pretty small. He claimed 40,000, but it looked smaller than 20,000 to me on TV. At any rate, it was clearly a far cry from his huge march around the city and rally in front of city hall on Golden Weekend 2014.

The last stages of election campaigns almost always pull voters back somewhat to their partisan roots. In this year’s condensed campaign season, I think that is happening even more dramatically than normal. If KMT and DPP voters are really drifting back to Ting and Yao, Ko is in trouble.

One of the more notable facets of the last month is that Ko has disappeared somewhat. For four years, Ko has pretty much been able to get high profile media coverage whenever he wanted. However, over the past month, the public discourse has been dominated by Han Kuo-yu and the resulting struggle between the two major parties. Ko has been pushed off the front pages. What happens to a charismatic outsider who suddenly can’t dominate the news? It’s an entirely new experience for Ko, and it couldn’t come at a worse time. (Aside: Can we try this with Trump?)

Yao has waited too long to make his move. Yao should have started his attack earlier, he should have mobilized a huge rally two weeks ago, and he should have resigned his seat in the legislature two or three weeks ago. It was always unlikely that he would win, but I originally thought he might be able to push Ko into third place. However, that scenario required that Yao could show the electorate a poll with him at least close to Ko. Now that the blackout period has started, we have no polling data to inform us. Most voters will continue to assume that the race is still basically what the final polls said it was: a close race between Ko and Ting. Still, I think Yao may have done enough to convince most of the DPP base to remain loyal to the DPP, and that will probably be enough to throw the election to Ting. If you want a guess, I’ll go with Ting 45, Ko 38, and Yao 17.

Of course, I’ve been wrong about Ko Wen-je before.