Published on
Image description

The Chiang-Shen debate on October 31 was a highlight of the campaign so far.

Taiwan's local elections are now only about two weeks away. I've been somewhat haphazardly updating the election previews that I wrote in August -- see:

  1. Previewing the Races
  2. The Big Six
  3. North and Central 
  4. South, East, and Offshore Islands 
  5. Summing Up.

But I thought it'd be useful to provide a more comprehensive update of the election campaigns in one place. 

Ratings Changes

Here are the races which have shifted significantly since August. 


Taipei: Likely KMT --> Leans KMT. 

I still think Chiang Wan-an has the edge. But this has turned into a much more interesting race than I, at least, expected. And Puma Shen / Shen Po-yang deserves a lot of credit for running an energetic, creative and issues-driven campaign. Rating change on October 17 to Leans KMT. 


Tainan: Likely DPP --> Safe DPP

This race was probably decided back in early September when no serious pan-green rivals registered for the race. That left Chen Ting-fei in a head-to-head green versus blue matchup with Hsieh Lung-chieh. In Tainan, that's a recipe for a blowout. The national political environment and the limited polling results here also point in the direction of a comfortable DPP win. Rating change on November 2 to Safe DPP. 


Keelung: Leans KMT --> Toss-Up. Incumbent mayor Hsieh Kuo-liang has has soft support numbers for months. DPP challenger Tung Tzu-wei's latest achievement is getting James Soong to appear on stage with him. A poll from Mirror Media from October 18-21 actually put Tung slightly in the lead, 37.0%-36.6%. Rating change on October 17 to Toss-Up. 


Hsinchu County: Leans KMT --> Toss-Up. A late change here: I've gone back and forth on this one. On the one hand, Hsinchu County is very blue (B+11), and Hsu Hsin-ying is a known quantity and sitting legislator from the county. On the other hand, the breakdown of KMT-TPP cooperation in the county, and the fight between the two parties over the Jhubei City nomination, could drag down support for Hsu. Cheng's Chao-feng's Hakka background and deep family connections in the county may also scramble the usual partisan patterns. Polls also are all over the place, but the latest (from Era News) has Cheng leading, and it's not the only one to show that. And even national KMT figures are sounding the alarm that they are at grave risk of losing here. Add all that up and this should be in toss-up territory. Rating change on November 14 to Toss-Up.  


Miaoli County: Likely KMT --> Safe KMT. This race has gotten overshadowed by the fireworks in Hsinchu County and the Taipei campaign. Without a third-party candidate in the race, Chung Tung-chin appears safely on his way to re-election now in a deep blue county. Rating change on November 10 to Safe KMT. 


Changhua County: Leans DPP --> Likely DPP. The few polls of this race show Chen Su-yue still firmly in the lead, and Wei Ping-cheng hasn't raised his support much from where he started. Perhaps more important, the incumbent KMT magistrate Wang Huei-mei still hasn't endorsed Wei or campaigned for him. Unless something dramatic happens soon, this is setting up to be a surprisingly easy DPP win -- one that Cheng Li-wun is going to bear a lot of the blame for inside the KMT. Rating change on November 10 to Likely DPP.     


Chiayi City: Leans DPP --> Likely DPP. Wang Mei-hui has declined to participate in a debate with Chang and the third-party challenger Huang. That's classic front-runner strategy. This race hasn't gotten much attention in the national media, and that also is to Wang's benefit. There's no new reliable polling here, but without anything to shake up the race this is trending towards a comfortable victory for Wang and the DPP. Rating change on November 10 to Likely DPP. 


Chiayi County: Likely DPP --> Safe DPP. Like the race in Chiayi City, this one has not gotten much national attention. Wu Pin-jui needs something splashy to move away from the baseline case here, which is an easy DPP win for . She hasn't gotten it so far, and there won't be a debate. Rating change on November 10 to to Safe DPP.


The overall picture now looks like this: 


Safe KMT (4): Miaoli, Nantou, Kinmen, Lienchiang

Likely KMT (1): Taoyuan 

Leans KMT (5): Taipei, Taichung, Yunlin, Hualien, Taitung

Toss-up (4): New Taipei, Keelung, Hsinchu County, Penghu

Leans DPP (2): Yilan, Pingtung 

Likely DPP (3): Kaohsiung, Changhua, Chiayi City

Safe DPP (2): Tainan, Chiayi County  

Leans IND (1): Hsinchu City


Assigning the leaners to their respective camps would put the KMT at 10, DPP at 7, with one IND (Kao in Hsinchu City), and four toss-ups. The KMT could still pull out a decent result if it holds the races it's leading in now, and wins the toss-ups -- especially if it wins New Taipei. But this is a significantly improved picture for the DPP, and a worse one for the KMT, than in August.


So why the shift?  

The Macro Picture: The Fundamentals

Public opinion is trending in the DPP's favor. MyFormosa's monthly polls show President Lai's approval rating has continued to improve over the last six months, and it's now above 50%. That's not determinative -- Tsai Ing-wen's approval rate was even higher when the DPP had a terrible performance in the 2022 local elections -- but it certainly doesn't hurt the party's candidates to have a popular president they can associate themselves with.

Image description

President Lai's approval rating has gradually risen to over 50% over the past six months. / Credit: MyFormosa

The DPP is also leading in the traditional party ID measure that NCCU's Election Study Center has tracked for over 30 years. Their share of the electorate, at 33.8%, is near an all-time high for the party, and there are now more self-identified DPP partisans than KMT+TPP partisans combined.

Image description

ESC's partisan ID trend line now shows more DPP partisans (33.8%) than KMT + TPP (29.8%). / Credit: Election Study Center

The DPP is also much more popular than the other two. From MyFormosa's October poll, the favorability ratings (喜好度) for the parties:

  • DPP: +46.5% vs - 40.3%, for a net favorability rating of +5.2%
  • KMT: +29.4% vs -55.0%, net -25.6%
  • TPP: +25.8% vs -57.4%, net -31.6%

That gap between the DPP and KMT is, frankly, astounding, and represents a dramatic reversal in political fortunes over a short time span. In August 2025, the KMT's favorability rating (-9%) was significantly better than the DPP's (-24%); now the KMT has fallen 30 points (!) behind.

Image description

The KMT is now viewed negatively by 55% of respondents, and positively by less than 30%. / Credit: MyFormosa.

And somehow, the TPP is now even less popular than the KMT. They seem to have been saddled with all the negatives from associating closely with the KMT without benefitting from  being in the opposition. I say this in every set of remarks I give about Taiwan politics now: the TPP has not played its hand well since it won the critical swing seats in the legislature in January 2024. As a consequence, it is clearly past its peak: that ESC party ID table shows an inflection point at the end of 2023 and a steady decline thereafter, one that looks remarkably similar to the rise and fall of the New Party in the late 1990s and the PFP in the early 2000s. Ko Wen-je may become the next James Soong. 


(To be fair, the TPP still rates better than the Chinese Communist Party, which has a favorable - unfavorable rating of 6.1%-76.1%, or a 70 point spread. So at least they still have that going for them. /Waves tiny flag.) 

Image description

KMT party chair Cheng Li-wun is now extremely unpopular; only 15.5% have a favorable view of her. / Credit: MyFormosa

Party Chair Cheng Li-wun is dragging the KMT down. There's an interesting analytical question about why the DPP and KMT's relative popularity has reversed so dramatically in such a short period of time. Some of it is probably reversion to the mean: August 2025 was the low point for President Lai's approval ratings, and the DPP's mass recall gambit had just failed spectacularly, while the KMT was riding high after their unexpected success. 


But there's another significant change that happened shortly after: in October 2025, Cheng Li-wun won the party chair race. The decline in the KMT's standing since then has been gradual enough, and with enough volatility, that the long-term trend has been somewhat obscured. But the latest MyFormosa data now show it clearly: the KMT's favorability ratings have dropped 10 points since Cheng assumed the top role. Additional evidence comes from a separate MyFormosa poll in late October, which asked about views of eight leading political figures. In that poll, Cheng's favorability rating was far and away the worst, at only 15.5% of respondents. (Tsai Ing-wen was at the top with 61.7% favorability; President Lai was a middling 49.7%; and Taichung Mayor Lu Shiow-yen and Taipei Mayor Chiang Wan-an were the most popular KMT figures at 36.9% each). 


Cheng had a personal triumph of sorts when she traveled to mainland China in April 2026 and was granted an audience with Xi Jinping. But over the longer run, that doesn't seem to have helped either her personal standing with the electorate or the KMT's public appeal. Her subsequent visit to the United States in June generated more controversy than positive headlines, and her intervention in some of the nomination decisions for the local mayor's races has added to grumbling among KMT elites about her leadership of the party.


She also hasn't managed expectations very well for the local elections. She noted in an interview in July that the party's candidates would struggle to hold Chiayi City and Changhua County, but expressed confidence about competing in Kaohsiung and optimism about holding Yilan and Hsinchu Counties. If the party loses all five of those races plus New Taipei, that's an underperformance even by her own benchmarks. If on top of that they lose one of the other special municipalities -- Taipei is the most prominent, but Taichung would be equally shocking, and even in Taoyuan the most recent poll shows a tightening race -- then Cheng is going to be under a lot of pressure to resign. And she probably will deserve a lot of the blame for the losses, whatever she might say otherwise. 

Image description

Taiwan's nominal quarterly GDP growth has hit highs not seen in almost 40 years.

The Taiwanese economy is roaring. Thanks to the AI boom and data center build-out, which have supercharged demand for leading-edge logic chips, semiconductor companies and high-tech adjacent businesses in Taiwan are making money hand over fist. Estimated nominal GDP growth in the 3rd quarter is over 14%, and estimates of real annualized quarterly growth rates (accounting for inflation) are now above 12%. These are numbers Taiwan's economy hasn't hit since the 1980s. Most Taiwanese who own stocks are feeling a lot richer these days, as well: the Taiwan stock exchange has surged about 60% over the last year.


It's true that these gains aren't shared evenly, and they have exacerbated an already large wealth gap between investors and hourly wage workers. It's not any easier than it was two years ago to buy a house if you're young with limited savings and not working in the tech industry. This unequal distribution of economic gains is also the motivation behind opposition-sponsored legislation to provide cash payouts to all citizens (which the DPP government relented to after the recalls failed last year) and to eliminate health care premiums for people over 65 (which the premier has refused to countersign).  


Nevertheless, the DPP still has a strong economic record to run on right now. If retrospective voting models have any predictive power, then we should expect that Taiwan's stellar economic performance will provide an extra boost to DPP campaigns across Taiwan, and it could override the historical tendency of voters to favor opposition candidates (re: those not from the president's party) in local races. 

The Micro Picture: Shen Po-Yang has Shaken Up the Campaign

Image description

The Taipei Race Has Caught Fire. At the level of individual races, the big news over the last two months has been Shen Po-yang's surprising surge in the Taipei mayor contest (which his supporters have branded the Ocean Current / 洋流 in a play on Shen's name). Sometime in September his campaign caught fire: he got a bunch of positive press for going on hostile talk shows and charming the hosts, huge crowds started turning out to see him speak during his campaign events, and other DPP candidates across the island started appealing for him to join them at their rallies. Those qualitative indicators were backed up by a series of polls that came out in September and October, which all showed a tightening race. The one that really jolted political observers was conducted by Tai Li-an and sponsored by the Taiwan Trust Association, which showed support for incumbent mayor Chiang Wan-an and Shen almost level. 


Shen's personal profile and political skills have recast Chiang's supposed political strengths in a less favorable light by comparison: Shen is also young (ish), good-looking, US-educated, speaks good English, etc. If these are really the reasons voters should support Chiang, then Shen is offering just as good or better. One of the ironies of a competitive campaign is that lots of people reexamine the incumbent, and stuff that has been in broad daylight for years suddenly appears problematic. Chiang's not a deep policy wonk, for instance, while Shen has a strong reputation as a good public speaker and a quick thinker, and he's emphasized his grasp of local policy details during many of his campaign events.


Faced with an alarming slide in the polls, Chiang has stepped up his campaign activities, and he finally agreed to debate Shen after dodging the opportunity. The debate was held on October 31, and I think it was pretty high stakes for both candidates, but especially for Chiang. It was an important moment for him to reset the narrative that had been developing about his shallowness and over reliance on aides to run the city government and handle questions and criticisms. For two hours, Chiang was alone on stage with Shen without any aides to lean on, and I thought he held his own.


Here's where the expectations game is so important: to hear some green partisans describe him, Chiang is a sheltered, privileged politician who memorizes rather than thinks deeply and has spent his career "being walked to first base." So the fact that he had well-prepared answers and went on the attack against Shen, I think helped him dispel some (not all) of that line of critique. I know I'm a bit contrarian on the Chiang-Shen race, but I think it's still Chiang's election to lose at this point. 


But...What Looks Like a "Yang-Liu" Taipei Spillover Effect Is Probably Just The Fundamentals. I'm also feeling contrarian on another point. There's a prominent narrative among the green camp that the DPP is going to do well in the local election campaigns are mostly because of Shen Po-yang: through sheer force of personality and clever campaigning, he's created a sea change (SWIDD?) in the electorate, somewhat like the Han Kuo-yu phenomenon in 2018. Some pundits have argued that Shen's popularity is spilling over into other races in northern Taiwan -- New Taipei, Keelung, and even Taoyuan, and that's the main thing helping the DPP candidates improve their standing there, too.   


Now, I think Shen has run a good campaign, and he's already succeeded in making Taipei a real contest. He's also been a massive hit among the green camp -- there was even an election rally for him in Cupertino last week. But there's a common tendency among political pundits to ascribe electoral outcomes to campaign effects, when something more general may actually be driving shifts in voting behavior. I've noted above that presidential approval rating, party ID, and economic growth all favor the DPP right now. In addition, the KMT is defending a lot of mayor's positions, including in places that lean green or are tossups under normal conditions. And in the north, Taipei (B+3), Taoyuan (B+4) and Keelung (B+7) aren't actually that much "bluer" than the national average. So if the fundamentals favor the DPP, we'd expect these races to become more competitive. What looks like a Yang-Liu effect may in fact be these races shifting based on general national factors.


Something to debate more as we head into the final stretch of this campaign... 


0 Comments

About Me

I am a political scientist with research interests in democratization, elections and election management, parties and party system development, one-party dominance, and the links between domestic politics and external security issues. My regional expertise is in East Asia, with special focus on Taiwan.

Archives

Categories