KMT troubles in Changhua 3

Apparently Cheng Ju-fen 鄭汝芬 has decided not to accept the KMT’s nomination to run for re-election in Changhua 3. She specifically cited the effect of Hung Hsiu-chu’s “One China with a common interpretation” position, saying it was too far away from public opinion. [Edit: She stated that the KMT has drifted too far away from public opinion. She did not specifically cite Hung Hsiu-chu; I confused her statements with those of KMT legislator Chang Chia-chun.]

I have been watching Changhua 3 for a few years now. This district is very similar in many ways to neighboring Yunlin. Compared to the rest of Changhua, the southwestern corner is less densely populated, more agricultural, more exclusively Min-nan, and has lower education levels. The biggest town (Erlin) is also a noted center of organized crime, similar to Huwei in Yunlin 1. Both areas were traditionally KMT strongholds, as the local factions’ grip on the voters kept the DPP at bay. This changed in Yunlin about 10 years ago, and the county suddenly shifted dramatically into the DPP’s column. However, the Changhua factions managed to hold on. The KMT incumbent in Changhua 3 is one of the prime reasons. She is from a political family that has dominated the area for at least three generations, and she easily won the legislative races in 2008 and 2012. Nevertheless, the KMT is under increasing pressure.

To illustrate, I have concocted very, very crude indicators for ethnic identity and party ID for each electoral district. The typical survey has about 1000 cases, so each of the 73 legislative districts will only have about 15 cases. That is far too few to estimate a parameter. I have combined several surveys from the past four years, and this has yielded about 200 cases for each district. 200 is still small, implying a margin of error of about 7% for a 95% confidence level. Still, it is sufficient to get a crude picture. Further, some of these surveys are a little older, so I have weighted the more recent surveys more heavily. This is anything but definitive.

On party ID in Changhua 3, my estimate is that the DPP now has a lead over the KMT, by 29.5-22.5%. Even considering the small sample size, a gap that large probably reflects an underlying DPP advantage. On ethnic ID, 71.5% think they are Taiwanese, 4.7% think they are Chinese, and 22.1% think they are both. The 71.5% figure is the highest point estimate of any district in the country. Even considering the large margin of error, Changhua 3 certainly has an overwhelmingly exclusive Taiwanese identity. Hung Hsiu-chu’s China discourse is probably highly unpopular here.

If Cheng Ju-fen isn’t representing the KMT in this district, I think it is highly likely to become a DPP pickup. I’m expecting Tsai Ing-wen to win this district by a healthy margin. If Cheng Ju-fen stays in the race, she might possibly be strong enough to overcome the partisan deficit, similar to the way Chang Chia-chun 張嘉郡 and Wong Chung-chun 翁重鈞 were able to hold Yunlin 1 and Chiayi County 1 in 2012. However, the combination of the KMT’s swing toward a pro-China stance and the entrance of a PFP candidate into the Changhua 3 race might have been sufficient to convince Cheng that it was becoming an unwinnable race.

If the localist rebellion in the KMT is confined to Yunlin and southern Changhua, it is bad but not necessarily catastrophic. These districts were probably going to flip anyway. However, if it expands further north, the KMT should probably start panicking.