Three cities and counties experienced a third pattern, that of outright partisan conflict. As in the second pattern, the choice facing the LEC was not trivial. The LEC had substantively different alternatives to choose from and selected the one most advantageous to the local governing party. Unlike in the second pattern, in these three cases the other camp was able to settle on a fundamentally different alternative. Whereas the plans in the lottery in the second pattern differed by less than 4%, the alternatives in this pattern shifted 15-20% of the population to a new district.
The critical feature separating the second and third patterns was that, in the third pattern, the alternative plan did not significantly harm the interests of anyone in the local opposition camp. All the major stakeholders were either better off or indifferent. As seen in the second pattern, this is a significant threshold, and the difficulty of achieving this threshold increases with size. Perhaps it is not surprising that in two of the three cases in this category, there were effectively only two seats to be decided. Both Taichung City and Pingtung County were apportioned three seats, but in both cases, there was consensus on one of the seats. However, the third case demonstrates that complexity was not a deterministic factor. Taoyuan County had six seats, and there was a consensus around only one of these seats. That left five seats to fight over. Nevertheless, the DPP was able to converge on an alternate plan. The plan not only made
the DPP better off from the standpoint of the whole party, but more importantly, it did not make any incumbent substantially worse off.19 Thus, while generally speaking, the third pattern was more likely to occur in medium sized cities and counties, this is a useful reminder that size is not a deterministic factor. The political skill needed to find plans that everyone in the camp could agree on also mattered.
Significantly, these three cases are arguably the three cases in which the LEC plans most overtly pursued partisan advantage. As noted above, the LEC plans in both Taoyuan County and Taichung City violated the CEC’s guidelines. The Pingtung LEC plan was within the CEC’s guidelines, but only barely (see Figure 3). The fact that these LEC plans were so overtly partisan may have played a role in helping the opposition to unite around an alternative plan.
Source: Data compiled by author from newspaper reports and Central Election Commission.
Figure 3 Map of Pingtung County
19 The only DPP incumbent whose vote share in his best district declined under the DPP plan was Peng Tian-fu. However, this drop was very small, from 9.1% to 8.9%.
DPP plan districts township borders
KMT plan D1 KMT plan D2 KMT plan D3
Pingtung City
Wandan
The third pattern is illustrated with the case of Pingtung County (see Figure 3). Pingtung had a DPP local executive and was apportioned three seats. Both parties agreed on D3 in the southern part of the county, so the question was how to divide the northern part into the remaining two districts. The main controversy dealt with the more urban areas centered on Pingtung City and Wandan Township. These areas are more favorable to the blue camp than the rural areas, possibly because Pingtung’s Mainlander population is disproportionately located in the urban areas. The plan proposed by the LEC split the urban areas and combined them with surrounding rural areas to form D1 and D2. This created one district (D1) in which the DPP had a very small advantage, one (D2) in which it had a moderate advantage, and one (D3) in which it had a very large advantage. The partisan intent of this plan is clear from the uneven population distribution.
The average district should have 281,126 people, and Pingtung City alone had 212,753. In an effort to negate the blue camp’s advantage in Pingtung City, the LEC plan added as many rural townships as possible to it to form D1. The result was a district that had 319,312 people, 13.6%
over the mean, and a very slight green camp advantage. The LEC plan also helped the DPP by placing its incumbents in different districts (see Table 4).20
Table 4 Pingtung County Redistricting Plans
Population +/- Blue Camp Green Camp
KMT Plan
D1 293,608 4.4 42.6 55.1
D2 278,875 -0.8 50.3 46.2
D3 270,896 -3.6 41.7 56.9
DPP Plan
D1 319,312 13.6 47.8 48.8
D2 264,718 -5.8 44.5 53.3
D3 259,349 -7.7 41.5 57.0
Source: Data compiled by author from newspaper reports and Central Election Commission.
Notes: Total population=843,379; mean district population=281,126. The last two columns show the percentage of the vote won by all Blue and Green Camp candidates, respectively, in the 2004 legislative elections.
The CEC approved the LEC plan, but blue camp legislators proposed an alternative in the legislature. This plan, which I label the KMT plan, packed the urban areas together to form
20 Lin Yu-sheng was clearly stronger in D1 than in D2 (11.5% to 8.6%), and Cheng Tsao-min was clearly stronger in D2 than in D1 (13.0% to 10.4%).
one blue-leaning district (D2) and two districts with very large green majorities. Moreover, the KMT plan threw the DPP incumbents’ plans into disarray. Under the KMT plan, two DPP incumbents had roughly the same support levels in both D1 and D2.21 However, it is not sufficient to demonstrate that the KMT plan was better for the blue camp as a whole; it is necessary to show that it was also better for all the major stakeholders. In 2004, the blue camp elected three legislators in Pingtung. One of these, Liao Wan-ju, was based in D3 and would have been indifferent to the two plans.22 Wu Jin-lin clearly benefitted under the KMT plan. Wu was from Wandan township and enjoyed strong support in Pingtung City. The KMT plan put his best two areas together in a district (D2) that favored the blue camp.23 The third blue camp incumbent, independent Tsai Hau, is the critical one. Tsai was based in D1, a district with a significantly larger green camp advantage in the KMT plan than in the LEC plan. There are two reasons that Tsai was happy with this arrangement, and both stem from the fact that he was an independent rather than a KMT member. On the one hand, Tsai had to worry about competition from the KMT. Because the blue camp was so weak in D1, the KMT was more willing to forego nominating its own candidate and to yield the district to Tsai. On the other hand, as an independent, Tsai was just as concerned about his own vote share, which increased from 14.0%
in D1 in the LEC plan to 15.2% in D1 in the KMT plan, as he was about the overall blue camp vote share.
With the blue camp unanimously supporting the KMT plan and the green camp equally united around the LEC plan, there was no compromise to be found. The party leaders supported their respective plans all the way to the lottery, in which the KMT plan was chosen.