Does the Legislative Yuan Need More Legislators?

The allocation of at least one district seat to every county results in mal-apportionment that in many cases favors KMT candidates
Timothy Rich

In 2005 the Legislative Yuan altered how legislators are elected by discarding the single non-transferable vote (SNTV) system for a mixed system combining single member districts (SMD) like congressional districts in the U.S. with seats allocated by proportional representation (PR). [1] At the same time, Taiwan took the unusual step of cutting the number of legislative seats, from 225 to 113, citing in part poor public opinion of legislators in general.

Recently, Tsai Ing-wen (蔡英文) and members of the opposition Democratic Progressive Party (DPP) have proposed increasing the number of seats in Legislative Yuan, an idea that Legislative speaker Wang Jin-pyng (王金平) of the Chinese Nationalist Party (KMT) has been open to discussing. Several pathologies of small legislatures, from filling committees to capturing constituency demands, have been raised. However, without a comparative framework, evaluating these claims is difficult.

Only 72 countries have smaller legislatures than Taiwan, a country with 23 million people. Many of these 72 are microstates. To put things in perspective, the average population of these countries is 2.6 million, dropping to 1.2 million if restricted just to those countries labeled “Free” by Freedom House. Taiwan currently has a ratio of roughly 203,000 citizens per legislator, the 15th highest ratio across all countries with legislatures, regardless of whether they are democratically elected. [2] This is far above the average across the globe of roughly 81,000, a number that drops to 72,000 if restricted to countries that are labeled “Free.” Five “Free” countries have higher ratios: India (1st at nearly 1.48 million citizens per legislator), the U.S. (2nd), Indonesia (5th), Brazil (9th), and Peru (12th). Of the nine “Free” countries using a mixed legislative system, Taiwan’s ratio is roughly double the average (104,000). Cursory cross-national evidence thus suggests that increasing the number of seats would push Taiwan to be more in line with other democracies.

Even if broad agreement among parties to increase the number of seats emerges, how these seats would be elected is likely to be contentious. Currently, only 34 of the 113 seats are filled by proportional representation. Yet, since districts considerably outnumber the party list seats, the results of winner-take-all competition in districts can lead to a very disproportional overall seat distribution. The allocation of at least one district seat to every county, including sparsely populated Penghu, Kinmen, and Lienchang (all with populations under 100,000), results in mal-apportionment that in many cases favors KMT candidates.

This disproportionality was most evident in the 2008 legislative election, where the KMT captured 51.2% of the party list vote, but received 72.57% of seats by winning a disproportionate number of district seats. Meanwhile the DPP captured 36.9% of the party vote, but only 23.9% of seats. While 2008 may be viewed as an anomaly, the results in the 2012 election still show a disproportionate boost for the KMT. Whereas the party list vote share differed little between the KMT and DPP (44.55% vs. 43.56%), the total seat share favored the KMT (56.64% vs. 35.4%).

Nor are such results uncommon in similar mixed legislative systems where district seats outnumber party list seats, as evidenced by elections in Japan and South Korean. Thus, one should not be surprised when Tsai supported increasing the number of party list seats, which would produce a more proportional overall distribution of seats. While smaller parties in particular would benefit from increasing the number of party list seats, making it more likely that a party can clear the 5% vote threshold to receive any of these seats, such a change would likely worry the KMT as it not only likely cuts into their winner’s bonus, but may embolden their smaller coalitional partners.

Increasing the party list seats does sidestep the more thorny issue of redistricting. Following the halving of the seats after the 2005 election, districts averaged slightly over twice the population per legislator compared to the pre-reform era. Four counties saw no difference in this ratio after reform, essentially creating over-representation (Taitung, Penghu, Kinmen, and Lienchang) while six areas (Hsinchu City and County, Keelung, Tainan, Yilan County, and Yunlin County) have roughly three times the population per legislator as under the old SNTV system. Determining new district boundaries threatens incumbents in particular who would in most cases have a vested interest in the status quo.

While increasing the number of seats in the Legislative Yuan would potentially resolve many of the problems with the current structure, adding seats in the absence of deliberation and explicit goals in mind could nevertheless be just as shortsighted as the earlier reduction in seats.

1. In addition six seats are reserved for two three-seat aboriginal districts.
2. Population estimates are derived from the CIA Factbook (2009).

Timothy S. Rich is an assistant professor in political science at Western Kentucky University. His main research focuses on the impact of electoral reforms in Taiwan, South Korea, and Japan compared to similar legislative systems (e.g. Germany, New Zealand). His broader research interests include electoral politics, domestic and international politics of East Asia, and qualitative and quantitative methods.

Comments are welcome, but will be moderated. Remarks containing abusive language, personal attacks or self-promotion will not be published. We encourage healthy discussion and, above all, tolerance of other's views.