The concept of vertical accountability entails the election of representatives by citizens through democratic processes. This mechanism provides individuals with the opportunity to choose their leaders and contribute to the formulation of governmental policies. The presidential and parliamentary elections in Taiwan are regarded as being conducted in a fair and competitive manner (Murkowski 2016). In its 2022 report, Freedom House rated Taiwan as free in both political rights and civil liberties (Freedom House 2022). As the function of vertical accountability is sound in general, this paper will examine some of the issues that can be further addressed to improve democratic quality or prevent a regression in this regard.

Taiwan has adopted a semi-presidential systеm, in which the president is directly elected and may serve up to two consecutive four-year terms. Additionally, the president has the authority to nominate and replace the premier. In Taiwan, the President is elected through a direct popular vote systеm. The election is held every four years, concurrently with the legislative Yuan member election. The candidate who obtains a majority of the votes cast is elected President. The President-elect serves a four-year term and may be re-elected for one consecutive term.

In Taiwan, members of the Legislative Yuan, the country’s unicameral parliament, are elected through a mixed-member majoritarian systеm. The Legislative Yuan is comprised of 113 seats, including 73 single-member district seats, 34 party-list proportional representation seats, and six aboriginal seats. In single-member districts, the legislator is elected through a first-past-the-post (FPTP) systеm. The candidate who receives the highest number of votes wins the seat. Each voter casts two ballots: one for the district seat and another for the party-list seats. The distribution of party-list seats is based on the proportion of the overall popular vote received by each party. Subsequently, political parties submit lists of candidates for these at-large seats, and the seats are allocated to the respective parties based on their proportion of the total votes cast. In order to gain representation in the Legislative Yuan, political parties are required to secure a minimum of 5% of the total valid votes cast or win at least three district seats to qualify for at-large seats. The six aboriginal seats are elected through a single non-transferable vote (SNTV) systеm.

As of 2024, Taiwan has held eight presidential elections and ten parliamentary elections. In 2016, Taiwan experienced its third transfer of executive authority between political parties. The previous two transfers occurred in 2000 and 2008. The 2016 election also marked the first parliamentary majority for the Democratic Progressive Party (DPP), one of the two main parties in addition to the Kuomintang (KMT). In 2024, the DPP continued to secure the presidency, yet it suffered a setback in terms of parliamentary representation, with the opposition parties, namely the KMT and Taiwan People’s Party (TPP).

The concurrent election of the president and the legislators mitigated the potential for a divided government. In addition, the change in the electoral formula for legislative elections, from a SNTV to a single-member district, has also been observed to increase the number of seats held by the president’s party. Since 2008, both the KMT and the DPP have held the majority of seats in parliament. In such circumstances, the president’s party is better positioned to exert control over both the executive and legislative branches. In essence, the systеm is closer to a presidential systеm with a unified government. Similar to the practices of numerous semi-presidential countries, in instances of a lower approval rate or policy failures, the president, who is the ultimate decision-maker behind the main policies, has the authority to replace the premier to address public discontent. Thus, due to the fixed term of the presidency, the citizenry is unable to hold the ultimate decision-maker accountable between elections.

In V-Dem, vertical accountability is comprised of three essential components. The initial aspect concerns the quality of elections, which involves an overall evaluation of the integrity, fairness, and transparency of the electoral procedures. The subsequent element pertains to the percentage of the eligible population that participates in the electoral process. This component assesses the inclusiveness of the democratic systеm by considering the proportion of individuals with voting rights who exercise them. The third aspect assesses the method used to select the chief executive, specifically whether this is achieved through a direct or indirect approach.

It is pertinent to conduct a comparative analysis of Taiwan’s performance in the context of countries currently undergoing the third wave of democratization like Indonesia and the Philippines, as well as in comparison to established Asian democracies, including Japan and India. By comparing Taiwan’s performance in these aspects to that of other nations, we can gain a more nuanced comprehension of the advancements it has made in democracy. The present study employs data from the year 2021 to demonstrate the aforementioned pattern. The graph below provides a visual representation of these comparisons, revealing Taiwan’s strengths in vertical and horizontal accountability.

As seen from Figure 1, Taiwan exhibits a notably high level of vertical accountability, indicating the efficacy of electoral processes and party competition. In terms of vertical accountability, Taiwan’s performance is superior to that of the majority of emerging and East Asian democratic countries. The high scores can be attributed to three factors.

Firstly, the election in Taiwan is perceived to be fair and transparent. The electoral commission, which is responsible for delineating electoral districts and admіnistering elections, is largely autonomous. Furthermore, the judiciary in Taiwan operates with a high degree of autonomy, with court regulations largely insulated from political or inappropriate influences. The courts, which frequently adjudicate cases pertaining to vote buying, defamation, and violations of election law, are not partisan in their decision-making. Secondly, Taiwan has an automatic registration systеm, whereby all citizens residing in Taiwan are automatically sent election notification letters several days prior to the election. This allows them to cast their ballots in a voting booth located within walking distance of their residence. The election date is typically set for a Saturday. Therefore, the cost of voting is minimal. Thirdly, the President and members of parliament are directly elected by the Taiwanese people. Subsequently, the president is responsible for nominating the prime minister and has the authority to dismiss them at their discretion.

The concept of horizontal accountability pertains to the maintenance of a systеm of checks and balances among the executive, legislative, and judicial branches of government. In comparison to other newly established democracies, Taiwan’s performance in horizontal accountability is approximately moderate. However, compared to South Korea, and several Latin American countries, its scores are relatively lower. The absence of robust horizontal and diagonal accountability mechanisms may result in the deterioration of a democracy into an electoral democracy. Fortunately, judicial systеm in Taiwan is characterized by a notable degree of independence. This aligns with the general perception that Taiwan’s judicial systеm has become relatively independent following democratization and is not unduly influenced by the executive branch.

Diagonal accountability refers to the oversight exerted by citizens, social groups, and the media on government actions that fall outside the purview of the representative political systеm (Malena et al. 2004). This mechanism ensures that government actions and decisions are subject to scrutiny by civil society, thereby promoting transparency and preventing the misuse of power.

The concept of diagonal accountability encompasses a range of actions and mechanisms that can be employed by citizens, civil society organizations, and an independent media to ensure governmental accountability. The measurement in V-Dem includes four aspects: media freedom, civil society characteristics, freedom of expression, and the extent of citizen engagement in political affairs. As illustrated in Figure 2, Taiwan exhibits a comparable level of performance in diagonal accountability to that observed in vertical accountability. The high performance of diagonal accountability in Taiwan denotes the capacity and active participation of civil society organizations, which contribute to the strengthening of vertical accountability. In the case that the government seeks to weaken vertical accountability, mass media and civil society organizations can collaborate to prevent the government from making wrong decisions.

Figure 3 depicts the change of Taiwan’s vertical accountability index. Following the Second World War, Taiwan’s score was relatively low due to the implementation of martial law and the holding of regular elections only at the county level. The data clearly illustrate a significant upswing in Taiwan’s horizontal accountability score in the late 1960s, coinciding with the gradual expansion of elected members of parliament. In the early 1990s, a further notable increase was observed, along with the country’s transition to democracy. Notably, this score has exhibited a commendable level of stability since that time. Taiwan’s remarkable increase and stability in vertical accountability since its democratic transition in the early 1990s, as depicted in the graph, underscores the country’s commitment to democratic principles.

As illustrated in Table 1, the voter turnout for presidential and legislative elections in Taiwan has exhibited a consistently high level of participation, with over 70% of the eligible population casting their ballots in most elections. In 2016, however, due to a significant poll gap between the two leading candidates, voter turnout declined to 66%. Following the 2008 electoral reform, the presidential and parliamentary elections are held concurrently. Consequently, the voter turnout for these two elections is almost identical. Therefore, we limit our analysis to the presidential election turnout.

	President election turnout	percentage of candidates who are below 40	percentage of male candidates	percentage of female candidates
2008	76%	19%	71%	29%
2012	74%	12%	68%	32%
2016	66%	18%	66%	34%
2020	75%	19%	62%	38%
2024	72%	17%	59%	41%

Table 1 also shows the age and gender distribution of parliamentary candidates across years. It can be observed that the rate of youth participation has not increased significantly, remaining below 20%. This means that candidates under 40 years of age constitute less than 20% of all candidates, which is not proportionate to the proportion of young people. With regard to the gender difference, while male candidate participation remains relatively high, female participation has been increasing across years, from 29% in 2008 to 41% in 2024, denoting the rising female participation in politics.

Furthermore, the election participation rate of young voters and gender groups can be examined. In order to facilitate a comparative analysis of electoral participation in Taiwan with that in other Asian countries, we utilize the fifth wave of the Asian Barometer Survey (ABS).

First, we use the following question: “Thinking of whether you voted or not ever since you became eligible for voting, how would you describe yourself – have you voted in every election, voted in most elections, voted in some elections or hardly ever voted?” This was done in order to capture political participation. The respondents who indicated that they had voted in every election and in most elections were classified as frequent voters, while those who responded that they had voted in some elections and had hardly ever voted were classified as infrequent voters. As can be seen from Figure 4, more than eighty percent of respondents reply that they are frequent voters. The electoral participation rate in Taiwan is not significantly higher than that observed in most other democracies in the Indo-Asia region. Only Thailand, Indonesia, and Australia exhibit significantly higher rates of electoral participation. The response to this question is inherently susceptible to social desirability bias, which presents a significant challenge in cross-country comparisons.

The subsequent section examines the election participation rate, disaggregated by gender. As seen from Figure 5, the voting rate is approximately equivalent between male and female voters in Taiwan. In certain countries, such as Japan, Korea, and Malaysia, male citizens exhibit a higher turnout. In the Philippines, however, women have a higher election participation rate than men.

Next, we look at the election participation rate, breakdown by age groups, as shown in Figure 6. The voting rate is considerably lower among young voters below the age of 40 in Taiwan. In the majority of countries within this region, young citizens are considerably less likely to be frequent voters. The discrepancy between the voting rates of young and adult citizens can reach 30 to 40 percentage points in countries such as Japan, South Korea, Taiwan, Vietnam, and Malaysia. In contrast, in some countries, such as Thailand, Indonesia, Australia, and India, the election participation rate gaps are much smaller between age groups.

In the recent general elections, the main candidates focused primarily on issues related to national defense, security, and sovereignty, with limited discussion on other public policy issues. The issue of sovereignty is largely rooted in political identity and can be employed as a means of consolidating support from core supporters in an effective manner. Conversely, a number of significant socio-economic concerns in Taiwan, including the potential insolvency of the labor pension scheme, severe demographic challenges, global warming, labor shortages, and the fiscal costs of energy subsidies, have not been subjected to sufficient scrutiny and debate. Instead, the three candidates and the media have primarily focused on discrediting their opponents based on minor flaws in their real estate holdings.

For example, in the 2024 presidential election, the three principal candidates did sparsely propose certain policies pertaining to these critical issues. However, there has been a lack of comprehensive discourse and sufficient public attention. Moreover, the three candidates frequently eschew addressing the fundamental causes of these issues by proposing measures that could potentially disadvantage specific groups. For example, concerning the mounting deficits in the labor pension scheme, all three candidates refrain from contemplating the potential of augmenting the contributions of laborers. In lieu of proposing alternative solutions, the candidates have pledged to maintain the current approach of relying on government funding to address the pension fund deficits.

With regard to their respective election platforms and campaign trails, the three sets of candidates demonstrate no discernible differences in their policy positions on a range of socio-economic issues, including industrial, labor, education, housing, and health policies. Political parties are generally aware of the appropriate policies beneficial to society as a whole; however, they often prioritize those that will not jeopardize their chances of winning elections. With regard to the energy issue, the KMT and TPP support the continued utilization of nuclear energy, whereas the DPP advocates for its immediate phase-out. The most notable divergence in political positions among candidates lies in cross-strait relationship.

Regarding cross-strait relations, the candidates all advocate for a close Taiwan-American alliance, increasing Taiwan’s defense deterrence, and engagement in dialogue with the mainland on the basis of equality and dignity. The U.S. academic and media Anticipated that the election of any of the three candidates will not result in any significant alterations to the U.S.-China-Taiwan relationship. All candidates have visited the United States, met with scholars from universities and think tanks, and held discussions with American officials, ensuring a clear understanding of their positions on the relationships.

However, the concepts of “equality and dignity” is interpreted differently by the three parties. Hou You-yi accepts a “One China with Different Interpretations” framework, which encompasses both the Republic of China and the People’s Republic of China. Additionally, he is opposed to Taiwanese independence and has cautioned against the risk of war. Similarly to his predecessors, Lai Ching-te does not accept the “One China” policy, stating that it is equivalent to “one country, two systеms.” The KMT has accused Lai Ching-te of pro-Taiwan independence and of inciting military conflicts. In turn, Lai Ching-te has characterized the KMT’s policy as tantamount to surrender. Ko Wen-je’s stance is more ambiguous. Additionally, in terms of cross-strait economic relations, Lai Ching-te emphasizes the strong linkage between economic and national security. It would be prudent for Taiwan to reduce its reliance on the Chinese market while simultaneously seeking to strengthen its economic ties with its democratic allies. In contrast, the KMT advocates for the establishment of more robust economic ties with mainland China.

In comparison to previous candidates from the KMT, Hou demonstrates a notable shift towards a more robust emphasis on democracy and Taiwan’s sovereignty, accompanied by a discernible expression of skepticism towards the Beijing government. They advocate a diplomatic policy for Taiwan that is pro-American, friendly with Japan, and engaging with China. The DPP’s policy is characterized by a pro-American, pro-Japanese, and anti-communist China stance.

In recent years, candidates have frequently elevated the stakes of general elections to a level where the outcome has been seen as a choice between the survival of democracy, the preservation of sovereignty, or the future of the country. Against the backdrop of China’s rising threat of force over Taiwan, the ruling party seized the opportunity to exploit the perceived threat. During the election campaign, the DPP candidate cast himself as the defender of Taiwan’s sovereignty, claiming that if the opposition candidate was elected, Taiwan would be forced to surrender to China, democracy would collapse, and Taiwan’s sovereignty would be weakened. In response to a multitude of criticisms levied against its domestic policies, the DPP has sought to deflect attention by attributing the blame to China, such as the dissemination of misinformation. The victory of the DPP in this election is undoubtedly influenced by the ongoing threat from China, coupled with the ruling party’s exploitation of the existential threats. However, the DPP garnered merely 40% of the vote, suggesting that the rhetoric of existential threat may not resonate profoundly with the majority of Taiwanese voters. Some voters may prefer more dialogues with mainland China, thereby reducing the potential for miscalculation and the risk of military confrontation.

On the other side, the KMT also leveraged the China threat by framing the election as a choice between war and peace. They asserted that the s pro-Taiwan independence stance and confrontational approach of Lai would contribute to the escalation of conflict, forcing young people to go to the battlefield. Lai has repeatedly stated that he has no intention of modifying the official name of the Republic of China or amending the constitution.

The external threats to Taiwan’s liberal democracy originate directly from China. Additionally, it is attempting to leverage its political and economic influence to shape public opinion and influence the policies of political parties in Taiwan. The dissemination of misinformation represents one of numerous strategies that China is utilizing to exert influence over Taiwan. One objective is to exert influence over the competition between political parties in Taiwan. Such narratives tend to denigrate the performance of the DPP government, impugning its ability to maintain peace in the Taiwan Strait. This strategy aims to diminish the approval rating of the ruling party and boost support for the opposition parties that are more amenable to China’s policies. Such actions have the effect of distorting the competitive landscape between political parties.

Conversely, in recent years, China has provided financial support to individual politicians with the objective of assisting candidates and political parties that are aligned with China’s interests in winning presidential elections. Subsequently, several of them have been subjected to investigation and prosecution. The scope of these activities is, in general, quite limited and does not extend to the mainstream political parties.

The second threat arises from the potential for supposedly neutral government agencies to intervene for political gain. By strategically emphasizing the significance of the election, the government provides itself with a rationale for its actions. Such circumstances have the potential to compromise the fairness of the electoral process and erode public confidence in the democratic process. For example, in the days preceding the election, China launched a satellite that passed over Taiwanese airspace. The Defense Ministry issued an air raid alert, stating that it was a Chinese missile test (Central News Agency 2024-01-09). The dissemination of misinformation contributed to an increased perception of national security threats and potentially enhanced support for the ruling party. The government has also intervened more frequently in the electoral process by instructing law enforcement agencies to investigate and prosecute political opponents and citizens for disseminating false information and aiding China’s infiltration before the election. In recent years, Taiwan has passed the Social Order Maintenance Act to tackle fake news and the Anti-Infiltration Act to counter Chinese influence in elections and politics.

While China is known for engaging in cognitive warfare and infiltration, the DPP also exploited this threat. In the context of fake news, the act of criticizing the government or commenting on government policies may be perceived as cognitive warfare, which could result in legal consequences and a chilling effect on free speech (Wu 2023). Additionally, the Ministry of Justice proposed more rigorous examination of new immigrants from China to counter foreign influence, sparking controversy due to its targeted focus on a specific group.

Another issue concerns the willingness to abide by the fundamental rules. Such actions are not exclusive to a particular political party. The acceptance of the result of democratic competition represents a fundamental tenet of democratic governance. Taiwan has undergone eight presidential elections and several party turnovers. Taiwan has successfully surpassed the two-turnover test of democratic consolidation proposed by Samuel Huntington. However, there have been instances where candidates have not adhered to the established regulations during the primary phase. While this phenomenon is most prevalent at the local primary level, it has been manifested in this electoral cycle as well. During the DPP primary, Tsai Ing-wen postponed the primary and modified the electoral rules on multiple occasions following a decline in her poll ratings due to a challenge from her former premier. Subsequently, she was able to reclaim her victory following Xi Jinping’s speech, which prompted Taiwanese citizens to rally behind her. Concurrently, Terry Kuo, the chair of Foxconn, who participated in the primary as the KMT candidate, similarly declined to endorse the winner of the election after placing second in 2020 and again in 2024.

Because of the alleged failure of representative democracy to address the needs of the people, the New Power Party (NPP) also promotes direct democracy. In 2018, the NPP and the DPP government proposed amendments to the Referendum Act, lowering the threshold of eligible voters required to initiate a referendum proposal (from 0.5 percent to 0.01 percent) and the threshold required to put a referendum proposal to a vote (from 5 percent to 1.5 percent). Furthermore, the amendments reduced the quorum required to pass a proposal. Before the amendment, it was necessary for at least 50 percent of the electorate to cast ballots, with an absolute relative majority of valid votes being required. The current legislation requires that at least 25 percent of the electorate cast ballots and that a relative majority of valid votes be cast. Furthermore, the recently enacted legislation has eliminated the necessity for a review committee, which was previously empowered to reject a referendum proposal. As a direct consequence of the alteration to the quorum, the number of citizen initiatives has mushroomed significantly. In the 2018 municipal and county elections, ten referendum proposals were put to a vote, encompassing a range of issues including same-sex marriage, nuclear power, and air pollution, as well as changing the name used for Olympic and international competitions from Chinese Taipei to Taiwan.

In its review of the referendum law, the NPP proposed an extension to include the possibility of deciding constitutional clauses by referendum. One of the most important issues is the designation of the nation’s official title and territory. These are particularly contentious issues, given Taiwan’s unique position within the global power structure. The NPP also proposed amending the law to require that any cross-strait political negotiations be subjected to a referendum prior to their commencement. The term “political negotiations” is quite vague. Following the Sunflower Movement, the implications of trade agreements have also become a matter of significant political and national security concern. Ultimately, these two proposals were rejected by other major political parties.

As a consequence of the recent amendment to the referendum law, which reduced the quorum requirement, ten referendum cases were put to a vote in the 2018 local elections. The sheer number of cases is overwhelming for the public to understand, digest, and ultimately make informed decisions about. Furthermore, the proposal, deliberation, and voting processes for such cases are conducted within a relatively short timeframe of just two months. The current practice of direct democracy does not allow thorough social deliberation that can be found in some mature democracies. Taiwan is a society that is divided along national identity lines and in its views of cross-strait political and economic relationships. The format of a referendum often presents a binary choice, which may limit the potential for compromise. In the case of identity-related issues, it would be preferable to engage in deliberation and pursue compromise within the framework of representative institutions.

After two years, in 2020, the DPP government reached the conclusion that the implementation of multiple referendums proposed by opposition parties could potentially have an adverse impact on its electoral prospects. Consequently, the ruling party opted to amend the referendum law. The revised law restricts the impact of referendums on presidential elections by scheduling them on a biannual basis, rather than concurrently with presidential elections. ■

Central News Agency. 2024. “Wrong Translation of National Missile Raid Alert Department of Defense Apologize.” [國家級警報衛星誤譯成飛彈國防部致歉] January 9. https://www.cna.com.tw/news/aipl/202401095003.aspx (Accessed January 28, 2024)

Malena, Carmen, Reiner Forster, and Janamejay Singh. 2004. Social Accountability: An Introduction to the Concept and Emerging Practice. Social development papers: No. 76. Washington, D.C.: World Bank.

Murkowski, Lisa. 2016. “International Election Observation Mission, 2016 - Taiwan.” Congressional Record 162, 33.

Wu, Chin-en. 2023. “Taiwan’s Civic Space Threatened by Chinese Misinformation and the Government’s Worrisome Legislative Responses.” ADRN Issue Briefing. February 10. http://adrnresearch.org/publications/list.php?idx=294 (Accessed January 28, 2024)

■ Chin-en Wu is an Associate Research Fellow at the Institute of Political Science at Academia Sinica, Taiwan.

■ Edited by Hansu Park, Research Associate
For inquiries: 02 2277 1683 (ext. 204) | hspark@eai.or.kr