Perception of China

The rise of China has led many to question whether the American era is coming to an end (Nye Reference Nye2020). Washington’s policy analyses frequently portray China as a significant challenger to US national interests. In fact, the 2021 US Annual Threat Assessment identifies China’s push for global power as the top threat to US national security (Office of the Director of National Intelligence 2021). The idea that China poses a major challenge to the US is so widely accepted by US lawmakers that the media has even concluded a hardline stance on China is “one of the few truly bipartisan sentiments in the deeply divided US Congress” (Zengerle and Martina Reference Zengerle and Martina2021).

The view that China is a major challenger of the US has significant implications for how US lawmakers deal with Taiwan. For many lawmakers, Taiwan may be the first and most vulnerable target of China due to its geographical closeness to Beijing, especially given China has not ruled out the use of force to achieve unification and the expanding military imbalance across Taiwan Strait. House Representative Michael McCaul considers Taiwan as “a neighbor to China and stalwart ally of the United States,” “facing the most immediate threat from an increasingly aggressive CCP” (McCaul Reference McCaul2019). Additionally, Taiwan is a critical part of US strategy in East Asia, whose goal, according to some, is to contain China. Former US Assistant Secretary of State for East Asian and Pacific Affairs Daniel Russel was outspoken about Taiwan’s importance to the US, saying that, for many Americans, Taiwan is an “offshore rebuke to the PRC” and a “blunt instrument used to harass and discomfit the Chinese Communist Party” (Russell Reference Russell2020). The Biden administration’s Indo-Pacific Strategy explicitly stated that the efforts to build a free and open Indo-Pacific region require strengthening relationships with partners such as Taiwan (White House 2022). Furthermore, by committing to supporting Taiwan, the US indicates to the world the strength in its capability and resolve to protect its allies and the international system (Rigger Reference Rigger2011). As Richard Bush replied when asked why Taiwan matters: “how the Taiwan Strait issue is resolved is an important—perhaps the most important test—of … how the US will play its role as the guardian of the international system” (Rigger Reference Rigger2011, 193).

Different perceptions of China, such as those of economic competitor, non-democracy and human rights violator, or a security threat, can result in varying levels of support for Taiwan. As Representative Rick Larsen aptly notes: “In Congress there are national security hawks on China, trade hawks on China, and human rights hawks on China” (Girard Reference Girard2019). Each group may harbor negative sentiments toward China for different reasons, and these can shape their stance on Taiwan. Legislators who view China primarily as an economic rival may not have a strong incentive to support Taiwan as a solution to economic frictions with China. This is because Taiwan, like China, runs a trade surplus with the US. In 2020, the US goods trade deficit with Taiwan increased by $7.4 billion, reaching $29.9 billion, which represents a 32.8 percent increase compared to 2019.Footnote ³ In 2020, the US Department of Treasury added Taiwan to its monitoring list of major trading partners due to a significant expansion in Taiwan’s trade account surplus (Sutter Reference Sutter2020). In 2021, the Treasury initiated talks to develop a plan with specific actions aimed at addressing the underlying causes of Taiwan’s currency undervaluation. In addition, US exporters often face difficulties with Taiwan’s trade barriers, including tariffs, non-tariff barriers, intellectual property protection, and investment barriers. Of these barriers, the exportation of US pork and beef is especially contentious (Glaser, Bush, and Green Reference Glaser, Bush and Green2020). The Office of the United States Trade Representative (USTR) suspended trade talks with Taiwan under the Trade and Investment Framework Agreement in 2007 and did not resume them until 2013. Such trade talks stopped again in 2016 and did not occur at all during the Trump administration. In addition, many Taiwanese products exported to the US are manufactured in China.Footnote ⁴ Supporting Taiwan by, for example, expanding US–Taiwan economic ties may further contribute to China’s trade surplus with the US.

The perception of China as a human rights violator and non-democracy may prompt US legislators to explicitly support Taiwan. Friedberg (Reference Friedberg2005) points out that if the US is more likely to be hostile toward China because it is not a democracy, it is also more inclined to assist perceived democratic polities threatened by China, even if such action may not align with a realpolitik calculation of its interests. An example of the US linking China’s undemocratic behavior to support for Taiwan is when a group of pro-democracy activists in Hong Kong were arrested in 2021, and the State Department announced plans to send the US Ambassador to the United Nations to visit Taiwan, calling it “a reliable partner and vibrant democracy that has flourished despite CCP efforts to undermine its great success” (US Department of State 2021). Today, US officials widely perceive Taiwan as a democracy. However, the contrast in regime types between China and Taiwan raises concerns about the survival of Taipei. House Representative Donald Manzullo refers to Taiwan as “one of the few beacons of freedom and democracy in a very dangerous neighborhood” and says that “[US] support for Taiwan must remain strong and steadfast.”Footnote ⁵ Moreover, supporting Taiwan may prove to the world, especially to those who live in mainland China, that democracy is compatible with Chinese culture (Christensen Reference Christensen2002). At a committee hearing on the twenty-fifth anniversary of the Taiwan Relations Act in 2004, House Representative James Leach commented that Taiwan’s democratic transition is of great importance to residents in mainland China “who have yet to enjoy the political freedoms many Taiwanese now take for granted.”Footnote ⁶ This approach could potentially pave the way for democratization in China, especially considering previous studies have shown that economic sanctions have not effectively improved the human rights situation in the country (Drury and Li Reference Drury and Li2006). Lastly, Taiwan’s democratic success is a powerful example of the resilience of democratic values amidst competition between US-led democracy and Beijing’s autocracy.

Perceiving China as a security threat should also be influential on US lawmakers’ support for Taiwan. When viewed as a non-democracy and human rights violator, China’s intentions to affect Taiwan’s democracy are concerning to lawmakers. However, it is China’s growing security ambition and capability that leads lawmakers to believe that China could actually carry out such actions. Moreover, the increasing military imbalance across the Taiwan Strait raises concerns that Taiwan’s survival is in jeopardy. Lawmakers often point to China’s expanding military strength as a reason for advocating for greater US support for Taiwan and for abandoning the “strategic ambiguity” approach.Footnote ⁷ For example, House Representative Ted Yoho argues that the strategic ambiguity approach “has failed to deter the China of today from building up an immense military presence along the Taiwan Strait and repeatedly threatening military confrontation” (Yoho Reference Yoho2020). Recently, multiple House Representatives introduced the Taiwan Peace and Security Act in the 117th Congress (2021–2022), which aims to enhance US support for Taiwan on a range of issues, including bolstering deterrence against a cross-strait conflict. Representative Steve Chabot, one of the original cosponsors, stated that he is “particularly concerned about the military dimension, and the increasing likelihood of a PRC miscalculation or, even worse, an invasion” (Bera Reference Bera2021).

Lawmakers’ concerns about the increasing military disparity between Taiwan and China and its potential impact on Taiwan’s security are consistently reflected in the annual National Defense Authorization Act (NDAA) and other national security legislation. For instance, the 2015 NDAA required the Secretary of Defense to report to Congress on the Chinese People’s Liberation Army’s potential impact on Taiwan’s maritime and territorial security. The 2021 NDAA stated that it is US policy that China’s increasingly coercive and aggressive behavior towards Taiwan runs counter to the expectation of the peaceful resolution for Taiwan’s future. The 2018 Asia Reassurance Initiative Act explicitly stated that the President should conduct transfers of defense articles to Taiwan regularly, which are customized to address the current and future threats from China. Therefore,

Partisanship

Taiwan is essentially a foreign policy issue. Drawing from issue ownership literature, I argue that Republican lawmakers are expected to care more about Taiwan because their party “owns” the issue. According to Egan (Reference Egan2013), issue ownership “describes the long-term positive associations between political parties and particular consensus issues in the public’s mind-associations created and reinforced by the parties’ commitments to prioritizing these issues with government spending and lawmaking” (156). His empirical analysis shows that during a four-decade period, unified governments under Republican control enacted a greater number of important laws than Democrats on foreign policy. In addition, Republican control of the government is accompanied by a net increase in federal funding on foreign policy. While why Republicans have formed this tradition is beyond the scope of this paper, Egan’s research shows that Republicans are more likely to prioritize foreign policy issues than Democrats.Footnote ⁸

Egan’s focus on the partisan priorities when defining issue ownership is supported by the legislative record, as Table A1 shows that of the 65 Taiwan-specific pieces of legislation introduced from 2007 to 2020, 54 were introduced by Republicans and only 11 by Democrats.

Other scholars have also observed a positive association between the Republican party and foreign affairs in the minds of voters (Craig and Cossette Reference Craig and Cossette2020) and party elites (Fagan Reference Fagan2021). The association can have different meanings in different contexts, such as the amount of attention given to foreign policy by different parties or which party is perceived as being better able to handle foreign policy.

By emphasizing foreign policy issues such as Taiwan through the introduction of relevant legislation, Republicans can strengthen their reputation as a party that prioritizes US foreign policy. This could potentially be an effective strategy during election campaigns. When attacking their Democratic opponents, Republican candidates may criticize their lack of foreign policy proposals as a sign of weakness on foreign policy. Therefore,

Measuring lawmakers’ support for Taiwan

There are various ways for lawmakers to show support for Taiwan. I focus on the number of bills and resolutions that lawmakers sponsor and cosponsor during a two-year term in Congress, with a higher count indicating greater levels of support. Only legislation with “Taiwan” or “One China Policy” in their titles were considered as they indicate a direct and strong interest in the issue.Footnote ⁹ While some other legislation may include “Taiwan” in their texts, the Taiwan-specific articles may only be a small portion, making it difficult to determine lawmakers’ level of interest in the issue. I only examined the House of Representatives, recognizing that support for Taiwan in the Senate may differ from that in the House. Table A1 lists all Taiwan-specific legislation from the 110th to 116th House of Representatives. Table A2 shows the majority party of each House of Representatives and the Taiwan-specific bills sponsored by Republicans and Democrats. Table A2 reveals that Republicans have proposed more Taiwan-specific bills than Democrats in each Congress, regardless of which party is in the majority in the House.

Roll-call votes are often seen as an intuitive way to determine the position of lawmakers, but they have several limitations. First, voting on bills related to Taiwan is uncommon. In Table A1, only 16 out of 65 Taiwan-specific measures received House floor action. Of those 16, only 4 bills received a roll-call vote, and out of those, only one bill had an opposing vote.Footnote ¹⁰ Therefore, there is little variation to be explained between legislators. The remaining 12 bills received a voice vote, making it difficult to count the number of “yeas” and “nays.”

Second, voting restricts lawmakers from choosing other than “yea” or “nay,” a pair of dyadic alternatives that most legislators have no part in shaping (Schiller Reference Schiller1995; Talbert and Potoski Reference Talbert and Potoski2002). It is important to recognize that not all lawmakers who vote “yea” on a bill necessarily have the same level of support for the issue. For instance, imagine three legislators who all voted “yea” on a Taiwan-specific bill. Legislator A might not have a strong interest in the Taiwan issue but voted “yea” simply to avoid being one of the few dissenting voices. Legislator B may prefer a bill that expresses stronger support for Taiwan but voted “yea” because they think the current bill is better than having no bill at all. Finally, legislator C voted “yea” because they genuinely believe that the bill aligns with their stance on how the US should deal with the issue. Thus, the dichotomous nature of voting may mask the nuanced motivations of lawmakers.

To overcome the limitations of roll-call voting, I used the number of Taiwan-specific bills and resolutions that lawmakers sponsor and cosponsor during a two-year term of one Congress as a measure of their support for Taiwan. In Congress, only one legislator can claim sponsorship of a measure, but the number of cosponsors is unlimited. Sponsoring and cosponsoring are better indicators than roll-call votes. First, they are less constrained by party leaders and more directly reflect lawmakers’ preferences (Alemán et al. Reference Alemán, Calvo, Jones and Kaplan2009). Additionally, lawmakers can introduce an unlimited number of bills in a single congressional term, which allows for a better reflection of the level of support they have for Taiwan. Thus, sponsoring and cosponsoring can not only indicate a lawmaker’s stance on an issue but also demonstrate the intensity of their preferences (Rocca and Gordon Reference Rocca and Gordon2010).

Some may argue that most bills and resolutions are introduced by lawmakers for symbolic purposes and are not intended to pass, which undermines the value of studying them. However, previous research shows that sponsors provide a significant push on the majority of proposed measures (Krutz Reference Krutz2005). Another worry is that cosponsoring is cheaper than sponsoring, and lawmakers do not take it seriously. Although cosponsoring is “less costly, it is not costless” (Wichowsky and Weiss Reference Wichowsky and Weiss2021, 641). Members of Congress have limited resources to advance their goals (Hall and Deardorff Reference Hall and Deardorff2006). They face an opportunity cost for every bill they participate in proposing as they cannot spend the time writing other bills that might help to build a reputation. Furthermore, lawmakers are selective in cosponsoring, with each legislator only cosponsoring an average of 396 out of 11,488 bills and resolutions introduced in the 116th House. Finally, legislators face repercussions if they break their cosponsoring agreements (Bernhard and Sulkin Reference Bernhard and Sulkin2013). In sum, studying introduced measures provides valuable insights into lawmakers’ preferences and priorities.

Sponsoring and cosponsoring data have been used to gauge legislators’ preferences on many issues. For example, Rosenson, Oldmixon, and Wald (Reference Rosenson, Oldmixon and Wald2009) use the number of sponsorship and cosponsorships of Israel-related bills to examine senators’ support for Israel. Kleinberg and Fordham (Reference Kleinberg and Fordham2013) investigate whether House Representatives’ trade interests in their electoral districts influence the number of hostile measures to China that they sponsor and cosponsor. Meanwhile, Heaney and Rojas (Reference Heaney and Rojas2015) delved into how lawmakers’ partisanship and ideology affected the count of antiwar legislation that they sponsor and cosponsor.

Figure 1 provides insight into the level of support for Taiwan among members of Congress from 2007 to 2020 (110th–116th Congress). The density distribution plots illustrate the number of measures sponsored and cosponsored, revealing a long tail that indicates a high level of activity among certain members. While many remained relatively inactive, others proposed as many as six, seven, or more bills to support Taiwan. Table 1 identifies these “active Taiwan supporters” and lists their names alongside the number of bills they sponsored and cosponsored.

Figure 1.

The number of measures legislators sponsored and cosponsored to support Taiwan (2007–2020).

Table 1.

Active Taiwan supporters in House of Representatives (110^th–116th Congress)

Note: The table shows the names of the House Representatives and the number of Taiwan-specific bills and resolutions they sponsored and cosponsored.

Independent variables

To test H1, I measured lawmakers’ perception of China based on the number of China-related bills and resolutions they sponsored and cosponsored. The data selection is based on the following procedure. First, on the website Congress.gov, I applied two filters: “Congress” (“110th-116th”) and “legislation type” (“All House”). Second, I limited search results to measures that have at least one of the following words in their titles or summaries: China, Chinese, People’s Republic of China (PRC), Communist, Tibet, Hong Kong, Macau, Dalai, Uyghur, Huawei, and Zhongxing (ZTE). All these words are highly relevant to China. I excluded bills and resolutions that deal with budgets, including authorization and appropriations measures. These omnibus measures are typically lengthier than traditional legislation and address a wide range of issues, and the proportion of China-related content within them is minimal, making it difficult to judge the views of legislators regarding China-related matters in these measures. I then read the text of each one of the bills and further excluded ones that do not portray China as a challenger to the US. First, I excluded proposals that have nothing to do with the Chinese government, such as those congratulating US athletes on their performance during the 2008 Olympics in Beijing. Second, I excluded bills that express kindness or promote cooperation with China, such as expressing condolences to the victims of the 2010 earthquake in China. Finally, I also excluded the bills that contain Taiwan in the abstract or the main text because it is unclear whether these bills aim to support Taiwan, oppose China, or have a combination of both. In other words, they can be coded as independent variables or dependent variables. Therefore, I removed them from consideration. Table A3 in the Appendix provides a complete list of all the China-related legislation I selected.

To categorize China-related bills in a way that accurately reflects their content, I followed Wichowsky and Weiss (Reference Wichowsky and Weiss2021) and grouped China-related measures into four categories: economy, human rights and democracy, security, and others. These categories are not mutually exclusive, as a bill may cover multiple issue areas. For each legislator, I then coded the number of relevant bills they sponsored and cosponsored across the three variables of China economic bills, China human rights bills, and China security bills.

The sub-categories for each major issue are as follows:

• • Economy: financial markets, trade, intellectual property rights, supply chain, aid, investment, debt, tariffs, China’s market economy status, consumer products safety, WTO Government Procurement Agreement, economic espionage, currency manipulation, acquisition, bonds

• • Human rights and democracy: Hong Kong, Chinese human rights activists, Tibet, Uyghurs in Xinjiang, China’s role in human rights abuses in foreign countries (e.g., Sudan and North Korea)

• • Security: South/East China Sea, Syria, People’s Liberation Army, cybersecurity, Indo-Pacific region, Chinese political influence in the US and overseas, arms sales, telecommunications industry (5G), natural resources security, security of US allies, fentanyl

• • Others: climate change, Covid-19 pandemic, and others

To test H2, I assigned a value of 1 to Democrats and 0 to all other legislators (Democrat).Footnote ¹¹

Control variables

I accounted for several potential drivers that may shape lawmakers’ support for Taiwan.

Personal interests

First, since Taiwan is primarily a foreign policy issue, it is reasonable to expect that members of Congress who sit on the Foreign Affairs Committee would be particularly invested in this issue. According to Adler and Lapinski (Reference Adler and Lapinski1997), such lawmakers may be more interested in foreign policy due to their personal backgrounds or the preferences of their constituents. I assigned a value of 1 to members who served on the Foreign Affairs Committee and 0 to those who did not (Foreign Affairs Committee).

Second, the Congressional Taiwan Caucus (CTC) has been actively promoting US–Taiwan relations since its founding. It is reasonable to expect that CTC members are more likely to propose Taiwan-specific legislation to improve ties between the US and Taiwan. Therefore, I coded CTC members as 1 and others as 0 (CTC). Data are from the Formosan Association for Public Affairs website.

Third, I included the total number of bills and resolutions (divided by 100) sponsored and cosponsored by each lawmaker (Active lawmaker). The rationale behind this is that the more active a lawmaker is, the more likely they are to support Taiwan by proposing relevant measures.

Finally, the ideologies of lawmakers may also influence their support for Taiwan. Those who hold strong liberal or conservative beliefs may have unique positions on foreign policy issues. Compared to other lawmakers, they are more likely to propose legislation aimed at modifying current US policy towards Taiwan. To code each lawmaker’s ideology, I used the first dimension of the DW nominate score (Poole and Rosenthal Reference Poole and Rosenthal2001) and converted the original score to its absolute value (Ideology).Footnote ¹²

Constituents’ interests

In his seminal book Congress: The Electoral Connection, Mayhew offers a simple assumption: members of Congress are “single‐minded seekers of reelection” (Mayhew Reference Mayhew2004, Reference Mayhew5).Footnote ¹³ Given the importance of taking constituents’ interests into account for reelection, I consider several factors that may shape lawmakers’ support for Taiwan.

First, lawmakers who represent districts with strong economic ties to China may be less supportive of Taiwan-specific bills, as, for instance, Chinese investment may provide job opportunities to their constituents.Footnote ¹⁴ To account for this, I included a control variable for the proportion of Chinese investment in total local household income (Chinese investment), using data from the American Enterprise InstituteFootnote ¹⁵ and the American Community Survey five-year estimate.Footnote ¹⁶

As US–China economic ties deepen, politicians in Washington have expressed concerns on the potential negative consequences. Autor, Dorn, and Hanson (Reference Autor, Dorn and Hanson2013) find that rising imports from China lead to higher unemployment, lower labor force participation, and reduced wages in US local labor markets that house import-competing manufacturing industries. Similarly, Kuk, Seligsohn, and Zhang (Reference Kuk, Seligsohn and Zhang2018) find that members of Congress who voted against China after 2003 were more likely to come from districts that were adversely impacted by import competition. Legislators from districts with a high concentration of manufacturing industries may be more likely to be tough on China-related issues due to concerns about the impact of imports from China on local labor market. To account for this, I included a control variable for lawmakers who come from districts where manufacturing is a major industry (Manufacturing industry), using data from Politics in America edited by Congressional Quarterly Staff.Footnote ¹⁷

Finally, lawmakers’ electoral landscape can affect their decision on sponsoring or cosponsoring bills to support Taiwan. Polls indicating voters rank the importance of foreign policy issues behind domestic issues are abundant (Drezner Reference Drezner2019).Footnote ¹⁸ Previous research also finds that during economic recessions, even dramatic international crises may not divert the public’s attention away from the economy (Heffington, Park, and Williams Reference Heffington, Park and Williams2019). Therefore, lawmakers who come from competitive districts may focus more on introducing legislation that deals with domestic instead of foreign policy issues. I controlled for lawmakers’ vote share in the last congressional election (Vote share) and expect it to be positively associated with support for Taiwan. Data come from MIT Election Data And Science Lab (2017).

Lobbying influence

Lawmakers’ support for Taiwan may also be influenced by lobbying organizations and interest groups. To account for this, I first examined the impact of the Formosan Association of Public Affairs (FAPA), the first Taiwanese American interest group to exert systematic efforts in congressional lobbying (Chen Reference Chen2007). As of 2021, FAPA has 44 local chapters with more than 2,500 active members around the US. Each chapter holds its own events with the local communities. In 2019, members who attended the conference visited 86 congressional offices.Footnote ¹⁹ It is reasonable to expect that legislators from states with local FAPA chapters might be more supportive of Taiwan. I thus included the number of FAPA chapters in the state where the member of Congress comes from as a control variable. Data are from FAPA’s website.

Furthermore, Taiwan ranks among the top recipients of US arms sales despite the lack of a formal bilateral defense treaty. The US defense industry might lobby lawmakers to support Taiwan, especially in arms sales. I thus controlled for Defense industry contribution, measured by the proportion of campaign contributions from the defense industry in the total amount of money raised by each legislator’s campaign. I anticipate that legislators’ support for Taiwan will be positively associated with the proportion of campaign contributions they receive from the defense industry.

My analysis covers 2,957 observations from the 110th to the 116th House of Representatives (2007–2020), with the individual legislator by congressional term as the unit of analysis. I selected these seven congresses because they span a period when executive and legislative power shifted between the two major parties in both the US and Taiwan, and they provide a good mix of US and Taiwanese leadership.Footnote ²⁰ In addition, throughout much of the George W. Bush Administration, US foreign policy primarily focused on counterterrorism, while relations with China remained relatively smooth (Dumbaugh Reference Dumbaugh2006). After the financial crisis, there has been a noticeable increase in Congress’s hostility towards China (Kuk, Seligsohn, and Zhang Reference Kuk, Seligsohn and Zhang2018). Moreover, some control variables, such as Chinese investment, CTC, and Manufacturing industry, are not available before 2007. I excluded legislators who left office before the end of their two-year term, shifted parties in the middle of a congressional term, were elected in special elections, or those from outside the 50 US states. Table 2 summarizes the independent and control variables, their hypothesized direction, and the descriptive statistics.

Table 2.

Descriptive statistics

Model specification

I tested the effect of the independent and control variables on two dependent variables: a dummy variable of whether a legislator sponsored and cosponsored any Taiwan-specific bills in a given Congress and a count variable representing the number of Taiwan-specific measures legislators sponsored and cosponsored in a given Congress. For the dummy dependent variable, I ran multilevel logistic regressions with random intercepts varying by legislators. For the count dependent variable, because Figure 1 suggests there are excessive zeroes in my data, I thus ran a multilevel zero-inflated negative binomial regression with random intercepts varying by legislators. All models include Congress fixed effects. I used a fully Bayesian approach due to its benefit of accounting for the uncertainty in the parameter estimates.Footnote ²¹ When CTC and Manufacturing industry are included as additional control variables, the sample size reduces to 2,112 (111th–115th Congress) due to limited data coverage.