Delayed cooperation: political systems, elections, and the outcomes of trade negotiations

Abstract

Why do some trade negotiations conclude with preferential trade agreements, while others drag on indefinitely? Departing from prior focuses on established cooperation, this study emphasizes the trade negotiation process and accounts for the unconcluded talks. I argue that domestic uncertainties induced by political systems and elections will influence negotiation lengths and outcomes. On the one hand, political systems’ cohesion and efficiency in decision-making may affect the progress of trade negotiations. On the other hand, the unpredictability of electoral outcomes and calendars may prolong trade negotiations. To have a complete picture of trade negotiations, I collected an original Trade Bargaining Dataset, which contains information on both concluded and unconcluded talks. To test the hypotheses, I use a sample of 157 negotiations from 1980 to 2016, in which at least one negotiating country is from the Indo-Pacific region. The results from the Cox proportional hazard models show that parliamentary systems make negotiations smoother than presidential systems. Additionally, negotiating governments are not likely to sign agreements in election years, particularly in the elections of parliamentary and democratic countries. This study contributes to the trade cooperation literature by highlighting the variations in negotiation outcomes and offering a novel approach to investigating the negotiation process.

Keywords:

• preferential trade agreements
• trade negotiations
• political systems
• election
• uncertainty

Acknowledgments

I am very grateful for the excellent comments and suggestions from Katja Kleinberg, Ryan Yu-Lin Liou, three anonymous reviewers, and the editors. Earlier versions of this paper were presented at ISA and MPSA in 2021 and 2022 as well as the Political Science Workshop at National Taiwan University in 2021. I thank the participants for their valuable feedback.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Funding

No funds were acquired for this paper.

Data availability statement

Replication materials for this paper are available on the author’s website (https://sites.google.com/view/hueijyunye/papers).

Notes

1 One exception takes a comparative approach to examine Singapore’s negotiations and concludes that resource constraints affect negotiation outcomes (Hamanaka, 2020).

2 One special case is the Turkey-Jordan Free Trade Agreement, which took effect in 2011 but was suspended by Jordan in 2018.

3 If a candidate changes her previous support for trade agreements before an election, such a turn may not be convincing to voters. For example, before the 2016 US presidential election, it was widely believed that Hilary Clinton would flip her objection to the Trans-Pacific Partnership if she had been elected.

4 After the newly elected President Widodo was inaugurated in October 2014, Indonesia and the EFTA signed the agreement in December 2018.

5 The 23 countries are Australia, Bangladesh, Bhutan, Brunei, Cambodia, China, India, Indonesia, Japan, Laos, Malaysia, Maldives, Mongolia, Myanmar, Nepal, New Zealand, the Philippines, Singapore, South Korea, Sri Lanka, Taiwan, Thailand, and Vietnam.

6 These 115 countries demonstrate substantial variations in political institutions. Appendix Table A1 and A2 break down the composition of the 23 Indo-pacific countries and their partners to display the distributions of regime types and political systems.

7 Based on the World Bank’s region classification, I combined “East Asia and Pacific” and “South Asia” to create a new category “Indo-Pacific” so that the geographic coverage is similar to that of the TBD. I also combined “North America” and “Latin America and the Caribbean” to create the category “America.” The number of countries in each region is: Sub-Saharan Africa (47), Indo-Pacific (28), Middle East & North Africa (20), America (27), Europe & Central Asia (48).

8 Although the TBD covers negotiations from 1980 to 2020, most of the negotiations took place after 2000.

9 I relied on the Free Trade Agreement Database of the Asia Regional Integration Center supported by the Asian Development Bank. As for the agreements with European or American partner countries, I also referred to the websites of the Foreign Trade Information System maintained by the Organization of American States, the European Commission of the EU, and the Eurasian Economic Commission of the Eurasian Economic Union.

10 I consulted the Asia-Pacific Trade and Investment Agreement Database by the United Nations Economic and Social Commission for Asia and the Pacific, the Global Preferential Trade Agreements Database of the World Bank, and the Regional Trade Agreements Database of the WTO. One thing worth noting is that member countries of these international organizations only report the signed treaties, and thus they do not have information on the unfinished negotiations.

11 There are 63 intra-regional negotiations and 140 cross-regional negotiations in which the 23 countries have negotiated with partners outside of the Indo-Pacific.

12 The trends of individual countries are presented in Appendix Figure A1.

13 Because of the data coverage limit, it is possible that some negotiations continued and were completed after 2016. In this case, their negotiation conclusions are not counted toward the number of events in the analysis sample.

14 To avoid confusion, I term the unit as negotiation-year rather than as PTA-year because the sample data include unfinished negotiations.

15 An alternative to account for electoral timing is to count the time to election (see Kleine and Minaudier (2019), Pervez (2015), and Rickard and Caraway (2014)). However, the countdown to an election differs in each negotiating country. As the analysis sample aggregates individual negotiating members’ information to the negotiation-year level, it is less viable to keep the country-wise electoral schedules for the testing. Additionally, negotiating countries may hold elections in turn in the same year, which causes difficulty in counting down the time to elections. Therefore, I use the number of elections rather than the approaching time as a tradeoff for measuring the election-induced uncertainty.

16 The data on Taiwan’s GDP, GDP per capita, and trade openness are derived from the National Statistics and the Bureau of Foreign Trade of Taiwan.

17 Contiguity denotes whether all negotiating countries are contiguous on land or across within 150 miles of water.

18 The results of the Schoenfeld residuals tests are presented in Appendix Table A4.

19 In Appendix Figure A4, the survival curve for bilateral negotiations plummets in a shorter period as opposed to negotiations involving more countries.

Additional information

Notes on contributors

Huei-Jyun Ye

Huei-Jyun Ye is a PhD candidate of Political Science at Binghamton University, SUNY. Her research interests focus on international political economy and international cooperation. Specifically, she delves into the negotiation process that leads to the formation of trade agreements. To improve the understanding of the negotiation process, she has collected the Trade Bargaining Dataset (TBD). She also has co-authored projects on how economic sanctions affect the survival strategies of target governments and on how IMF programs influence human rights.