On the Path to Trade Liberalisation: Political Regimes in Trade Negotiations

• DOI: http://doi.org/10.1111/twec.12280
• Published date: 01 July 2016
• Date: 01 July 2016

On the Path to Trade Liberalisation:

Political Regimes in Trade Negotiations

Florian M€

olders

International Finance Corporation, Washington, DC, USA

1. INTRODUCTION

DESPITE the World Trade Organization’s (WTO) effort to promote multilateral trade lib-

eralisation, there has been a surge in bilateral and multilateral free trade agreements

(FTAs) instead. Consequently, the proliferation of these agreements encouraged the WTO to

frame FTAs in terms of coherence instead of co-existence (WTO, 2011). While we have an

idea of what sort of countries eventually enter into trade agreements (see Mansfield et al.,

2002; Baier and Bergstrand, 2004; Martin et al., 2012), there is little known about what deter-

mines the duration prior to actual enforcement.

1

This is particularly important if the expecta-

tions of forward-looking firms (exporters) rely on credible information on the date of tariff

removal to initiate investment and exporting operations. In 2010, when the Nobel Prize was

awarded to Liu Xiaobo, China postponed trade talks on the bilateral China–Norway FTA (The

Guardian, 2010). This is an indication of the sensitivity of trade negotiations to the types of

political regimes and forms a central part of this paper.

Durations from the start of the negotiations through enforcement differ considerably:

ranging from just under one year in the case of the Pan-Arab Free Trade Area to over

10 years for the bilateral trade initiative between Panama and Chile. Yet, the literature has

so far missed out on the dynamics that go hand in hand with trade policy; specifically,

the anticipation of trade liberalisation offers an abundance of research questions. Burstein

and Melitz (2011) analyse transition dynamics at the firm level in the course of trade lib-

eralisation by focusing on innovation activities that determine the decision to export and

the trade volume prior trade liberalisation. In a similar vein, Bergin and Lin (2012) empir-

ically show that trade increments are triggered well ahead of the enforcement of the Euro-

pean Monetary Union (EMU). In both papers, the authors refer to firm behaviour as the

leading driver to a trade increases, mainly triggered by a shift in expectations with respect

to future profits in the foreign market. In this context, policy uncertainty is an issue in the

international economics literature that deserves further attention as it adds a new perspec-

tive to the interaction between the private sector and the policy level. Handley and Lim~

ao

(2012) contribute to this debate by assessing Portugal’s accession to the European Union

The views expressed in this paper are those of the authors and do not necessarily represent those of the

IFC, IFC management or the World Bank Group. All errors and omissions are sole responsibility of the

author.

1

The enforcement of a trade agreement in this study refers to the end of the ratification process,

whereas implementation is not necessarily to coincide with an agreement in force (entry into force is a

necessary but insufficient condition for implementation). The end point of the ratification stage is defined

as the legal enforcement of the agreement in all participating countries.

©2015 John Wiley & Sons Ltd

890

The World Economy (2016)

doi: 10.1111/twec.12280

The World Economy

(EU). Even though trade barriers were lifted well ahead of the formal accession to the

EU, trade-creating effects only emerged after Portugal officially joined the bloc.

2

Economic actors adjust their expectations according to new information. Forward-looking

firms react to news on countries willing to form a trade agreement as this directly impacts

their cost function. However, little is known about what country characteristics shape the pro-

cess towards an agreements’ enforcement. The contribution of this paper is therefore twofold:

(i) it focuses on duration rather than just successful conclusion of agreements and (ii) distin-

guishes the two phases of cooperation, negotiation and ratification, with a focus on driving

factors from the political spectrum. The results allow a better evaluation of the announc e-

ments of trade initiatives and thereby may help to reduce uncertainty involved in the interac-

tion between firms and policymakers. Using a data set with time-varying covariates, the

determinants of the duration of FTA negotiations are found to be multidimensional. Political

aspects, especially the levels of democratisation and political constraints, matter.

3

Controlling for various influences on the negotiation and ratification durations, my findings

suggest that bilateral trade agreements (BTA) reflect significant faster negotiation processes as

compared to agreements with more than two countries involved. The same effect on the nego-

tiation process is observed for countries that are more reliant on international trade. I find sup-

port for the claim that political regime types play a major role in determining the duration of

the pre-enforcement process: delays during the negotiation phase are more likely if countries

have less democratic political systems. This is measured through a composite polity index

(Polity2) and other indicators that provide information on the levels of democracy and politi-

cal constraints. The increasing application of additional trade provisions is taken into accou nt

by including the number of WTO-X and WTO+provisions that mirror the depth of a trade

agreement. It is found that these provisions are associated with significant increases in the

duration of negotiations and ratifications. The implications for trade policy from this research

are manifold: first, if the fast finalisation of trade negotiations with a single country is a prior-

ity, countries should seek a bilateral agreement instead of a multilateral one. Second, political

constraints (which often go hand in hand with a country’s democratisation) come at a cost:

they are associated with longer negotiation periods.

4

Third, the pre-enforcement process of

deeper agreements needs more time. Reductions in non-tariff barriers often intervene with

national policies and policymakers may therefore be more sensitive regarding changes in the

respective sectors.

5

The paper is structured as follows: the next section reviews the respective literature on

international trade and cooperation, while Section 3 outlines the determinants of interest in

this study for the duration of negotiation and ratification stages. The data and the econometric

specifications are described in Sections 4 and 5, respectively. Section 6 presents the results on

the duration of the negotiation and ratification processes before Section 7 concludes.

2

Handley and Lim~

ao (2012) interpret this as a sign of revealed uncertainty in the private sector as firms

had to take into account a potential policy reversal.

3

I refer to all types of trade agreements as FTAs, including customs unions and preferential trade agree-

ments. In the regression analysis of Section 6, the results only change marginally in the magnitude of

the coefficients, if customs unions, for example, are excluded.

4

Yet, there is supportive evidence that more time spent on negotiating is associated with faster ratifica-

tions.

5

See also an article by Economist (2013) which reports on the potential free-trade deal between the

USA and EU. Accordingly, it argues that ‘[e]asy deals [reductions in tariff levels] have mostly been

done; what is left is complicated’.

©2015 John Wiley & Sons Ltd

POLITICAL REGIMES IN TRADE NEGOTIATIONS 891

2. RELATED LITERATURE

In the following, I review the literature on (i) anticipatory trade effects and (ii) timing

and duration of investment and trade policies. The former relates to the empirical analysis

in this paper as it bridges the outcome of the analysis (what factors are associated with

the duration of trade negotiations and ratifications?) with the implications this might have

for private sector companies assessing the timing until trade barrier removal. Exporting

firms, for example, invest and trade in anticipation of trade and investment policies, based

on information on the timing of the enforcement of these policies. The second part of this

review then discusses papers directly related to duration models applied to trade and

investment policies.

a. Anticipatory Trade Effects

The results of the majority of papers on the trade-creating effects of FTAs lead to the

conclusion that trade is triggered by the implementation of an agreement (examples

include Frankel, 1997: Glick and Rose, 2002). While the literature on the ex post trade

effects of FTAs is exhaustive, there is substantial room for contributions that aim at inves-

tigating the accompanying effects of trade agreements. In this section, I first give a brief

overview on the discussion about anticipatory trade effects of FTAs before I summarise

the relevant body of literature that deals with duration analysis in the context of

international treaties.

Since Freund and McLaren (1999) pointed out anticipatory effects of trade agreements,

the notion of pre-implementation effects has been included in successive analyses. Accord-

ing to their results, trade adjustment led to anticipatory trade effects four years before

countries officially joined the EU. Other contributions to the literature seem to affirm the

presence of anticipatory trade effects of trade agreements. Results by Magee (2008) show

that there is substantial anticipation prior to an agreement’s implementation, if controlled

for leads in a gravity model analysis. In a similar vein, M€

olders and Volz (2011) analyse

pre-enforcement effects from a different angle by focusing on the consecutive stages of

enforcement that an FTA runs through. Their results point towards the existence of

anticipatory trade effects in the context of East Asian bilateral FTAs during the negotia-

tion stage of an FTA. Handley and Lim~

ao (2012) examine Portugal’s accession to the

EU. Prior to its formal inclusion into the trade bloc, trade preferences were already at the

free-trade level for industrial goods. However, there was a substantial increase in trade

flows after the accession was formally concluded. This stresses the importance of formal

trade agreements for increasing credibility and exploiting the full scope of the trade

potential.

With respect to the impact of regime types on cooperation, Mansfield et al. (2002) find that

states which are classified as being more democratic have a good chance of successfully con-

cluding FTAs. In a similar vein, Mansfield and Milner (2012) re-emphasise this claim, but

focus on the role of veto players in the ratification period of a trade agreement. When policy-

makers have to decide on whether to form an FTA, they compare the potential benefits (e.g.

better economic performance via free trade) against the costs arising during the negotiation

and ratification stages. Democratic countries are more likely to form an agreement,

although the associated political constraints increase the costs (decrease the likelihood) of the

respective ratification.

©2015 John Wiley & Sons Ltd

892 F. M€

OLDERS