Instruments of Coercion: International Institutions and

the Sites of Power in International Relations

Allison Carnegie∗

August 15, 2013

∗Allison Carnegie is Senior Research Fellow, Niehaus Center for Globalization and Governance, Prince-ton University, Princeton, NJ; Assistant Professor, Department of Political Science, University of Chicago,Chicago IL. An earlier version of this paper was prepared for the 2012 International Political EconomyConference (IPES). Helpful comments from the participants of the 2013 APSA conference, as well as PeterAronow, Cameron Ballard-Rosa, Thad Dunning, Nikhar Gaikwad, Susan Hyde, Giovanni Maggi, Ken Scheveand Rory Truex, are greatly appreciated. All remaining errors are my own.

Abstract

The international institutions literature suggests that institutions foster co-

operation by allowing states to credibly commit to specific policies. However,

these commitments increase the costs associated with some foreign policy instru-

ments, causing members to substitute towards less constrained options. This

paper argues that by altering the tools that states wield to influence other states,

international institutions shift the sites of power in international relations. The

analysis is focused on a particular institution, demonstrating that because WTO

membership curtails states’ use of trade policies for foreign policy leverage, mem-

bers employ others levers of coercion instead. After presenting detailed case study

evidence in support of the theory, I test these claims systematically, showing that

once states join the WTO, their trade flows become less correlated with political

tensions, while other policy domains, such as foreign aid allocation and unilateral

preference program eligibility, become more responsive to these political issues.

Keywords: foreign aid, GATT, GSP, human rights, international institutions,

policy substitution, trade, WTO

“As a member of the WTO, Zimbabwe is able to enjoy the rights of allmembers....WTO members cannot discriminate against Zimbabwe’s trade.”

The WTO director for information and external relations, Explaining the U.S. and EU’s 2001 decision to respond to

Zimbabwe’s human rights violations with travel bans, asset freezes, and aid cuts rather than tariff increases

“Turkey has obligations. The Turkish state can’t do this given the WTOand Customs Union rules.”

The Turkish Ambassador to France, Explaining Turkey’s 2011 decision to not to retaliate against a controversial French

vote through trade restrictions, and to instead pursue diplomatic avenues of retaliation

“Since Burma is a member of the WTO....EU policy in the field ofrestrictive measures will have to take into account the international

commitments of the EU, in particular WTO agreements.”The EU Commission, Explaining the EU’s 1997 decision to respond to Burma’s human rights violations with foreign aid cuts

and asset freezes rather than tariff increases

The idea that membership in international institutions can enhance cooperation between

states is central to international relations. However, this article presents a revision and exten-

sion of this claim, demonstrating that while international institutions can bolster cooperation

in the specific policy domains they govern, they can simultaneously reduce cooperation in

areas outside of their purview. Despite the large body of work on the impact of international

institutions, these indirect effects of institutional membership have been largely overlooked

in favor of a focus on direct effects. I argue that by failing to consider the unintended con-

sequences of institutional membership, scholars have been left with an incomplete picture

of the role of international institutions in interstate dynamics. In particular, I show that by

decreasing states’ flexibility with respect to some types of policies, membership in interna-

tional institutions increases political maneuvering in other, less constrained policy domains.

By altering the tools that states select to influence other states, these institutions thereby

shift the sites of power in international relations.

How do international institutions politicize policies outside of the areas they govern?

Institutions permit states to credibly commit to specific policies by imposing a variety of

reputational, domestic, and economic costs to the state’s deviation from agreements.1 Mem-

ber states are thereby limited in their abilities to extend or revoke policy concessions within

1See, for example, Busch, Raciborski and Reinhardt (2009); Davis (2006, 2012); Keohane (1984).

1

the policy domains regulated by institutions. Since states often offer (or retract) foreign

policy concessions to influence other states, such as lowering trade protection, increasing

foreign aid, or granting security concessions, the rigidity in policy induced by these institu-

tions constitutes a loss of leverage over other members. However, I argue that rather than

abandoning their attempts to influence the policies of other states, members of international

institutions replace policies made more costly by institutional membership for less costly

policy options. Indeed, a key factor shown to influence foreign policy substitution is cost:

leaders choose between economic, military and diplomatic levers of influence as a function of

the costs of the policies available to them (Clark, Nordstrom and Reed 2008; Most and Starr

1989). But while scholars have proposed a variety of domestic and economic factors which

shape the costs imposed by certain policies, the role played by international institutions

remains under-theorized. This paper demonstrates that by limiting members’ options in

specific policy domains, institutions lead these states to attempt to influence their partners

through a range of more flexible tools. International institutions thus create ripple effects

across policy arenas, bolstering cooperation in some areas, while politicizing others.

To test this theory, the analysis is focused on a particular institution; namely, how mem-

bership in the World Trade Organization (WTO) and its predecessor, the General Agreement

on Tariffs and Trade (GATT),2 affects the means by which states attempt to coerce their

partners. As one of the largest and most influential institutions in international relations,

the WTO represents an ideal setting in which to test this theory. Indeed, the WTO is

considered a success-story in terms of liberalizing trade relations among its 159 members,

spawning a large literature on its inner workings and effects (Liu 2009; Rose 2004; Subrama-

nian and Wei 2007; Goldstein, Rivers and Tomz 2007). A central finding of this body of work

is that, on average, WTO membership increases trade flows between members.3 However,

2In 1995, the WTO replaced the GATT. Henceforth, I use the term WTO to refer to both the GATTand the WTO unless otherwise specified.

3The WTO can increase trade for a variety of reasons. For example, the WTO can resolve terms of trade

2

this predominant emphasis on economic gains has obscured the importance of the WTO’s

political impact, which has led scholars to overlook many unintended consequences of WTO

membership.

I demonstrate that because the WTO constrains the ability of its members to condition

their trade policies upon their political relationships with their trading partners, members

instead manipulate policies that are not regulated by international institutions in order to

pursue their political objectives. In what follows, I first describe the international context

of the theory, detailing donors’ available instruments of coercion and explaining how the

WTO constrains policy leverage in the trade domain. Next, I present case study evidence

showing that because WTO membership curtails states’ use of trade policies for foreign

policy leverage, members use other policy arenas to exercise power. I then test these claims

systematically, demonstrating that once states join the WTO, their trade flows become less

correlated with political tensions, while other levers of influence become more responsive to

foreign policy issues.

The WTO and Sites of Power

This article presents a theory of how international institutions cause their members to en-

gage in foreign policy substitution, altering the tools that their members select to influence

other states. Indeed, states seek a variety of policy concessions from partner states, such as

democratization, respect for human rights, intelligence cooperation, technology sharing, ac-

cess to military bases, etc. To entice their partners to make concessions in these areas, states

often offer economic or political benefits in exchange, and withhold preferential treatment if

their requests are ignored. This practice, known as foreign policy conditionality, is exercised

externalities (Bagwell and Staiger 1999), allow countries to avoid succumbing to domestic political pressuresfor protection (Buthe and Milner 2008; Maggi and Rodriguez-Clare 1998; Mansfield and Reinhardt 2008),and remedy political hold-up problems (Carnegie 2013).

3

frequently in international relations.

While states have many policy areas to choose from, I argue that their selection is shaped

by the broader institutional environment. Specifically, membership in a particular interna-

tional institution raises the cost of policy manipulation in areas governed by the institution.

For instance, because the WTO governs trade relations between states, it limits its members’

use of trade policy conditionality; while non-members often use their trade policies to punish

or reward other states, the WTO explicitly prohibits unilateral trade policy discrimination.

Indeed, WTO members must agree to accord all members the same low tariff rates, known

as Most Favored Nation (MFN) tariff levels,4 which are enforced through the WTO’s sophis-

ticated Dispute Settlement Body (DSB). The DSB adjudicates WTO disputes, providing

WTO participants with “a guarantee for the right to negotiate, a common standard for eval-

uating outcomes, the option for several countries to join a dispute, and incentives for states

to change a policy found to violate trade rules” (Davis 2006). By specifying which states are

non-compliant, the DSB impacts members’ reputations, which provides states with strong

incentives to cooperate (Tomz 2007). If countries are known to violate agreements frequently,

their partners may punish them in a variety of ways, including removing trade concessions,

exhibiting unwillingness to form agreements, or reducing cooperation in other areas (Maggi

1999). WTO rulings can also provide countries with domestic political cover for adhering

to agreements (Allee and Huth 2006; Staiger and Tabellini 1999) and the ability to develop

domestic reputations and norms for compliance (Mansfield, Milner and Rosendorff 2002).5

4There are several exceptions to this rule. Under the GATT, exceptions include: Article I:2-4 based onHistorical Preferences, Article IV(c) for Cinematographic Films, Article XX for General Exceptions such asthose relating to morals or the environment, Article XXIV:3 for Frontier Traffic, Article XXIV:5 for PTAsand Customs Unions, Article XXI for Security Exceptions, the 1979 Enabling Clause, and the MarrakeshAgreement Article IX:3 Waiver. Many of these exceptions, such as the security clause, are rarely used,while others, such as the PTA exception, are used frequently. Similar MFN exceptions exist under theGeneral Agreement for Trade in Services (GATS) and Trade-Related Aspects of Intellectual Property Rights(TRIPS). These exceptions are discussed in detail in the appendix due to space constraints; however, I notethat they are limited and do not allow WTO members much leeway for trade policy manipulation. This isdiscussed further in the supplemental appendix.

5I further elaborate on the enforcement capabilities of the WTO in the appendix due to space constraints.

4

By constraining the use of trade policies, the WTO therefore causes states to substitute

towards levers of influence that are not governed by institutions, altering the sites of pressure

within the international system.6 Indeed, a variety of alternative levers are available to WTO

members. First, states often turn to bilateral foreign aid, which is assistance given to other

countries in the form of grants or subsidized loans, as a means to pressure their partners.

Although the primary goal of foreign aid policies is typically economic growth, donors have

long sought to use foreign aid to obtain political influence, as well. When states comply with

donors’ demands, the donors often provide additional aid, but when recipients ignore their

requests, donors withhold aid.7

Second, unilateral preference programs, permitted under the Enabling Clause of the

WTO, function as an alternative source of political leverage. Unilateral preference programs

specify a variety of goods on which tariffs are reduced for all countries that meet specific

criteria, which typically include both political and economic factors.8 The most common

preference program is the Generalized System of Preferences (GSP). Most developed coun-

tries, including the United States, the EU, Canada, Japan, and Australia, offer some form of

Note that states may still use trade policy in conjunction with other instruments; the argument is simplythat they will lessen the use of trade as a political tool and increase the use of other tools in their arsenal.

6Indeed, interviews with government and policy officials support the contention that raising tariffs aboveWTO mandated levels is often much more costly than altering other types of policy commitments. Accordingto an official with extensive experience in U.S. trade policy, when a country joins the WTO, “companiesbelieve that they have a legally secure agreement. Violating WTO principles hurts companies by disruptingthe stability and security of their business environment.” However, “the effects of disruptions to the financialsector tend to be more opaque and distant. Unilateral grants...can be taken away and given back much moreeasily...Financial flows are an easy spigot to turn on and off.” The official pointed out that internationalcourts do not become involved when, for example, a bilateral foreign aid agreement is violated, as thereare no binding multilateral obligations created when one country agrees to provide bilateral assistance toanother state. Interview by author. February 22, 2012. As an official of USAID explained, when choosingan instrument of coercion, states “look around at what tools [they] have...Once [a country] is in the WTOit’s harder to use trade for influence.” Official of USAID. Interview by author. February 10, 2012.

7For example, the EU incorporates conditionality in many of its foreign aid agreements: The 1989 LomeIV Convention linked human rights, democracy, and good governance with aid allocation to 70 African,Caribbean and Pacific (ACP) countries (Holland 2002). In addition, the 2000 Cotonou Convention thatreplaced the Lome agreements provided only a first tranche of aid, holding back the second tranche untilcompliance with conditionality was confirmed (Smith 2003). Similarly, many U.S. programs, such as theMillennium Development Program, make conditionality a crucial part of aid allocation.

8See Onguglo (1999) for an overview of existing preference programs.

5

GSP. While GSP eligibility requirements vary by donor, they typically include many types

of political and economic factors, and if countries do not meet these qualifications, tariff

preferences may be withdrawn at any time. For example, eligibility for the U.S.’s GSP pro-

gram may be revoked if a recipient violates human rights, harbors terrorists, nationalizes

U.S. property, experiences high economic growth, or places undue pressure on American

producers that compete with goods covered by the program.9

Third, a variety of financial restrictions exist which may be employed as leverage: For

example, foreign direct investment (FDI), or direct investment into production or invest-

ment in another country, is often desired by recipient countries to bolster economic growth.

Although FDI is a tool of private economic actors in many countries, states can use FDI for

political leverage by threatening to place restrictions on its flow unless recipients make con-

cessions.10 Additionally, some governments invest in the host country themselves, in which

case FDI is more readily used as political leverage.11 States may also employ other financial

tools, such as the provision of capital through official export credit agencies, development

finance institutions, or trade agencies.12 Alternatively, states may gain political leverage

9If an interested party such as a firm, labor union, or NGO believes a recipient has violated one of theeligibility conditions, the party may request an investigation, which may be accepted or rejected by the Officeof the U.S. Trade Representative (Jones 2013). Because accepting an investigation constitutes a threat ofGSP revocation, these investigations are often used as leverage. For example, in Congressional testimony,then-U.S. Trade Representative Robert B. Zoellick stated, “The threat of loss of GSP...has proven to be aneffective point of leverage with some of our trading partners (Zoellick 2002). Further, the U.S. Ambassadorof the Brazilian Embassy stated that Brazil had made improvements “on piracy over the past 18 months,under the threat of U.S. revocation of GSP,” acknowledging that “GSP can be a powerful tool” (Sobel 2006).

10For example, a 2008 Congressional Research Service report describes the “good governance require-ments, human rights conditions, approved-project restrictions, and environmental quality regulations thatcharacterize U.S. and other Western government investments” (Dumbaugh 2008).

11For example, China often requires host countries to adopt a “one-China” policy, and to use Chinesecompanies, suppliers, and banks, as a condition of receiving FDI.

12For instance, the United States’ Export-Import (Exim) Bank loans money to other countries at lowinterest rates to allow them to buy products from the United States. The ExIm bank requires human rightsand political clearances by the State Department before it will agree to lend (Report 1994). Further, indeciding whether to accept an application for Ex-im Bank programs, the Bank may consider the extent towhich a nation has been helpful in combating terrorism (Jackson 2003). Other U.S. tools include the OverseasPrivate Investment Corporation (OPIC), which is the U.S. government’s development finance institution,and the United States Trade and Development Agency (TDA).

6

through their treatment of foreign assets. Many countries have considerable assets invested

in the financial institutions of their partners, and these partners may render such assets

inaccessible in order to attain political influence.

Fourth, diplomatic levers may allow states to influence their partners. States may punish

each other by recalling ambassadors, boycotting conferences and events, or refusing to allow

participation in international fora. For instance, states often make entry into international

institutions or bilateral agreements contingent on the fulfillment of specific criteria, such

as economic or political reforms. Perhaps the most well-known instance of conditionality

applied to accession agreements are those that provide a pathway to membership in the EU.

Although reforms have been uneven, a large literature demonstrates that the desire for Eu-

ropean integration has successfully promoted liberalization in Eastern Europe (Kelley 2004;

Vachudova 2005; Levitz and Pop-Eleches 2010; Schimmelfennig 2005). Similarly, another

institution that has demanded many stringent reforms prior to entry is the North Atlantic

Treaty Organization (NATO), a multilateral alliance.13

Thus, once states join the WTO, a variety of alternative levers of influence are available

that do not fall under the institution’s purview. But how do states choose between their

various policy options? Scholars have shown that policies are selected due to their associated

costs and benefits, which depend on a variety of factors including:14 the welfare of partner

countries, national security concerns, costs to specific domestic groups, retaliation by the

recipient, or the interests of lobbyists and constituents.15 While the precise considerations

13For example, Senator Bond noted that “Poland, Hungary, and the Czech Republic as well as othercountries in Central and Eastern Europe that aspire to join NATO, have worked to alleviate historicalgrievances and build relationships with their neighbors based on mutual trust, respect and cooperation”(The Congressional Record, 105th Congress 1998). Similarly, Secretary Perry claimed that “NATO is thecarrot encouraging reforms” (The Congressional Record 1996a).

14Although conditionality has been criticized for donors’ inconsistent applicability and conflicting demands(Crawford 2001), it is still frequently practiced by many countries (Dunning 2004; Bueno De Mesquita andSmith 2009; Bearce 2010). Perhaps the most straightforward motivation for applying conditionality is self-interest, as obtaining concessions may enhance a country’s standing in the international community, orimprove the administration’s electoral prospects at home (Crawford 2001).

15Businesses, investors, non-governmental organizations (NGOs), domestic consumer groups, and others

7

involved in selecting a particular set of policies likely vary by state, I argue that WTO

membership, overall, results in a reduction in the reliance on trade policies as a means of

coercion.

Existing Accounts of Institutions and Policy Substitu-

tion

By offering an account of how institutions affect interstate relations in areas outside of the

institution’s purview, this article contributes to the large literature analyzing the impact of

international institutions on international cooperation. Beginning with the seminal work of

Keohane (1984), theories of international institutions have typically argued that these in-

stitutions benefit all members through increased cooperation. Some scholars have criticized

this claim, suggesting that such benefits accrue most to the powerful states that created the

institutions, potentially at the expense of weak states (Krasner 1991), or serve merely as

“arenas for acting out power relationships” (Mearsheimer 1994, 13). However, I argue that

these accounts are incomplete, since in addition to the direct effects highlighted by existing

theories, international institutions also impact policy areas outside of those that they govern

in ways which fundamentally alter interstate relations. Since institutional membership rep-

resents a credible commitment not to use certain policies for foreign policy leverage, I show

that countries attempt to extract concessions using alternative, less costly tools.

Further, by showing that institutional membership induces foreign policy substitution,

my theory also contributes to the literature examining the determinants of states’ foreign

policy selection. While the motivations for choosing certain foreign policies over others has

often call for governmental responses to policies enacted by other states. For instance, a senior advisor toa U.S. Senator on the Foreign Relations Committee recalls that aid and trade concessions to Burma werecut partly in reaction to lobbying activities by U.S. citizens who resented the influx of Burmese immigrants.Interview by author. February 23, 2012.

8

been explored in a variety of areas,16 I offer new insights on how foreign policy substitution is

conditioned by membership in international institutions,17 contributing to our understanding

of the role of specific policies in relation to the broader international context in which they are

used (McGinnis 2000; Palmer, Wohlander and Morgan 2002). Having suggested that states

select policy levers based on their associated costs and benefits, I show that institutions

increase the cost of particular tools, leading members to rely more heavily on alternative

policy options.

In addition, while many scholars have investigated the determinants of foreign policy

selection, their analyses tend to remain outside of the broader institutional context. For

instance, while scholars have considered many potential donor motivations for aid giving,18

and have conducted innovative analyses of variation in aid institutions (Milner 2006; Schnei-

der and Tobin 2010), the indirect effects of institutional membership remain under-explored.

Similarly, empirical studies of unilateral preference programs typically focus on their eco-

nomic causes and consequences,19 though some scholars recognize that these programs may

be used for political purposes (Johnson 1967; Orbie 2011; Patterson 1965), and that domestic

politics may impact their efficacy.20 Yet despite the large literature examining the average

16See Bennett and Nordstrom (2000) for enduring rivalries, Diehl (1994); Milner and Tingley (2012);Morgan (2000); Palmer and Morgan (2010) for alliances and aid, Regan (2000) for interventions, Clark andReed (2005); Lektzian and Sprecher (2007) for sanctions, and Moore (1998) for repression.

17Foreign policy substitution should not be confused with the interesting work conducted on issue linkagein the context of institutions (Davis 2004, 2009), on institutional membership as a means of allowing statesto receive bribes (Dreher, Sturm and Vreeland 2009; Kuziemko and Werker 2006), or on the effects of bothmultiple and overlapping institutions (Busch 2007; Davis 2009; Tobin and Busch 2009).

18See Bermeo (2010, 2007); Lumsdaine (1993); Bueno De Mesquita and Smith (2009) for humanitarian anddevelopment concerns, Alesina and Dollar (2000) for security issues, Dollar and Levin (2006) for influenceover policy-making in the recipient country, De Mesquita and Smith (2007) for coalition sizes, and Fleck andKilby (2006); Lumsdaine (1993); Milner (2006); Noel and Therien (1995); Therien and Noel (2000); Tingley(2010, 2009) for domestic pressures.

19For recent studies, see Aiello, Cardamone and Agostino (2010); Borchert (2009); Brenton (2003); Herzand Wagner (2011); Keck and Lendle (2012); Lederman and Ozden (2004); Nilsson (2002, 2007); Ozdenand Reinhardt (2005); Stevens and Kennan (2005); Subramanian and Wei (2007); Verdeja (2006). Also seeDe Benedictis and Salvatici (2011) and Hoekman and Ozden (2005) for overviews.

20For example, GSP may reduce incentives for exporters to lobby for liberal trade policies (Ozden andSharma 2006), and can therefore result in trade protection and higher export/GDP ratios (Ozden andReinhardt 2004). However, note that this interpretation is contested, as GSP benefits may also help countries

9

effects of these programs, I am unaware of any study examining GSP as a substitute lever of

influence. Other tools, such as diplomacy, accession, and lending behavior, are also typically

explored without reference to the larger institutional environment.

Evidence of Institution-Driven Foreign Policy Substitu-

tion

I now investigate whether states behave in the manner predicted by my theory: first by

examining detailed, real-world examples, and then by conducting a systematic empirical

investigation. To make this challenge more tractable, I focus my analysis on a particular

policy area: human rights. Since the middle of the 20th century, increased respect for

human rights has been an important policy goal of many powerful actors in the international

community.21 In particular, members of the Organization for Economic Cooperation and

Development (OECD) have long expressed concern with their partners’ human rights records

and levels of democracy. Indeed, OECD countries espouse a moral commitment to human

rights, stating that their mission is to “promote policies that will improve the economic

and social well-being of people around the world” (OECD 2011). Accordingly, many of

these states condition their foreign policy concessions upon recipients’ respect for human

rights and democratization (Dobbins 2008). A reflection of the significant role that respect

for human rights plays in OECD foreign policy decisions is the fact that each of the foreign

policy instruments discussed above, when employed, is often explicitly linked to human rights

promotion by the country in question.

Due to the salience of respect for human rights in interstate relations, the role of the

transition to liberal trade policies before they graduate from the program (Conconi and Perroni 2004).21For example, the United Nations’ Universal Declaration of Human Rights promises to protect the “equal

and inalienable rights” of all people. Similarly, the EU’s founding 1957 Treaty of Rome states that theinstitution’s goal is “developing and consolidating democracy and the rule of law, and....respecting humanrights and fundamental freedoms” (Article 177). See Simmons (2009) for an overview.

10

international community in advancing rights and freedoms has attracted a good deal of

scholarly attention. Scholars have demonstrated that international commitments have se-

cured significant improvements in the human rights records of many states (Conrad and

Moore 2010; Hafner-Burton 2005, 2009; Sikkink 2011; Simmons 2009).22 International in-

stitutions, in particular, can promote such improvements (Hafner-Burton 2012b, 2013) by

pressuring states to improve their human rights records in order to obtain material bene-

fits, helping them overcome domestic barriers to political liberalization, or fostering liberal

norms (Checkel 2005; Pevehouse 2002). But while scholars have analyzed how institutions

can stimulate respect for rights, as well as how concerns about human rights drive policy

decisions,23 their effects on the policy tools used to promote rights remain opaque.24

WTO Membership and U.S. Foreign Policy

I investigate whether membership in international institutions causes states to substitute to-

ward other levers of influence by first examining the U.S.’s selection of foreign policy before

and after its trading partners join the WTO, and then considering the policies of OECD

countries more systematically. As the most powerful state in the international system, the

United States provides an instructive example of institution-driven foreign policy substitu-

tion. By closely following a particular state’s interactions with multiple trade partners, I

22See Hafner-Burton (2012a) for an overview. But see Hollyer and Rosendorff (2011) for an opposing view.In addition, the international community may influence rights simply by trading more (Compa and Vogt2000).

23For example, some scholars argue that human rights concerns impact foreign aid provision (Berthelemy2006; Cingranelli and Pasquarello 1985; Poe 1992; Scott and Steele 2011), though some disagree (Barratt2007; Carleton and Stohl 1987; Dollar and Levin 2006; Easterly 2007; McCormick and Mitchell 1988; Neu-mayer 2003a,b,c; Schraeder, Hook and Taylor 1998; Schoultz and Schoultz 1981). Still others assert thathuman rights only matter sometimes, or at least play a small role (Alesina and Dollar 2000; Apodaca andStohl 1999; Carey 2007; Lai 2003; Nielsen 2012; Payaslian 1995; Poe and Sirirangsi 1994; Poe et al. 1994;Svensson 1999; Tomasevski and Tomasevski 1997).

24See also Lebovic and Voeten (2009) for the argument that international organizations can shame recip-ients, providing justification for lenders to reduce spending, and Hafner-Burton and Montgomery (2008) foran argument that PTAs can sometimes increase, and sometimes decrease sanctions by raising competitionbut also providing trade benefits.

11

am able to demonstrate that institution-driven policy substitution represents a pattern of

behavior, rather than a few isolated incidents. Thus, I investigate all cases in which the U.S.

expressed strong resistance to providing WTO benefits for a particular trade partner.25 I

show that before these trade partners entered the WTO, the U.S. focused primarily on using

trade for leverage over its partners’ human rights policies. However, once these states joined

the WTO, the U.S. became constrained in its ability to use trade policy to extract reforms

through the use of trade policy and switched to other tools such as asset freezes, foreign aid

cuts, unilateral preference programs, and NATO membership conditionality.

When the U.S. Congress disapproves of policies enacted by trading partners who are non-

WTO members, it often offers the renewal of MFN status in exchange for policy concessions.

As explained above, this treatment entitles countries to nondiscriminatory trade policies such

that a country cannot lower tariffs for a partner with MFN status without lowering tariffs for

all partners with MFN status. While all WTO members must grant each other this status,

extending it to non-WTO members is optional. Indeed, the United States frequently used

the designation of MFN status to attempt to extract concessions from partner countries who

are non-WTO members. In particular, 1972 marked the passage of a bill which included the

“Jackson-Vanik amendment,” blocking MFN status for countries that restrict emigration.

The amendment requires the president to grant yearly MFN waivers to countries disallowing

emigration and for Congress to renew MFN agreements every three years. The amendment

has been consistently applied as an effort to incite political reforms, particularly in the area

of human rights (Lilley and Willkie 1994). For example, MFN status was taken from Poland

after it became more repressive, but was then restored in 1987 when Poland implemented

liberalizing reforms (Pregelj 2005). Similarly, MFN status was rescinded from Afghanistan

in response to the death of Ambassador Adolph Dubs, who was kidnapped and killed in 1979

(Gladstone 2001). It was also revoked from Serbia and Montenegro in 1992 due to “war-like

25I exclude GATT accession cases due to space constraints.

12

activities” (Pregelj 2005).26 However, when countries join the WTO, the U.S. relinquishes its

ability to revoke MFN status without violating WTO laws, since the WTO requires members

are required to grant each other permanent MFN status, also known as permanent normal

trade relations (PNTR) status. Thus, once these states are given PNTR status (along with

WTO membership), the U.S. switches to alternative levers of influence to obtain human

rights concessions. Several specific cases of this practice are detailed below.

Vietnam

During the period between the normalization of U.S.-Vietnam relations in the mid-1990s and

Vietnam’s 2006 WTO accession, the United States relied primarily upon its trade policies

to pressure Vietnam to improve its human rights record. In 1998, President Clinton granted

Vietnam conditional MFN trade status, which was subject to a yearly renewal dependent

upon Vietnam’s human rights record.27 Congress consistently threatened to revoke trade

concessions due to human rights concerns, and in response, Vietnam made several human

rights reforms such as signing an agreement on religious freedom in which it pledged to

improve conditions for religious citizens (Manyin 2006).28

In 2001, the United States signed a bilateral trade agreement (BTA) with Vietnam which

was the last major hurdle to Vietnam’s WTO accession (Manyin 2001). The U.S. Congress

intensely debated the passage of the BTA, as it recognized that Vietnam’s near-certain WTO

accession would eliminate the U.S.’s ability to credibly threaten to terminate Vietnam’s

MFN status if Vietnam violated human rights. For instance, Rep. Loretta Sanchez argued

for continuing to tie human rights to MFN renewal, rather than pass the BTA, because,

26The prospect of revoking MFN status was similarly used to try to extract policy concessions fromAlbania, Georgia, Laos, and Ukraine. It has been denied to Cuba and North Korea due to political frictions(Pregelj 2005).

27See the discussion in the Congressional Record each year the waiver is granted. For example, in the1998 act entitled “Disapproving Extension of Waiver Authority with Respect to Vietnam,” there are manyarguments that the waiver should not be granted due to human rights violations.

28The agreement has not been released publicly.

13

“if we insist that Vietnam improve its human rights record as a condition to trading with

America, we would gain human rights advances in Vietnam, so I think it is a tragic mistake

for the United States to decline to use this tool” (The Congressional Record 2001).

To compensate for losing leverage over Vietnam in the realm of trade, the U.S. Congress

sought to tie Vietnam’s aid allocation and GSP status to its human rights practices instead.

On the same day that Congress passed the BTA, the House approved a companion bill,

H.R. 1587, by a vote of 410-1, which sought to prohibit non-humanitarian increases in aid

for Vietnam unless the Vietnamese government “made substantial progress toward releasing

all political and religious prisoners from imprisonment, house arrest, and other forms of

detention...respecting the right to freedom of religion.... respecting the human rights of

members of ethnic minority groups.....[and was not] complicit in a severe form of trafficking

in persons.” Congressional debate emphasized that the bill’s purpose was to connect aid to

respect for human rights because of the probable decoupling of MFN status and human rights

due to the passage of the BTA. Representative Henry Hyde (R-IL) stated, “This bill makes

clear that progress towards freedom and democracy will continue to be a central theme of

U.S. foreign policy toward Vietnam. It uses forms of leverage other than trade sanctions to

promote this objective, such as conditions on non-humanitarian foreign assistance” (Hyde

2001).29

Congress also sought to tie Vietnam’s GSP status to its human rights record. Senator

Barbara Boxer introduced the Vietnam Human Rights Act of 2008, which would “prohibit

Vietnam’s entry into the U.S. GSP program unless Vietnam’s labor rights regime is certified

as making improvements in certain areas” (Boxer 2008). The bill proposed that Congress be

provided with a new lever of influence, because “Since Vietnam’s accession to the WTO on

January 11, 2007, the Government of Vietnam arbitrarily arrested and imprisoned numerous

29Perhaps under pressure from the U.S., other countries also moved to tie their aid allocations to Vietnam’shuman rights record around this time, such as Japan in 2004 (AFP 2004).

14

individuals for their peaceful advocacy of religious freedom, democracy, and human rights”

(Boxer 2008). However, Congress did not recommend revoking Vietnam’s MFN status;

instead, it substituted toward linking the country’s GSP status and foreign aid allocation to

its ability to demonstrate respect for human rights.

Cambodia

Prior to granting PNTR, the precursor to WTO accession, to Cambodia in 1996, the U.S.

frequently conditioned its trade policies upon Cambodia’s human rights performance. Since

1975, the U.S. had maintained a trade embargo against Cambodia as punishment for its

frequent civil and political rights violations. The embargo was lifted after Cambodia held

elections and improved its human rights record (Lum 2009). These reforms led the United

States to accord Cambodia PNTR status, which was viewed by Congress as the last major

step to its WTO entry (Lum 2009). However, the U.S. remained concerned about respect for

human rights in Cambodia. Indeed, John McCain stated: “It is true that the Cambodian

people have a freely elected government, freedom of speech and freedom of association. It is

also true, however, that each of these democratic institutions has at one time or another come

under attack from the coalition government. The Senate is today approving unconditional

most-favored-nation status for Cambodia....But...granting MFN to Cambodia should not be

interpreted as disinterest in the course of Cambodian democracy. The United States Senate

is committed to helping democracy and human rights to flourish in Cambodia. Our efforts

will not end with this vote” (Mccain 1996).

Due to the concern over potential future human rights abuses, the U.S. House of Repre-

sentatives unanimously passed the 1996 House Resolution 345, which linked foreign aid to

Cambodia’s human rights record. Congressman Roth explained, “Recently, Congress sent to

the President H.R. 1642, a bill to extend permanent most-favored-nation tariff treatment to

Cambodia...The resolution we have passed today is meant to send a parallel message- that

15

the United States Senate remains deeply concerned about problems in Cambodia” (Roth

1996). The Congressman went on to detail the “problems” he was referring to, listing a

variety of human rights issues in Cambodia that included corruption, politically motivated

deaths, and lack of political freedom. The resolution also urged other aid donors to condi-

tion Cambodia’s foreign aid allocation upon its human rights record, stating, “The Secretary

of State should encourage Cambodia’s other donors...to raise concerns with the RGC over

Cambodia’s record on human rights” (Relative to Cambodia Human Rights Record 1996).

Similarly, Senator Thomas argued that “the United States and other donor nations should

reconsider the amount and extent of our financial aid” (Thomas 1996). Indeed, each year

since PNTR status was granted, international donors have met to set political conditions for

the receipt of foreign aid (Lum 2009).

Since Cambodia received PNTR status, the U.S. has relied on forms of leverage other

than trade policy. For example, in 1997, Hun Sen staged an armed take-over of the Cambo-

dian government, which involved considerable political violence and human rights violations.

The U.S. acknowledged that “Vietnam’s overall human rights record...has deteriorated since

Vietnam joined the WTO” (Human Rights Act 2009), but, as Congress had pointed out dur-

ing the debates over PNTR, it was unable to use MFN status for leverage, and thus shifted

toward other tools. In particular, the U.S. “announced it was suspending for 30 days all

assistance provided to the Cambodian Government....until the democratically elected Gov-

ernment of Cambodia [was] restored” (Biden 1997). The U.S. also pressured ASEAN not to

allow Cambodia to enter the organization until it improved rights and governance. Congress

congratulated ASEAN for following the U.S.’s request; for instance, Senator Biden stated,

“I applaud the decision taken by the Association of Southeast Asian Nations [ASEAN] to

delay Cambodia’s membership in that organization” (Biden 1997). Further, the Senate intro-

duced a resolution which “call[ed] upon the Secretary of State to impose visa restrictions on

members of the Cambodian National Assembly and their families” who were found guilty of

16

human rights offenses (Brownback 2005). Thus, once Cambodia had made significant strides

toward joining the WTO, the U.S. Congress moved to link aid disbursements, ASEAN mem-

bership, and visa applications to Cambodia’s human rights record, but refrained from using

trade policy conditionality.

Romania

Although Romania entered the WTO in 1971, the U.S. had invoked its right under the

GATT Article XXXV to withhold PNTR until Congress chose to grant it. Once Congress

grants PNTR to a nation, it cannot be revoked without violating WTO law.30 Prior to the

decision of whether to grant PNTR to Romania, renewal of MFN status was the primary tool

employed by the U.S. to elicit improvements in Romania’s human rights record, particularly

with regard to religious freedom and the release of political prisoners (Kirk and Raceanu

1994). In response, Romanian President Ceausescu sent a personal envoy to the U.S. to

convince the government that Romania was respecting human rights, and asked the U.S. to

grant MFN status in return (Kirk and Raceanu 1994, 7). However, in 1988 Romania became

increasingly repressive, and as punishment MFN status was not renewed. It was subsequently

renewed in 1993, after Romania held democratic elections and provided greater rights and

freedoms to its citizens, prompting the U.S. Assistant Secretary of State to remark that the

U.S. had used its MFN leverage to secure distinct improvements in Romania’s respect for

human rights (Kirk and Raceanu 1994, 71).

In 1996, the U.S. Congress debated whether to extend PNTR to Romania. Members

of Congress were concerned about relinquishing the ability to influence Romania through

trade policy. Although Romania had made many human rights reforms by this time, they

30This treatment is very rare; the U.S.-Romania case is one of only a handful of instances in which WTOand PNTR were not granted virtually simultaneously. Previously, 14 countries did not accord PNTR toJapan when it entered the WTO in 1995, Cuba did not grant PNTR to acceding states that restricted sugarimports, and India and Pakistan withheld PNTR from the Union of South Africa in protest of apartheid(Suzumura 1997).

17

were reluctant to lose the option of using MFN as a bargaining tool in the future. However,

many members pointed out that the U.S. would retain influence over Romania due to the

latter’s aspirations to enter NATO and the EU. Marshall F. Adair, the U.S. Deputy Assis-

tant Secretary of European and Canadian Affairs, stated that MFN renewal “was a useful

kind of leverage to try and move Romania onto the right path.” However, he argued that

Congress should extend PNTR to Romania because the U.S. could rely on other forms of

influence, stating, “We do have a lot of leverage with them to continue working with them, to

continue encouraging democratic reform, human rights reform, and economic reform. They

are desperate to become full-fledged members of the Western community of nations. They

want to get into NATO, they want to get into the EU” (The Congressional Record 1996b).

Senator Grassley concurred, arguing that Congress should capitalize on Romania’s “desire

to become integrated into the European community and to become a member of NATO”

(The Congressional Record 1996c). Congressman Lantos similarly advocated a focus on “Ro-

mania’s quest to join NATO, which, once Romania fulfills all of the obligations for NATO

membership, [he would] be pleased to champion” (The Congressional Record 1996d). The

PNTR issue therefore served to shift the locus of power from trade policy to Romania’s

accession agreements.

China

In 1979, China signed a bilateral trade agreement with the United States granting China

conditional MFN status, which had to be renewed yearly by the U.S. After the 1989 Tianan-

men Square protests, the U.S. Congress repeatedly threatened China with tariff increases

unless it addressed the United States’ concerns regarding human rights (Lilley and Willkie

1994).31 China made many policy concessions in response: between 1990-93, it released

881 Tiananmen prisoners, lifted martial law, released information on high-profile political

31Congress sometimes attached other requirements to MFN renewal as well, such as measures to reduceweapons proliferation (Lilley and Willkie 1994).

18

prisoners, agreed not to export products made with prison labor, allowed the Red Cross to

visit prisoners, sent two human rights delegations to the U.S., and gave passports to many

families of political exiles (Lilley and Willkie 1994).

In 2000, the U.S. commenced a debate over whether to accord PNTR status to China,

the last major hurdle to its WTO entry. The U.S. recognized that once China entered the

WTO, any leverage the country through trade would be dramatically curtailed. For example,

U.S. Rep. Smith advocated denying PNTR status to China, stating, “This is a real vote-

the dictatorship will actually lose something they want. Deny China’s PNTR today- require

them to move in the direction of reform and the protection of human rights (The Congres-

sional Record 2000). However, other members of Congress argued that while PNTR would

result in a decreased ability to pressure China using trade policy, the U.S could promote

human rights there through other levers. Indeed, Congress introduced the “Levin Bereuter

proposal,” which would allow it to monitor China’s human rights performance. Congress-

man Underwood explained, “Once China is a member of the WTO, the United States still

can impose sanctions on China but they have to be ‘WTO consistent’...There are many

other processes to affect this ‘leverage’ over China. For example, the U.S. could use the

power of the Export-Import Bank, TDA and OPIC to apply pressure on China. Finally, the

Levin-Bereuter language...will annually grant this body the opportunity to investigate and

criticize China’s abuse in these areas” (Underwood 2000). Similarly, Congressman Hoyer

claimed that monitoring China’s human rights performance would allow Congress to main-

tain diplomatic pressure on China, arguing that “the Levin Bereuter proposal...would estab-

lish a congressional executive commission on China....A China commission will be a more

effective mechanism for maintaining pressure on China on human rights, worker rights, and

rule of law issues than our brief annual reviews” (Hoyer 2000). China’s impending WTO

entry thus led the U.S. to switch its focus from trade policy conditionality to other levers of

influence.

19

Russia

In August of 2012, the U.S. Congress debated whether to extend PNTR status to Russia,

which would pave the way for its WTO accession. Members worried that Russia’s WTO entry

would end Congress’s long-time practice of using U.S. trade policies as a tool of influence

over Russia’s human rights policies (Columbus 2003). For example, Congressman Devin

Nunes argued against granting PNTR status to Russia because, “The 1974 Jackson-Vanik

amendment [MFN conditionality] effectively pressured the Soviet Union over its appalling

human rights record. It was an important part of America’s decades-long effort to contain

and ultimately bring down an evil empire” (of Congress 2012a). Similarly, Congressman

Peter DeFazio warned against approving PNTR status for Russia, stating, “We are giving

up the tools we have to try and push Russia...We have more power today with the capability

of depriving them of a normal trade relation status with the United States. If we want to

use our clout, we should vote this bill down” (of Congress 2012a). Congressman Bill Pascrell

also opposed the bill, stating, “I’d like to use trade as leverage” (of Congress 2012a).

Because Congress worried that Russia’s eventual WTO entry would remove its leverage

over the country, it attached additional human rights provisions to the bill granting PNTR

to Russia. These provisions were derived from the Magnitsky Act, which was named after a

lawyer who had been imprisoned in Moscow due to his investigation of tax fraud by Russian

officials. The provisions prohibit the officials blamed for the incident, as well as any other

individuals involved in violations of human rights, from entering the U.S. and using the

U.S. banking system.32 Members of Congress, as well as other U.S. news sources, readily

acknowledged that the bill would substitute one form of pressure– trade– with another form

of pressure, asset freezes. Congressman Howard Berman stated: “Just as Jackson-Vanik

[MFN conditionality] became a tool to deal with one aspect of a horrible set of policies

32The EU, too, implemented visa bans and asset freezes against violators of human rights (of Congress2012a). Canada, too, considered implementing a similar bill (Kara-Murza 2012).

20

by the Soviet Union during the Cold War, we now, using the Magnitsky legislation, deal

with some very serious human rights issues remaining in Russia, but not in the context of

restricting trade” (of Congress 2012a). Senator Cardin added, “Today we close a chapter

in the U.S. history on the advancing of human rights with the repeal basically of Jackson-

Vanik [MFN conditionality]. It served its purpose. Today, we open a new chapter in U.S.

leadership for human rights with the Sergei Magnitsky Rule of Law Accountability Act”

(of Congress 2012b).33 Similarly, Congressman Robert Brady stated that because PNTR

would eliminate a source of leverage over Russia, “This legislation also creates important

new tools to continue to pressure Russia to make progress on the important issue of human

rights” (of Congress 2012a).

The bill’s intent was summed up in a Washington Post opinion article that was read to

the U.S. Congress. The article, written by a member of the federal council of the Repub-

lican Party of Russia, stated: “The prospect of losing access to the West and its financial

systems...may well be, for now, the only serious disincentive to corruption and human rights

violations by Russian officials....the adoption of the Magnitsky Act has been tied to the re-

peal of the antiquated Jackson-Vanik Amendment, thus replacing trade sanctions against a

nation with personal sanctions against specific criminals” (Kara-Murza 2012). As this article

and the surrounding Congressional testimony make clear, Russia was given PNTR, but only

after the U.S. had substituted toward using Russian assets as leverage over the country’s

human rights policies.

33Congressman Dave Camp concurred, stating, “I also fully share the concerns of many of my colleagueson Russia’s abysmal human rights record, and that’s why I support adding the Magnitsky legislation to thisbill” (of Congress 2012a).

21

Empirical Strategy

The preceding section provides evidence that WTO membership can cause foreign policy

substitution, but it could be argued that these represent isolated cases. I now provide

empirical evidence that the theory holds systematically across OECD donors. Indeed, the

logic of institution-driven policy substitution yields several testable hypotheses, which I now

examine. First, as detailed above, respect for human rights is a primary and longstanding

goal of OECD members, which they often promote by extending or revoking trade privileges.

However, WTO membership curtails OECD countries’ abilities to do so by increasing the

cost of trade policy discrimination. OECD countries should therefore be more likely to

condition trade concessions upon rights and freedoms when partner countries are not WTO

members. Accordingly, I anticipate the following:

Hypothesis 1 OECD countries should condition their imports upon the human rights records

of non-WTO members to a greater extent than they condition their imports upon the human

rights records of WTO members.

However, OECD countries are not expected to abandon their efforts to promote respect

for human rights among WTO members. Instead, states should substitute towards other

policy levers in rewarding respect for human rights and punishing human rights violations.

Thus, OECD donor countries should be more likely to condition other policy levers upon

human rights practices when partner countries are WTO members.

Hypothesis 2 OECD countries should condition non-trade policy instruments upon the hu-

man rights records of WTO members to a greater extent than they condition non-trade policy

instruments upon the human rights records of non-WTO members.

I now test these hypothesis using data from three areas of foreign policy: international

trade, bilateral foreign aid, and GSP. I select the foreign aid and GSP policy arenas to test

22

hypothesis two for two primary reasons. First, the examples above showed that both foreign

aid allocations and GSP eligibility are frequently used tools of influence, particularly once

recipients join the WTO. Second, for a long period of time and across many donors, respect

for human rights has been an important criterion in determining the allocations of these

policies, as I explain further below.

Testing Hypothesis 1

To test whether OECD countries apply conditionality to trade policies with non-WTO mem-

bers to a greater extent than they with WTO members, I estimate a gravity model of bilateral

trade flows. I adopt the traditional work-horse model in the trade literature, the log-linear

gravity model (Dutt and Traca 2010; Eicher and Henn 2011; Goldstein, Rivers and Tomz

2007; Gowa 1995; Herz and Wagner 2006; Mansfield and Bronson 1997; Mansfield, Milner

and Rosendorff 2000; Rose 2004; Tomz, Goldstein and Rivers 2007). The sample consists of

directed dyads, composed of OECD importers and their trade partners who are also eligi-

ble for foreign aid. I focus on trade partners who may receive foreign aid for two reasons:

First, since I am interested in foreign policy substitution, concentrating the analysis on aid

recipients ensures that the importers are able to substitute toward other policies. Second,

using the same sample of states to test both hypotheses allows for comparability and con-

sistency.34 The model is estimated using OLS with robust standard errors, clustered at the

directed dyad level. All specifications include year and directed dyad fixed effects, as fixed

effects are robust to many types of misspecification and endogeneity concerns. The data

contain 31 OECD states and 165 recipient states from 1981-2003.35

The dependent variable, Log Imports, is the log of imports of the first country in a pair

34Note that the results hold when the full sample of exporters is used, and also when the sample of donorsis restricted to members of the Development Assistance Committee (DAC).

35Trade data was generously supplied by Liu (2009), who carefully accounts for observations of zero tradeflows, as well as country formation and disintegration. This data is only available through 2003.

23

from the second (in constant 1995 U.S. dollars) and comes from the IMF’s “Direction of Trade

Statistics.”36 I select OECD imports as the dependent variable rather than OECD exports

because when an OECD member uses its trade policies to punish a trading partner for human

rights violations, it usually reduces imports on goods from that partner by increasing trade

protection. In other words, imports, rather than exports, are typically conditioned upon a

partner’s respect for human rights.

A key independent variable is respect for Human Rights, which is measured by the

Cingranelli-Richards (CIRI) Human Rights Project (Cingranelli and Pasquarello 1985) and

is scored on a 14-point scale, ranging from least to most respect for human rights. The

CIRI index is complied using annual reports from Amnesty International and the U.S. State

Department, which record specific instances of human rights violations. Because the CIRI

index is measured using actual violations, instead of expert opinions, it may be more credible

than alternative measures.37 For ease of interpretation, I rescale Human Rights such that it

ranges from 0, indicating no respect for rights, to 1, indicating full respect for rights. Further,

the theory implies that the effect of Human Rights on trade flows differs for WTO members

versus non-WTO members. To account for the conditional effect of rights, I interact Human

Rights with Both in WTO, an indicator of joint WTO membership.38

I also include several variables which have been shown to impact trade flows. First,

countries may trade more with states that accord them special trading privileges. This is

36Since dropping observations of zero trade leads to bias, the dependent variable is specified aslog(imports + 1) (McCallum 1995; Raballand 2003). Summary statistics for all variables are presentedin the appendix.

37This variable is constructed using the following indicators: Foreign Movement, Domestic Movement,Freedom of Speech, Freedom of Assembly and Association, Workers’ Rights, Electoral Self-Determination,and Freedom of Religion.

38I focus my analysis on formal members of the WTO, though Tomz, Goldstein and Rivers (2007) note theexistence of informal members as well, which are entitled to the majority of the rights and responsibilitiesgranted to WTO members. However, these informal members include the following groups: colonies, whichdid not have an independent foreign policy and therefore do not fit the theory; de facto members, who couldnot access the DSB and thus likely found it difficult to enforce agreements; and provincial members, whoonly obtained rights from members who had agreed to grant them. However, I note that the results arerobust to the inclusion of informal members.

24

captured by RTA, which indicates pairs that are part of a regional trade agreement (RTA),

and GSP, which indicates whether an importing country extends GSP tariff preferences to

its partner.39 Further, for each state in a pair, I control for Log GDP and Log GDP Per

Capita, measured in logged constant 1995 U.S. dollars.40 Note that many variables often

included in gravity models such as the distance between states, an indicator of a shared

border, the number of islands in a pair, etc. are collinear with the fixed effects.

TABLE 1 ABOUT HERE

Column 1 of Table 1 presents results which comport with the hypothesis. The coefficient

on Human Rights is positive and statistically significant (p < 0.001) and indicates that

among non-WTO trading partners, OECD countries import 317% more from regimes that

fully respect rights relative to those that do not respect rights. The estimated effect of

respect for human rights on OECD imports is much smaller for WTO members and is not

statistically significant (p = 0.252); the coefficient corresponding to respect for human rights

among WTO members is only 0.211 (calculated by adding the coefficients on Human Rights

and Human Rights X WTO), indicating that OECD members trade 23% more with WTO

members who fully respect rights than with WTO members that do not.

In addition to the results presented above, I demonstrate the robustness of my findings to

a variety of alternative empirical specifications. In all specifications, I find strong support for

the theory. First, we might worry about many time-varying factors specific to the importing

country. For example, importers’ strategic goals, political environments, policy priorities,

and a variety of other factors may change overtime. To account for these possibilities,

column 2 of Table 1 presents results that also include time-varying importer fixed effects (in

39None of the exporters in the analysis extend GSP status to OECD importer countries.40Trade and GDP data were supplied by Liu (2009). Many gravity models include an indicator of whether

only one state in a pair is a WTO member. However, since virtually all OECD members are WTO membersfor the sample period, this variable is collinear with the fixed effects. Additionally, since WTO members arenot bound by WTO laws in their trade relations with non-member partners, only Both in WTO is interactedwith the key independent variables.

25

addition to the year and dyad fixed effects). Second, column 3 controls for other variables

that could impact whether states use trade policy to punish or reward their partners for

human rights violations: Log(Foreign Aid) measures the foreign aid given by the importer

to the exporter, Post Cold War is an indicator of the post Cold War period (beginning in

1989), and Alliance is an indicator of a strategic military alliance.41 Finally, columns 4 and 5

use alternative measures of human rights: column 4 employs a measure of civil liberties and

column 5 includes a measure of political rights, as recorded by House (2006). Civil liberties

are defined as policies that “allow for the freedoms of expression and belief, associational

and organizational rights, rule of law, and personal autonomy without interference from the

state,” while political rights “enable people to participate freely in the political process.”

These variables are again rescaled such that they range from 0, indicating no government

respect for rights, to 1, indicating full government respect for rights. Since these measures

are available for a longer time span than the measure used in the baseline analysis, the

model is now estimated for the period 1972-2009. Additionally, since colonizers often accord

special trade privileges to their colonies, I include an indicator of whether the importer is

a Current Colonizer of its partner. This indicator was collinear with the fixed effects in

previous specifications, but can now be included due to the larger time frame.

Testing Hypothesis 2

To test whether OECD countries condition non-trade policy instruments upon the human

rights records of WTO members to a greater extent than they do for non-WTO members,

I examine two specific policy levers: first, I explore whether the hypothesis holds in the

domain of foreign aid, and second, I test whether it helps to explain unilateral preference

program eligibility.

Foreign aid serves as an ideal setting in which to test my hypothesis, as respect for hu-

41Alliance data comes from Leeds et al. (2002), and aid data is from OECD (2011).

26

man rights is commonly included in OECD criteria governing aid allocation. For example,

the EU has often incorporated human rights conditionality into its foreign aid-giving, using

instruments such as the The Lome IV Convention and the Cotonou Convention,42 and the

U.S. conditions much of its aid upon human rights through programs such as the Millennium

Challenge Account. More broadly, human rights are emphasized as an important determi-

nant of aid allocation in many OECD countries; for instance, the Millennium Development

Goals for foreign aid list explicit human rights objectives.

Similarly, GSP eligibility criteria often state that recipients must respect human rights,

particularly labor and worker rights. The U.S. in particular places strong emphasis on this,

leading Ambassador Barshefsky to testify to the U.S. Congress that, “the GSP program has

been particularly valuable in furthering...worker rights” (U.S. Congress 1995). Indeed, if

U.S. GSP recipients are thought to have violated rights, interested parties may file petitions

requesting an investigation into a state’s eligibility.43 The vast majority of petitions are

filed by the American Federation of Labor and Congress of Industrial Organizations (AFL-

CIO), the largest federation of unions in the U.S., in response to perceived labor violations.

These investigations represent a threat to revoke GSP, and are therefore frequently used

in an attempt to induce changes in labor practices. For example, the U.S. Ambassador to

Bangladesh stated, “The GSP review process can help motivate Bangladesh to reenergize

its efforts on labor issues” (Embassy Dhaka 2007). Similar examples abound.44 Due to the

42The Lome IV Convention provided financial aid to 70 African, Caribbean and Pacific countries thatcould be cut off from receiving aid if these countries violated the terms of the conditionality (Holland 2002).Similarly, the 2000 Cotonou Convention that replaced the Lome agreements provided only a first tranche ofaid while holding back the second tranche until compliance with conditionality was confirmed (Smith 2003).

43The U.S.’s program currently has 128 beneficiary developing countries (BDC) that face no tariffs onabout 4,650 tariff lines; 43 beneficiaries are designated as least-developed countries and pay no tariffs on anadditional 1,750 tariff lines.

44For instance, when the U.S. commenced an investigation of Guatemala’s labor practices in 1992, theexecutive director of the U.S./Guatemala Labor Education Project (U.S./GLEP), stated: “If Guatemalafails to respect basic worker rights, it will increasingly find...ever-increasing efforts to deny Guatemala accessto the U.S. market.” In response to the investigation, Guatemala made many reforms to its labor practicesincluding union recognition and increased penalties for violations of labor laws (Multinational Monitor 1993).In another case, in response to a GSP investigation in 1995, a representative of the Dominican Association of

27

emphasis placed on promoting respect for human rights in both GSP eligibility and foreign

aid allocation, these policy areas represent instructive domains in which to test the theory.

Foreign Aid

I investigate whether OECD members substitute towards foreign aid conditionality once their

partners become WTO members. I test this hypothesis using data on the foreign allocations

of OECD donor states from 1981 to 2009. As above, the sample includes 31 OECD member

states and 165 aid recipients. To estimate the model, I use OLS with year and directed

dyad fixed effects in all specifications. The dependent variable, Log Aid Disbursements,

is the logged net official development assistance (ODA) disbursements from OECD donor

countries, measured in constant U.S. dollars.45 The key independent variables are Human

Rights, measured by Cingranelli and Pasquarello (1985), and its interaction with Both in

WTO, all of which are defined above.

The foreign aid literature indicates that recipient need, recipient merit, and donor interest

are important factors in determining aid allocation; thus, measures of these elements should

be included in any such model (Hoeffler and Outram 2011). Because recipient merit is

typically measured using respect for human rights, one of my key independent variables,

I focus on accounting for recipient need and donor interests. To measure recipient need,

I follow Hoeffler and Outram (2011) and include measures of recipient size and wealth:

Log(GDP and Log(GDP per Capita), measured in logged constant 1995 U.S. dollars, as

well as Log(Population). I also include the Infant Mortality rate (per 1,000 Live Births), a

Free Zones testified before the U.S. Congress that the government had undertaken many labor rights reformsto prevent the loss of GSP, including: “the adoption of a New Labor Code in June 1993 (and implementingregulations in October 1993), the establishment of significant new protections for union organizers, theformation of new Labor Courts, the formation of a tripartite (government, business, and labor) OversightCommission for the free zones, and the appointment of a new, reform-minded Minister of Labor” (U.S.Congress 1995). The representative expressed hope that these reforms would satisfy the U.S. governmentand would therefore reduce investigations.

45The OECD defines ODA as non-military grants and net disbursements of loans which contain at least25% in the form of a grant. The data come from OECD (2011).

28

commonly used measure of need (Boone 1996; Feyzioglu, Swaroop and Zhu 1998), as well

as the Government’s Share of GDP, which is often used as an indicator of aid dependence

(Alesina and Weder 1999). Because recipient need also may depend on the amount of aid

the recipient receives from other donors, I control for Log(Other Aid), defined as the log of

the foreign aid that a recipient receives from sources other than the donor. Similarly, states

receiving large amounts of FDI Inflows may have less need for foreign aid.46

In addition, I include several measures of donor interest. Since donor motivations for aid

giving often depend on the donor’s strategic environment, I include an indicator of the Post

Cold War period (beginning in 1989), as the geopolitical landscape changed dramatically at

the end of the Cold War (Dunning 2004). I also include an indicator of the presence of a

strategic military Alliance between the donor and recipient (Leeds et al. 2002). Additionally,

since donor interests may depend on the donor’s trade relations with the recipient, I include

the Log(Imports) from the recipient. Many other common indicators of donor interest, such

as the donor’s colonial relationship with the recipient, global economic shocks, and geography

are accounted for by the presence of dyad and year fixed effects.

TABLE 2 ABOUT HERE

Column 1 of Table 2 presents the results of the analysis, which show strong support for

the hypothesis. Among non-WTO members, the relationship between human rights and

foreign aid is weak and insignificant, as states that fully respect rights are only allocated

about 1% more aid than states that do not respect rights. However, WTO members that

show full respect for human rights receive a large and statistically significant increase in

aid relative to WTO members that do not respect rights, obtaining 109% (p < 0.001) more

aid than repressive states receive. Figure 1 displays these results graphically, plotting the

marginal effect of human rights on logged aid for WTO members versus non-members, along

46All measures are from World Bank (2011) except for Government’s Share of GDP, which is from Heston,Summers and Aten (2009).

29

with the associated 95% confidence intervals.

Further, I show that these results are robust to a variety of alternative model specification

and coding decisions. First, column 2 accounts for any time-varying factors affecting donor

strategy or aid priorities. For example, donors’ aid preferences, domestic political environ-

ment, ideology, or other factors that influence aid giving may change overtime in ways that

are difficult to measure. To capture these possibilities, I include time-varying donor fixed

effects in addition to the dyad and year fixed effects. Second, column 3 presents the results

from re-specifying the dependent variable, which was done by averaging the dependent vari-

able over three periods. Fourth, column 4 adds other levers of influence that donors might

employ instead of foreign aid conditionality: an indicator of GSP status and an indicator of

an RTA. Column 4 also includes an indicator of whether a state is a Democracy, as another

potential measure of recipient merit. Finally, as above, columns 5 and 6 use alternative

measures of human rights: House (2006)’s measure of civil liberties (column 5), and House

(2006)’s measure of political rights (column 6).47 These measures are available for a wider

range of years than the CIRI score, so the analysis ranges from 1972-2009.

Unilateral Preference Programs

A second area in which I expect states to substitute towards foreign aid conditionality once

their partners become WTO members is in the domain of unilateral preference programs. In

particular, GSP is a wide-ranging, prominent program which is explicitly used to promote

human rights.48 Because GSP investigations are frequently used to demand improvements

in recipient’s human rights records, my theory implies that these investigations should occur

more often once states join the WTO. To test this hypothesis, I collected original data of all

instances of U.S. GSP investigations due to worker rights concerns, which I obtained directly

47These variables are again rescaled such that they range from 0, indicating no government respect forrights, to 1, indicating full government respect for rights.

48See Compa and Vogt (2000) for an overview.

30

from the USTR. The dataset begins in 1985, the year in which worker rights were added

as a criterion for GSP eligibility, and ends in 2012, the most recent year for which data

is available. While I only have data on the U.S.’s program, GSP eligibility across OECD

countries is correlated, such that U.S. policy is a good indicator of the policies of many of

those countries (Ozden and Reinhardt 2005). This correlation is due to several factors: First,

U.S. GSP eligibility is influenced by “the extent to which other major developed countries

are undertaking a comparable effort” (USTR 1999), and, second, many OECD countries

condition GSP eligibility upon respect for human rights.49

Using this data, I first present some descriptive statistics: I compute the number of years

that GSP recipients were under investigation in the area of workers’ rights, and compare

the number of investigations that occurred in the five years prior to a recipient’s WTO

accession to the number of investigations that occurred in the five years after that recipient

joined the WTO. Thus, the sample is restricted to GSP recipients that both joined the WTO

between 1985-2012, and were under investigation within a 10-year window of WTO entry;

this includes 16 countries. As shown in Table 3, in the five years prior to joining the WTO,

investigations occurred in only three country-years, whereas in the five years after joining,

investigations occurred in 16 country-years.

TABLE THREE ABOUT HERE

While the large difference in the number of investigations conducted before versus after

WTO entry is striking, I now investigate whether such a relationship between WTO mem-

bership and human rights investigations holds more systematically. The dependent variable,

GSP Investigation, is an indicator of the presence of a GSP investigation on the basis on

workers’ rights. The dataset contains a total of 190 country-years during which an inves-

tigation was underway. Since this variable is binary, I conduct the analysis using logistic

49For example, the EU also conditions GSP eligibility on workers’ rights and, consequently, investigatesmany of the same countries that the U.S. investigates, such as Pakistan in 1997, El Salvador in 2008, andmany others (Orbie 2011).

31

regression. All specifications include year and county fixed effects.

The key independent variable is an indicator of whether the recipient is a member of the

WTO. I also control for several other variables that could impact the decision to investigate

a country’s GSP eligibility, specifically where this concerns workers’ rights. First, such

a decision should depend on the country’s workers’ rights record. I therefore control for

Workers’ Rights, which is measured by Cingranelli and Pasquarello (1985) on a scale from

1, indicating no respect for rights, to 3, indicating full respect. Second, a recipient may be

more to be investigated if it is less likely to meet eligibility criteria for other reasons. The

GSP guidelines explicitly state that eligibility depends on economic development; therefore,

I include measures of Log(GDP) and Log(GDP per Capita). The guidelines also highlight

the importance of trade openness, so I control for a recipient’s total Imports and Exports,

which are common measures of openness.50 I also include Terms of Trade, the ratio of the

price of exportable goods to that of importable goods, since this is an observable indicator

often used to evaluate GSP eligibility (Grossman and Sykes 2005; JOC staff 2013).51

TABLE FOUR ABOUT HERE

Table 4 demonstrates that the hypothesis is borne out by the data. WTO membership

increases the likelihood of investigation strongly and significantly: the coefficient on WTO,

1.462, indicates that for WTO members, the odds of an investigation increases by 4.314

relative to non-WTO members. Reassuringly, improvements in workers’ rights strongly and

significantly decreases the likelihood of a GSP investigation based on violations of workers’

rights.

Column 2 adds additional variables that may impact whether an investigation occurs.

First, I include measures that may reflect the U.S.’s economic relationship with the recipient,

including U.S. Aid, Total Aid, U.S. Imports, and FDI Inflows. I also control for the Post

50Imports, exports, and GDP are used to calculate Deardorff’s “openness index.”51All economic data come from World Bank (2011).

32

Cold War period (beginning in 1989), as this may impact the strategic environment. Finally,

I include a measure of Property Rights Enforcement, which is sometimes considered when

determining GSP eligibility, along with Economic Growth. The results remain statistically

and substantively significant.

Conclusion

This article claims that a deeper understanding of the indirect effects of institutions is critical

for determining how these institutions shape the behavior of political actors. The argument

advanced here suggests a reexamination of a central tenet in international relations: that

international institutions benefit all members by enhancing cooperation. While members

may gain in the policy area governed by the institution, ascertaining whether membership is

advantageous overall requires a thorough understanding of the institution’s effects in policy

domains outside of its purview. I have demonstrated one such effect, showing that because

institutional membership constrains some policies, states substitute towards the use of more

flexible means of foreign policy leverage. Current explanations of the impact of international

institutions do not acknowledge that while institutions can improve cooperation in some

areas, they may politicize relations in others. These accounts might therefore provide an

overly optimistic view of their role in interstate relations.

Analysis of the unintended consequences of institutions is not only important for a more

complete understanding of their general benefits, but also for discerning the distribution of

those benefits. For example, this article focuses on the effect of WTO membership on foreign

policy selection, finding that membership politicizes other levers of influence. An implication

of this result is that the indirect effects of the WTO differ for powerful and weak states. For

powerful states, who frequently seek to influence their partners, the WTO restricts the use

of trade policies as a lever of influence, causing substitution toward potentially less desirable

33

tools. For weak states, the WTO increases the political requirements necessary to receive

benefits such as foreign aid, unilateral trade preferences, loans, FDI, etc. These two types

of states therefore face distinct trade-offs from WTO membership.

Further, recognition of the substitutability of levers of influence contributes to the large

literature examining the motivations underlying foreign policy selection. This article sug-

gests that policy selection depends on the entire bundle of available policies, which is shaped

by membership in international institutions. Incorporation of this important source of het-

erogeneity provides a more complete picture of the calculus by which states select policies

from an array of policy options. This paper therefore demonstrates that cognizance of the

larger institutional environment within which states operate is essential for understanding

interstate behavior.

34

References

AFP. 2004. “Japan Lays Out New Aid Policy for Vietnam.” Channel NewsAsia .

Aiello, F., P. Cardamone and M.R. Agostino. 2010. “Evaluating the impact of nonreciprocal

trade preferences using gravity models.” Applied Economics 42(29):3745–3760.

Ala’i, P. 2000. “A Human Rights Critique of the WTO: Some Preliminary Observations.”

Geo. Wash. Int’l L. Rev. 33:537.

Alesina, A. and D. Dollar. 2000. “Who Gives Foreign Aid to Whom and Why?” Journal of

Economic Growth 5(1):33–63.

Alesina, Alberto and Beatrice Weder. 1999. Do corrupt governments receive less foreign aid?

Technical report National Bureau of Economic Research.

Allee, T. L. and P. K. Huth. 2006. “Legitimizing Dispute Settlement: International Legal

Rulings as Domestic Political Cover.” American Political Science Review 100(2):219.

Apodaca, C. and M. Stohl. 1999. “United States Human Rights Policy and Foreign Assis-

tance.” International Studies Quarterly 43(1):185–198.

Appleton, Arthur E. 2000. “The World Trade Organization: Implications for Human Rights

and Democracy.” Thesaurus Acroasiuml 19:415–460.

Bagwell, K. and R. W. Staiger. 1999. “An Economic Theory of GATT.” The American

Economic Review 89(1):215–248.

Barratt, B. 2007. Human Rights and Foreign Aid: For Love or Money? Vol. 2 Psychology

Press.

Bearce, David H. 2010. “Why Foreign Aid May Be Less Effective at Promoting Economic

Growth in More Democratic Countries.” Unpublished Manuscript .

35

Bennett, D.S. and T. Nordstrom. 2000. “Foreign Policy Substitutability and Internal Eco-

nomic Problems in Enduring Rivalries.” Journal of Conflict Resolution 44(1):33–61.

Bermeo, Sarah Blodgett. 2007. “Foreign Aid, Foreign Policy, and Development: Sector

Allocation in Bilateral Aid.” Unpublished Manuscript .

Bermeo, S.B. 2010. “Development and Strategy: Aid Allocation in an Interdependent

World.” Manuscript, Duke University, Durham, NC .

Berthelemy, J.C. 2006. “Bilateral Donors Interest Vs. Recipients Development Motives in

Aid Allocation: Do All Donors Behave the Same?” Review of Development Economics

10(2):179–194.

Bhagwati, J. 1998. “Trade Linkage and Human Rights.” The Uruguay Round and beyond:

essays in honor of Arthur Dunkel 1998:241.

Biden, Joseph. 1997. URGENT CALL FOR RESTORATION OF DEMOCRACY IN CAM-

BODIA. Senate.

Boone, Peter. 1996. “Politics and the effectiveness of foreign aid.” European economic review

40(2):289–329.

Borchert, I. 2009. “Trade diversion under selective preferential market access.” Canadian

Journal of Economics/Revue canadienne d’economique 42(4):1390–1410.

Boxer, Barbara. 2008. S. 3678 (110th): Vietnam Human Rights Act of 2008. 110th Congress,

20072009.

Brenton, P. 2003. Integrating the least developed countries into the world trading system:

The current impact of EU Preferences under Everything but Arms. Vol. 3018 World Bank

Publications.

36

Brownback, Samuel. 2005. CALLING FOR THE GOVERNMENT OF CAMBODIA TO

RELEASE CHEAM CHANNY FROM PRISON, AND FOR OTHER PURPOSES. Senate

Resolution 65.

Bueno De Mesquita, B. and A. Smith. 2009. “A Political Economy of Aid.” International

Organization 63(2):309–340.

Busch, M. L. and E. Reinhardt. 2003. “Developing Countries and General Agreement on

Tariffs and Trade/World Trade Organization Dispute Settlement.” Journal of World Trade

37(4):719–736.

Busch, M. L. and E. Reinhardt. 2004. “The WTO Dispute Settlement Mechanism and

Developing Countries.” Trade Brief, Swedish International Development and Cooperation

Agency .

Busch, M. L., R. Raciborski and E. Reinhardt. 2009. “Does the Rule of Law Matter? The

WTO and US Antidumping Investigations.” Working Paper .

Busch, Marc L. 2007. “Overlapping institutions, forum shopping, and dispute settlement in

international trade.” International Organization pp. 735–761.

Buthe, T. and H. V. Milner. 2008. “The Politics of Foreign Direct Investment into Developing

Countries: Increasing FDI Through International Trade Agreements?” American Journal

of Political Science 52(4):741–762.

Carey, S.C. 2007. “European Aid: Human Rights Versus Bureaucratic Inertia?” Journal of

Peace Research 44(4):447–464.

Carleton, D. and M. Stohl. 1987. “The Role of Human Rights in US Foreign Assistance

Policy: A Critique and Reappraisal.” American Journal of Political Science pp. 1002–

1018.

37

Carnegie, Allison. 2013. “States Held Hostage: Political Hold-Up Problems and the Effects

of International Institutions.” Working Paper .

Checkel, Jeffrey T. 2005. “International institutions and socialization in Europe: Introduc-

tion and framework.” International organization 59(04):801–826.

Cingranelli, D.L. and T.E. Pasquarello. 1985. “Human Rights Practices and the Distribution

of US Foreign Aid to Latin American Countries.” American Journal of Political Science

pp. 539–563.

Clark, David H. and William Reed. 2005. “The Strategic Sources of Foreign Policy Substi-

tution.” American Journal of Political Science 49(3):609–624.

Clark, D.H., T. Nordstrom and W. Reed. 2008. “Substitution Is in the Variance: Resources

and Foreign Policy Choice.” American Journal of Political Science pp. 763–773.

Cleveland, S.H. 2001. “Human Rights Sanctions and the World Trade Organization.” Envi-

ronment, human rights and international trade .

Cohn, M. 2000. “World Trade Organization: Elevating Property Interests Above Human

Rights, The.” Ga. J. Int’l & Comp. L. 29:427.

Columbus, Frank. 2003. Russia in Transition: Volume 2. Nova Science Publishers Inc.

Compa, Lance and Jeffrey S Vogt. 2000. “Labor rights in the generalized system of prefer-

ences: a 20-year review.” Comp. Lab. L. & Pol’y J. 22:199.

Conconi, P. and C. Perroni. 2004. “The economics of special and differential trade regimes.”.

Conrad, Courtenay Ryals and Will H Moore. 2010. “What stops the torture?” American

Journal of Political Science 54(2):459–476.

38

Crawford, G. 2001. Foreign Aid and Political Reform: A Comparative Analysis of Democracy

Assistance and Political Conditionality. Palgrave Macmillan.

Davis, C. L. 2004. “International Institutions and Issue Linkage: Building Support for

Agricultural Trade Liberalization.” American Political Science Review 98(1):153–169.

Davis, C. L. 2006. “Do WTO Rules Create a Level Playing Field? Lessons from the Expe-

rience of Peru and Vietnam.” Negotiating Trade: Developing Countries in the WTO and

NAFTA pp. 219–256.

Davis, C. L. 2009. “Linkage Diplomacy: Economic and Security Bargaining in the Anglo-

Japanese Alliance, 1902-23.” International Security 33(3):143–179.

Davis, Christina L. 2012. Why Adjudicate?: Enforcing Trade Rules in the WTO. Princeton

University Press.

De Benedictis, L. and L. Salvatici. 2011. The Trade Impact of European Union Preferential

Policies: An Analysis Through Gravity Models. Springer.

De Mesquita, B.B. and A. Smith. 2007. “Foreign Aid and Policy Concessions.” Journal of

Conflict Resolution 51(2):251.

Diehl, P.F. 1994. “Substitutes or Complements?: The Effects of Alliances On Military

Spending in Major Power Rivalries.” International Interactions 19(3):159–176.

Dobbins, J. 2008. Europe’s Role in Nation-Building: From the Balkans to the Congo. Vol.

722 Rand Corp.

Dollar, D. and V. Levin. 2006. “The Increasing Selectivity of Foreign Aid, 1984-2003.” World

Development 34(12):2034–2046.

39

Dreher, A., J. E. Sturm and J. R. Vreeland. 2009. “Global Horse Trading: IMF Loans for

Votes in the United Nations Security Council.” European Economic Review 53(7):742–757.

Dumbaugh, K. 2008. “China’s Foreign Policy: What Does It Mean for U.S. Global Interests?”

Congressional Research Service .

Dunning, T. 2004. “Conditioning the Effects of Aid: Cold War Politics, Donor Credibility,

and Democracy in Africa.” International Organization 58(2):409–423.

Dutt, Pushan and Daniel Traca. 2010. “Corruption and bilateral trade flows: extortion or

evasion?” The Review of Economics and Statistics 92(4):843–860.

Easterly, W. 2007. “Are Aid Agencies Improving?” Economic Policy 22(52):633–678.

Eicher, Theo S and Christian Henn. 2011. “In search of WTO trade effects: Preferential trade

agreements promote trade strongly, but unevenly.” Journal of International Economics

83(2):137–153.

Embassy Dhaka. 2007. “BANGLADESH: LABOR-GSP PETITION – STAKEHOLDER

VIEWS.”.

Feyzioglu, Tarhan, Vinaya Swaroop and Min Zhu. 1998. “A panel data analysis of the

fungibility of foreign aid.” The World Bank Economic Review 12(1):29–58.

Fleck, R.K. and C. Kilby. 2006. “How Do Political Changes Influence US Bilateral Aid

Allocations? Evidence From Panel Data.” Review of Development Economics 10(2):210–

223.

Gladstone, Cary. 2001. Afghanistan: history, issues, bibliography. Nova Science Pub Inc.

Goldstein, J. L., D. Rivers and M. Tomz. 2007. “Institutions in International Relations:

Understanding the Effects of the GATT and the WTO on World Trade.” International

Organization 61:37–67.

40

Gowa, J. 1995. Allies, Adversaries, and International Trade. Princeton, NJ: Princeton

University Press.

Grossman, Gene M and Alan O Sykes. 2005. “A preference for development: the law and

economics of GSP.” World Trade Review 4(01):41–67.

Guide to GATT Law and Practice (Analytical Index). 1995. World Trade Organisation:

Geneva .

Hafner-Burton, E. 2009. Forced to Be Good: Why Trade Agreements Boost Human Rights.

Cornell University Press.

Hafner-Burton, E.M. 2005. “Trading human rights: How preferential trade agreements

influence government repression.” International Organization pp. 593–629.

Hafner-Burton, E.M. and A.H. Montgomery. 2008. “Power or Plenty: How Do International

Trade Institutions Influence the Proliferation of Economic Sanctions?” Journal of conflict

Resolution 52(2):213–242.

Hafner-Burton, Emilie. 2013. Making Human Rights a Reality. Princeton University Press.

Hafner-Burton, Emilie M. 2012a. “International regimes for human rights.” Annual Review

of Political Science 15:265–286.

Hafner-Burton, Emilie M. 2012b. “International Regimes for Human Rights.” Annual Review

of Political Science 15:265–286.

Herz, B. and M. Wagner. 2011. “The Dark Side of the Generalized System of Preferences.”

Review of International Economics 19(4):763–775.

Herz, Bernhard and Marco Wagner. 2006. “Do the World Trade Organization and the

Generalized System of Preferences foster bilateral trade?” Available at SSRN 1002846 .

41

Heston, Alan, Robert Summers and Bettina Aten. 2009. “Penn world table version 6.3.”

Center for International Comparisons at the University of Pennsylvania .

Hoeffler, Anke and Verity Outram. 2011. “Need, Merit, or Self-InterestWhat Determines the

Allocation of Aid?” Review of Development Economics 15(2):237–250.

Hoekman, B. and C. Ozden. 2005. “Trade preferences and differential treatment of developing

countries: a selective survey.” World Bank policy research working paper (3566).

Holland, M. 2002. THe European Union and the Third World. Palgrave.

Hollyer, James and Bryan Rosendorff. 2011. “Why do authoritarian regimes sign the con-

vention against torture? Signaling, domestic politics and non-compliance.” Signaling,

Domestic Politics and Non-Compliance (June 1, 2011) .

House, F. 2006. “Freedom in the World: Political Rights and Liberties 1972-2005.” Dataset

available at www.freedomhouse.org .

Howse, R. and M. Mutua. 2000. Protecting Human Rights in a Global Economy: Challenges

for the World Trade Organization. Rights & Democracy, International Centre for Human

Rights and Democratic Development.

Hoyer, Steny. 2000. Authorizing Extension of Non-Discriminatory Treatment (Normal Trade

Relations Treatment) to People’s Republic of China. Congressional Record.

Human Rights Act. 2009. “Vietnam Human Rights Act of 2009–Introduced in Senate.”.

Hyde, Henry. 2001. Viet Nam Human Rights Act. Congressional Record.

Jackson, James K. 2003. Export-Import Bank: Background and Legislative Issues. Congres-

sional Research Service, Library of Congress.

JOC staff. 2013. “GSP Renewal Legislation Introduced in US House.”.

42

Johnson, H.G. 1967. Economic policies toward less developed countries. Brookings institution

Washington, DC.

Jones, V.C. 2013. Generalized System of Preferences: Background and Renewal Debate.

DTIC Document.

Jones, Vivian C. 2008. “Generalized System of Preferences: Background and Renewal De-

bate.” CRS Report for Congress RL33663.

Kara-Murza, Vladimir V. 2012. “The Kremlin’s Blacklist.” The Washington Post .

Keck, A. and A. Lendle. 2012. “New evidence on preference utilization.” World Trade

Organization Staff Working Paper No. ERSD-2012-12 .

Kelley, Judith. 2004. “International Actors on the Domestic Scene: Membership Condition-

ality and Socialization by International Institutions.” International Organization 58:425–

457.

Keohane, R. 1984. “0.(1984) After Hegemony.” Cooperation and Discord in the World Po-

litical Economy, Princeton NJ .

Kirk, R. and M. Raceanu. 1994. Romania Versus the United States: Diplomacy of the Absurd,

1985-1989. Palgrave Macmillan.

Krasner, S. D. 1991. “Global Communications and National Power: Life on the Pareto

Frontier.” World Politics: A Quarterly Journal of International Relations 43(3):336–366.

Kuziemko, I. and E. Werker. 2006. “How Much Is a Seat On the Security Council Worth?

Foreign Aid and Bribery at the United Nations.” Journal of Political Economy 114(5):905–

930.

43

Lai, B. 2003. “Examining the Goals of US Foreign Assistance in the Post-Cold War Period,

1991-96.” Journal of Peace Research 40(1):103–128.

Lebovic, J.H. and E. Voeten. 2009. “The Cost of Shame: International Organizations and

Foreign Aid in the Punishing of Human Rights Violators.” Journal of Peace Research

46(1):79–97.

Lederman, D. and C. Ozden. 2004. “US trade preferences: All are not created equal.”

Documentos de Trabajo (Banco Central de Chile) (280):1.

Leeds, B., J. Ritter, S. Mitchell and A. Long. 2002. “Alliance Treaty Obligations and

Provisions, 1815-1944.” International Interactions 28:237–260.

Lektzian, D.J. and C.M. Sprecher. 2007. “Sanctions, Signals, and Militarized Conflict.”

American Journal of Political Science 51(2):415–431.

Levitz, Philip and Grigore Pop-Eleches. 2010. “Why No Backsliding? The European Union’s

Impact on Democracy and Governance Before and After Accession.” Comparative Political

Studies 43(4):457–485.

Lilley, J. R. and W. L. Willkie. 1994. Beyond MFN: Trade with China and American

Interests. American Enterprise Institute.

Lim, H. 2001. “Trade and Human Rights What’s at Issue?” Journal of World Trade

35(2):275–300.

Liu, X. 2009. “GATT/WTO Promotes Trade Strongly: Sample Selection and Model Speci-

fication.” Review of International Economics 17(3):428–446.

Lum, T. 2009. Cambodia: Background and U.S. Relations. DTIC Document.

44

Lumsdaine, D.H. 1993. Moral Vision in International Politics: The Foreign Aid Regime,

1949-1989. Princeton Univ Pr.

Maggi, G. 1999. “The Role of Multilateral Institutions in International Trade Cooperation.”

The American Economic Review 89(1):190–214.

Maggi, G. and A. Rodriguez-Clare. 1998. “The Value of Trade Agreements in the Presence

of Political Pressures.” Journal of Political Economy 106(3):574–601.

Mansfield, E. D., H. V. Milner and B. P. Rosendorff. 2000. “Free to Trade: Democracies,

Autocracies, and International Trade.” American Political Science Review 94(2):305–321.

Mansfield, E. D., H. V. Milner and B. P. Rosendorff. 2002. “Why Democracies Cooperate

More: Electoral Control and International Trade Agreements.” International Organization

56:477–513.

Mansfield, E. D. and R. Bronson. 1997. “Alliances, Preferential Trading Arrangements, and

International Trade.” The American Political Science Review 91(1):94–107.

Mansfield, Edward D. and Eric Reinhardt. 2008. “International Institutions and the Volatil-

ity of International Trade.” International Organization 62(4):pp. 621–652.

Manyin, M. E. 2006. “Vietnam PNTR Status and WTO Accession: Issues and Implications

for the United States.”.

Manyin, M.E. 2001. The Vietnam-US Bilateral Trade Agreement. Congressional Research

Service, Library of Congress.

Marceau, G. 2002. “WTO Dispute Settlement and Human Rights.” European Journal of

International Law 13(4):753–814.

45

Mccain, John. 1996. EXTENDING MOST-FAVORED-NATION TREATMENT FOR CAM-

BODIA. Senate.

McCallum, J. 1995. “National Borders Matter: Canada-US Regional Trade Patterns.” The

American Economic Review 85(3):615–623.

McCormick, J.M. and N. Mitchell. 1988. “Is US Aid Really Linked to Human Rights in

Latin America?” American Journal of Political Science 32(1):231–239.

McGinnis, M.D. 2000. “Policy Substitutability in Complex Humanitarian Emergencies.”

Journal of Conflict Resolution 44(1):62.

Mearsheimer, J. J. 1994. “The False Promise of International Institutions.” International

Security 19(3):5–49.

Milner, H. and D. Tingley. 2012. “Sailing the Water’s Edge: Where Domestic Politics Meets

Foreign Policy.” Working Paper .

Milner, H.V. 2006. “Why Multilateralism? Foreign Aid and Domestic Principal-Agent

Problems.” Delegation and agency in international organizations pp. 107–39.

Moore, W.H. 1998. “Repression and Dissent: Substitution, Context, and Timing.” American

Journal of Political Science pp. 851–873.

Morgan, Dan. 2000. Merchants of Grain. iUniverse.

Most, B.A. and H. Starr. 1989. Inquiry, Logic, and International Politics. Univ of South

Carolina Pr.

Multinational Monitor. 1993. Challenging Guatamala ’s Labor Abuses. multinationalmoni-

tor.org.

46

Neumayer, E. 2003a. “Do Human Rights Matter in Bilateral Aid Allocation? A Quantitative

Analysis of 21 Donor Countries*.” Social Science Quarterly 84(3):650–666.

Neumayer, E. 2003b. “Is Respect for Human Rights Rewarded-An Analysis of Total Bilateral

and Multilateral Aid Flows.” Hum. Rts. Q. 25:510.

Neumayer, E. 2003c. The Pattern of Aid Giving: The Impact of Good Governance On

Development Assistance. Vol. 34 Psychology Press.

Nielsen, Rich. 2012. “Rewarding Human Rights? Selective Aid Sanctions Against Repressive

States.” International Studies Quarterly .

Nilsson, L. 2002. “Trading relations: is the roadmap from Lome to Cotonou correct?”

Applied Economics 34(4):439–452.

Nilsson, L. 2007. “Comparative effects of EU and US trade policies on developing country

exports.” The European Union and Developing Countries: Trade, Aid, and Growth in an

Integrating World pp. 49–70.

Noel, A. and J.P. Therien. 1995. “From Domestic to International Justice: The Welfare

State and Foreign Aid.” International Organization 49(03):523–553.

OECD. 2011. http://www.oecd.org.

of Congress, Library. 2012a. RUSSIA AND MOLDOVA JACKSON-VANIK REPEAL AND

SERGEI MAGNITSKY RULE OF LAW ACCOUNTABILITY ACT OF 2012, House of

Representatives. Library of Congress.

of Congress, Library. 2012b. RUSSIA AND MOLDOVA JACKSON-VANIK REPEAL AND

SERGEI MAGNITSKY RULE OF LAW ACCOUNTABILITY ACT OF 2012, Senate.

Library of Congress.

47

Onguglo, B. 1999. “Developing countries and trade preferences.” Trade rules in the making:

challenges in regional and multilateral negotiations pp. 109–133.

Orbie, J. 2011. “Promoting labour standards through trade: normative power or regulatory

state Europe?”.

Ozden, C. and E. Reinhardt. 2004. “First do no harm: The effect of trade preferences on

developing country exports.” Emory University manuscript, April .

Ozden, C. and E. Reinhardt. 2005. “The perversity of preferences: GSP and developing

country trade policies, 1976–2000.” Journal of Development Economics 78(1):1–21.

Ozden, C. and G. Sharma. 2006. “Price effects of preferential market access: Caribbean basin

initiative and the apparel sector.” The World Bank Economic Review 20(2):241–259.

Palmer, G., S.B. Wohlander and T.C. Morgan. 2002. “Give or Take: Foreign Aid and Foreign

Policy Substitutability.” Journal of Peace Research 39(1):5–26.

Palmer, G. and T.C. Morgan. 2010. A Theory of Foreign Policy. Princeton Univ Pr.

Patterson, G. 1965. “Would Tariff Preferences Help Economic Development?” The Interna-

tional Executive 7(3):1–2.

Payaslian, S. 1995. “US Foreign Economic and Military Aid: Reagan and Bush Administra-

tions.” Recherche 67:02.

Petersmann, E.U. 2000. “From “Negative” to “Positive” Integration In The WTO: Time

for Mainstreaming Human Rights Into WTO Law?” Common Market Law Review

37(6):1363–1382.

Pevehouse, Jon C. 2002. “Democracy from the Outside-In? International Organizations and

Democratization.” International Organization 56(3):pp. 515–549.

48

Poe, S., S. Pilatovsky, B. Miller and A. Ogundele. 1994. “Human Rights and US Foreign

Aid Revisited: The Latin American Region.” Hum. Rts. Q. 16:539.

Poe, S.C. 1992. “Human Rights and Economic Aid Allocation Under Ronald Reagan and

Jimmy Carter.” American Journal of Political Science pp. 147–167.

Poe, S.C. and R. Sirirangsi. 1994. “Human Rights and US Economic Aid During the Reagan

Years.” SOCIAL SCIENCE QUARTERLY-AUSTIN- 75:494–494.

Pregelj, V. N. 2005. Normal-Trade-Relations (Most-Favored-Nation) Policy of the United

States. CRS Report for Congress.

Qureshi, A.H. 1998. “International Trade and Human Rights From the Perspective of the

WTO.” International Economic Law with a Human Face pp. 159–173.

Raballand, G. 2003. “Determinants of the Negative Impact of Being Landlocked on Trade:

An Empirical Investigation Through the Central Asian Case.” Comparative Economic

Studies 45(4):520–536.

Regan, P.M. 2000. “Substituting Policies During US Interventions in Internal Conflicts.”

Journal of Conflict Resolution 44(1):90–106.

Relative to Cambodia Human Rights Record. 1996. Congressional Record.

Report, GAO. 1994. International Trade: U.S. Government Policy Issues Affecting U.S.

Business Activities in China. United States General Accounting Office.

Rice, Condoleeza. Dec 10, 2007. “Demarche On Gambling - WTO Article XXI Procedure.”

wikileaks.org/cable/2007/12/07STATE165472.html.

Rose, A. K. 2004. “Do We Really Know That the WTO Increases Trade?” The American

Economic Review 94(1):98–114.

49

Roth, Toby. 1996. Relative to Cambodia Human Rights Record. Congressional Record.

Schimmelfennig, Frank. 2005. “Strategic Calculation and International Socialization: Mem-

bership Incentives, Party Constellations, and Sustained Compliance in Central and Eastern

Europe.” International Organization pp. 827–860.

Schneider, C.J. and J.L. Tobin. 2010. “Tying the Hands of Its Masters? Interest Coalitions

and Multilateral Aid Allocation in the European Union.” Unpublished Working Paper .

Schoultz, L. and L.G. Schoultz. 1981. Human Rights and United States Policy Toward Latin

America. Princeton University Press Princeton, NJ.

Schraeder, Peter J., Steven W. Hook and Bruce Taylor. 1998. “Clarifying the Foreign Aid

Puzzle: A Comparison of American, Japanese, French, and Swedish Aid Flows.” World

Politics 50(2):294–323.

Scott, J.M. and C.A. Steele. 2011. “Sponsoring Democracy: The United States and Democ-

racy Aid to the Developing World, 1988–20011.” International Studies Quarterly 55(1):47–

69.

Sikkink, Kathryn. 2011. The Justice Cascade: How Human Rights Prosecutions Are Chang-

ing World Politics (The Norton Series in World Politics). WW Norton & Company.

Simmons, Beth A. 2009. Mobilizing for human rights: international law in domestic politics.

Cambridge University Press.

Smith, K.E. 2003. European Union foreign policy in a changing world. Polity Cambridge.

Sobel, Clifford. 2006. “BANGLADESH: LABOR-GSP PETITION – STAKEHOLDER

VIEWS.”.

50

Staiger, R. W. and G. Tabellini. 1999. “Do GATT Rules Help Governments Make Domestic

Commitments?” Economics and Politics 11(2):109–144.

Stevens, C. and J. Kennan. 2005. “Making trade preferences more effective.” IDS Briefing.

Institute for Development Studies, Brighton .

Stirling, P. 1996. “Use of Trade Sanctions As an Enforcement Mechanism for Basic Human

Rights: A Proposal for Addition to the World Trade Organization, The.” Am. UJ Int’l L.

& Pol’y 11:1.

Subramanian, A. and S.J. Wei. 2007. “The WTO promotes trade, strongly but unevenly.”

Journal of International Economics 72(1):151–175.

Suzumura, Kotaro. 1997. “Japan’s Accession to the GATT: A Teacher by Positive or Negative

Examples?” Hitotsubashi Journal of Economics 38:101–123.

Svensson, J. 1999. “Aid, Growth and Democracy.” Economics & Politics 11(3):275–297.

The Congressional Record. 1996a. Quoted by Secretary William Perry.

The Congressional Record. 1996b. Quoted by Marshall F. Adair.

The Congressional Record. 1996c. Quoted by Senator Chuck Grassley.

The Congressional Record. 1996d. Quoted by Rep. Tom Lantos.

The Congressional Record. 2000. Quoted by Rep. Christopher Smith.

The Congressional Record. 2001. Quoted by Rep. Loretta Sanchez.

The Congressional Record, 105th Congress. 1998. Quoted by Senator Samuel Bond.

Therien, J.P. and A. Noel. 2000. “Political Parties and Foreign Aid.” American Political

Science Review pp. 151–162.

51

Thomas, Craig. 1996. DEMOCRACY IN CAMBODIA. Senate.

Tingley, D. 2009. “Beliefs and Preferences Over Foreign Economic Aid: An Experimental

Study.” Manuscript, Department of Politics, Princeton University .

Tingley, D. 2010. “Donors and Domestic Politics: Political Influences On Foreign Aid Effort.”

The Quarterly Review of Economics and Finance 50(1):40–49.

Tobin, Jennifer L and Marc L Busch. 2009. “A BIT is better than a lot: Bilateral Investment

Treaties and Preferential Trade Agreements.” World Politics 62(1):1.

Tomasevski, K. and K. Tomasevski. 1997. Between Sanctions and Elections: Aid Donors

and Their Human Rights Performance. Cassell.

Tomz, M. 2007. Reputation and international cooperation: sovereign debt across three cen-

turies. Princeton Univ Pr.

Tomz, M., J. L. Goldstein and D. Rivers. 2007. “Do We Really Know That the WTO

Increases Trade? Comment.” The American Economic Review 97(5):2005–2018.

Underwood, Robert. 2000. Authorizing Extension of Non-Discriminatory Treatment (Normal

Trade Relations Treatment) to People’s Republic of China. Congressional Record.

U.S. Congress. 1995. “CAMBODIA AND BULGARIA MFN TRADE STATUS; RENEWAL

OF THE GSP PROGRAM; AND TRADE AGENCY FY 1996 BUDGETS HEARING

BEFORE THE SUBCOMMITTEE ON INTERNATIONAL TRADE OF THE COM-

MITTEE ON FINANCE.”.

USTR. 1999. US Generalized System of Preferences Guidebook. Technical report Office of

the U.S. Trade Representative, Executive Office of the President, Washington, DC.

52

Vachudova, Milada. 2005. Europe Undivided: Democracy, Leverage, and Integration After

Communism. New York: Cambridge.

Verdeja, L. 2006. “EUs Preferential Trade Agreements With Developing Countries Revis-

ited.” Unpublished, University of Nottingham, School of Economics .

Wilson, B. 2007. “Compliance by WTO Members with Adverse WTO Dispute Settlement

Rulings: The Record to Date.” Journal of International Economic Law 10(2):397–403.

World Bank. 2011. World Development Indicators, CD-ROM. Washington, DC: World Bank

.

WTO. 2012. http://www.wto.org. Accessed April 25, 2012 .

Zagel, G. M. 2004. “The WTO and Trade-Related Human Rights Measures: Trade Sanctions

Vs. Trade Incentives.” Austrian Review of International and European Law .

Zoellick, Robert. 2002. “A LEGACY AND A FUTURE – INITIAL OBSERVATIONS ON

BRAZIL - U.S RELATIONS.”.

53

Tables

Table 1: Effects of WTO and Human Rights on Imports1 2 3 4 5

Human Rights X WTO -1.217*** -1.204*** -1.223*** -0.173** -0.111*(0.337) (0.331) (0.337) (0.062) (0.048)

Human Rights 1.428*** 1.478*** 1.434*** 1.158*** 0.538(0.343) (0.340) (0.343) (0.336) (0.278)

Both in WTO 0.659** 0.640** 0.669** 0.695** 0.517*(0.251) (0.245) (0.251) (0.230) (0.204)

Log(GDPi) 8.070*** 8.461*** 11.169*** 11.123***(1.750) (1.813) (1.472) (1.473)

Log(GDPj) 1.465* 1.658** 1.371* 1.000 1.028(0.634) (0.627) (0.638) (0.532) (0.536)

Log(GDP per Capitai) -8.204*** -8.668*** -11.924*** -11.890***(1.893) (2.079) (1.963) (1.627)

Log(GDP per Capitaj) 0.629 0.475 0.717 1.556** 1.545**(0.574) (0.567) (0.702) (0.575) (0.506)

RTA 0.236 0.300* 0.247 -0.138 -0.163(0.139) (0.143) (0.140) (0.119) (0.119)

GSP 0.291 0.271 0.314 0.593*** 0.593***(0.181) (0.204) (0.184) (0.176) (0.177)

Currently Colonizer -6.087*** -6.111***(1.483) (1.506)

Alliance -0.612*(0.277)

Log(Foreign Aid) 0.007(0.006)

Post Cold War -0.767*(0.369)

Constant -81.071*** -83.113*** -106.764*** -106.242***(18.535) (19.894) (15.406) (15.418)

R-Squared 0.755 0.7935 0.755 0.731 0.730N 37320 37844 37320 50546 50546

Notes: Estimates from OLS regression. The unit of observation is the directed dyad-year and the dependentvariable is the natural log of (imports +1). The data in columns 1-3 cover the years 1981-2003, and the datain columns 4-5 cover the years 1972-2003. Robust standard errors, clustered by directed dyad, appear inparentheses. All models include year and directed dyad fixed effects, which are not shown. “*”, “**”, and“***” denote p < 0.05, p < 0.01, and p < 0.001 respectively.

54

Table 2: Effects of WTO and Human Rights on Foreign Aid1 2 3 4 5 6

Human Rights X WTO 0.725* 0.726* 0.795* 0.888* 0.105*** 0.079***(0.368) (0.350) (0.382) (0.372) (0.028) (0.023)

Human Rights 0.013 0.053 0.035 -0.071 0.450** 0.302*(0.322) (0.308) (0.317) (0.323) (0.160) (0.133)

Both in WTO -0.167 -0.107 -0.494 -0.355 -0.119 -0.039(0.263) (0.249) (0.272) (0.266) (0.099) (0.087)

Log(GDP) 2.662*** 2.232*** 2.329** 2.244** 2.463*** 2.255***(0.726) (0.631) (0.779) (0.748) (0.389) (0.389)

Log(GDP per Capita) -2.644*** -2.214*** -2.321** -2.214** -2.486*** -2.276***(0.727) (0.632) (0.781) (0.749) (0.389) (0.388)

Government Share of GDP -0.005*** -0.005*** -0.004*** -0.005*** - 0.004*** -0.004***(0.001) (0.001) (0.001) (0.001) (0.000) (0.000)

FDI Inflows 0.004 0.004 0.005* 0.003 0.003 0.003(0.003) (0.003) (0.002) (0.003) (0.002) (0.002)

Infant Mortality Rate -0.021*** -0.016*** -0.019*** -0.022*** -0.014*** -0.014***(0.005) (0.005) (0.006) (0.005) (0.002) (0.002)

Log(Other Aid) 0.090*** 0.093*** 0.070*** 0.094*** 0.106*** 0.104***(0.017) (0.017) (0.016) (0.018) (0.007) (0.007)

Log(Imports) 0.021** 0.002 0.017* 0.023** 0.005 0.006(0.008) (0.007) (0.008) (0.008) (0.004) (0.004)

Post Cold War -0.151 -1.064*** -0.899** 0.687 0.733(0.269) (0.286) (0.281) (0.571) (0.571)

Log(Population) 3.654*** 2.870*** 2.365* 3.189*** 2.305*** 2.056***(0.890) (0.747) (0.936) (0.917) (0.400) (0.399)

Alliance -0.502 -0.039 -0.335 -0.307 -0.275 -0.298(0.780) (0.787) (0.786) (0.752) (0.155) (0.155)

Democracy -0.101(0.110)

GSP 0.973*(0.388)

RTA -0.072(0.352)

Constant 21.486*** 13.996* 20.250*** 9.829*** 9.570***(2.107) (6.350) (5.768) (1.614) (1.614)

Year and Dyad Effects Yes No Yes Yes Yes YesR-Squared 0.597 0.688 0.606 0.557 0.557N 49625 49625 41040 46664 59013 59013

Notes: Estimates from OLS regression. The unit of observation is the directed dyad-year and the dependentvariable is the natural log of (ODA +1). The data in columns 1-4 cover the years 1981-2003, and thedata in columns 4-6 cover the years 1972-2003. Robust standard errors, clustered by directed dyad, appearin parentheses. All models include year and directed dyad fixed effects, which are not shown. Dummyindicating missing import data is also not shown. “*”, “**”, and “***” denote p < 0.05, p < 0.01, andp < 0.001 respectively.

55

Table 3: Years Under GSP Investigation Before and After WTO Entry

5 Years Pre 5 Years PostWTO Entry WTO Entry

Years UnderInvestigation 3 16

56

Table 4: Effects of WTO on GSP Human Rights Investigations1 2

Both in WTO 1.462* 1.546*(0.658) (0.751)

Log(GDP) 6.800* 6.406(3.039) (3.357)

Log(GDP per Capita) -5.821* -4.629(2.822) (3.347)

Exports 0.069* 0.072*(0.033) (0.036)

Imports -0.088** -0.094*(0.033) (0.037)

Terms of Trade -0.008 -0.011(0.006) (0.008)

Workers’ Rights -1.056*** -1.231***(0.280) (0.307)

Economic Growth -0.005(0.041)

FDI Inflows -0.031(0.070)

Log(U.S. Aid) -0.109**(0.034)

Total Aid -0.000(0.000)

Post Cold War -15.127*(6.498)

Log(U.S. Imports) 0.284(18.839)

Security of Property Rights -0.445(0.236)

Log likelihood -157.715 -141.121Chi Sq 73.960 100.860N 593 577

Notes: Estimates from logistic regression. The unit of observation is the directed dyad-yearand the dependent variable is an indicator of the presence of a GSP investigation. The datacover the years 1985-2009. All models include year and directed dyad fixed effects, whichare not shown. Dummies indicating missing import data and missing property right data isalso not shown. “*”, “**”, and “***” denote p < 0.05, p < 0.01, and p < 0.001 respectively.

57

Figures

-0.5

0.0

0.5

1.0

1.5

2.0

WTO Member

Log

Trad

e

No Yes

-0.5

0.0

0.5

1.0

1.5

2.0

WTO Member

Log

Aid

No Yes

Effect of Human Rights

Figure 1: OLS regression estimates of the effect of full respect for human rights relativeto no respect for human rights on logged imports and logged aid, conditional on WTOmembership. Estimates presented with 95% confidence intervals.

58

Supplemental Appendix

WTO and Cooperation

The WTO in particular represents a large, multilateral trading arrangement whereby all

states agree to provide their trading partners with the same low tariff rates, known as Most

Favored Nation (MFN) treatment. WTO members may not discriminate by offering some

trading partners better tariff treatment than others, although several exceptions exist to

these rules, which I discuss below. Commitments made to other states are enforced by

the WTO’s Dispute Settlement Body (DSB), which adjudicates disputes between members.

Potential violations of WTO laws are often identified by industries that have a stake in

parties’ compliance with those laws, after which the government decides whether to take

the case to the WTO. Once one party has initiated a complaint against another, the two

parties begin consultations. If they are unable to reach an agreement, a panel is formed,

consisting of three expert judges chosen from the member states and nominated by those

states.52 Third parties may also participate in a case if they have a “substantial interest”

in the outcome. The DSB may authorize retaliation against the offending party, after which

it monitors compliance with the ruling. In this way, the court can resolve uncertainties

about the law, provide the authority to retaliate, and allow states to develop reputations for

compliance.

An important feature of the WTO’s impact on states’ reputations is that it ensures

transparency in compliance with the law in several ways. First, it publicizes the cases

brought before it along with the rulings, so that all members learn which states were found

to be in violation, and whether they complied with the court. Second, it regularly monitors

compliance with international law through trade policy reviews. Reviews are conducted

52The DSB was strengthened when the GATT became the WTO in 1995, such that an appellate bodywas formed, strict timetables were established, and parties could no longer veto the adoption of a panel.

A1

every two years on the trading practices of the biggest traders, the U.S., EU, China and

Japan: every four years for the next 16 largest traders: and every six years for the smallest

traders. Third, the WTO requires states to provide notification of changes in their trade

policies that would affect other members.

The DSB therefore provides WTO participants with “a guarantee for the right to ne-

gotiate, a common standard for evaluating outcomes, the option for several countries to

join a dispute, and incentives for states to change a policy found to violate trade rules”

(Davis 2006). By developing norms and behaviors that states are expected to follow and

by specifying whether states are compliant or non-compliant, the DSB impacts members’

reputations, which serves as an effective restraint in many settings (Tomz 2007). Developing

a reputation for cooperation encourages continued cooperation in the future. Conversely, if

countries are known to be frequent violators, their partners may reduce cooperation with

them by withdrawing trade concessions, becoming more reluctant to enter agreements, or

becoming less cooperative in related areas (Maggi 1999). Additionally, the WTO provides

systemic incentives for cooperation: if a country often breaches WTO law, what is to prevent

its trading partners from recurrent violations? This could lead to a breakdown of the system

as a whole– an outcome which many states have a strong incentive to avoid. Finally, WTO

rulings can provide countries with domestic political cover for adhering to agreements (Allee

and Huth 2006; Staiger and Tabellini 1999) and an ability to develop domestic reputations

for compliance (Mansfield, Milner and Rosendorff 2002).

The loss of reputation before the large audience of WTO members appears to be a

strong deterrent, as there are many examples of the WTO helping to uphold agreements

between powerful and weak states. Powerful states have shown high rates of compliance

with WTO rulings on cases brought by weak states (Busch and Reinhardt 2003, 2004; Davis

2006; Wilson 2007). Additionally, Busch, Raciborski and Reinhardt (2009) shows that large

WTO members reduce protectionist practices against small members in particular. Further,

A2

evidence shows that stock prices respond to WTO rulings, indicating that investors anticipate

compliance (Busch, Raciborski and Reinhardt 2009).53 Often, the mere threat of a WTO

case can lead to compliance with WTO law. For example, Japan threatened to file a case

against the U.S. in 1995 in response to U.S. threats regarding Japanese automobiles, after

which the U.S. backed down (Davis 2012). By improving members’ abilities to commit to

agreements, the WTO can help countries to limit political hold-up problems.

Indeed, members certainly seem to believe that the WTO can enable their partners to

commit to agreements. Consider the U.S. Congress’ discussion of the potential benefits from

Russia’s recent WTO accession. Congressman Sandy Levin argued that Russia’s entry would

benefit the U.S. because “now that they’re in the WTO...when they violate the requirements,

there’s a dispute settlement system that can be enforced.” Similarly, Congressman Howard

Berman argued that without the constraints imposed by the WTO, “we have no idea how

Russia will treat our U.S. exports and we will have no way to hold them accountable”

(of Congress 2012a). Senator Portman stated, “By joining [the WTO], Russia did agree to

abide by a certain set of common rules, and when they break those rules, other countries

can then take them to court and help hold their feet to the fire” (of Congress 2012a).

WTO and Conditionality

While the previous section discussed incentives for adhering to WTO law, this section now

discusses WTO law with regard to policy conditionality. The WTO limits trade policy con-

ditionality by restricting unilateral discrimination against specific trade partners. Outside of

the WTO, states often use their trade policies to punish or reward trading partners.54 How-

53An example is the movement in the stock price of Chiquita bananas after the WTO ruling that the EUhad to lower tariffs on them.

54Note that there is a large legal debate regarding WTO members’ abilities to restrict trade due to humanrights concerns (Ala’i 2000; Appleton 2000; Bhagwati 1998; Cleveland 2001; Cohn 2000; Howse and Mutua2000; Lim 2001; Marceau 2002; Petersmann 2000; Qureshi 1998; Stirling 1996). However, I simply claimthat the WTO makes restricting trade for these types of reasons more costly, as explained below.

A3

ever, WTO members must agree to accord all other members MFN tariffs. Several exceptions

to this rule exist, which either allow countries to implement additional trade restrictions, or

enable them to offer concessions beyond those guaranteed by WTO membership. Consider

the exceptions which permit additional trade restrictions. Although Articles I and XI of the

GATT prohibit discriminatory trade restrictions, exceptions to the rule are given by Article

XXI, which states that GATT laws are subordinate to the United Nations (UN) Charter.

If the UN votes to sanction a specific country, WTO members may abide by the decision.

Importantly, unilateral action is not covered by this exception. UN decisions are multilateral

in nature, and therefore do not permit individual countries to sanction partners without UN

approval (Zagel 2004). Securing approval can be difficult and time-consuming, resulting in

many instances where a country would like to administer sanctions but is not able to obtain

UN agreement.

Article XXI also contains a national security exception stating that “nothing in this

Agreement shall be construed to...prevent any contracting party from taking any action

which it considers necessary for the protection of its essential security interests.” At first

glance, this seems like an extremely permissive exception. Yet countries have been very

cautious about invoking the security exception due to fears about its potential to undermine

the trading system. For example, in 2007, the U.S. sought to use Article XXI to clarify that

it does not have WTO gambling commitments, for which the EU sought compensation. The

U.S. and EU reached an agreement, and in a declassified cable from then-Secretary of State

Condoleeza Rice, the U.S. stated that it “shared the systemic concerns expressed by the EU

and a number of other claimants about setting any precedent that might encourage misuse

of Article XXI by WTO Members. The United States therefore agreed that in order to

address the overall systemic concerns, it would make an offer for compensatory adjustment”

(Rice Dec 10, 2007). Countries have clearly exercised considerable restraint regarding the

use of the security exception, as there have been very few cases where the exception has been

A4

invoked.55 According to an official with extensive experience in U.S. trade policy, “The bar

is high for the national security exemption. If the bar is too low, the [WTO] system falls

apart. Countries must believe in a stable, predictable environment.”56 Thus, although this

tool is technically available to justify discriminatory policies, countries must carefully weigh

the gains of invoking the exception against the risks posed to the WTO as a whole.

Other exceptions allowing trade restrictions are found in Article XX of the GATT, which

condones deviation from GATT rules in cases involving “the protection of public morals,

the protection of human, animal or plant life or health, measures relating to prison labor,

and the conservation of exhaustible natural resources.” Again, however, these exceptions

place strict limits on policy discrimination. For example, a country can invoke the public

morals clause, which allows it to decline to import goods that conflict with the morals of

its society, such as pornography or gambling. However, it must refuse to import the good

from all WTO members, which can be very costly, due both to the costs borne by domestic

consumers and the anger that such a ban would elicit from trading partners. Similarly, a

country could place restrictions on goods from a country using prison labor, but would have

to first prove that the prison labor had actually been used and then place restrictions on all

other countries using prison labor to produce that good as well. Otherwise, trade protection

would constitute “arbitrary and unjustifiable discrimination between countries where the

same conditions prevail” (Article XX).

Exceptions also exist which permit preferential trade schemes. Under the Enabling

Clause, countries may grant each other extra concessions as part of a preferential trade

55The United States used the exception to defend its import ban on Czechoslovakia during the secondyear of the GATT; Sweden tried to use it to justify a 1975 import quota on footwear but instead revoked thequota; the EEC, Australia and Canada used it to defend a 1982 two-month ban on Argentinian goods afterArgentina’s annexation of the Falkland Islands, although no dispute panel was convened; the U.S. used it tojustify a 1985 trade and transactions ban on Nicaragua, although the panel refused to rule on the dispute;and the U.S. invoked it to justify the 1996 Helms-Burton Act embargo on Cuba, though the parties reachedan agreement and the WTO proceedings were dropped (Guide to GATT Law and Practice (Analytical Index)1995).

56Interview by author. February 22, 2012.

A5

agreement, and may provide programs such as Generalized System of Preference schemes

and EU development agreements, which offer lower tariffs for developing countries (Hafner-

Burton 2009). While these types of trade concessions may be conditioned upon a partner’s

policies, such as its human rights record, they often provide limited leverage over partner

countries since the maximum penalty that a country can exact unilaterally is the repeal of

the preference program. In other words, the WTO sets an upper bound on states’ abilities to

punish partners, as states can withhold preferential treatment, but must grant MFN status

regardless (Jones 2008).57 Because the WTO limits the ability of states to lower their tariffs

below MFN levels, these states often substitute towards other forms of conditionality.

57Additionally, the criteria given for preference program eligibility cannot discriminate between countries.For instance, the EU offered trade privileges to specific developing countries who showed progress in com-bating drug trafficking. This program was disputed by India, a developing country that was not eligiblefor the program, arguing that its exclusion represented unlawful discrimination. The EU lost the case andrepealed the drug trafficking program (WTO 2012). Similarly, the EU undertook a total reevaluation ofits development aid under the Lome agreements, preferential trade concessions offered to 53 EU tradingpartners which could be suspended for violations of human rights, because the program was challenged fordiscriminating between developing countries (Holland 2002). However, this limitation often does constrainstates in practice, since often the criteria for program eligibility can be defined very narrowly such that itonly applies to a specific country.

A6