A partnership with academia

Building knowledge for trade and development

Vi Digital Library - Text Preview

The 'emulator Effect' of the Uruguay Round on United States Regionalism

Case study by Fugazza, Marco, Robert-Nicoud, Frédéric, 2011

Download original document (English)

Using a detailed data set at the tariff line level, we find an emulator effect of multilateralism on subsequent regional trade agreements (RTAs) involving the United States. We exploit the variation in the frequency with which the United States grants immediate duty free access (IDA) to its RTA partners across tariff lines. A key finding is that the United States grants IDA status especially on goods for which it has cut the multilateral most favoured nation (MFN) tariff during the Uruguay Round the most. Thus, the Uruguay Round (multilateral) “concessions” have emulated subsequent (preferential) trade liberalization. We conclude from this that past liberalization may sow the seeds of future liberalization.



UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT







POLICY ISSUES IN INTERNATIONAL TRADE AND COMMODITIES


STUDY SERIES No. 51





THE ‘EMULATOR EFFECT’ OF THE URUGUAY ROUND
ON UNITED STATES REGIONALISM




by



Marco Fugazza
UNCTAD, Geneva



and



Frédéric Robert-Nicoud


University of Geneva; CEP and SERC,
London School of Economics, United Kingdom;


CEPR, London











UNITED NATIONS
New York and Geneva, 2011





ii


Note

The purpose of this series of studies is to analyse policy issues and to stimulate discussions
in the area of international trade and development. The series includes studies by UNCTAD staff
and by distinguished researchers from academia. This paper represents the personal views of the
authors only, and not the views of the UNCTAD secretariat or its member States.

The designations employed and the presentation of the material do not imply the
expression of any opinion whatsoever on the part of the United Nations Secretariat concerning the
legal status of any country, territory, city or area, or of its authorities, or concerning the
delimitation of its frontiers or boundaries.

Material in this publication may be freely quoted or reprinted, but acknowledgement is
requested, together with a reference to the document number. It would be appreciated if a copy of
the publication containing the quotation or reprint could be sent to the UNCTAD secretariat at the
following address:



Marco Fugazza


Trade Analysis Branch
Division on International Trade in Goods and Services, and Commodities


United Nations Conference on Trade and Development
Palais des Nations, CH-1211 Geneva 10, Switzerland


Tel: +41 22 917 5772; Fax: +41 22 917 0044
E-mail: marco.fugazza@unctad.org





Series Editor:
Victor Ognivtsev
Officer-in-Charge


Trade Analysis Branch
DITC/UNCTAD







UNCTAD/ITCD/TAB/52






UNITED NATIONS PUBLICATION


ISSN 1607-8291









© Copyright United Nations 2011
All rights reserved





iii


Abstract



Using a detailed data set at the tariff line level, we find an emulator effect of


multilateralism on subsequent regional trade agreements (RTAs) involving the United States. We
exploit the variation in the frequency with which the United States grants immediate duty free
access (IDA) to its RTA partners across tariff lines. A key finding is that the United States grants
IDA status especially on goods for which it has cut the multilateral most favoured nation (MFN)
tariff during the Uruguay Round the most. Thus, the Uruguay Round (multilateral) “concessions”
have emulated subsequent (preferential) trade liberalization. We conclude from this that past
liberalization may sow the seeds of future liberalization.





Key words: Regionalism, multilateralism, stumbling bloc, Uruguay Round

JEL Classification: F13, F14, F15, N70




 
 








iv






Acknowledgements



We are especially grateful to Alessandro Nicita, Marcelo Olarreaga and Emanuel Ornelas for
detailed feedback and comments as well as to Richard Baldwin, Emily Blanchard, Caroline
Freund, Jaya Krishnakumar, Ugo Panizza, Esteban Rossi-Hansberg and participants in
seminars at the World Trade Organization (WTO), Johns Hopkins, Simon Fraser, Glasgow
University and the Workshop on “The New Political Economy of Trade” at the European
University Institute for useful discussions and suggestions. Part of this paper was written
when Robert-Nicoud was a Peter B. Kenen Visiting Scholar at Princeton University; the
hospitality of its International Economics Section is gratefully acknowledged.



Any mistakes or errors remain the authors’ own.










v


Contents
 




1 Introduction..........................................................................................................................1


2. Related literature .................................................................................................................3

2.1. Are RTAs and MTAs substitutes or complements?.....................................................3
2.2. Are past and current liberalization episodes complements or substitutes? ..................4
2.3. Relations with the empirical literature .........................................................................4


3. Definition of variables, data and summary statistics ........................................................5


4. Estimation strategy and estimation results ........................................................................8

4.1. Evidence at the good level: Logit.................................................................................9
4.2. Evidence at the good-partner level: Logit ..................................................................13


5. Identification of the “Emulator Effect” ...........................................................................16

5.1. Non-tariff measures....................................................................................................16
5.2. Unused rules of origin................................................................................................17
5.3. The role of intermediate goods ..................................................................................18
5.4. IV estimation..............................................................................................................20


6. Sensitivity analysis .............................................................................................................22

6.1. Evidence at the good level: Alternative Logit............................................................22
6.2. Evidence at the good level: Poisson...........................................................................22
6.3. Evidence at the good level: Hurdle ............................................................................23
6.4. Interaction between CUT and MFN...........................................................................23
6.5. Zero Imports...............................................................................................................24
6.6. Implementation period ...............................................................................................24


7. Summary and concluding remarks .................................................................................25


References ....................................................................................................................................26

Appendix tables ...............................................................................................................................29





vi


List of figures


Figure 1. United States Tariffs (Simple Means)..........................................................................2
Figure 2. Tariff lines in RTAs.....................................................................................................7
Figure 3. Estimated Marginal Effects with Quadratic CUTg Term...........................................12





List of tables


Table 1. Descriptive Statistics....................................................................................................6
Table 2. LOGIT “Seven” ........................................................................................................11
Table 3. p-g LOGIT .................................................................................................................14
Table 4. g-Logit on Partner-Specific Sub-sample....................................................................15
Table 5. Non-tariff Measures (NTM) ......................................................................................17
Table 6. Unused Rules of Origin (RoO) ..................................................................................18
Table 7. LOGIT “Seven” by Type of Goods ...........................................................................19
Table 8. IV Estimation.............................................................................................................21


 





1


1. Introduction 


Many preferential trade agreements came to light since the completion in 1994 of the
Uruguay Round of multilateral trade negotiations under the auspices of the General Agreement on
Tariffs and Trade (GATT). The United States is no exception. These agreements involving the
United States vary in scope – the number of goods included in the agreement varies across
agreements – and breadth – the United States tariff on some goods goes to zero immediately upon
implementing the agreement but the imports of many others are fully liberalized only gradually. In
this paper, we shed light on the causes of these cross-good variations and show that they are best
thought of as the continuation of a process that includes multilateral liberalizations. Specifically,
we find that the imports of goods that the United States liberalizes swiftly the most frequently on a
preferential basis are also the goods for which it granted the boldest tariff cuts during the Uruguay
Round and/or those for which the current MFN tariff levels are low. Both findings are robust to a
variety of specifications. The quantitative effects are also quite large. We interpret the former
finding as evidence that past trade agreements are dynamic complements, or emulator, to
consecutive agreements. The latter finding is consistent with the idea that the benefits of new
regional trade agreements are especially large when multilateral tariffs are low.


By design, the paper addresses on an important and little studied question in international
trade: whether multilateral trade agreements (MTAs) drive, in any way, the proliferation of
regional trade agreements (RTAs). Since there is some concern and evidence that RTAs block or
slow down the formation of MTAs (Limão, 2006), it is important to understand whether the
success of one MTA may actually lead to the failure of the following MTA negotiations as a result
of increased regionalism. Several theoretical papers have demonstrated that a multilateral reduction
in tariffs can increase the formation and the self-enforceability of RTAs (Ethier, 1998; Freund,
2000a). Other papers have shown in a dynamic framework that existing trade agreements erode the
resistance to liberalization and hence pave the way for further liberalization in the future (Staiger,
1995; Maggi and Rodriguez-Clare, 2007). These papers are motivated by the fact that in many
countries tariffs are declining over time (figure 1 illustrates this pattern for the United States).1 We
take this feature of the data seriously in our analysis.


While existing empirical papers on the subject focus on the determinants of RTA
formation (e.g. Baier and Bergstrand, 2004; Egger and Larch, 2008; Mansfeld and Reinhardt,
2003), the current paper strives to explain the impact of an MTA on the characteristics of RTAs
that follow it. Our contribution is threefold. We start by examining which products are liberalized
most swiftly in an RTA, taking its existence as given. In particular, as we explain in section 3, we
focus on RTAs signed by the United States after the Uruguay Round in 1994 and, in section 4, we
show that the products that were most likely to be liberalized quickly in an RTA were precisely the
ones that had the largest tariff reduction in the Uruguay Round and, for a given Uruguay Round
tariff reduction, those that had high Uruguay round tariff levels. The former finding supports the
claim that MTAs and RTAs are dynamic complements and it constitutes our first contribution. The
latter finding provides original evidence that is consistent with Freund’s (2000b) theoretical
argument suggesting that RTA liberalization is more likely when MFN tariffs are low.2 This is our
second contribution. Our third contribution is to analyse the effect of MTAs on RTAs using



1 This theory is usually cast in a two-country framework and thus is silent about the multilateral vs.
preferential liberalization issue. It can thus guide the dynamic flavour of our empirical analysis.
2 A lower MFN tariff expands imports from trading partners to which the United States applies the MFN
tariff and reduces its imports from all others. As a result, the new RTA optimal tariff is lower. Since
institutional constraints require the actual RTA tariff to be either 0 (if the good is excluded from the RTA) or
equal to the MFN tariff (if the good is included in the RTA), the empirical counterpart of this theoretical
effect is to increase the likelihood of this good being included in the list of goods that are liberalized swiftly.





2


differences rather than levels, thus avoiding the assumption of stationary tariffs (made previously
in the literature), which is clearly incorrect for United States tariffs during that period (see figure
1).




Figure 1: United States Tariffs (Simple Means)3


2
3


4
5


6
(%


)


1989 1995 2007
year


MFN Applied Tariff Effectively Applied Tariff



Note: At the tariff line level, the effectively applied tariff corresponds to the lowest available tariff.
Whenever it exists, the lowest preferential tariff is the effectively applied tariff. Otherwise it is the
MFN applied tariff.




Section 5 then undertakes to establish the causal effect of MTA on RTAs. This is achieved
by, first, choosing only RTAs signed by the United States whose negotiations were initiated well
after the Uruguay Round of 1994 (a total of seven agreements signed between 2001 and 2006).
Second, we add sector dummies to control for any political economy or other unobserved factors
common to all goods in a sector which can affect the swiftness of liberalization across sectors.
Third, we demonstrate that goods with high trade barriers due to non-tariff measures had no
significant impact of MTA tariff-reduction on the swiftness of RTA liberalization. Fourth, we show
that the emulator effect is weaker for goods with preference margins that do not bind because of
prohibitive rules of origin. Fifth, we classify the goods in the sample according to the different
stages of production in the value chain. Following the logic of the “Protection for sale” framework
(Grossman and Helpman, 1994), downstream sectors oppose tariffs in upstream sectors from which
they source (Gawande, Krisha and Olarreaga, 2009); conversely, upstream sectors support tariffs in
downstream sectors to which they sell. Consistently, we find some evidence that the positive
impact of MTA liberalization on the swiftness of RTA liberalization is weaker for Consumption
and stronger for Equipment and Intermediate goods. Finally, we use hypothetic Uruguay Round



3 In Figure 1, the "effectively applied tariff" series is a simple average of MFN and preferential tariffs across
tariff lines. For institutional reasons specific to the United States, most of the preferential tariffs are zero.





3


tariff cuts to instrument for actual cuts: the overall aim of this Round was to achieve a 30 per cent
cut in tariffs on manufacturing goods. Taken together, our results provide strong evidence that
higher MTA tariff cuts increase the likelihood of immediate RTA liberalization and demonstrate
that the effect is far from linear and apparently stronger for products that were swiftly liberalized in
all seven RTAs considered in the paper versus those that were liberalized swiftly in six or less.


While we are able to demonstrate a clear link between greater MTA tariff reduction and a
higher probability of immediate duty-free trade in more RTAs, we also exploit other RTA
characteristics related to the timing of liberalization which could potentially be just as interesting.
This is done, among other extensions, in section 6. For instance, we find that short MTA
implementation periods are associated with swift RTA liberalizations: this is another manifestation
of the emulator effect.




2.  Related literature 

Our findings speak to two different theoretical arguments put forth in the literature.


2.1.  Are RTAs and MTAs substitutes or complements? 


The first class of models studies the welfare effects of preferential versus multilateral trade
liberalization and, on the positive side, whether liberalizing on a preferential basis first, by
changing the status quo ante, undermines multilateralism. Answering such questions is important,
not least because several scholars fear that regionalism is a dynamic substitute, or stumbling block,
to multilateral free trade and a menace to the multilateral trading system incarnated by the
GATT/WTO (Bhagwati, 1991; Grossman and Helpman, 1995; Levy, 1997; Bagwell and Staiger,
1998; Krishna, 1998; Cadot, De Melo and Olarreaga 1999; McLaren, 2002; Saggi, 2006; Limão,
2007).4 Limão (2006) finds empirical support for the stumbling block hypothesis for the United
States case; Estevadeordal, Freund and Ornelas (2008) find a “building block” effect in a sample of
10 Latin American countries; Freund and Ornelas (2010) provide an excellent review of this
abundant literature.5 We complement it by asking the causality question in the opposite direction,
as Ethier (1998) and Freund (2000a), but from an empirical angle.6


The models in this literature are essentially static: the supply side of the economy is
exogenously given and the only dynamic thought experiment is an application of the agenda-
setting game, a classic in political science. Aghion, Antràs and Helpman (2007) study this
canonical game in a trade liberalization context explicitly. Freund (2000b) emphasizes that the

4 Also, not one month elapses without the economic press worrying about this issue. Editorial lines
predominantly echo the “stumbling block” hypothesis. For economic and political mechanisms consistent
with the “building block” hypothesis, see e.g. Kennan and Riezman (1990), Richardson (1993), Bagwell and
Staiger (1999) and Ornelas (2005a).
5 Limão and Karakaovali (2008) find a stumbling block effect for the EU. Baldwin and Seghezza (2008) find
a negative correlation between MFN tariffs and preference margins in their sample of 23 large countries.
They conclude from this that the stumbling block mechanism, if it exists, is not of first order importance.
6 Ethier (1998) analyses whether a multilateral trade agreement between developed countries promotes
bilateral agreements between a developed country, which is part of the multilateral agreement, and a
developing country, which is not part of the multilateral agreement. However, in the present analysis, all
subsequent partners of the seven bilateral agreements are also members of the World Trade Organization
(WTO), so Either’s analysis is not suited to guide our empirical work.





4


same type of logic also entails that the incentives to form an RTA are shaped by the state of
multilateral tariff levels. In an oligopolistic setting, she finds that the profit-shifting effect of
regionalism, whereby discriminatory trade agreements expand output and profits in the
participating countries at the expense of the countries left out, is especially strong when
multilateral tariffs are low. She concludes from her analysis that “each Round of multilateral tariff
reduction should lead to a new wave of RTAs” (Freund, 2000a: 359). Our results vindicate her
conclusion. In a “Protection for sale” setting, Ornelas (2005a) points out that preferential trade
liberalization erodes the rents from protection, which encourages participating countries to lower
their external tariff. Insofar as this line of reasoning also applies in the opposite direction, our
results are consistent with Ornelas’ theoretical findings. A similar line of analysis asks whether the
conditions under which RTAs are enforceable are affected by the multilateral trading environment
(Freund, 2000b and Ornelas, 2005b). In these models, the static costs and benefits from protection
are time-invariant by construction, so that natural solution to this kind of dynamic problem is a
stationary tariff. In this, these papers are no different from existing theoretical studies on the
complements-vs.-substitutes issue. Yet, if anything, tariffs fall over time and hence this line of
explanation misses an important dimension of the real world.




2.2.  Are past and current liberalization episodes complements 
or substitutes? 


The “Juggernaut theory” of trade liberalization implies that current liberalization, by
eroding protectionist forces and hence resistance to future trade reforms, is sowing the seeds of
future liberalization (Baldwin, 1994; Staiger, 1995; Maggi and Rodrìguez-Clare, 2007; Baldwin
and Robert-Nicoud, 2007). Our regression results provide (to the best of our knowledge: original)
evidence consistent with the Juggernaut theory. A central insight in these papers recognizes that
some sector-specific factors of production like (human) capital depreciate gradually over time; as a
result, the politically optimal tariff is thus also decreasing over time. Freund (2000b) and McLaren
(2002) also combine dynamic aspects of trade liberalization with the regionalism versus
multilateralism issue but their focus (the hysteretic effects of preferential trade barriers) is
different.




2.3.  Relations with the empirical literature  


From an empirical point of view, the main strand of the literature that relates to our
research is on the determinants of RTAs formation. Several papers study the economic
determinants of RTAs. The main identifying assumption remains that RTA-related trade gains are
closely linked to the standard gravity covariates. Baier and Bergstrand (2004) find that the
likelihood of an RTA is larger, the closer the two countries are to each other, the more remote they
are from the rest of the world, the larger their GDPs, and so on. Building on Baier and Bergstrand
(2004), Egger and Larch (2008) find evidence consistent with Baldwin’s (1995) Domino theory of
regionalism, whereby pre-existing RTAs increase the likelihood that two countries participate in a
common RTA. In a separate but no less interesting line of research, Martin, Mayer and Thoenig
(2009) find that multilateralism causes peace-motivated regional trade agreements.7 The macro-
level empirical evidence in these papers complements our micro-level evidence.



7 The logic goes as follows: countries that have fought wars in the distant past tend to sign RTAs as a way of
increasing the opportunity cost of a bilateral war, thereby reducing the probability that possible bilateral





5


Importantly, whereas we take the existence of the Free Trade Agreement as given, and aim
to find out which tariff lines are liberalized the most swiftly, the three aforementioned papers aim
to explain the formation of RTAs.




3.  Definition of variables, data and summary statistics 


In the case of the United States (and others), the legally binding and the applied MFN
tariffs coincide exactly (by definition the latter may not be higher than the former), so we refer to
them as the MFN tariff for short (World Tariff Profiles 2007). All United States MFN tariffs are
non-increasing in the post-Uruguay Round period. Our key explanatory variable is a good-specific
measure of the intensity of multilateral trade liberalization. We denote it by CUTg with the
subscript g referring to good g. CUTg is defined as the (non-negative) difference (or tariff “cut”)
between the Tokyo and Uruguay MFN rates, i.e. Uruguayg


Tokyo
gg MFNMFNCUT −≡ . The stated


aim of the Uruguay Round was to cut tariffs by about 30 per cent for industrial goods and bind the
MFN tariff rate for all agricultural goods but in the end Canada, the EU, Japan, and the United
States achieved a larger reduction on average (Baldwin, 2009).


Our main sources are the UNCTAD Trade Analysis and Information System (TRAINS)
and the WTO Consolidated Tariff Schedules (CTS) Bound Duty Rates databases. Both databases
provide information at the legal tariff line level (8-digit in the Harmonized System (HS)
nomenclature), what we refer to as goods. They do not include goods subjected to non-ad valorem
duties.8 This leaves 9,303 goods. The WTO-CTS database provides information on bound rates
negotiated at both the Tokyo and the Uruguay Rounds. Hence, CUTg corresponds to the effective
reduction in bound tariffs negotiated during the Uruguay Round. The database also provides
information on the implementation period of bound tariff reductions that were negotiated during
the Uruguay Round


In our analysis, we want to understand to what extent past multilateral trade liberalization
is a factor towards current regional trade liberalization. A measure of the intensity of the regional
trade liberalization similar in spirit to CUTg is the preference margin PMg,p, defined as the (non-
negative) difference between the MFN tariff and the preferential tariff, or PMg,p ≡ UruguaygMFN –
PTg,p, where PTg,p is the good- and partner-specific preferential tariff. We exclude tariff lines for
which the Uruguay MFN tariff was already zero, since no preference margin can be granted to such
goods by definition. This leaves 7,419 goods in our reference sample.


The UNCTAD-TRAINS database includes MFN applied rates and preferential rates. The
informed period is 1996–2008. This exhaustive database covers 15 trade agreements, from which
we exclude trade agreements that were negotiated before the end of the Uruguay Round (1994) so
as to eliminate an obvious source of reverse causality bias from our regressions (more on this in the
next section); we also exclude unilateral trade agreements, for the focus of our analysis is on
preferential trade liberalization or RTAs, not on unilateral ones. We are thus left with seven RTAs:
Jordan (2001), Chile (2004), Singapore (2004), Morocco (2006), Bahrain (2006), Australia (2005)

conflicts might escalate into wars. In previous work (Martin et al. 2008), the same authors show that
multilateral trade reduces the opportunity cost of a bilateral war. Taken together, this line of reasoning and
these results imply that an increase of multilateralism raises the probability of bilateral war among old foes
and they then enter bilateral or regional trade deals as an endogenous response to the threat it poses to
bilateral peace.
8 Such tariff lines account for around 8 per cent of the HS-6 subheadings of the World Tariff Profiles (2007).





6


and the Central American-Dominican Republic Trade Agreement (2006).9 In our analysis, an
“observation” is a good-and-partner entry for PTg,p. Our reference sample has 51,814 observations,
which is slightly lower than 7 x 7,419 = 51,933, because not all goods are included in all RTAs.
Table 1 (panel a) breaks down the number of tariff lines included in our reference sample by
partner. Table 1 (panel b) presents the summary statistics of our quantitative variables. For
instance, the sample mean of CUTg is 4,22 percentage points and the sample mean of


Uruguay
gMFN is 6.2 percentage points.





Table 1: Descriptive Statistics


Panel (a) Tariff Lines in Trade Agreements
Tariff Lines Status


Partner
Immediate
duty free


Gradual
duty free


Total
included Excluded



Australia 5,319 1,591 6,910 509
Bahrain 5,306 2,113 7,419 None
Chile 6,651 733 7,384 35
Jordan 4,420 2,557 6,977 442
Morocco 5,397 1,979 7,376 43
Singapore 5,033 1,735 6,768 651
CAFTA 5,394 2,025 7,419 None

Panel (b) Variables

Mean Median


Standard
Deviation Min Max



MFN tariff CUT, in
pp (Tokyo minus
Uruguay)


4.22 2.1 4.34 0 31.5


MFN tariff rate, in
pp (Uruguay) 6.2 4.19 5.02 0.1 48


Share of imports
(total) from RTA
partners


.45 .23 .51 .005 1.31


Share imports
(tariff line)
from RTA partners


.21 0 2.63 0 100


Share imports from
NAFTA partners 13.15 .73 24.09 0 100


Share exports to
RTA partners .91 .44 .89 .04 2.25


Note: All shares are calculated for the year 2000.

9 That is, we exclude the Generalized System of Preferences (1976), Israel (1985), the Caribbean Basin
Economic Recovery Act (1986), the Andean Trade Preference Act (1992), NAFTA (1994), the Generalized
System of Preferences (GSP) for Least Developed Countries (1997), the African Growth and Opportunity
Act (2000, 2001 and 2002), and the Caribbean Basin Trade Partnership Act (2000). See Romalis (2007).





7


It turns out that, in the United States case, each RTA is in fact a free trade agreement
(FTA) de jure, namely, all tariffs on included goods eventually go to zero.10 However, there is
considerable variation in the timing of the implementation of this free trade policy about both
goods and partners: overall, 69 per cent of our observations are fully liberalized at the start of the
implementation of the RTA, whereas goods that are included in any of the RTAs but that are
liberalized only gradually represent 27 per cent of our observations; the rest consists of good-
partner pairs that are excluded from the corresponding RTA altogether (fewer than 4 per cent of
observations).


Figure 2 illustrates various cross-RTA features of the sample. No tariff line has been
included in fewer than four RTAs and the majority of them is part of all agreements (dark bars).
Variation is clearly higher when considering the implementation of duty-free access (light bars).
Many tariff lines (35 per cent) are set to zero on the date of entry into force of each and every trade
agreement. Conversely, 6 per cent of some tariff lines are set to zero only gradually in all trade
agreements. The remaining tariff lines are set to zero gradually in at least one but fewer than seven
trade agreements.



Figure 2: Tariff lines in RTAs




0
2,


00
0


4,
00


0
6,


00
0


8,
00


0
#


of
T


ar
iff


L
in


es


0 1 2 3 4 5 6 7
Frequency


RTA IDA



Note: The RTA histograms refer to the number of tariff lines included in an RTA by frequency; “frequency”
refers to the number of RTAs in which a given tariff line is being included. The IDA histograms refer to
number of tariff lines granted IDA (Immediate Duty-free Access) status (i.e. tariff lines that are liberalized as
an RTA enters into force).



10 In a separate and fascinating line of research, Conconi, Fachini and Zanardi (2008) dig deeper into another
peculiarity of the United States trade policy institutional setting: the fast track authority.





8


We also use the information available in the TRAINS database for non-tariff measures
(NTM). We focus on NTMs classified as Technical Measures in the UNCTAD Coding System of
Trade Control Measures (chapter 8). This covers inter alia both sanitary and phyto-sanitary (SPS)
and technical barriers to trade (TBT) type of measures. Data are available only for the year 1999.
Our control variables include imports at the tariff line; this information is also provided by
UNCTAD-TRAINS. Table 1 (panel b) reports the summary statistics of the share of imports at the
tariff line level that are covered by a preference margin as well as of the other controls.




4.  Estimation strategy and estimation results 


At a very general level, we would ideally like to regress the preference margin on both the
multilateral CUTg (in first differences) and the most-favoured-nation tariff rate MFNg (in levels),
that is, estimate an equation of the form


, 1 2 ,g p g g g pPM CUT MFNα β β ε= + + + . (1)
The null hypothesis is that RTA liberalization is independent of MTA liberalization


( 1 0β = ) and of MFN tariff levels ( 2 0β = ). Our alternative hypothesis on β1, which we dub as the
“emulator effect”, predicts a positive coefficient, whereas 1 0β < would be consistent with a
dynamic version of the competing “money-left-on-the-table hypothesis”.11 Freund (2000a) guides
our alternative hypothesis on β2 (positive) and this competes with a static version of the “money-
left-on-the-table hypothesis” ( 2 0β < ) whereby there is more room to include a tariff line in an
RTA if the MFN rate is relatively high to start with. Let us emphasize that MFNg is orthogonal to
CUTg (the correlation is -.01 in our reference sample) so our empirical analysis is able to
discriminate between competing hypotheses in both levels and first differences. This somewhat
surprising feature of the data is also helpful for our identification strategy and we return to it in
section 5.4.


The problem with a naïve estimation of the intensive margin of the emulator effect in (1)
is that the United States institutional setting is such that a Preferential Trade Agreement is de jure a
Free Trade Agreement. This makes using the intensive margin of preferential trade liberalization as
the dependent variable problematic (at the end of the implementation period PTg,p = 0, hence PMg,p
boils down to UruguaygMFN by definition). For this reason we exploit instead its extensive margin
and the timing of the preferential liberalization. Our first cut through the data is to set goods that
are granted duty free access to the United States market immediately upon implementation of each
of the seven RTAs in the sample apart from other goods. The idea is that these goods turned out to
be the easiest to liberalize on a preferential basis and we want to understand the dimensions that
make such goods special. Inspection of Figure 2 also shows that the most frequent number of times
a good is granted “immediate duty-free access” (IDA) to the United States market is the maximum
(seven). For these reasons, we create a binary variable for each good g, SEVENg, with SEVENg = 1
if good g is granted IDA status in the seven RTAs and 0 otherwise (i.e. if the good is granted only
gradual duty-free access in, or excluded altogether from, at least one RTA); formally,



11 Many commentators believe that the surge of regional trade agreements is the result of the stalling of the
Doha Round: frustrated parties conclude bilateral agreements to substitute for the lack of a multilateral deal.
The weekly The Economist is a major proponent of this thesis.





9


{ }7 ,I # : 0implg g pSEVEN p PT≡ = , where impl denotes the implementation year and I7{.} denotes
an indicator function that takes value 1 if its component is equal to seven and 0 otherwise.12 We
also create two additional measures along those lines, the binary variable ONEg that takes value 1 if
good g gets IDA status in at least one RTA and 0 otherwise and the count variable NTLg that counts
the number of RTAs in which good g gets IDA; these being mostly robustness checks, we
postpone the regression results for ONEg and NTLg to Section 6.


As our second measure of the extensive margin of preferential trade liberalization, we
define a good- and partner-specific measure of preferential trade liberalization for our central
specification that takes value 1 if imports of good g from partner p are granted the IDA status upon
implementation of the RTA in question and zero otherwise.


 


4.1.  Evidence at the good level: Logit 


We start by running the following logit:


( )( ) 1 2Pr{ 1} Uruguayg G g g gSEVEN f CUT MFNβ β= = Λ + + + g,pX β , (2)

where [ ]( ) exp( ) / 1 exp( )Λ ⋅ ≡ ⋅ + ⋅ is the logistic cumulative distribution function, fG(p) is a sector
dummy, CUTg is the reduction in the MFN tariff negotiated over the course of the Uruguay Round
(in percentage points), UruguaygMFN is the ad-valorem Uruguay MFN tariff rate (in percentage
points) and Xg,p is a set of additional controls; β1 is our coefficient of interest. Denote the set of all
goods by { }1,..., gNΓ = ; then G is a partition of Γ and we use ( )G g to denote the HS-2 sector in
which good g is classified. Thus, G is also a mapping : good sectorG → .


Though we view (2) as a reduced form relationship between SEVENg and CUTg, we must
assume that gCUT is exogenous in order to obtain consistent and unbiased estimates of the
coefficients. Our strategy to rid ourselves of the reverse causation bias rests partly on the timing of
events. We limit our sample to the seven RTAs that entered into force after the conclusion of the
Uruguay Round in 1994. This sample selection is expected to eliminate any reverse-causality bias
for two main reasons: first, no new multilateral trade agreements had been implemented by the
United States between 1994 and 2000. This buffer is likely to be long enough to ensure that these
trade agreements to come did not influence the Uruguay Round trade negotiators. The second
reason reinforces this point: no trade agreement signed in the post-Uruguay Round period had
actually been negotiated during the pre-Uruguay Round period. The Clinton administration did
undertake talks to form a Free Trade Area of the Americas (FTAA) and to sign a trade agreement
with the Asian Pacific Economic Cooperation (APEC) country members in 1994. However, no
agreement has yet been reached in the context of FTAA negotiations. In addition, the APEC forum
held in Bogor in 1994 signed a declaration to work toward free trade in the region by 2010 for



12 A comment about goods-partner pairs that do net get the IDA status is in order here. Goods g that are
included in the RTA p but that are liberalized only gradually and goods that are excluded from that RTA
altogether are both coded the same way. This is because the frequency of the latter in the data is very low
(less than 5 per cent of good-partner pairs). Our qualitative results do not change if we drop these
observations from the sample.





10


developed countries and by 2020 for all member-countries. A 16-year time frame makes any
influence of those talks on tariff cuts defined the Uruguay Round quite implausible.13 Note that the
absence of correlation between CUTg and UruguaygMFN is also helpful: it implies that the past
determinants of trade liberalization (at the good level) that cumulated to give rise to the Tokyo
tariff level are different from those that led to the Uruguay Round tariff cut: in line with the
Juggernaut hypothesis, this suggests that the sectoral determinants of tariffs are not as long-lived as
one might think. However, if an omitted variable affects SEVENg and CUTg simultaneously, then
regressing the former on the later will cause a spurious correlation. We thus introduce sector
dummies ( )G gf in (2) to capture sector invariant sources of unobserved heterogeneity such as the
political economy determinants of tariffs or the sources of comparative advantage.14 Insofar as such
unobserved shocks are common to goods within sectors, then including ( )G gf in (2) corrects for
this source of omitted variable bias in our cross section exercise.15 Together, these three working
assumption constitute our maintained identification hypothesis. We complement them with
additional approaches in Section 5.


We use sector fixed effects at a relatively high degree of aggregation so that our sample
has a large number of observations for each partner p and for each sector G; as a result, the β’s in
the conditional logit in (3) are consistently estimated.


Table 2 presents the results. We report odds ratios throughout (standard errors clustered at
the tariff line in parenthesis). The odds ratio associated to βj is defined as exp βj (j = 1,2,...) and has
the meaning that a one extra percentage point in CUTg raises the probability of granting IDA status
to all partners for the good in question by a factor exp βj relative to not including the tariff line or
delaying setting this preferential tariff to zero. The two independent variables of interest, CUTg and


Uruguay
gMFN , are significant beyond the one per cent level in all specifications and the results are


stable across specifications. The regression in Column (1) includes the two independent variables
and Column (2) adds sector dummies. The findings are consistent with the emulator hypothesis: the
odds ratio implies that one extra percentage point of CUTg raises the probability that good g gets
IDA treatment for all of the United States’ RTA partners by almost a fourth (1.227 – 1 = .227)
relative to getting it only for a subset of those. By contrast, the “money-left-on-the-table
hypothesis” is rejected by the data: raising UruguaygMFN by one percentage point decreases the odds
that good g gets IDA status by a third (1 – .657 = .343). This result is thus empirical evidence in
favour of Freund (2002a).



13 What is usually recognized is that the APEC summit together with NAFTA helped “squeeze the European
Union to complete the Uruguay round of GATT” in the words of Robert Zoellick’s (2001) statement as
United States Trade Representative.
14 The chosen level of disaggregation corresponds to standard practice in the literature (Limão, 2006).
However, we also check the robustness of our benchmark specifications using HS-4 sectoral dummies.
Coefficients vary only slightly indicating that HS-2 sectoral level dummies are sufficient to capture major
sources of unobserved heterogeneity. We opted for the HS2- level dummies in order to gain computational
flexibility in implementing our sensitivity analysis presented in Section 6.
15 See also Broda, Limão and Weinstein (2008) on this.





11


Table 2: LOGIT “Seven”


Dependant variable: SEVEN
(Probability that tariff line g is granted IDA to United States market


to all 7 partners)

(1) (2) (3) (4) (5)



Tariff CUT 1.140a 1.227a 1.330a 1.331a 1.313a


(Tokyo minus Uruguay) (0.00826) (0.0109) (0.0158) (0.0158) (0.0159)


MFN 0.668a 0.657a 0.612a 0.612a 0.611a
tariff rate (0.0127) (0.0165) (0.0174) (0.0175) (0.0175)



DIFF0 (no Uruguay 4.375a 4.378a 4.253a


Round cut) (0.459) (0.459) (0.446)


Share imports 1.019 1.010
from RTA partners (0.0351) (0.0341)



Share imports 0.992a


from NAFTA partners (0.00162)


Sector FE No Yes Yes Yes Yes
Observations 7419 6822 6822 6822 6822
Pseudo R2 0.209 0.294 0.318 0.318 0.321
LL -3815.2 -3206.3 -3099.7 -3099.5 -3085.6



Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors in parentheses.
a p < 0.01, b p < 0.05.




In Column (3), we add a good-specific dummy DIFF0g that takes value DIFF0g = 1 if the
United States did not liberalize good g during the Uruguay Round (i.e. if CUTg = 0) and zero
otherwise.16 That is, we estimate


{ } ( )( ) 1 2 3Pr 1 0β β β= = Λ + + +Uruguayg G g g g gSEVEN f CUT MFN DIFF .
The fact that goods that were not liberalized during the Uruguay Round – because these


sectors are better organized and successfully fought to be left out of the Uruguay Round entirely,
say – might be quite different from other goods motivates this specification. The coefficient β3 is
positive at the 1 per cent level, implying that goods that were not liberalized at the multilateral
level were more likely to be liberalized at the preferential level: on its own, this result is consistent
with a dynamic version of the “money-left-on-the-table hypothesis”. Adding this control also raises
the odds ratio of CUTg to 1.33. Overall, the effect of CUTg on IDA treatment thus seems to be non
monotonic: the United States grants IDA status more frequently for goods for which the Uruguay
Round tariff cut was zero as well as for those that had a large CUTg


In order to quantify this non-monotonic effect, we replace DIFF0 by a quadratic term to
(2) and we compute the marginal effect of CUTg for all observations/goods. Specifically, we first
run



16 This is verified for 21.8 per cent of the tariff lines in our reference sample.





12


{ } ( )2( ) 0 1 2Pr 1 β β β= = Λ + + + Uruguayg G g g g gSEVEN f CUT CUT MFN
and we obtain that the odd ratios associated with β0, β1 and β2 are 0.997 (0.007), 1.144 (0.012) and
0.931 (0.004), respectively (t-statistics in parenthesis), which confirms the non-monotonicity
uncovered in the previous specification. In order to quantify this non-monotonic effect, we
compute the marginal effect as


{ } ( )' 0 1Pr 1 2β β∂ ⎡ ⎤= = Λ ⋅ +⎣ ⎦∂ g g gg SEVEN CUTCUT ,
where ( )'Λ ⋅g is the density of the logistic distribution Λ(.) evaluated at the explanatory variables
pertaining to observation g. Figure 3 plots the estimated values of the marginal effect as well as the
95 per cent confidence interval against CUTg. As is obvious from the figure, the dynamic version
of the “money-left-on-the-table hypothesis” is rejected in 99.99 per cent of cases (i.e. for all but
seven observations out of 51,814). By contrast, the data are consistent with the emulator effect (i.e.
the net effect is statistically significantly positive) in 99.41 per cent of cases. We conclude from
this that the data provide strong support for the emulator hypothesis and reject the “money-left-on-
the-table hypothesis”. To sum up, the average, median and net effects are all consistent with the
emulator hypothesis.





Figure 3: Estimated Marginal Effects with Quadratic CUTg Term



Note: The net marginal effect of the CUT variable is significantly positive for 99.4% of observations. It is
not significantly different from zero for 0.7 per cent of the observations.




<1% of observations


>99% of obs.


-.0
6


-.0
4


-.0
2


0
.0


2
.0


4
M


ar
gi


na
l E


ffe
ct


0 5 10 15 20 25 30 35 40 45 48
MFN CUT (in percent)


Estimated Low95
Up95





13


The results reported in Columns (4) and (5) show that these qualitative findings are robust
to the inclusion of several controls. Column (4) introduces the import share of all seven partners in
United States total imports of good g, defined as , /g g p gpSM M M≡∑ (where M denotes the
value of imports observed in the year 2000), to control for the possibility that the United States
might be granting IDA access to prominent exporters more easily. The estimated coefficient in Col.
(4) is statistically insignificant: thus, the United States does not seem to discriminate between large
and small exporters when granting IDA status.17


Column (5) adds SNAFTAg to the set of controls, with SNAFTAg being defined as the good-
specific import share of NAFTA products in year 2000, i.e. , /g g NAFTA gSNAFTA M M≡ . Its
coefficient is statistically negative at the one-percent level (its odds ratio is lower than unity),
implying that the United States is less likely to grant IDA status from markets that NAFTA already
penetrates widely. This suggests that NAFTA and ensuing RTAs are substitutes, that is, NAFTA
worked as a ‘stumbling block’ to post-Uruguay Round regionalism. Col. (5) forms our baseline
specification henceforth.




4.2.  Evidence at the good‐partner level: Logit 


The evidence so far indicates that CUTg and UruguaygMFN influence the extensive margin
of preferential trade liberalization. The evidence portrayed is at the good level. However, we can
address a more demanding question to the data: given some other good characteristics (observable
or not), how do CUTg and UruguaygMFN influence the likelihood that the United States grants IDA
status to partner p’s exports of good g to the United States? For this purpose, we create a good-
partner indicator variable { }, ,I 0implg p g pIDA PT≡ = that takes value 1 if partner p gets immediate
duty-free access to the United States market for good g and zero otherwise. We then estimate the
following logit:


{ } ( ), ( ) 1 2Pr 1 ,Uruguayg p p G g g gIDA f f CUT MFNβ β= = Λ + + + + g,pX β (3)

where fp is a partner dummy and the other right-hand side variables are as in (2).18 Running (3) is
similar to running (2) at the good-partner level. The implicit assumption in (3) is that the functional
form that maps the right-hand-side variables into IDAg,p is symmetric for each partner. As we shall
see, though, the effect of CUTg on IDAg,p is non-linear. For this reason, we consider running (3) as
a conservative robustness check that provides a lower bound for the emulator effect.


With this caveat in mind, turn to Table 3, which reports the results (standard errors
clustered at the tariff line in parenthesis). The qualitative results are in line with those of Table 2.
The coefficients for CUTg , UruguaygMFN , DIFF0g and SNAFTAg are still precisely estimated and
they have the expected sign.



17 We use a qualitative variable to discriminate between goods with zero imports and those with positive
imports in Section 6.5 below.
18 Preferential trade agreements can be motivated by non trade objectives as argued in Limão (2007). The
inclusion of partner dummies in specification (3) absorbs any effect possibly related to such non trade
objectives. Also, our set up is cross-sectional while trade agreements were signed in different years; the
inclusion of partner dummies also absorbs factors that may cause a specific sequence in bilateral trade talks.





14


Table 3: p-g LOGIT



Dependant variable: Pr{IDA = 1}


(Probability that tariff line g is granted IDA to United States market
to partner p)


(1) (2) (3) (4) (5) (6)



Tariff CUT 1.064a 1.099a 1.125a 1.126a 1.115a 1.115a


(To. minus Ur.) (0.0162) (0.0197) (0.0221) (0.0220) (0.0212) (0.0213)


MFN tariff 0.922a 0.931a 0.926a 0.925a 0.930a 0.930a
level (0.0119) (0.0125) (0.0139) (0.0136) (0.0134) (0.0134)



DIFF0 (no Uruguay 1.683a 1.688a 1.623a 1.623a


Round cut) (0.316) (0.316) (0.296) (0.298)


Partner’s 1.039a 1.039a 1.041a
share of Mg (0.0144) (0.0152) (0.0128)



Share imports 0.996a 0.996a


from NAFTA partners (0.00103) (0.00103)


SALL: Partner’s 0.951
share of United States X+M (0.160)



Sector FE No Yes Yes Yes Yes Yes
Partner FE No Yes Yes Yes Yes No


Observations 51814 51814 51814 51814 51814 51814
Pseudo R2 0.044 0.115 0.119 0.120 0.085 0.086
LL -29248.8 -27064.3 -26942.2 -26909.6 -28003.2 -27973.3



Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors (clustered by tariff line) in


parentheses. a p < 0.01, b p < 0.05.


Running (3) enables us to control explicitly for partner and good-partner characteristics.
Thus, let , , /g p g p gSM M M≡ define the share of good-g imports that are sourced in country p.
What are our priors on the sign of its coefficient? In Grossman and Helpman’s (1994) ‘protection
for sale’ (PFS) framework, keeping the elasticity of imports and the domestic production constant
(both vary across goods but are constant across partners), protection decreases in the volume of
imports (which does vary across partners) in organized sectors. In non-organized sectors, the
opposite is true. Estimation of


{ } ( ), ( ) 1 2 3 4 ,Pr 1 0Uruguayg p p G g g g g g pIDA f f CUT MFN DIFF SMβ β β β= = Λ + + + + +


includes neither domestic production nor import elasticities. The former omission is harmless: for
each good, there are several import sources (the partners) and possibly a different preferential tariff
for each of them; this enables us to estimate β4 via the cross-sectional variation of SMg,p along the
p-dimension. The latter, however, introduces measurement error in the estimation of β4. Also, the
left-hand side of the structural PFS model is different from the LHS of (3). With these caveats in





15


mind, the estimated coefficient in column (5) of Table 3 is statistically positive at the one-percent
level. This is consistent with the PFS qualitative prediction for organized sectors. This finding is
important for the interpretation of the emulator effect as evidence of the juggernaut mechanism.
The estimated odds ratio corresponding to β4 is equal to 1.04, which implies that an increase in the
import penetration ratio of the pair (g, p) by 1 per cent increases the odds of the United States
granting IDA status to p’s exports of good g by 4 percentage points. In other words, the United
States grants IDA status disproportionately to important import sources. The estimated coefficient
is stable across specifications.


We might also expect the United States to grant tariff-free access to important trading
partners as part of broader foreign and trade policy objectives. To check whether this intuition is
verified in the data, we introduce the Partner’s share of imports across all tariff lines as a an
additional control in (3), namely , /p g pgSMALL M M≡∑ , as well as the United States’ share of
exports towards p, defined as , /p g pgSXALL X X≡∑ , where X denotes exports observed in the
year 2000. In the same spirit, we also create pSALL as , ,( ) / ( )p g p g pgSALL M X M X≡ + +∑ as
an overall measure of the importance of p as a trading partner for the United States. SALLp,
SMALLp and SXALLp are defined at the partner level, so we drop the partner dummy in these
regressions. Column (6) reports the results for SALLp (the results for SMALLp and SXALLp are
similar and so we omit them). The estimated coefficient is statistically indistinguishable from zero,
rejecting the hypothesis that the United States grants free access to its markets disproportionately
to large partners.




Table 4: g-Logit on Partner-Specific Sub-sample


Dependant variable: Pr{IDA = 1}
(Probability that tariff line g is granted IDA to United States market


to partner p)
(AUS) (BHR) (CHL) (JOR) (MAR) (SGP) (CAFTA)



Tariff CUT 1.075b 1.261a 1.120a 1.197a 1.090b 1.175a 1.273a


(To. minus Ur.) (0.0313) (0.0411) (0.0448) (0.0318) (0.0369) (0.0309) (0.0449)


MFN 0.815a 0.956a 0.895a 0.687a 0.878a 0.640a 0.968
tariff rate (0.0342) (0.0142) (0.0277) (0.0418) (0.0282) (0.0720) (0.0207)



2.110a 2.440a 1.862 2.902a 3.097a 2.389b 2.410a DIFF0 (no Uruguay


Round cut) (0.577) (0.715) (0.710) (0.997) (1.099) (0.817) (0.715)


1.017 38.49 0.971b 1.083 1.057 0.998 1.019b Share imports
from RTA partners (0.0176) (115.5) (0.0112) (0.151) (0.0351) (0.00926) (0.00970)



0.995b 0.995 0.997 0.992a 0.997 0.995b 0.996 Share imports from


NAFTA partners (0.00210) (0.00242) (0.00306) (0.00222) (0.00220) (0.00211) (0.00256)


Sector FE Yes Yes Yes Yes Yes Yes Yes
Observations 6929 7287 6420 7332 6474 6771 7246
Pseudo R2 0.463 0.180 0.207 0.343 0.453 0.341 0.184
LL -2278.5 -3589.8 -1845.3 -3254.6 -2006.0 -2889.9 -3494.1



Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors in parentheses.
a p < 0.01, b p < 0.05.





16


Finally, we re-run (3) for each partner separately (more precisely, the specification
corresponding to Table 3, Col. 5). Table 4 reports the results. The coefficients of CUTg and


Uruguay
gMFN have the expected signs. The emulator effect is economically and statistically weakest


for Australia and Morocco and especially large for CAFTA. The “money-left-on-the-table
hypothesis” is rejected in all cases, albeit only in a weak sense in the case of CAFTA.19





5.  Identification of the “Emulator Effect” 


The “emulator” effect seems to be a robust feature of the data, unlike the “money-left-on-
the-table” argument. We have so far relied mostly on the timing of events to identify the effect. In
this section, we use the interaction between our variable of interest (CUTg) and non-tariff measures
(Section 5.1), the rules of origin (Section 5.2) or the type of goods (Section 5.3) to interpret the
positive correlation between CUTg and IDA in a causal way. Finally, we instrument for CUTg
(Section 5.4).




5.1.  Non‐tariff measures 


We start by controlling for the presence of non-tariff measures, or “NTM”, at the tariff
line.20 The idea is that the presence of such non-tariff measures should weaken the effect of CUTg
on preferential liberalization: a multilaterally agreed tariff cut is less effective if the imports of that
good are impeded by other measures. We thus expect the CUTg coefficient to be larger for NTM-
free goods than for goods with some NTM. To test this idea, we create a dummy variable NTMg
that takes value one if the tariff line g has some NTM and zero if good g is NTM-free.


We first re-run (2), adding the NTMg dummy and its interaction with CUTg. Table 5, Col.
(2) reports the results; these have to be compared with Col. (1), which reports the odds ratios of our
baseline specification (Table 2, Col. 5). As expected, the CUTg coefficient for NTM-free goods is
(much) larger than for NTM goods; the difference is significant at any conventional level. The
coefficient for CUTg in goods with non-tariff measures is insignificant (the odds ratio is one). This
finding is exactly what we should expect if multilateral and preferential tariff cuts are dynamic
complements and if the presence of NTMs prevents the emulator effect from playing its role. We
repeat this exercise for the good-partner specification (3) and the results, reported in Table 5, Col.
(4), do not affect these conclusions.21 These findings thus vindicate our emulator hypothesis
further. By contrast, the odds ratio of MFN is reduced in this specification, weakening further the
“money-left-on-the-table hypothesis”.








19 We also run (3) on the full sample and with country specific CUT coefficients to be estimated. Results
indicate that they are pairwise statistically different. The larger estimate of the odds ratio is obtained for the
CAFTA agreement (1.20), followed by Bahrain (1.18), Jordan (1.14), Australia (1.11), Morocco (1.09),
Chile (1.08) and Singapore (1.06).
20 There are 19 per cent of tariff lines with an NTM in our reference sample.
21 Table 5, Col. (3) reproduces Table 3, Col. (5) to ease comparison.





17


Table 5: Non-tariff Measures (NTM)


Dependant variables:
SEVEN Pr{IDA = 1}
(1) (2) (3) (4)


Tariff CUT 1.313a 1.115a
(To. minus Ur.) (0.0159) (0.0212)



1.010 0.993 NTM * cutMFN


(0.0375) (0.00689)


1.310a 1.140a (1-NTM) * cutMFN
(0.0155) (0.00455)



MFN 0.611a 0.603a 0.930a 0.924a


tariff rate (0.0175) (0.0173) (0.0134) (0.00261)


4.253a 4.173a 1.623a 1.700a DIFF0 (no Uruguay
Round cut) (0.446) (0.431) (0.296) (0.0583)



NTM dummy No Yes No Yes



PartnerFE N.A. N.A. Yes Yes



Observations 7419 7419 51814 51814
Pseudo R2 0.327 0.329 0.124 0.129
LL -3056.2 -3046.0 -26810.9 -26652.3



Notes: Coefficients: Exponentiated; Robust standard errors in parentheses. a p < 0.01, b p < 0.05. All


regressions include sector dummies and the controls Share imports from RTA partners and Share
imports from NAFTA partners. The dummy NTM takes value one whenever a NTM is applied at
the tariff line. NTM*CUT represents the interaction between the NTM dummy and the variable
Tariff CUT.




5.2.  Unused rules of origin 


It is well known that the compliance costs of rules of origin (RoO henceforth) can be
prohibitive (Krishna, 2006). Specifically, when the preference margin is low, foreign exporters
might not bother with complying with rules of origin. In our setting, the preference margin is the
MFN tariff rate. If the emulator effect is the manifestation of an actual economic mechanism
whereby trade agreements are dynamic complements, then we expect the coefficient of CUTg to be
higher for the goods where the rules of origin are actually exploited by foreign exporters.
Preference margins are irrelevant when below 2 to 3 percentage points (Estevadeordal et al., 2008).
To identify this differential effect in the data, we construct a dummy variable RoOg that takes value
1 if MFNg > 2.5 (when foreign exporters are expected to use the preference and thus to comply
with the rules of origin) and zero otherwise and we re-run (2) and (3) with this dummy as an
additional control variable. We expect the CUTg coefficient to be larger for RoO-goods than for
goods that have irrelevant rules of origin.






18


Table 6: Unused Rules of Origin (RoO)


Dependant variables:
SEVEN Pr{IDA = 1}
(1) (2) (3) (4)


Tariff CUT 1.321a 1.120a
(Tokyo minus Uruguay) (0.0165) (0.00411)



RoO * CUT 1.374a 1.169a


(0.0181) (0.0107)


(1-RoO) * CUT 1.309a 1.113a
(0.0328) (0.00425)


MFN 0.551a 0.553a 0.927a 0.928a
tariff rate (0.0216) (0.0228) (0.00270) (0.00269)



DIFF0 (no Uruguay 1.636a


Round cut)
4.358a
(0.453)


4.239a
(0.439)


1.666a
(0.0580) (0.0571)



RoO dummy Yes Yes Yes Yes



PartnerFE N.A. N.A. Yes Yes


Observations 6822 6822 51814 51814
Pseudo R2 0.329 0.329 0.121 0.122
LL -3049.1 -3046.0 -26876.9 -26861.0


Notes: Coefficients: Exponentiated; Robust standard errors in parentheses. a p < 0.01, b p < 0.05. All
regressions include sector dummies and the controls Share imports from RTA partners and Share imports
from NAFTA partners. The dummy RoO takes value 1 when MFN values are above or equal to the 2.5%
threshold and zero otherwise. RoO*CUT represents the interaction between the RoO dummy. and the
variable Tariff CUT.




Table 6, Col. (2) reports the results for (2), which have to be compared with those of the
baseline specification, reproduced in Col. (1). The results are supportive of the emulator
hypothesis: as expected, the CUTg coefficient is larger for the goods for which it matters than for
goods with an irrelevant preference margin. By contrast, the coefficient and the odds ratio for


Uruguay
gMFN shrink noticeably, rejecting the “money-left-on-the-table hypothesis” further.


Table 6, Col. (4) reports the results for (3), which have to be compared with those of Col.
(3). Here, the results are as again supportive; the Wald statistics rejects the hypothesis that the
coefficients are the same at the one-percent level. We have also re-ran (2) and (3) with 2 and 3
percentage points as thresholds (results not reported); the qualitative results were not affected.


In sum, the differential effect of CUTg on granting IDA status for goods affected by rules
of origin or non-tariff measures that we find in the data confirms this set of predictions of the
emulator hypothesis.





5.3.  The role of intermediate goods 


As we shall see in Section 6, the emulator effect is non-linear. Specifically, the largest
emulator effect is between granting this preferential access to all partners or not, rather than
between some partners or none. This in turn suggests that the type of goods might be more





19


important than the partners’ characteristics; also, when we include sector dummies in our
regressions, the coefficients of interest tend to rise in a significant way, suggesting that unobserved
sector-invariant characteristics are indeed important. Therefore, we split the sample among the
following categories of goods that correspond to different stages of production in the value chain:
Basic manufacturing, Consumption goods, Equipment goods, Intermediate goods, Mixed products
and Primary goods and we estimate one β1 for each category in our baseline regression (with


Uruguay
gMFN and DIFF0g as controls).


22 Table 7 reports the results. The estimated coefficients are
positive and significant at the one-percent level in all cases but for consumption and primary
goods, for which it is insignificant. It is particularly strong for equipment and intermediate goods
and weakest for consumption and primary goods.




Table 7: LOGIT “Seven” by Type of Goods


Dependant variable: SEVEN
(Probability that tariff line g is granted IDA to United States market


to all 7 partners)


Basic- manufacturing
Consumption-


goods
Equipment-


goods
Intermediate-


goods
Mixed-


products Primary


Tariff CUT 1.423a 1.181a 1.306a 1.343a 1.404a 1.061
(To. minus Ur.) (0.0433) (0.0572) (0.0426) (0.0404) (0.0613) (0.102)



MFN 0.561a 0.494a 0.838a 0.445a 0.808a 0.201a


tariff rate (0.0301) (0.0407) (0.0368) (0.0344) (0.0353) (0.0632)


DIFF0 (no Uruguay 18.62a 1.675 3.080a 2.493a 5.951a 2.53e-09
Round cut) (5.180) (0.529) (0.667) (0.711) (1.785) (0.00031)



Share imports 1.018 1.085 1.366 0.676 0.679b 1.257b


from RTA partners (0.0716) (0.103) (0.260) (0.229) (0.125) (0.121)


0.996 0.986a 0.976a 0.994 0.995 0.990 Share imports
from NAFTA partners (0.00352) (0.00468) (0.00519) (0.00434) (0.00385) (0.0134)



Sector FE Yes Yes Yes Yes Yes Yes


Observations 1598 1031 859 1029 691 132
Pseudo R2 0.313 0.480 0.226 0.361 0.222 0.669
LL -726.4 -335.9 -457.6 -437.3 -335.2 -28.68


Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors in parentheses. a p < 0.01, b p < 0.05.




These results are less helpful in the quest of identifying the emulator hypothesis. To see
why, recall that in our interpretation of the dynamic complementarity between trade agreements,
past trade liberalization in a given sector undermines its current resistance to trade openness
because trade liberalization decreases the (quasi) rents associated with the (quasi) fixed factors that



22 The most represented categories of goods are Equipment, Consumption and Intermediate goods categories.
The goods that are liberalized the most systematically belong to the Equipment goods followed by Basic and
Primary goods. The largest average MFN cut is obtained for Primary goods followed by Basic and
Equipment goods. When considering only the set of goods included in all seven RTAs (SEVENg=1),
Equipment, Intermediate and Basic goods categories show the largest average MFN cuts. Source: United
Nations statistical division (2007).





20


fight for protection. Over time, these factors depreciate and with them the resistance to trade
liberalization. By the same logic, downstream sectors oppose tariffs in upstream sectors from
which they source, and this opposition increases as downstream tariffs fall; also, upstream sectors
that sell domestically have an interest in keeping downstream tariffs high (Gawande, Krishna and
Olarreaga, 2009). As a result, we expect the emulator effect to be strongest in upstream sectors, i.e.
for primary products, intermediate goods and capital goods (“equipment goods”), and weakest in
downstream sectors (“consumption goods”).23 With the noticeable exception of primary products,
the results in Table 7 are in line with those priors.




5.4.  IV estimation 


Finally, we use some exogenous sources of variation for our key right-hand side variable,
CUTg, to identify the causal effect of CUTg on the dependent variable (SEVENg or Pr{IDAg,p = 1},
depending on the specification). We have initially experimented with three instruments: the
corresponding EU’s MFN tariff cut, the share of EU and Japanese exports in United States imports
at the tariff line and a theoretical MFN cut. Standard redundancy tests led to the exclusion of the
first two instruments. Their exclusion is likely to lead to more reliable estimation (Hahn and
Hausman, 2002).


In constructing our theoretical MFN tariff cut, we exploit the objective of the Uruguay
Round which was to obtain an overall reduction target of thirty per cent in non-agricultural
products (Baldwin 2009). We thus construct a hypothetical CUTg variable, denoted by HCUTg,
where the base tariff rate of each industrial product is reduced by thirty per cent (thus HCUTg ≡ 0.7
x Tokyo MFN).24 HCUTg is a valid instrument for CUTg insofar as it is correlated with CUTg and
does not influence SEVENg or Pr{IDAg,p=1} directly: the correlation between the actual and the
hypothetical CUTg variables is equal to 0.67; the latter hypothesis is warranted because this 30 per
cent reduction was across the board, i.e. it was meant to affect all manufacturing goods and all
countries. We then run (2) and (3) by TSLS (two-stage least squares) and by two-step IV Probit
with the actual CUT being instrumented by its hypothetical value. The second-stage results are
reported in Table 8 in the (IV) columns. Note that the TSLS and IV-Probit coefficients are not
readily comparable to those of the logit regressions so far. Therefore, we also run (2) and (3) by
OLS and Probit as a benchmark and the results are reported in Table 8, in the (Non Instrumented)
columns.25 Two results are noteworthy. First, the coefficients of the variable of interest, CUTg, are
positive, which is in line with our findings so far in all specifications. Second, except for the IV
Probit estimation results obtained in the good-specific case and reported in panel (a), the IV and
Non-IV coefficients are quantitatively similar, suggesting that the endogeneity bias is not severe in
the first place. More importantly, we conclude that the effect of bold Uruguay Round tariff cuts on
the likelihood of posterior (preferential) trade liberalization is causal.





23 “Basic manufacturing” is a mixed-bag category: it includes beverages, spirits and vinegar as well as iron,
steel and other base metals inter alia.
24 Note that the correlation between HCUT and MFN is almost zero. That is to say, and perhaps surprisingly,
the Tokyo and Uruguay MFN rates are almost uncorrelated.
25 The estimated coefficients could be compared to those obtained by Logit estimation by using a simple rule
of thumb as suggested by Cameron and Trivedi (2005), that is, OLSLogit ββ ˆ4ˆ ≅ and obitLogit Prˆ6.1ˆ ββ ≅ .
Applying that rule of thumb to our estimates we find that the correspondence is small enough to relate the IV
findings to the Logit ones.





21


Table 8: IV Estimation


Panel (a): Good Specific



Dependant variable: SEVEN


(Probability that tariff line g is granted IDA to United States market
to all 7 partners)


Linear Probability Probit
(IV) (Not Instrumented) (IV) (Not Instrumented)


Tariff CUT 0.0307a 0.0362a 0.0379a 0.159a
(Tokyo minus Uruguay) (0.00202) (0.00147) (0.010) (0.00695)



MFN -0.0162a -0.0168a -0.2184a -0.253a


Tariff rate (0.00160) (0.00147) (0.0105) (0.0147)


DIFF0 (no Uruguay 0.0982a 0.129a 0.0420 0.825a
Round cut) (0.0181) (0.0158) (0.0844) (0.0622)


Observations 6822 6822 6822 6822
R2 0.299 0.301 - -
LL - - - -26952


Notes: Robust Standard errors in parentheses. a p < 0.01, b p < 0.05.
All regressions include sector dummies and the controls ‘Share imports from RTA partners’ and “Share
imports from NAFTA partners”.


Panel (b): Country-Good Specific



Dependant variable: Pr{IDA = 1}


(Probability that tariff line g is granted IDA to United States market
to partner p)


Linear Probability Probit
(IV) (Not Instrumented) (IV) (Not Instrumented)


Tariff CUT 0.0185a 0.0204a 0.0683a 0.0651a
(Tokyo minus Uruguay) (0.00122) (0.000926) (0.0026) (0.00324)



MFN -0.0125a -0.0127a -0.0484a -0.0439a


Tariff rate (0.000957) (0.000894) (0.0021) (0.00229)


DIFF0 (no Uruguay 0.0845a 0.0952a 0.320a 0.283a
Round cut) (0.0109) (0.00949) (0.0345) (0.0298)


Observations 51814 51814 51814 51814
R2 0.097 0.097 - -
LL - - - -3132


Notes: Robust Standard errors in parentheses (clustered by tariff line). a p < 0.01, b p < 0.05.
All regressions include sector and partner dummies as well as the additional controls of Table 3. We carried
out the usual series of tests (Hansen-J and Kleibergen-Paap rk ML statistics) that assess the validity of the
instrumental variables and none of these tests indicates a problem at the usual confidence levels. The
Kleibergen-Paap rk Wald F-statistics for weak instruments in the presence of clustered standard errors
indicate with 95% confidence a maximum TSLS size well below 10%, implying that our instrument is
strong. The values of the test are 314,96 and 251,12 for column (1) of panels (a) and (b) respectively. The
10% maximal IV size test critical value is 19.93. With IV-Probit estimation, the set of available tests to
assess the validity of instrumental variables remains limited. First-step statistics point to a very strong
predictive power of our excluded instrument. Exogeneity is strongly rejected according to the standard Wald
test in the IV probit specifications reported in panel (a). In panel (b) specification, it is only rejected at the
10-percent level.





22


6.  Sensitivity analysis 


In this section we subject our results to a variety of robustness checks. We start by running
alternative specifications to (2); we further test the relevance of the “money-left-on-the-table
hypothesis”; we account for the high incidence of zero imports, for the influence of stalling
multilateral trade talks and for the length of the implementation period of tariff cuts. As we shall
see, these essentially establish that the emulator effect is non-linear.




6.1.  Evidence at the good level: Alternative Logit 


In our quest for the effects of CUTg on the IDA status of goods, specification (2) with
SEVENg as the dependent variable is quite conservative insofar as it lumps together goods that are
excluded from all RTAs altogether with goods that are granted IDA status in all but one RTA.
Other categorizations of the data are possible.


Our first robustness check is to run a logit that is the mirror image of (2):


{ } ( )( ) 1 2Pr 1 Uruguayg G g g gONE f CUT MFNβ β= = Λ + + + g,pX β , (4)
where ONEg takes value one if the specific good gets IDA status into the United States market in at
least one RTA and zero otherwise (i.e. { }0 ,1 I # : 0implg g pONE p PT≡ − = , where I0{.} denotes an
indicator function that takes value 1 if its component is equal to zero and value 0 otherwise).


We report the results in Table A1, which is symmetric to Table 2 (same set of controls,
same estimator). Qualitatively, all the findings are similar to those of Table 2. Quantitatively, the
positive effect of CUTg and the negative effects of UruguaygMFN , DIFF0g and SNAFTAg in (4) are
smaller in absolute value than in (2). The odds ratio corresponding to the coefficient of interest β1
is ranges from 1.13 in the baseline specification to 1.17 with the DIFF0g, SMg and SNAFTAg
controls, implying that an additional one-percentage point multilateral tariff cut is associated with a
13–17 per cent increase in the odds of including the good in the group of IDA goods. Though
quite strong, the effect of CUTg on ONEg is weaker than its effect on SEVENg. This suggests that
the domestic resistance to preferential trade liberalization is decreasing in the number of IDA
statuses being granted at the margin.




6.2.  Evidence at the good level: Poisson 


A natural alternative to (2) and (4) is to regress the number of times good g is being
granted IDA status, defined, as ,#{ : 0}


impl
g g pNTL p PT≡ = , on our list of control variables. This


alternative measure of the extensive margin of the ‘emulator effect’ is a count variable, so we run
the constant semi-elasticity model (Poisson regression)


( )( ) 1 2E , , expUruguay Uruguayg g g G g g gNTL CUT MFN f CUT MFNβ β⎡ ⎤ = + + +⎣ ⎦g,p g,pX β X β (5)
with one observation per good g.





23


Table A2 presents our findings. The results are consistent with those of Tables 2 and 8.
Columns (1) and (2) report the results of specification (5), respectively excluding and including the
sector dummies fG(g), excluding any other control. The coefficients are precisely estimated. In
column (2), the Poisson incidence rate ratio (PIRR = exp β1) is equal to 1.02, implying that an extra
one percentage point CUTg increases the expected number of times that the good in question is
granted IDA status by 2 per cent. The PIRR rises to 1.03 when we add the additional controls of
columns (3) and (4) (our preferred specification). The effect is not strong quantitatively but it is
statistically significant and robust. Again, this suggests that the domestic resistance to preferential
trade liberalization is decreasing in the number of IDA statuses being granted at the margin. We
confirm this in the immediate sequel.




6.3.  Evidence at the good level: Hurdle 


We verify that the effect of CUTg on the extensive margin of preferential trade
liberalization as captured by the IDA status is non-linear by implementing a two-stage Hurdle
regression. The first step is a logit that is the mirror image of (2),


( )( ) 1 2Pr{ 0} Uruguayg G g g gSEVEN f b CUT b MFN= = Λ + + + g,pX b , (6)
and the second step is the conditional Poisson regression:


( )( ) 1 2E 7 0; exp .Uruguayg g G g g gNTL SEVEN f c CUT c MFN⎡ ⎤− = ⋅ = + + +⎣ ⎦ g,pX c (7)
For instance, b1 informs us about the extent to which one extra percentage point of CUTg


for good g is associated with a reduction of the likelihood of that good of being granted IDA status
to all seven partners and, failing this, c1 says how this extra percentage point cut reduces the
likelihood of good g being included in one extra RTA. In line of our previous findings, we expect
b1 to be negative (and b2 to be positive).


The results of the first step (6) are reported in Table A3, panel (a). As expected, the
exponentiated coefficients are the mirror image of those of Table 2 (the values of 1jβ − in tables
2 and 5 are comparable for all j = 1,2,…) and thus require no further discussion. Likewise, the
results for the second step (7) – reported in Table A3, panel (b) – are comparable to those of (5).
They also confirm our priors, in line with our earlier finding, that most of the emulator effect is
captured by SEVENg. The economic significance of the coefficients is small (though all
coefficients are statistically significant at the one per cent level with the exception of SMg, which is
significant at the five per cent level).


Taken together, the findings of Tables 2 and Appendix tables A1 to A3 imply that the
manifestation of the emulator effect is non-linear and most strongly felt between granting 7 IDA
statuses and 6 IDA statuses or fewer.




6.4.  Interaction between CUT and MFN 


In order to further distinguish between the “money-left-on-the-table hypothesis” and the
emulator effect, we now interact CUTg with UruguaygMFN in all the above specifications. The
motivation for doing this is the following. If the dynamic complementarity between past





24


(multilateral) cuts and current (preferential) liberalization that we have uncovered so far hid a static
substitution between multilateralism and regionalism, then we should expect the effect of CUTg on
IDA treatment to be stronger where there is more room for manoeuvre, that is, where UruguaygMFN
tariff rates are larger. This is not what we find, however.


Table A4 reports the results, two of which are noteworthy. First, the coefficient of
Uruguay
gMFN *CUTg is strongly negative (its odds ratio is lower than unity), which rebukes the


hypothesis whereby multilateralism and regionalism are substitutes. Second, the addition of this
interaction term increases the coefficient on CUTg and reduces the coefficient on UruguaygMFN .
Results obtained with the Hurdle estimation strategy largely confirm these patterns.


We interpret all these results as adding extra pieces of evidence if favour of the emulator
hypothesis and against the alternative “money-left-on-the-table hypothesis”.




6.5.  Zero Imports 


The share of zero imports is a prominent feature in the dataset (see Table 1). We exploit
this feature of the data as an additional test for the emulator hypothesis in two ways that are in line
with political economy arguments presented above. First, the emulator effect is economically
meaningful only if it holds for goods for which import competition bites. We thus introduce a
dummy ZMg to signal zero imports (ZMg = 1 if imports are zero and ZMg = 0 if imports are strictly
positive) and we expect the coefficient of the interaction term (1-ZMg)*CUTg to be positive.
Second, we expect the coefficient of ZMg*CUTg to be larger than the former coefficient because
zero imports are synonymous with no competitive threat by foreign firms in the import competing
segment. In mercantilist terms, granting preferential access to such imports is a cheap “concession”
to make.


The results are reported in Table A5. In line with our priors, the coefficient of CUTg
remains positive and the coefficient of the interaction term ZMg*CUTg is larger than the coefficient
on (1-ZMg)*CUTg. That is to say, the emulator effect is stronger for goods where imports are zero
than for goods where imports are strictly positive. Import liberalization is thus likely to take place
more systematically for products not exposed to competition from the partner country exporters.




6.6.  Implementation period 


Tariff cuts negotiated during the Uruguay Round have not been implemented uniformly
either across countries or across products. Only 40 per cent of United States tariff cuts for
industrial products have been implemented the year after the end of the Round while 30 per cent of
tariff cuts took 10 years to be implemented.26 Differences in implementation periods are likely to
add to the non-linearity of the emulator effect. They also provide a different metric to capture the
boldness of multilateral trade liberalization. We thus control for the implementation period of the
MFN tariff cuts using two alternative approaches. The first one consists in adding a control
variable, (Implementation period)g, defined as the number of years taken for implementing the



26 The proportions are different for agricultural products: 30 per cent of the tariff cuts have been implemented
within the first year following the completion of the Uruguay round it took six years to implement the
remaining 70 per cent of tariff cuts.





25


cuts. The second approach combines the two measures of the boldness of multilateral
liberalization, CUTg and (Implementation period)g, into a new one, (Speed of CUT) g, defined as the
ratio of the two. The results, reported in Table A6, are all consistent with the emulator hypothesis:
the coefficient of (Implementation period)g in the first two columns is negative (the odd ratio is
smaller than unity), implying that a good for which the tariff cut takes more time to be effective is
less likely to be granted IDA status later on; note that the coefficient of our initial measure of
multilateral liberalization, CUTg, remains statistically positive and of the same order of magnitude.
Consistently, the coefficient of the (Speed of CUT)g variable indicates that the emulator effect
increases with the speed of the implementation of the MFN tariff cut.27




7.  Summary and concluding remarks 


This paper investigates the empirical relationship between cuts in MFN bound rates
negotiated during the Uruguay Round of the GATT (1986–1994) and the depth and breadth of
Preferential Trade Agreements signed in the aftermath of its completion. Our empirical
investigation focuses on the United States using official tariff line level data. To the best of our
knowledge, our paper is unique in looking at the causal relationship from multilateralism to
regionalism. The existing empirical literature is looks exclusively at the relationship running the
other way. This line of research is motivated by the view expressed in numerous theoretical
contributions that regionalism may have a “stumbling block” effect on multilateral trade
liberalization (Bhagwati, 1991). If the stumbling block hypothesis is correct, then the proliferation
of RTAs involving at least one WTO member is guilty of slowing down and threatening the Doha
Round of negotiations at the GATT/WTO.


The main findings of the paper are that (a) the imports of goods that the United States
liberalizes swiftly the most frequently on a preferential basis are also the goods for which it granted
the largest MFN tariff reductions during the Uruguay Round; (b) this effect is robust qualitatively
but varies across the types of goods, being stronger for intermediate and capital goods; (c) it holds
only for goods that have no alternative import restrictions in the form of Non Tariff Measures; (d)
it is weaker for goods with prohibitively costly Rules of Origin.


We interpret these findings as evidence that multilateral tariff “concessions” are dynamic
complements to preferential treatment of RTA partners. Such dynamic complementarities between
sequential Rounds of trade liberalization are consistent with the “Juggernaut” theory of trade
liberalization. This theory stresses the role of domestic sluggish adjustments to account for the
systematic, monotonically decreasing trade barriers of the modern trading system.


Crossing our results with those of Limão (2006) – who finds that preferential trade
liberalization prior to the completion of the Uruguay Round acted as a stumbling block to
multilateralism in the United States case – we may thus conclude that the difficulties encountered
by the Doha Round might in part be the indirect result of the success of the Uruguay Round.





27 We also ran a regression with (Speed of RTA liberalization)g,p as the dependant variable (and defined in a
similar way as (Speed of CUT)g) on the controls of columns 3 and 4 of Table A6. The results (not reported)
are also strongly consistent with the emulator hypothesis.





26


References 


 
Aghion P, Antràs P and Helpman E (2007). Negotiating Free Trade. Journal of International


Economics 73: 1–30.


Bagwell K and Staiger RW (1998). Will Preferential Agreements Undermine the Multilateral
Trading System? Economic Journal 108, 1162-1182.


Bagwell K and Staiger RW (1999). Regionalism and Multilateral Tariff Cooperation, in: Piggott, J.
and Woodland, A. (Eds.), International Trade Policy and the Pacific Rim, St. Martin’s
Press, New York.


Baier SL and Bergstrand JH (2004). Economic Determinants of Free Trade Agreements. Journal of
International Economics 64 (1), 29-63.


Baldwin RE (1994). Towards an Integrated Europe. London, United Kingdom, CEPR.


Baldwin RE (1995). A domino theory of regionalism, in: Baldwin, R.E., Haaparanta, P., Kiander,
J. (Eds.), Expanding Membership in the European Union. Cambridge University Press.


Baldwin RE (2009). Trade negotiations within the GATT/WTO framework: A survey of successes
and failures. Journal of Policy Modelling 31, 515-525.


Baldwin RE and Robert-Nicoud F (2007). A Simple Model of the Juggernaut Effect of Trade
Liberalization. CEPR Discussion paper No.6607.


Baldwin RE and Seghezza E (2008). Are Trade Blocs Building or Stumbling Blocks? New
Evidence. CEPR Discussion paper No. 6599.


Bhagwati J (1991). The World Trading System at Risk. Princeton University Press, Princeton, New
Jersey.


Broda C, Limão N and Weinstein D (2008). Optimal Tariffs and Market Power: The
Evidence. American Economic Review 98(5), 2032-65.


Cadot O, de Melo J and Olarreaga M (1999). Regional Integration and Lobbying for Tariffs against
Non-Members. International Economic Review 40, 635-657.


Cameron AC and Trivedi PK (2005). Microeconometrics: Methods and Applications. Cambridge
University Press, New York.


Conconi P, Facchini G and Zanardi M (2008). Fast track authority and international trade
negociation. CEPR discussion paper No. 6790.


Egger P and Larch M (2008). Interdependent Preferential Trade Agreement Memberships: An
empirical analysis. Journal of International Economics. 76(2), 384-399.


Ethier WJ (1998). Regionalism in a Multilateral World. Journal of Political Economy. 106(6),
1214-1245.


Estevadeordal A, Freund C and Ornelas E (2008). Does Regionalism Affect Trade Liberalization
towards Non-Members? Quarterly Journal of Economics. 123(4), 1531-1575.






27


Freund C (2000a). Multilateralism and the Endogenous Formation of Preferential Trade
Agreements, Journal of International Economics. 52, 359-376.


Freund C (2000b). Different Paths to Free Trade: The Gains from Regionalism. Quarterly Journal
of Economics. 115, 1317-1341.


Freund C and Ornelas E (2010). Regional trade agreements. Annual Review of Economics. 2, 139-
167.


Gawande K, Krishna P and Olarreaga M (forthcoming). Lobbying Competition Over Trade Policy.
Journal of European Economic Association.


Grossman GM and Helpman E (1994). Protection for Sale. American Economic Review. 84(4),
833-850.


Grossman GM and Helpman E (1995). The Politics of Free-Trade Agreements. American
Economic Review. 85(4), 667-690.


Hahn J and Hausman J (2002). Notes on Bias in Estimators for Simultaneous Equation Models.
Economics Letters 75(2), 237-241.


Kennan J and Riezman R (1990). Optimal tariff equilibria with customs unions. Canadian Journal
of Economics. 23, 70-83.


Krishna K (2006). Understanding Rules of Origins, in: Cadot O, Estevadeordal O, Suwa-
Eisenmann A and Verdier T (eds.) The Origin of Goods: Rules of Origin in Regional
Trade Agreements. Oxford University Press, 2006.


Krishna P (1998). Regionalism and Multilateralism: A Political Economy Approach. Quarterly
Journal of Economics. 113, 227-252.


Limão N (2006). Preferential Trade Agreements as Stumbling Blocks for Multilateral Trade
Liberalization: Evidence for the United States. American Economic Review. 96(3), 896-
914.


Limão N (2007). “Are Preferential Trade Agreements with Non-trade Objectives a Stumbling
Block for Multilateral Liberalization?” Review of Economic Studies. 74(3), 821-855.


Limão N and Karacaovali B (2008). The Clash of Liberalizations: Preferential vs. Multilateral
Trade Liberalization in the European Union. Journal of International Economics. 74(2),
299-327.


Maggi G and Rodríguez-Clare A (2007). A Political-Economy Theory of Trade Agreements.
American Economic Review. 97, 1374-1406.


Mansfeld ED and Reinhardt E (2003). Multilateral Determinants of Regionalism: The Effects of
GATT/WTO on the Formation of Preferential Trading Agreements, International
Organization, 57(4), 829-862.


Martin P, Mayer T and Thoenig M (2008). Make trade not war? Review of Economic Studies.
75(3), 865-900.


Martin P, Mayer T and Thoenig M (2009). The Geography of Conflicts and Regional Agreements.
Processed, Sciences-Po Paris and University of Lausanne.





28


McLaren J (2002). A Theory of Insidious Regionalism. Quarterly Journal of Economics. 118(2),
571-608.


Ornelas E (2005a). Rent Destruction and the Political Viability of Free Trade Agreements.
Quarterly Journal of Economics. 120, 1475-1506.


Ornelas E (2005b). Endogenous Free Trade Agreements and the Multilateral Trading System.
Journal of International Economics. 67, 471-497.


Richardson M (1993). Endogenous protection and trade diversion. Journal of International
Economics. 34, 309-324.


Romalis J (2007). Market Access, Openness and Growth, NBER Working Paper 13048.


Saggi K (2006). Preferential Trade Agreements and Multilateral Tariff Cooperation. International
Economic Review. 47(1), 29-57.


Staiger R (1995). A Theory of Gradual Trade Liberalization, in: Deardorff, A., Levinsohn, J. and
Stern, R. (Eds.), New Directions in Trade Theory, University of Michigan Press.


WTO, UNCTAD, International Trade Centre (2007). World Tariff Profiles 2006. WTO, UNCTAD
and ITC, Geneva.


Zoellick RB (2001). 2001 Trade Policy Agenda and 2000 Annual Report. Office of the United
States Trade Representative, Washington D.C.






29


Appendix tables 


Table A1: LOGIT “One’”


Dependant variable: ONE
(Probability that tariff line is granted IDA to United States market


to at least one partner)
(1) (2) (3) (4) (5)



Tariff CUT 1.054a 1.133a 1.178a 1.178a 1.169a


(Tokyo minus Uruguay) (0.0124) (0.0179) (0.0226) (0.0227) (0.0234)


MFN 0.976a 0.954a 0.946a 0.947a 0.948a
tariff level (0.00644) (0.00543) (0.00581) (0.00581) (0.00590)



DIFF0 (no Uruguay 2.275a 2.279a 2.217a


Round cut) (0.378) (0.379) (0.371)


Share imports 1.037 1.031
From RTA partners (0.0671) (0.0675)



Share imports 0.995b


from NAFTA partners (0.00202)


Sector FE No Yes Yes Yes Yes



Observations 5756 5756 5756 5756 5756
Pseudo R2 0.019 0.132 0.140 0.140 0.141
LL -1662.1 -1355.6 -1343.0 -1342.8 -1340.6



Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors in parentheses.
a p < 0.01, b p < 0.05.





30


Table A2: POISSON regressions


Dependant variable: NTL
(Number of times that tariff line g is granted IDA to United States


market)
(1) (2) (3) (4) (5)


Tariff CUT 1.015a 1.021a 1.028a 1.028a 1.026a
(Tokyo minus Uruguay) (0.000949) (0.00102) (0.00129) (0.00129) (0.00133)



MFN 0.971a 0.975a 0.974a 0.974a 0.974a


tariff rate (0.00122) (0.00134) (0.00137) (0.00137) (0.00137)


DIFF0 (no Uruguay 1.152a 1.153a 1.150a
Round cut) (0.0152) (0.0152) (0.0152)



Share imports 1.011b 1.010b


from RTA partners (0.00500) (0.00494)


Share imports 0.999a
from NAFTA partners (0.000201)



Sector FE No Yes Yes Yes Yes


Observations 7419 7419 7419 7419 7419
Pseudo R2 0.029 0.045 0.048 0.048 0.048
LL -15775.5 -15505.6 -15469.7 -15468.0 -15459.7



Notes: Coefficients: Exponentiated (Poisson Incidence Rate Ratios, or PIRR); Robust standard errors in


parentheses. a p < 0.01, b p < 0.05.






31


Table A3: HURDLE regressions

Panel (a) Logit


Dependant variable: 1- SEVEN
(Probability that tariff line g is not granted IDA to United States


market to all 7 partners)
(1) (2) (3) (4) (5)



Tariff CUT 0.877a 0.815a 0.752a 0.751a 0.761a


(Tokyo minus Uruguay) (0.00636) (0.00727) (0.00892) (0.00894) (0.00924)


MFN 1.496a 1.522a 1.635a 1.635a 1.637a
tariff rate (0.0286) (0.0382) (0.0466) (0.0467) (0.0469)



DIFF0 (no Uruguay 0.229a 0.228a 0.235a


Round cut) (0.0240) (0.0240) (0.0247)


Share imports 0.981 0.990
from RTA partners (0.0338) (0.0334)



Share imports 1.008a


from NAFTA partners (0.00165)


Observations 7419 7419 7419 7419 7419
LL -12392.7 -11372.1 -11254.2 -11253.8 -11238.9


Notes: Coefficients: Exponentiated (odds ratios); Robust standard errors in parentheses.
a p < 0.01, b p < 0.05.


Panel (b) Conditional Poisson


Dependant variable: 7 – NTL, conditional on NTL < 7
(Number of times that tariff line g is not granted IDA to United


States market)
(1) (2) (3) (4) (5)



Tariff CUT 0.995 0.982a 0.977a 0.977a 0.977a


(Tokyo minus Uruguay) (0.00248) (0.00281) (0.00311) (0.00312) (0.00315)


MFN 1.004a 1.011a 1.012a 1.012a 1.012a
tariff rate (0.000331) (0.00144) (0.00150) (0.00150) (0.00151)



DIFF0 (no Uruguay 0.871a 0.871a 0.873a


Round cut) (0.0242) (0.0242) (0.0244)


Share imports 0.993 0.994
from RTA partners (0.00762) (0.00765)



Share imports 1.001


from NAFTA partners (0.000344)


Sector FE No Yes Yes Yes Yes
Observations 7419 7419 7419 7419 7419
LL -12392.7 -11372.1 -11254.2 -11253.8 -11238.9


Notes: Coefficients: Exponentiated (Poisson Incidence Rate Ratios, or PIRR); Robust standard errors in
parentheses. a p < 0.01, b p < 0.05.






32


Table A4: Interacting CUT and MFN


Specification:
LOGIT


Seven
p-g


LOGIT
LOGIT


One POISSON
HURDLE
I [*] (logit)


HURDLE II [*]
(trunc. poisson)



Tariff CUT 1.443a 1.172a 1.187a 1.033a 0.693a 0.983a


(To. minus Ur.) (0.0419) (0.0115) (0.0269) (0.00209) (0.0201) (0.00339)


MFN 0.669a 0.953a 0.970b 0.979a 1.494a 1.020a
tariff rate (0.0255) (0.00414) (0.0118) (0.00172) (0.0568) (0.00186)



MFN*Tariff


CUT
0.979a 0.993a 0.998 0.999a 1.021a 0.999a


(0.00541) (0.000880) (0.00117) (0.000260) (0.00564) (0.0000832)


DIFF0 (no
Uruguay


Round cut)


3.891a
(0.406)


1.567a
(0.0783)


2.126a
(0.348)


1.145a
(0.0151)


0.257a
(0.0268)


0.864a
(0.0242)



Share imports


from RTA
partners


1.012
(0.0331)


1.039a
(0.00818)


1.033
(0.0674)


1.010b
(0.00489)


0.988
(0.0323)




0.993
(0.00771)


Sector FE Yes Yes Yes Yes Yes Yes
Partner FE N.A. Yes N.A. N.A. N.A. N.A.


Observations 6822 51814 5756 7419 7419 7419
Pseudo R2 0.324 0.089 0.143 . -
LL -3072.3 -27870.2 -1338.2 -15450.5 -11215.5



Notes: Coefficients: Exponentiated ; Robust standard errors in parentheses. a p < 0.01, b p < 0.05.
All regressions include sector FE and “Share imports from NAFTA partners”. MFN*CUT represents the
interaction between the variable MFN tariff rate and the variable Tariff CUT. [*] Columns (5) and (6) report
results from Hurdle estimation and should then be considered jointly. Column (5) shows results obtained in
the first step (a logit estimation). Column (6) shows results obtained in the second step (a truncated Poisson
estimation). Note that we expect the coefficients of the Hurdle regressions to be the opposite of the
coefficients in Col. (1) to (4) because the Hurdle regressions are specified as the mirror image of the logit
and Poisson regressions.






33


Table A5: Zero Imports


Specification:
LOGIT Seven p-g LOGIT


(1-ZM)* Tariff CUT 1.211 a 1.068 a
(To. minus Ur.) (0.0181) (0.0188)



ZM * Tariff CUT 1.363 a 1.123 a


(0.0219) (0.00698)


MFN 0.599a 0.926 a
tariff rate (0.0175) (0.00387)



ZM 1.047 0.767 a


(Zero Imports) (0.13) (0.0397)


Controlsc Yes Yes
Sector FE Yes Yes


Partner FE No Yes
Observations 6822 51814
Pseudo R2 0.333 0.121
LL -3028.1 -26894.4



Notes: Coefficients: Exponentiated; Robust standard errors in parentheses. a p < 0.01, b p < 0.05; c all
regressions include “DIFF0” and “Share imports from NAFTA partners” as controls. ZM is a dummy variable
taking the value one if imports at the tariff line level are zeros. ZM*CUT represents the interaction between
the ZM dummy and the variable Tariff CUT.



Table A6: Implementation Period


Specification:
LOGIT Seven p-g LOGIT LOGIT Seven p-g LOGIT


Tariff CUT 1.211 a 1.096 a
(To. minus Ur.) (0.0157) (0.00749)



MFN 0.62 a 0.929 a 0.598 a 0.925 a


tariff rate (0.179) (0.00386) (0.0185) (0.00391)


Implementation period 0.773a 0.935 a
(Number of years) (0.0136) (0.00721)



CUT speed 1.361 a 1.181 a


(CUT over Impl. period) (0.0154) (0.00804)


Controlsc Yes Yes Yes Yes
Sector FE Yes Yes Yes Yes


Partner FE No Yes No Yes
Observations 6822 51814 6822 51814
Pseudo R2 0.343 0.125 0.362 0.144
LL -3028.1 -26894.4 -2897.6 -26190.8



Notes. Coefficients: Exponentiated; Robust standard errors in parentheses. a p < 0.01, b p < 0.05; c all
regressions include “DIFF0” and “Share imports from NAFTA partners” as controls. “Implementation
period” is defined as the number of years taken to implement the Tariff CUT. “CUT speed” is the Tariff
CUT divided by the number of years of the “Implementation period”.








35


UNCTAD Study Series on


POLICY ISSUES IN INTERNATIONAL TRADE
AND COMMODITIES






No. 1 Erich Supper, Is there effectively a level playing field for developing country
exports?, 2001, 138 p. Sales No. E.00.II.D.22.



No. 2 Arvind Panagariya, E-commerce, WTO and developing countries, 2000, 24 p. Sales


No. E.00.II.D.23.

No. 3 Joseph Francois, Assessing the results of general equilibrium studies of multilateral


trade negotiations, 2000, 26 p. Sales No. E.00.II.D.24.

No. 4 John Whalley, What can the developing countries infer from the Uruguay Round


models for future negotiations?, 2000, 29 p. Sales No. E.00.II.D.25.

No. 5 Susan Teltscher, Tariffs, taxes and electronic commerce: Revenue implications for


developing countries, 2000, 57 p. Sales No. E.00.II.D.36.

No. 6 Bijit Bora, Peter J. Lloyd, Mari Pangestu, Industrial policy and the WTO, 2000, 47 p.


Sales No. E.00.II.D.26.

No. 7 Emilio J. Medina-Smith, Is the export-led growth hypothesis valid for developing


countries? A case study of Costa Rica, 2001, 49 p. Sales No. E.01.II.D.8.

No. 8 Christopher Findlay, Service sector reform and development strategies: Issues and


research priorities, 2001, 24 p. Sales No. E.01.II.D.7.

No. 9 Inge Nora Neufeld, Anti-dumping and countervailing procedures – Use or abuse?


Implications for developing countries, 2001, 33 p. Sales No. E.01.II.D.6.

No. 10 Robert Scollay, Regional trade agreements and developing countries: The case of


the Pacific Islands’ proposed free trade agreement, 2001, 45 p. Sales No.
E.01.II.D.16.



No. 11 Robert Scollay and John Gilbert, An integrated approach to agricultural trade and


development issues: Exploring the welfare and distribution issues, 2001, 43 p. Sales
No. E.01.II.D.15.



No. 12 Marc Bacchetta and Bijit Bora, Post-Uruguay round market access barriers for


industrial products, 2001, 50 p. Sales No. E.01.II.D.23.

No. 13 Bijit Bora and Inge Nora Neufeld, Tariffs and the East Asian financial crisis, 2001,


30 p. Sales No. E.01.II.D.27.

No. 14 Bijit Bora, Lucian Cernat, Alessandro Turrini, Duty and quota-free access for LDCs:


Further evidence from CGE modelling, 2002, 130 p. Sales No. E.01.II.D.22.





36


No. 15 Bijit Bora, John Gilbert, Robert Scollay, Assessing regional trading arrangements in
the Asia-Pacific, 2001, 29 p. Sales No. E.01.II.D.21.



No. 16 Lucian Cernat, Assessing regional trade arrangements: Are South-South RTAs more


trade diverting?, 2001, 24 p. Sales No. E.01.II.D.32.

No. 17 Bijit Bora, Trade related investment measures and the WTO: 1995-2001, 2002.

No. 18 Bijit Bora, Aki Kuwahara, Sam Laird, Quantification of non-tariff measures, 2002,


42 p. Sales No. E.02.II.D.8.

No. 19 Greg McGuire, Trade in services – Market access opportunities and the benefits of


liberalization for developing economies, 2002, 45 p. Sales No. E.02.II.D.9.

No. 20 Alessandro Turrini, International trade and labour market performance: Major


findings and open questions, 2002, 30 p. Sales No. E.02.II.D.10.

No. 21 Lucian Cernat, Assessing south-south regional integration: Same issues, many


metrics, 2003, 32 p. Sales No. E.02.II.D.11.

No. 22 Kym Anderson, Agriculture, trade reform and poverty reduction: Implications for


Sub-Saharan Africa, 2004, 30 p. Sales No. E.04.II.D.5.

No. 23 Ralf Peters and David Vanzetti, Shifting sands: Searching for a compromise in the


WTO negotiations on agriculture, 2004, 46 p. Sales No. E.04.II.D.4.

No. 24 Ralf Peters and David Vanzetti, User manual and handbook on Agricultural Trade


Policy Simulation Model (ATPSM), 2004, 45 p. Sales No. E.04.II.D.3.

No. 25 Khalil Rahman, Crawling out of snake pit: Special and differential treatment and


post-Cancun imperatives, 2004.

No. 26 Marco Fugazza, Export performance and its determinants: Supply and demand


constraints, 2004, 57 p. Sales No. E.04.II.D.20.

No. 27 Luis Abugattas, Swimming in the spaghetti bowl: Challenges for developing


countries under the “New Regionalism”, 2004, 30 p. Sales No. E.04.II.D.38.

No. 28 David Vanzetti, Greg McGuire and Prabowo, Trade policy at the crossroads – The


Indonesian story, 2005, 40 p. Sales No. E.04.II.D.40.

No. 29 Simonetta Zarrilli, International trade in GMOs and GM products: National and


multilateral legal frameworks, 2005, 57 p. Sales No. E.04.II.D.41.

No. 30 Sam Laird, David Vanzetti and Santiago Fernández de Córdoba, Smoke and mirrors:


Making sense of the WTO industrial tariff negotiations, 2006, Sales No.
E.05.II.D.16.



No. 31 David Vanzetti, Santiago Fernandez de Córdoba and Veronica Chau, Banana split:


How EU policies divide global producers, 2005, 27 p. Sales No. E.05.II.D.17.

No. 32 Ralf Peters, Roadblock to reform: The persistence of agricultural export subsidies,


2006, 43 p. Sales No. E.05.II.D.18.






37


No. 33 Marco Fugazza and David Vanzetti, A South-South survival strategy: The potential
for trade among developing countries, 2006, 25 p.



No. 34 Andrew Cornford, The global implementation of Basel II: Prospects and outstanding


problems, 2006, 30 p.

No. 35 Lakshmi Puri, IBSA: An emerging trinity in the new geography of international


trade, 2007, 50 p.

No. 36 Craig VanGrasstek, The challenges of trade policymaking: Analysis, communication


and representation, 2008, 45 p.

No. 37 Sudip Ranjan Basu, A new way to link development to institutions, policies and


geography, 2008, 50 p.

No. 38 Marco Fugazza and Jean-Christophe Maur, Non-tariff barriers in computable general


equilibrium modelling, 2008, 25 p.

No. 39 Alberto Portugal-Perez, The costs of rules of origin in apparel: African preferential


exports to the United States and the European Union, 2008, 35 p.

No. 40 Bailey Klinger, Is South-South trade a testing ground for structural


transformation?, 2009, 30 p.

No. 41 Sudip Ranjan Basu, Victor Ognivtsev and Miho Shirotori, Building trade-relating


institutions and WTO accession, 2009, 50 p.

No. 42 Sudip Ranjan Basu and Monica Das, Institution and development revisited: A


nonparametric approach, 2010, 26 p.

No. 43 Marco Fugazza and Norbert Fiess, Trade liberalization and informality: New stylized


facts, 2010, 45 p.

No. 44 Miho Shirotori, Bolormaa Tumurchudur and Olivier Cadot, Revealed factor intensity


indices at the product level, 2010, 55 p.


No. 45 Marco Fugazza and Patrick Conway, The impact of removal of ATC Quotas on
international trade in textiles and apparel, 2010, 50 p.



No. 46 Marco Fugazza and Ana Cristina Molina, On the determinants of exports survival,


2011, 40 p.

No. 47 Alessandro Nicita, Measuring the relative strength of preferential market access,


2011, 30 p.

No. 48 Sudip Ranjan Basu and Monica Das, Export structure and economic performance in


developing countries: Evidence from nonparametric methodology, 2011, 58 p.

No. 49 Alessandro Nicita and Bolormaa Tumurchudur-Klok, New and traditional trade flows


and the economic crisis, 2011, 22 p.

No. 50 Marco Fugazza and Alessandro Nicita, On the importance of market access for trade,


2011, 35 p.





38


No. 51 Marco Fugazza and Frédéric Robert-Nicoud, The 'Emulator Effect' of the Uruguay
round on United States regionalism, 2011, 45 p.














































Copies of UNCTAD Study series on Policy Issues in International Trade and Commodities may be
obtained from the Publications Assistant, Trade Analysis Branch (TAB), Division on International
Trade in Goods and Services and Commodities (DITC), United Nations Conference on Trade and
Development, Palais des Nations, CH-1211 Geneva 10, Switzerland (Tel: +41 22 917 4644).
These studies are accessible on the website at http://www.unctad.org/tab.




Since 1999, the Trade Analysis Branch of the Division on International Trade in Goods and
Services, and Commodities of UNCTAD has been carrying out policy-oriented analytical work
aimed at improving the understanding of current and emerging issues in international trade and
development. In order to improve the quality of the work of the Branch, it would be useful to
receive the views of readers on this and other similar publications. It would therefore be greatly
appreciated if you could complete the following questionnaire and return to:


Trade Analysis Branch, DITC
Rm. E-8065


United Nations Conference on Trade and Development
Palais des Nations, CH-1211 Geneva 10, Switzerland


(Fax: +41 22 917 0044; E-mail: tab@unctad.org)


1. Name and address of respondent (optional):


2. Which of the following describes your area of work?


Government Public enterprise
Private enterprise institution Academic or research
International organization Media
Not-for-profi t organization Other (specify) _________________


3. In which country do you work? _________________________________________


4. Did you fi nd this publication Very useful Of some use Litt le use
to your work?


5. What is your assessment of the contents of this publication?
Excellent Good Adequate Poor


6. Other comments:


QUESTIONNAIRE


UNCTAD Study series on


POLICY ISSUES IN INTERNATIONAL TRADE
AND COMMODITIES


(Study series no. 51: The ‘Emulator Eff ect’ of the Uruguay Round
on United States regionalism)


Readership Survey




Login