domestic policies, national sovereignty, and...

DOMESTIC POLICIES NATIONAL SOVEREIGNTY ANDINTERNATIONAL ECONOMIC INSTITUTIONS

KYLE BAGWELL AND ROBERT W STAIGER

To what extent must nations cede control over their economic and socialpolicies if global efciency is to be achieved in an interdependent world Thisquestion is at the center of the debate over the future role of the WTO (formerlyGATT) in the realm of labor and environmental standards In this paper weestablish that the market access focus of current WTO rules is well equipped tohandle the problems associated with choices over labor and environmental stan-dards In principle with relatively modest changes that grant governments moresovereignty not less these rules can deliver globally efcient outcomes

I INTRODUCTION

To what extent must nations cede control over their economicand social policies if global efciency is to be achieved in aninterdependent world At a broad level this question probes thelimits of any international economic institution whether gearedtoward real or nancial concerns that is designed to promoteglobal efciency while respecting national sovereignty Naturallythe answer depends upon the particular problem that the insti-tution is meant to solve In other words the answer depends uponthe inefciency that would arise under unilateral policy choices

At a more specic level this question is at the center of thedebate concerning the appropriate scope of the World Trade Or-ganization (WTO formerly GATT) Recently member countrieshave considered ways to broaden the WTOrsquos orientation beyondconventional trade policy measures to include labor and environ-mental standards There are now initiatives to introduce theissue of labor standards directly onto the negotiating agenda ofthe WTO with the purpose of creating a WTO ldquosocial clauserdquo Thesocial clause would specify a set of minimum international laborstandards and then permit restrictions to be placed againstimports from countries not complying with these minimum stan-dards With regard to environmental policies a WTO Committeeon Trade and Environment has been established to identify therelationships between trade and environmental measures and to

We thank Donald Davis Scott Taylor an anonymous referee and seminarparticipants at the University of Illinois Iowa State University the New YorkFederal Reserve the University of Toronto the University of Washington and theUniversity of Wisconsin for helpful comments on an early version of this paper

copy 2001 by the President and Fellows of Harvard College and the Massachusetts Institute ofTechnologyThe Quarterly Journal of Economics May 2001

519

recommend necessary modications to the WTO To some degreethese labor and environmental initiatives are responsive to ldquorace-to-the-bottomrdquo concerns In the face of falling trade barriers withweak-standards countries it is feared that the labor and environ-mental standards of the industrialized world might be compro-mised in the name of international competitiveness But suchinitiatives encroach on traditional limits of national sovereigntyThey therefore raise difcult issues about the structure of inter-national economic relations among sovereign states

Motivated by the general question raised above and by therecent debate surrounding the scope of the WTO we ask here amore specic question how should the issue of domestic stan-dards be handled in the WTO We answer this question in asetting where governments choose both trade and domestic stan-dards policies and countries affect each other through their mar-ket interactions so that any externalities across countries arepurely pecuniary in nature By ruling out nonpecuniary external-ities from the start we are excluding ldquoglobal commonsrdquo issuesfrom our analysis and so countries have no reason to care abouteach othersrsquo standards choices directly We do not deny the im-portance of global commons concerns however we choose toexclude them from our analysis since the need to involve theWTO in such concerns is far from obvious1 But even in theabsence of such concerns countries may still care about eachothersrsquo standards choices indirectly because of the trade effectsthat such choices could imply Indeed it is the competitive pres-sures exerted by these trade effects that are often identied asfueling a race to the bottom And as these effects travel throughtrade they are inextricably intertwined with the business of theWTO Our paper considers the question of how labor and envi-ronmental standards should be handled in the WTO in light oftheir associated trade effects

We are of course not the rst to consider this question (seefor example the inuential volumes edited by Bhagwati andHudec [1996]) However analytical results are scarce and ofthese even fewer are concerned with the interaction betweennegotiated reductions in trade barriers and the choice of domestic

1 Except perhaps for reasons of enforcement but even here the case is notwithout qualication see for example Roessler [1998] Ederington [1999] andSpagnolo [1999] On global commons issues see Alesina and Wacziarg [1999]

520 QUARTERLY JOURNAL OF ECONOMICS

standards2 Yet it is from the backdrop of previous tariff reduc-tions that the case for adding labor and environmental standardsto the negotiating agenda of the WTO has been most forcefullymade Hence an understanding of the interaction between tariffnegotiations and the determination of domestic standards seemsa necessary starting point for assessing the claim that thesestandards will suffer as a result of trade liberalization and there-fore necessary as well for considering how labor and environmen-tal standards ought to be approached by the WTO3

We study this question within a general equilibrium frame-work in which two countries trade two goods and governmentsmake decisions over their trade policies (eg tariffs) and theirdomestic standards (eg labor and environmental policies) inpursuit of their own national objectives In modeling governmentdecisions we build on our earlier work [Bagwell and Staiger1999a] representing the objectives of each government as a gen-eral function of its local prices and terms of trade and we extendthis representation in order to incorporate the presence of localstandards The advantages of this approach are twofold First itis very general being consistent (as we later show) both with thetraditional view that governments maximize national income bytheir policy choices and with the view embodied in leading polit-ical-economy models that governments are concerned about thedistributional impacts of their policy choices as well Second byrepresenting government objectives in this way the channelthrough which one governmentrsquos policy choices affect anothergovernmentrsquos welfare is made transparent This helps us to iden-tify and interpret the inefciency associated with unilateral pol-icy choices and this in turn helps to clarify both the potentialproblems that arise when governments focus their negotiations

2 For example Brown Deardorff and Stern [1996] focus on the welfare andterms-of-trade effects of the imposition of labor standards in the presence of freetrade but do not consider the choice of tariff policy while Srinivasan [1996]considers whether diversity of labor standards alters the case for free trade but isnot concerned with whether trade liberalization might alter a countryrsquos choice oflabor standards

3 In this regard there is a small but growing formal literature on the relatedquestion of ldquoissue linkagerdquo in trade agreements (see for example Ederington[1999] and Spagnolo [1999]) in which this interaction is a central concern Thesepapers consider how to structure enforcement provisions when there is a range ofpolicies over which governments are attempting to cooperate In contrast in thispaper we abstract from questions of enforcement and consider instead the com-plementary question of how to structure negotiations when the scope of policiesover which governments could negotiate is potentially broad Adopting the specicfocus of labor standards and following a partial equilibrium approach Bagwelland Staiger [2000] also explore some of the general themes considered here

521DOMESTIC POLICIES AND NATIONAL SOVEREIGNTY

on tariffs alone and the manner in which various rules of nego-tiation may address these problems

As we noted at the outset an answer to the question weconsider requires an understanding of the inefciency associatedwith unilateral policy choices To characterize this inefciencywe begin by drawing a distinction between the level of marketaccess that a government grants to its trading partners and thepolicy mix with which it chooses to deliver this market accesslevel We dene the former to reect the position of a countryrsquosimport demand curve and the latter then captures the means bywhich the countryrsquos import demand curve is positioned throughits chosen mix of policies For example a given level of marketaccess that is implied by a low tariff and weak labor standardmight be implied as well by an alternative policy mix in which thelabor standard is strengthened and the tariff is raised

With this distinction made we may now report our rstresult the inefciency associated with unilateral policy choicesreects a problem with the level of market access not with thepolicy mix More specically we show that in the absence ofinternational negotiations market access levels would be inef-ciently low in light of the objectives of each government but giventhese levels of market access each government would choose anefcient mix of trade and domestic standards policies Put differ-ently in the absence of the GATTWTO governments wouldchoose their labor and environmental standards efciently theonly problem would be a market-access problem

The intuition for this result is simple The inefciencies as-sociated with unilateral policy choices are all traceable to thedesire to shift costs onto onersquos trading partner through the terms-of-trade effects of onersquos policies This cost-shifting is engineeredthrough the impact on exporter prices that market access levelsimply Hence conditional on a level of market access there issimply no reason for a government to distortmdashin light of its ownobjectivesmdashthe policy mix with which it delivers that marketaccess and this policy mix is irrelevant to the objectives of itstrading partner

Having identied the inefciency associated with unilateralpolicy choices we next consider how negotiations could addressthis inefciency Of course governments might negotiate directlyover all policy instruments including domestic labor and envi-ronmental standards But we are interested in asking whetheranything short of direct international negotiations over both tar-


iffs and domestic standards could solve this problem We there-fore suppose rst that governments agree to negotiate over tar-iffs but that they maintain policy autonomy over their domesticstandards In this case we show that an attempt to achievegreater levels of market access through negotiated tariff reduc-tions would lead governments subsequently to distort their do-mestic standards choices More specically our second resultmay now be stated market access negotiations that target tariffsalone cannot achieve efcient policy outcomes as these negotia-tions deect the unilateral incentive to restrict market access onto domestic policies such as labor and environmental standards

When viewed from the perspective of our rst two results theincentive to distort onersquos domestic standards derives from a singlesource the desire to reclaim unilaterally a portion of the marketaccess that onersquos negotiated tariff liberalization has granted Itmight then be tempting to conclude that fty-some years of ne-gotiated tariff liberalization under the GATTWTO has indeedcreated a problem with regard to the determination of labor andenvironmental standards This conclusion however would bepremature It is true that if left unchecked this incentive wouldindeed introduce inefciencies into domestic standards choicesthereby thwarting the efforts of governments to achieve efcientpolicy combinations through tariff negotiations But it is not truethat this incentive is left unchecked within the WTO In fact at abroad level the rules of the WTO exist precisely to provide gov-ernments with a legal framework within which to make marketaccess commitments that are secure against subsequent erosionby unilateral actions of this type As such these rules permit eachmember government to choose its own domestic standards with-out WTO involvement so long as the existing market access com-mitments it has made are not undermined by those choices Wetherefore turn in the remainder of the paper to evaluate WTOrules in more detail and we ask whether these rules might enablegovernments to achieve efcient policy combinations with tariffnegotiations alone

We focus on the right of redress that a government haswithin the WTO whenever it can show that market access com-mitments which it had previously negotiated are being system-atically offset by an unanticipated change in the policiesmdashanypolicies but including in principle labor and environmental stan-dardsmdash of another WTO member even if these policy changesbroke no explicit WTO rules The right to bring these ldquononviola-


tionrdquo complaints is provided in GATT Article XXIII which alsosets out procedures for ldquoviolationrdquo complaints The function ofnonviolation complaints can best be understood when viewedwithin the broader context of WTO rules under which govern-ments are not held rigidly to their negotiated market accesslevels but are expected to follow explicit procedures (contained inGATT Article XXVIII) to renegotiate their market access commit-ments if they so desire As Petersmann [1997] explains thefunction of nonviolation complaints in the WTO is to provide acheck on the domestic policy autonomy of member-countries ldquo and to prevent the circumvention of the provisions in GATTArticle XXVIII if a member rather than withdrawing a con-cession de jure in exchange for compensation or equivalent with-drawals of concessions by affected contracting parties withdrawsa concession de factordquo [Petersmann 1997 p 172]

Under a successful nonviolation complaint the complainingcountry is entitled to a ldquorebalancingrdquo of market access commit-ments wherein either its trading partner nds a way to offercompensation for the trade effects of its domestic policy change(typically in the form of other policy changes that restore theoriginal market access) or the complaining country is permittedto withdraw an equivalent market access concession of its own Inprinciple the prospect of nonviolation complaints therefore se-cures the balance of negotiated market access commitmentsagainst erosion as a result of future changes in domestic policies4

When viewed in the context of our rst two results this feature ofWTO rules is potentially well designed to enable governments toachieve efcient policy combinations with tariff negotiationsalone

To formally evaluate this possibility we construct a simpletwo-stage tariff negotiating game that captures the essence of theability of governments to bring nonviolation complaints underWTO rules Using this formal structure we establish two addi-tional results First the ability to bring nonviolation complaints

4 Nonviolation complaints have proved difcult to carry out in practiceFrom 1947 through 1995 only 14 out of the more than 250 Article XXIII proceed-ings have centered on such complaints and none of these explicitly involved laboror environmental standards (see for example Petersmann [1997 pp 135ndash176])This may in part reect the real-world difculties in determining the trade effectsof domestic policy changes Still the impact of the right to bring nonviolationcomplaints may not be well measured by the numbers of such complaints actuallybrought and in principle this right may restrain governments in their decisions toalter labor and environmental standards just as with domestic policies moregenerally


can indeed allow governments to achieve efcient combinations oftrade and domestic standards policies while negotiating overtariffs alone In essence governments may rst use their tariffnegotiations to achieve efcient levels of market access thenwith the prospect of nonviolation complaints securing marketaccess at the negotiated (efcient) levels governments may makeunilateral policy adjustments that achieve an efcient policy mixImportantly however this feat can only be accomplished if thesubsequent change in domestic standards that each governmentdesires would by itself reduce the market access that it afforded toits trading partner so that it would then be induced to makecompensating tariff reductions by the prospect of a nonviolationcomplaint If instead subsequent to tariff negotiations a govern-ment wished to change its domestic standards in a way thatwould effectively grant greater market access to its trading part-ner at existing tariff levels under WTO rules it would not havethe exibility to unilaterally raise its tariff so as to secure marketaccess at the negotiated level and so in this case efciency cannotbe achieved by tariff negotiations We show however that grant-ing this additional exibility would ensure that governmentscould achieve efcient trade and domestic policy outcomes withtariff negotiations alone and this is our nal result We concludethat with this modication which amounts to granting govern-ments more sovereignty not less WTO rules could thereforeenable governments to achieve efcient levels of labor and envi-ronmental standards while continuing to focus their trade nego-tiations on trade policy

More broadly we interpret our results as indicating that theprinciples of the WTO offer a compelling solution to a key chal-lenge that is now before the multilateral trading system This isnot to say that these principles are necessarily well reected incurrent WTO practice there may be desirable ways to bring WTOpractice more in line with WTO principles Nor would we neces-sarily advocate any changes in WTO rules with regard to laborand the environment such changes may open Pandorarsquos Box Butour results do offer formal support for the view that fundamentalchanges in the WTOrsquos approachmdashsuch as would be implied by aWTO social clausemdashare not required to handle the contentiousissues of labor and environmental standards

The rest of the paper proceeds as follows The next sectionpresents the basic model and derives efcient and noncooperativepolicy choices within this setting The noncooperative policy


choices are shown to be inefcient and the source of the inef-ciency is interpreted Sections III and IV then consider the ef-ciency properties of various approaches to negotiation In SectionIII we establish that international negotiations over tariffs alonelead to inefcient outcomes in the absence of any restraints ondomestic policy choices and we then formally model and evaluatethe impact of the restraints that WTO rules place on thesechoices In Section IV we consider how WTO rules could bemodied to achieve efcient outcomes Finally Section Vconcludes

II THE BASIC MODEL

In this section we develop a simple model of internationaltrade within which the essential inefciencies associated withunilateral choices of trade and domestic policies may be under-stood With the problems created by unilateral policy choicesidentied we then use this model in the remainder of the paperto characterize bargaining outcomes under alternative negotiat-ing structures

A The Economic Environment

We begin with a description of the economic environment Wework within a two-sector two-country perfectly competitive gen-eral equilibrium trade model modied to capture the issue ofnational standards In particular in addition to tariffs we allowthat a government may wish to impose standards of variouskinds possibly reecting ldquosocial concernsrdquo but possibly also toaddress externalities associated with the private production andconsumption decisions of its citizens We restrict the underlyingmotives for standards-setting to reect national issues Whereglobal (nonpecuniary) externalities arise international negotia-tions are clearly warranted but as we observed in the Introduc-tion these need not require WTO involvement Our analysis per-tains to standards issues that become an international concern togovernments as a result of their trading relationships

Two countries home (no ) and foreign () trade two goodsx and y taken to be normal goods in consumption and producedunder perfect competition Let x( y) be the natural import good ofthe home (foreign) country and dene p ordm pxpy ( p ordm pxpy) tobe the local relative price facing home (foreign) producers andconsumers Local relative prices may differ across the two coun-


tries as a result of the tariff policies of each government With t(t) representing the home (foreign) ad valorem import tariffwhich we take to be nonprohibitive and with t ordm (1 + t) and t ordm (1 + t) we have p = t pw ordm p( t pw ) and p = pw t ordmp( t pw ) where pw ordm pxpy is the ldquoworldrdquo (ie untaxed) relativeprice The foreign (home-country) terms of trade are then mea-sured by pw (1pw ) We interpret t gt 1 ( t lt 1) to be an import tax(import subsidy) and similarly for t 5

In the usual way each countryrsquos import demands and exportsupplies can be expressed as functions of its local relative priceand the terms of trade but we now also introduce the possibilitythat these functions may be affected by a countryrsquos choice ofstandards6 We denote by s the home-country standard with thestandard in the foreign country denoted by s

These standards could take the form of production stan-dards corresponding to a countryrsquos legal minimum (or maximum)working age its legal minimum real wage or the maximum legalemissions level per unit of output where any of these might beapplied to a particular sector or on an economywide basis Suchproduction standards could potentially alter the shape of a coun-tryrsquos production possibilities frontier and hence for given localprices its production choices7 A countryrsquos production standardscould also alter its consumption choices for given local and worldprices by affecting the level and distribution of factor income inthe economy And in altering its production and consumptionchoices for any given local and world prices a countryrsquos produc-tion standards may thereby affect its import demands and exportsupplies But as we have ruled out international nonpecuniaryexternalities by assumption the production standard set by onecountry will not affect directly (ie in a nonpecuniary fashion) theimport demands and export supplies of its trading partner theonly effects on these magnitudes travel through prices Alterna-tively these standards could take the form of consumption stan-

5 By the Lerner symmetry theorem trade taxes can be equivalently depictedas applying to exports or imports

6 In assuming that each countryrsquos import demands and export supplies arefunctions we are abstracting from the possibility of multiple equilibria at thenational level and are thus implicitly placing limits on the kind and degree ofdistortions that are allowed in each country We also abstract from the possibilitythat these functions may be nondifferentiable Our analysis can be extended in anatural way to handle these complications

7 An economyrsquos production decisions could also depend on world prices viaincome effects if factor supplies were endogenous For simplicity and in keepingwith most of the literature we abstract from this possibility here


dards corresponding to national restrictions on the consumptionof products made in a particular way or possessing a particularattribute (eg a ban on the consumption of hormone-treatedbeef) A countryrsquos consumption standards can thereby affect itsimport demands and export supplies as well But again in theabsence of international nonpecuniary externalities all interna-tional effects travel through prices and so the consumption stan-dard set by one country will have no direct effects on the importdemands and export supplies of its trading partner8 Hence s ands act as ldquoshiftrdquo parameters in the import demand and exportsupply functions of the home and foreign country respectively

To complete our characterization of the economic environ-ment we introduce notation for imports and exports so that thetrade balance and equilibrium conditions may be expressed Forthe home country imports of x are represented as Mx(sppw )while Ey(sppw ) denotes home-country exports of y Foreign-country imports of y My and exports of x Ex are similarlydened Home and foreign budget constraints imply that for anyworld price we have

(1) pwMx(sp( t pw)pw) 5 Ey(sp( t pw)pw)My(sp( t pw)pw) 5 pwEx(sp( t pw)pw)

where we now represent explicitly the functional forms of thelocal prices Finally the equilibrium world price pw ( t s t s) isdetermined by the x-market-clearing condition

(2) Mx(sp( t pw)pw) 5 Ex(sp( t pw)pw)with market clearing for good y then implied by (1) and (2)

In summary given national standards in each country and apair of tariffs the equilibrium world price is implied by (2) andthe equilibrium world price and the given tariffs then togetherdetermine the local prices In this way the national standardsand tariffs imply local and world prices and thereby the levels forproduction consumption imports exports and tariff revenue

8 In interpreting s as a consumption standard some care needs to be takennot to exceed the dimensionalityof the model since the process by which goods areproduced in one country may distinguish them from the point of view of theconsumption standard in the other Thus for example a national ban on con-sumption of hormone-treated beef in the domestic country could lead to theproduction of hormone-free beef for sale in the domestic-country market withhormone-treated beef produced for sale in the foreign country Hence each coun-try could potentially produce two kinds of beef for sale at two distinct prices Thiscould be handled in our two-good setting by letting good x be hormone-treated beefand good y be hormone-free beef


Finally we assume that the Marshall-Lerner stability conditionsare met so that an inward shift of the domestic (foreign) importdemand curve results in a lower (higher) equilibrium world priceWe further add the restrictions that dpd t gt 0 gt dpd t andshy pw shy t lt 0 lt shy pw shy t which ensure that the equilibrium pricesdo not exhibit the Lerner or Metzler paradoxes

B Government Objectives

We next describe government preferences While it is cus-tomary to represent a governmentrsquos payoff (ie welfare) directlyin terms of the underlying choice variables (ie tariffs and na-tional standards) we choose to represent government objectivesin a somewhat different manner extending the approach taken inBagwell and Staiger [1999a] in order to incorporate the presenceof national standards To this end we represent governmentpreferences over tariffs as preferences instead over the local andworld prices that the tariff choices imply for given standardslevels similarly we separate government preferences over stan-dards into direct preferences over national standards and prefer-ences over the world prices that standards choices imply for giventariff levels This approach to representing government objectivesenables us to isolate the terms-of-trade externality that tariff andstandards selections generate We thus represent the objectivesof the home and foreign governments by the general functionsW(sp( t pw )pw ) and W(sp( t pw )pw ) respectively

Notice that each government cares about the policy choices ofits trading partner only indirectly through the effects that thesechoices have on world prices This structure reects two under-lying features of the environment set out above First our exclu-sion of global social concerns and international nonpecuniaryexternalities implies that governments have no direct reason tocare about the policy choices of their trading partners And sec-ond the nature of international economic interaction ensuresthat all indirect effects of a governmentrsquos policy choices on theeconomy of its trading partner are channeled through worldprices

The only additional structure we place on W and W is thatholding its local price and its national standards xed eachgovernment achieves higher welfare when its terms of tradeimprove

(3) shy W(sppw) shy pw 0 and shy W(sppw) shy pw 0


We illustrate this structure with Figure I which depicts iso-local-price and iso-world-price loci as a function of home and foreigntariff levels given xed national standards in each country Withstandards levels held xed an initial tariff pair A ordm ( t t ) isassociated with a domestic iso-local-price locus p( A) reg p( A)and an iso-world-price locus pw ( A) reg pw ( A)9 Also depicted is asecond iso-world-price locus pw(C) reg pw (C) along which theworld price is lower than at point A indicating an improvedterms of trade for the home country A reduction in the worldprice that maintains the home-country local price is thus

9 Given the assumptions that Metzler and Lerner paradoxes are absent theiso-local-price locus exhibits negative slope and the iso-world-price locus is posi-tively sloped in Figure I

FIGURE IThe World- and Local-Price Effects of a Tariff Change


achieved (for xed standards) with the movement from point A toB corresponding to a higher (lower) home-country (foreign-coun-try) import tariff We assume only that the home-country govern-ment values the implied income transfer

To explore the generality of our representation of govern-ment objectives we note that the structure imposed on govern-ment preferences by (3) states that a government would alwaysstrictly prefer a terms-of-trade improvement which allowed it toprovide lump-sum distributions of additional income to its con-sumers if this terms-of-trade improvement could be accom-plished without altering any of the following (i) the economyrsquoslocal relative price and production standards faced by producersand therefore the economyrsquos production decisions (ii) the leveland distribution of factor income in the economy or (iii) theeconomyrsquos local relative price and consumption standards facedby consumers From a political economy perspective the assump-tion that a government would benet from a terms-of-trade im-provement of this nature seems benign in light of (ii) because thelevel and distribution of a countryrsquos factor income is being heldxed as its terms-of-trade improve under (3) Indeed we haveargued elsewhere [Bagwell and Staiger 1999a] that each of themajor approaches to the political economy of trade policy satisesan assumption of this nature10 This assumption would also seemto be satised in most environments where a government had adistinct reason to intervene in the production decisions of theeconomy as it might for example if pollution was a by-product ofthe production process because by (i) all production decisions inthe economy are being held xed as its terms-of-trade improveunder (3) We do note however that this assumption is perhapsmore restrictive when a government has a distinct reason tointervene in the consumption decisions of the economy sincenational consumption depends on pw through income effects Forinstance negative externalities associated with consumption of aparticular good could be exacerbated by the added consumption

10 Intuitively each of the approaches to the political economy of trade policyamounts to specifying government preferences over the levels and distributions offactor income that can be achieved with different tariff levels and the levels anddistributions of factor income are in turn determined by the local prices that agiven tariff level implies By leaving government preferences over local pricesunrestricted we thus ensure that our results apply regardless of the underlyingapproach to political economy that one prefers For further elaboration on thesepoints see Bagwell and Staiger [1999a]


opportunities that additional tariff revenue affords and if thiseffect is strong then (3) might be violated

C Efcient Policy Choices

We rst characterize efcient policy choices Any efcientcombination of policies ( t E sE t E sE ) solves

(I) maxt s t s

W(sp( t pw)pw)

such that W(sp(t pw)pw) $ W Ewhere W E ordm W(sE p( t E pw E )pw E ) and pw E ordm pw ( t E sE t E sE ) The set of efcient policy combinations is dened as theset of solutions to the rst-order conditions associated with (I)which with some manipulation can be represented as11

(4) Ws S 1shy pw shy s D 5 WpS pw

shy pw shy t D

(5) Ws S 1shy pw shy s D 5 Wp S 2 p t

shy pw shy t D

and

(6) (1 2 AWp)(1 2 AWp) 5 1

where A ordm (1 2 t l ) (Wp + l Wp w) and A ordm (1 2 l t )(Wp +l Wp w) and where with the Metzler and Lerner paradoxes ruledout

l ordm[ shy pw shy t ]

[dpd t ] 0 l ordm

[ shy pw shy t ]

[dpd t ] 0

Conditions (4) and (5) can be interpreted as ldquonationalrdquo ef-ciency conditions Condition (4) says that at an efcient policycombination any small changes in t and s which together leavethe equilibrium world price unchanged must leave home welfareunchanged as well12 Similarly condition (5) says that at anefcient policy combination any small changes in t and s whichleave the equilibrium world price unchanged must leave foreignwelfare unchanged as well

11 We assume throughout that policy choices correspond to interior solu-tions of the relevant maximization problems

12 Changes in s and t which keep pw xed must satisfy d t ds = ( 2 shy pw shy s)( shy pw shy t ) Efciency requires that no change in home-government welfare can beinduced by such policy changes or that Ws + Wp[ pw ( 2 shy pw shy s)( shy pw shy t )] = 0which yields (4)


Conditions (4) and (5) are illustrated in the top left and rightpanels of Figure II respectively Here and throughout we willillustrate our results for the case in which shy pw shy s gt 0 and shy pw shy slt 0 ie an increase in the national standard would worsen thecountryrsquos terms of trade In this case the iso-world-price locus ispositively sloped in each panel Note from our representations ofW and W that in each of these panels the iso-world-price locusis also the indifference curve of the other country This furtherclaries why (4) and (5) are necessary for efciency they indicatethat each government should set its own policies where it wouldnot be possible for it to benet from a small change in its policiesthat kept its trading partner indifferent Note also from (2) that

FIGURE IIEfcient Policy Choices


unilateral changes in policy mix which leave the equilibriumworld price unaltered must leave equilibrium trade volumes un-altered and hence the iso-world-price locus in each of the toppanels is also an iso-equilibrium-trade-volume locus Conditions(4) and (5) therefore ensure that each government is utilizing itspolicies efciently in light of its own preferences and the equilib-rium trade volume

Condition (6) now may be interpreted as the ldquointernationalrdquoefciency condition as it ensures that policies are set so that theequilibrium trade volume is indeed efcient The bottom panel ofFigure II illustrates a choice of t and t that satises condition (6)given that each country is choosing its policy mix to satisfy itsrespective national efciency condition

D Noncooperative Policy Choices

We next characterize the noncooperative Nash policy choicesIf governments do not cooperate over policies then for any set offoreign policy choices the domestic government chooses its poli-cies to solve

(II) maxt s

W(sp( t pw)pw)

Similarly for any set of domestic policies the foreign governmentchooses its policies to solve

(II) maxt s

W(sp( t pw)pw)

The Nash equilibrium choices are dened as a set of policies( t N sN t N sN ) which jointly satisfy the rst-order conditionsassociated with (II) and (II)

(7) WsS 1shy pw shy s D 5 2 [ t Wp 1 Wpw]

(8) Wp 1 l Wpw 5 0

(9) Ws S 1shy pw shy s D 5 2 F 1

t Wp 1 Wpw G

and

(10) Wp 1 l Wpw 5 0

To interpret these conditions consider (7) and (8) whichdene the home countryrsquos best-response policy choices as a func-tion of foreign country policies Observing that 2 [ t Wp + Wp w]


gives the impact on home government welfare of a small decreasein pw when the home tariff is held xed condition (7) dictatesthat the home government will set its national standard so thatthe direct effect on its welfare of a small change in its standard isjust offset by the indirect effect on its welfare that the inducedworld price movement would imply A similar interpretationapplied to the home governmentrsquos tariff choice holds for condition(8) which dictates that the home government will set its tariff sothat the welfare effect of a small change in the local price inducedby a change in its tariff is just offset by the indirect welfare effectthat the world price movement induced by this tariff changewould imply Note also that as l lt 0 and as Wp w lt 0 by (3)condition (8) implies Wp lt 0 so that the home government isinduced by the terms-of-trade effects of its policy choices to pro-vide greater protection to its import-competing sector (and there-fore a higher local relative price p) than it would choose to providebased on the local price effects of its tariff choice alone Similarlywith Wp lt 0 by (8) condition (7) implies that the home govern-ment will be induced by the world-price effects of its policy choicesto adopt national standards which are more favorable to its termsof trade than it would choose to adopt based on the direct impactof these standards on its welfare Analogous statements applywith respect to the interpretations of (9) and (10)

Consider now the efciency properties of the Nash equilib-rium Conditions (7) and (8) determine the best-response home-country policies to a set of foreign policies and these two condi-tions together imply that (4) is satised Likewise conditions (9)and (10) determine the best-response foreign policies to a set ofhome-country policies and these two conditions imply that (5) issatised Therefore conditional on the Nash trade volume eachgovernment is making efcient use of its policies That is eachgovernment is choosing a policy mix that satises its nationalcondition for efciency This is illustrated in the top two panels ofFigure III where at the Nash policies each governmentrsquos welfareis maximized with respect to choices over its own policies and soany small changes in its policiesmdashincluding those that preservethe equilibrium world pricemdashmust leave its welfare unchangedas the national efciency conditions (4) and (5) require

But conditions (8) and (10) violate (6) and therefore Nashpolicies are inefcient because the international condition forefciency is not met Hence Nash policies are inefcient because


of the inefcient equilibrium trade volumes they imply as thebottom panel of Figure III illustrates

In fact the inefciency can be interpreted as indicating thatNash policies result in ldquomarket accessrdquo levels that are too low Tomake this connection we dene the market access with which acountry provides its trading partner by the volume of imports itwould accept at a particular world price Thus for a particularworld price pw the domestic market access afforded by the do-mestic-country policy combination T ordm ( t s) is given byMx(sp( t pw )pw ) and similarly at pw the foreign market accessafforded by the foreign-country policy combination T ordm ( t s)is given by My(sp( t pw )pw )

FIGURE IIINash Policy Choices


We will say that a government secures additional marketaccess from its trading partner through negotiations if there ex-ists a world price such that the trading partnerrsquos negotiatedpolicy changes provide additional access to the trading partnerrsquosmarket (ie if the trading partnerrsquos import demand curve shiftsout for at least some world price) According to this denition ifthe domestic government failed to secure additional market ac-cess from its foreign trading partner through negotiations thenthe foreign import demand curve would shift in (weakly) at allworld price levels and given our stability assumptions the ne-gotiated foreign-country policy changes would contribute towarda rise (weakly) in the equilibrium world price pw

We now establish that beginning from the Nash equilibriumeach government must secure additional market access from itstrading partner in order to reach a mutually benecial agree-ment In this way we show that Nash market access levels areinefciently low

To see this consider the impact of foreign policy changes onthe domestic welfare

dWd t

1dWds

5 [ t Wp 1 Wpw] F shy pw

shy t 1

shy pw

shy s G

When the domestic-country government is on its reaction curvesas dened by (7) and (8) we have

(11)dWd t

1dWds

5 [1 2 t R(T) l ]Wpw F shy pw

shy t 1

shy pw

shy s G

Hence using (3) and (11) and recalling that l lt 0 along thedomestic governmentrsquos reaction curves any small change inforeign policies that fails to offer additional foreign-marketaccess must (weakly) reduce domestic government welfare asit (weakly) increases pw

Consider next an agreement that species the domestic andforeign policy vectors (T0 T0) and suppose that the foreignpolicy vector T0 fails to offer additional foreign-market accessrelative to TN the vector of Nash foreign policies Let W0 denotedomestic welfare at the policy vector (T0 T0) WN denote domes-tic welfare at the Nash policy vector (TN TN ) and WR(T)denote domestic welfare at the domestic best-response policyvector (TR (T)T) Then for any T0 specied in the agreementwe must have W0 WR (T0) WR (TN ) = WN so that thedomestic-country government must be worse off (weakly) under


any agreement which species the foreign policy vector T0 13 Asimilar argument holds with respect to the foreign governmentHence each government must secure additional market accessfrom its trading partner in order to reach a mutually benecialagreement14

Finally we may inquire into the reasons that governmentsare led through their unilateral decisions to restrict market ac-cess to inefciently low levels It should come as no surprise thatthe terms-of-trade consequences of unilateral policy choices rep-resent one source of inefciency However we now ask whetherthere are additional distortions in this setting that keep govern-ments from the efciency frontier when making unilateral policydecisions To explore this possibility we follow our earlier work[Bagwell and Staiger 1999a] and imagine a world in which gov-ernments are not motivated by the terms-of-trade implications oftheir policy choices and dene the resulting politically optimalpolicies ( t po spo t pospo) as the solution to

(12a) Ws 5 0 Wp 5 0and

(12b) Ws 5 0 Wp 5 0The political optimum corresponds to the decisions governmentswould have made if they had not been concerned with exploitingtheir power over the terms of trade But together (12a) and (12b)satisfy (4)ndash(6) and therefore politically optimal policies are in-deed efcient Figure IV illustrates the way in which politicallyoptimal policies satisfy these three efciency conditions As thebottom panel of the gure depicts at the political optimum smalladjustments in each countryrsquos tariff that preserve the world pricewill leave the welfare of each government unchanged Thusterms-of-trade manipulation is the problem that keeps Nash pol-icy choices from reaching the efciency frontier

13 The second inequality reects the following logic Beginning on the do-mestic reaction curve with policies (TR (T0 )T0) construct a policy path to theNash policies (TN TN) Along this path as T is adjusted set T along thedomestic reaction curve Envelope arguments ensure that the resulting changes inT have no rst-order impact on W The T changes however result (weakly) in anincrease in foreign-market access causing a (weak) reduction in the equilibriumworld price and thereby an increase (weak) in W (by (11))

14 If in addition it is assumed that policy changes shift the import demandfunction in the same direction for all world prices then a mutually benecial tradeagreement implies that each government secures additional equilibrium importvolumes from its trading partner


Of course the political optimum is just one point on theefciency frontier More generally any combination of policiessatisfying (4)ndash(6) is efcient and the efciency frontier is the setof all welfare pairs associated with policy combinations satisfying(4)ndash(6) We summarize with the following proposition

PROPOSITION 1 Nash policy choices are inefcient and the incen-tive to manipulate the terms of trade is the source of theinefciency This incentive does not distort the policy mixchosen by each government but Nash market access levelsare inefciently low and each government must secure ad-ditional market access from its trading partner in order toreach a mutually benecial agreement

FIGURE IVPolitically Optimal Policy Choices


As Proposition 1 indicates the terms-of-trade effects of uni-lateral policy choices are in fact the fundamental source of inef-ciency that governments can correct through international ne-gotiations in our formal setup and it is therefore fair to say thatcontending with the terms-of-trade motives of governments is thefocus of our subsequent analysis Yet real governments rarelydiscuss in any explicit way such abstract notions as the terms-of-trade consequences of their decisions and the attraction thesegovernments have to international trade negotiations seems inany event to reect a simple mercantilist desire for export mar-kets It is therefore worth pausing to interpret the terms-of-tradeeffects in more familiar terms lest it be concluded that ourframework while general is incapable of capturing the underly-ing forces at work in actual trade negotiations

In this regard it is important to observe that the terms-of-trade effects of a governmentrsquos policy choices refer simply to itsability to shift the costs of its policies onto trading partners Thiscost-shifting will occur provided only that some of the incidenceof a governmentrsquos policies are borne by foreign exporters Thusfor example when a domestic government offers protection to animport-competing industry some of the costs of that protectionare shifted abroad if foreign exporters accept lower (fob) pricesfor their sales in the domestic market When such cost-shiftingdoes occur it is natural to expect that governments distort theirpolicy choices as they do not bear the whole cost of their deci-sions Consequently when viewed from the perspective of cost-shifting terms-of-trade effects can be seen to represent a naturalsource of inefciency associated with unilateral policy decisions15

At the same time these effects can also help to provide an eco-

15 This logic is sometimes raised in discussions of standards and tradepolicy although it is not recognized as a terms-of-trade argument For example indiscussing the introduction of a new clean-air standard for gasoline Roessler[1998 p 222] observes ldquoA problem of WTO consistency would arise however ifthe domestic political constraints are such that a new standard would secure aparliamentary majority only if domestic gasoline is exempted from the standardfor ve years or to put the issue in political-economy terms if the cost of reducingpollution is initially borne only by nonvoting producers abroadrdquo On cost-shiftingmotives and their relation to GATT more generally see for example Jackson[1989 p 19] who observes ldquoMore subtle is the possibility that a national con-sensus could explicitly opt for a choice of policies that would not maximize wealth(in the traditionally measurable sense at least) but would give preference toother non-economic goals It can be argued that when a nation makes anldquouneconomicrdquo choice it should be prepared to pay the whole cost and not pursuepolicies which have the effect of unloading some of the burdens of that choice onto other nations In an interdependent world paying the whole cost is not ofteneasy to accomplishrdquo


nomic explanation for the mercantilist orientation of actual ne-gotiations for as Proposition 1 indicates they imply that eachgovernment is right to pin its hopes for a benecial outcome ofnegotiations on its ability to gain enhanced access for its export-ers to the markets of its trading partner (see Bagwell and Staiger[1999a] for an elaboration on these points)

Returning now to the results summarized in Proposition 1 itis clear that direct negotiations over ( t s t s) could allow gov-ernments to move to a point on the efciency frontier But candirect negotiations over tariffs alone be structured so as to gen-erate outcomes on the efciency frontier as well This is thequestion to which we now turn

III TARIFF NEGOTIATIONS DOMESTIC POLICIES AND WTO RULES

In this section we consider the properties of bargaining out-comes under various negotiating structures We begin in the nextsubsection by showing that negotiations over tariffs alone willlead to inefcient outcomes in the absence of any restraints ondomestic policy choices and we identify why this is so We thenturn in the following subsections to the task of formally modelingand evaluating the impact of the restraints that the WTOrsquos ex-isting rules place on these choices

A Unrestricted Sovereignty over National Standards

We rst consider ldquounrestricted sovereigntyrdquo over nationalstandards That is we suppose that governments negotiate overtariffs but that each government retains the unrestricted right tomake unilateral adjustments to its national standards in thefuture In effect governments cooperate over tariffs aware thatnational standards will then be set noncooperatively16

Here and throughout this section we will follow WTO prac-tice and depict the tariff commitments that countries make

16 Copeland [1990] formally explores a negotiating structure analogous toours In a setting where governments are assumed to be national-income maxi-mizers he shows that tariff negotiations can be benecial even if other instru-ments are ldquonon-negotiablerdquo Our emphasis here is different we simply wish toestablish that this negotiating structure cannot deliver governments to the ef-ciency frontier and to evaluate why this is so and we then move on to consider theefciency properties of alternative negotiating structures At a general levelBhagwati [1996 pp 23ndash24] has also observed that following a trade concessiona country may have incentive to impose offsetting changes to its domestic policiesand that this incentive could have important consequences for the properties ofvarious negotiating structures


through negotiated agreement as bindings The tariff bindingsthat a country agrees to in a WTO negotiation dene the maximaltariff levels that it can legally apply The legal right to set tariffsat an applied rate that is lower than the bound rate is importantlater in this section and so for consistency we allow for thispossibility in this subsection as well

Starting from any negotiated pair of tariff bindings ( tt)the domestic-country government makes use of its unrestrictedsovereignty over national standards to solve the followingproblem

(III) maxt s

W(sp( t pw)pw)

such that t tComparing (III) with (II) it is clear that if the negotiated tariffbinding t does not constrain the domestic governmentrsquos (applied)tariff choice then its policy choices satisfy the rst-order condi-tions dening its unconstrained best-response given by (7) and(8) Otherwise the domestic government sets t at its bound levelt and (7) implicitly denes sR(s t t )17

Similarly the foreign government will solve

(III) maxt s

W(sp( t pw)pw)

such that t tComparing (III) with (II) it follows that if the negotiated tariffbinding t does not constrain the foreign governmentrsquos (applied)tariff choice then its policy choices satisfy (9) and (10) the rst-order conditions dening its unconstrained best-response Other-wise the foreign government chooses to set t at its bound levelt and (9) implicitly denes sR (s t t)

With equilibrium domestic and foreign policy choices thendetermined by the joint solutions to the relevant best-responsefunctions we can now state the next proposition

PROPOSITION 2 When governments retain unrestricted sover-eignty over their national standards agreements to reducetariff levels create an incentive to restrict market access andmanipulate the terms of trade through domestic policy

17 This observation relies on a slight strengthening of our assumptions Weassume henceforth that W(sp( t pw )pw ) is globally concave in t with an analo-gous assumption for foreign-government welfare


choices and therefore tariff negotiations cannot achieve ef-cient outcomes

Proof Observe that the national standards choices for anytariffs satisfy (7) and (9) Efciency on the other hand requiresthat conditions (4) through (6) are satised But together with (7)efciency condition (4) implies that the Nash condition (8) mustalso be satised while together with (9) efciency condition (5)implies that the Nash condition (10) must hold Hence if ef-ciency conditions (4) and (5) are to hold all four Nash conditions(7) through (10) must hold as well But as established previouslythis implies that the nal efciency condition (6) must be vio-lated

QED

Intuitively any attempt to use tariff negotiations alone tomove from the inefcient Nash equilibrium to a point on theefciency frontier simply shifts governmentsrsquo incentives to ma-nipulate their terms of trade on to their national standardschoices which are then manipulated to reduce market accessFigure V illustrates a case in which governments attempt toimplement a set of efcient tariffs and national standards( t E sE t E sE ) through negotiations over tariffs alone As beforewe illustrate the case in which shy pw shy s gt 0 and shy pw shy s lt 0 sothat an increase in the national standard would worsen thecountryrsquos terms of trade The bottom panel of the gure illus-trates the international efciency condition (6) which must holdunder the efcient policy combination ( t E sE t E sE ) To achievethis policy combination the two governments might agree to bindtheir tariffs at the efcient levels ( tE = t E tE = t E ) in thehope that each would then follow through with the efcient re-spective national standards levels (sE sE ) satisfying (4) and (5)But with unrestricted sovereignty over national standardschoices this hope would be in vain having granted the additionalmarket access that mutually benecial negotiations demand(Proposition 1) each government would then have an incentive toreclaim unilaterally a portion of this market access with itsnational standards choice (Proposition 2) For example as illus-trated in the top left panel of the gure if the foreign governmentfollowed through with its efcient tariff and standards choicesthe domestic government would lower its standard below sE to itsbest response sR(sE t E tE ) where with its tariff bound at t E ithad reclaimed an optimal amount of market access by distorting


its national standards choice ie by violating the national ef-ciency condition (4) An analogous incentive exists for the foreigngovernment as illustrated in the top right panel of Figure V

Therefore if governments negotiate tariff agreements butare granted unrestricted sovereignty over their national stan-dards choices then inefciencies remain and these inefcienciesnow extend beyond issues of inadequate market access to includedistorted national standards choices

B WTO Rules Nonviolation Complaints

If the problem with unrestricted sovereignty over nationalstandards is that governments take advantage of their sover-

FIGURE VNegotiations over Tariffs Alone


eignty to distort market access to inefciently low levels why notsimply restrict their sovereignty to choices over policy combina-tions that do not reduce market access from the levels implied bytariff negotiations This is the essential logic behind the WTOrsquosrules as they apply to the domestic policy choices of membergovernments In this and the next subsection we explore hownegotiated tariff bindings the ability to renegotiate these bind-ings and the right to bring nonviolation complaints can interactto address the inefciencies identied above

We attempt to capture the implications of these rules for-mally in two steps First in this subsection we introduce a simpletwo-stage tariff negotiating structure in which the role of tariffbindings and the right to bring nonviolation complaints are high-lighted In this simple structure nonviolation complaints work toensure that the level of market access commitments implied bytariff negotiations is not eroded by subsequent changes in domes-tic policies This framework serves to illustrate our main pointsThen in the next subsection we describe an extension of thistwo-stage negotiating structure to three stages which allowsgovernments as well the opportunity to renegotiate tariff bind-ings as the WTO permits This extended three-stage negotiatingstructure captures an additional feature of WTO rules which isthat these rules prevent erosion of the balancemdash but not neces-sarily the levelmdashof market access commitments implied by tariffnegotiations We argue that our main results are preserved inthis extended model (and refer readers to our working paper for aformal development of this argument)

We begin with some denitions We say that for any recordedlevel of standards s0 and s0 a negotiated pair of tariff bindings( t t) implies a world price p w and a level of domestic and foreignmarket access commitments M x and M y respectively wherep w ordm pw ( t s0 ts0) M x ordm Mx(s0 p( t p w )p w ) and M y ordmMy(s0 p( tp w )p w ) We may now formally dene the Two-Stage Tariff Negotiation Game As tariff negotiations commencenational standards are initially recorded at their existing levelss0 and s0 Governments then proceed in two stages

Stage 1 Governments bargain over tariffs a pair of tariffbindings ( t t) is determined and a world priceand market access commitments p w M x and M yare implied

Stage 2 Each government is entitled to make unilateral ad-justments to its policy mix so long as (i) its tariff does


not exceed its bound level and (ii) its policy adjust-ments do not erode its implied market accesscommitments

Stage 1 corresponds to tariff negotiations under GATTrsquos Ar-ticle XXVIII bis eg a WTO ldquoRoundrdquo of tariff negotiations Stage2 reects the freedom governments have to adjust unilaterallytheir trade and domestic policies subsequent to tariff negotia-tions This freedom is constrained in two ways First the appliedtariff that each government implements cannot exceed its boundrate as determined in Stage 1 This reects the legal commitmentembodied in a WTO tariff binding And second governments areprevented from altering their policy mix in a way that woulderode their implied market access commitments from Stage 1This reects the constraint imposed on each government by thelegal right of its trading partner to bring a nonviolation complaintunder Article XXIII and seek redress if it alters its policies insuch a way as to reduce access to its markets below that impliedby the outcome of Stage 1 We use the denition of market accessintroduced earlier and evaluate changes in market access at theimplied world price Hence we explore here the consequences ofthe following supposition the prospect of nonviolation ArticleXXIII complaints prevents each government from adjusting its(Stage 2) policies in a way that would reduce market access at theinitial (Stage 1) world price18

To determine the properties of this two-stage procedure con-sider rst the problem that each government solves in the secondstage Starting from ( t s0 ts0) the domestic-country govern-ment is permitted to adjust its policy mix subsequent to tariffnegotiations to solve

18 This presumes that the right to bring Article XXIII complaints extendsbeyond trade volumes to prices as well If market access assurances were simplya matter of assured equilibrium import volumes then the strong propertiesassociated with nonviolation complaints that we establish below would have to bequalied (see Winters [1997] Srinivasan [1997] and Bagwell and Staiger[1999b]) However this presumption nds some support in the legal argumentssurrounding Article XXIII disputes For example a WTO Panel Report regardingthe Article XXIII (violation) complaint brought by the United States and othersagainst the EC ldquobanana regimerdquo stated ldquoThe Hawaiian producers had expressedtheir concerns that the EC banana regime was lowering the price of bananas inthe free market adversely affecting their ability to continue to produce andpotentially export bananasrdquo [WTO 1999a] Similarly a WTO Panel Report regard-ing the Article XXIII (violation) complaint brought by Thailand and others againstthe U S embargo on shrimp imports in the sea-turtle dispute stated ldquoIn shortthe embargo had two adverse trade effects it reduced the total volume and theaverage unit value of shrimp exported to the United Statesrdquo [WTO 1999b p 69]


maxt s

W(sp( t pw)pw)

(IV) such that (i) t t and

(ii) Mx(sp( t p w)p w) $ M x

The rst-order conditions associated with (IV) dene the domesticgovernmentrsquos best-response policy mix for any set of foreign pol-icies Likewise the foreign-country government is permitted toadjust its policy mix subsequent to tariff negotiations to solve

maxt s

W(sp(t pw)pw)


(ii) My(sp( t p w)p w) $ M y

The rst-order conditions associated with (IV) dene the foreigngovernmentrsquos best-response policy mix given a set of domestic-country policies Given a set of recorded standards and a Stage 1negotiation outcome the equilibrium Stage 2 domestic and for-eign policy choices are then determined by the joint solution tothese best response functions

Consider now the domestic countryrsquos best-response policies toany foreign policy pair ( tˆs) that meets the foreign marketaccess commitments exactly (ie ( tˆs) satises constraint (ii)with equality) In this case (1) and (2) imply that domesticpolicies ( t s) satisfy constraint (ii) with equality if and only ifpw ( t s tˆs) = p w And under our stability assumptions con-straint (ii) is satised with strict inequality in this case if and onlyif the associated domestic-country policy choices imply an equi-librium world price pw ( t s tˆs) which exceeds p w Hence incalculating its best response to any foreign policies ( tˆs) thatmeet the foreign market access commitments exactly the domes-tic governmentrsquos problem (IV) may be rewritten as

maxt s

W(sp( t pw)pw)

(IV 9 ) such that (i) t t and

(ii 9 ) pw( t stˆs) $ p w

Evidently in this case the prospect of an Article XXIII nonviola-tion complaint against it is sufcient to prevent the domesticgovernment from altering its policies subsequent to tariff negotia-


tions in a way that would improve its terms of trade Recallingnow from Proposition 1 that terms-of-trade manipulation is theunderlying source of inefciency that negotiations can correctand from Proposition 2 that an agreement on tariffs alone willshift the incentive to manipulate the terms of trade on to nationalstandards choices it may be anticipated that the ability to bringnonviolation complaints potentially has an important efciency-enhancing role19

To explore this role we ask whether points on the efciencyfrontier can be reached with appropriate Stage 1 outcomes inlight of the Stage 2 adjustments that may be anticipated Givenan existing set of recorded standards we will say that a policycombination can be implemented under tariff negotiations if thereexists a pair of tariff bindings such that this policy combinationcorresponds to a Nash equilibrium in Stage 2 of the Two-StageTariff Negotiation Game20

Let us consider then a combination of policies( t E sE t E sE ) satisfying (4)ndash(6) and ask whether given a set ofrecorded standards this efcient policy combination can be im-plemented under tariff negotiations Associated with this combi-nation are domestic and foreign import volumes and a worldprice which we respectively denote by Mx

E MyE and pw E

Suppose then that Stage 1 negotiations were to result in a pairof tariff bindings ( t E tE ) dened implicitly by M x = Mx

E andM y = My

E and note that these bindings would imply a worldprice p w = pw E Suppose further for the moment that tE $ t E

and tE $ t E so that these bindings are set above the tarifflevels associated with the efcient policy combination under con-sideration Then it follows from (IV) and (IV) that( t E sE t E sE ) can be implemented under tariff negotiationsprovided only that at this efcient policy combination we havedWd t lt 0 and dWd t lt 0 (ie provided that at this efcientpolicy combination each government would be hurt by a unilat-eral increase in the tariff of its trading partner) To see this note

19 In this regard the role of nonviolation complaints that we highlight herebears a relationship to the well-known results of Kemp and Wan [1976] In theirclassic paper Kemp and Wan showed how the membership of a customs unioncould always be increased in such a way as to raise the national income of membercountries without reducing the national income of any nonmember country Thecustoms union need only adjust its (common) external tariffs to neutralize theimpact of the addition of new members on the external terms of trade withremaining nonmembers

20 We do not explore conditions for the uniqueness of the Stage 2 Nashequilibrium


rst that it is feasible for the domestic government to select( t E sE ) and for the foreign government to select ( t E sE ) inStage 2 And second note that these selections are best-re-sponses since the resulting policy choices (a) are efcient (b)satisfy dWd t ordm [ t Wp + Wpw] shy pw shy t lt 0 and dWd t ordm[Wp t + Wp w] shy pw shy t lt 0 which implies in turn that [ t Wp +Wp w] lt 0 and [Wp t + Wp w] gt 0 and (c) meet the domestic andforeign market access commitments exactly Consequently to dobetter for itself a government would have to hurt its tradingpartner by (a) and therefore by (b) would have to worsen theterms of trade of its trading partner from p w but by (c) would beprevented from doing so as a result of constraint (ii 9 ) or itsforeign-government analogue

Hence any efcient combination of policies ( t E sE t E sE )satisfying the restrictions (a) dWd t lt 0 and dWd t lt 0 and(b) tE $ t E and tE $ t E can be implemented under tariffnegotiations We now consider these restrictions in more detailso as to better assess the limitations associated with attempts toimplement efcient outcomes with tariff negotiations in thissetting

Consider rst the restriction that dWd t lt 0 and dWd t lt0 It is straightforward to show that a sufcient condition for thisrestriction to be met by an efcient policy combination is that theefcient policies require each government to bind its tariff belowits best-response tariff21 This rst restriction is met then for allefcient policy combinations at which each government agrees torestrain the unilateral desire to raise its tariff Given the natureof WTO tariff bindings (ie dening a maximal tariff level) afocus on efcient policy combinations that satisfy this propertydoes not seem particularly limiting Therefore in light of theWTOrsquos approach to tariff bindings we will henceforth refer toefcient policy combinations that share this property as efcientcombinations of tariff bindings and standards policies and wewill focus on the feasibility of implementing such policy combina-tions under tariff negotiations for the remainder of the paper

Consider next the restriction that tE $ t E and tE $ t E Toexplore the nature of this additional restriction observe that tE

and tE must satisfy respectively

21 This can be seen by noting that if the domestic (foreign) government setsits tariff below its best-response tariff then we must have dWdt gt 0 (dWdt gt 0)and efciency then requires dWdt lt 0 and dWdt lt 0


M x ordm Mx(s0p( tEpwE)pwE) 5 Mx(sEp( t EpwE)pwE) ordm MxE

and

M y ordm My(s0p( tEpwE)pwE)

5 My(sEp( t EpwE)pwE) ordm MyE

where we have used the denitions of t E and tE and the fact thatthe implied world price satises p w = pw E As the rst expressionmakes clear if tE $ t E then existing domestic standards arebeing set in a way that encourages market access at pw E relativeto the efcient choice of standards A similar interpretation fromthe foreign countryrsquos perspective holds for the condition thattE $ t E

Whether the restriction that tE $ t E and tE $ t E is metdepends therefore on whether existing standards are being setin a way that encourages market access at pw E relative to theefcient choice of standards This in turn depends on wheregovernments are starting from (the existing standards levels) andwhere they wish to go (the particular point on the efciencyfrontier) But the circumstances under which this restriction willnot be met namely that governments face the prospect of tariffnegotiations from a starting point in which their existing stan-dards are set in a way that discourages access to their marketsrelative to efcient standards policy seem quite plausible Suchcircumstances would certainly be suggested by Proposition 2 Inthis event in order to achieve efcient trade volumes throughtariff negotiations each government must offer the efcient levelof market access to its trading partner by agreeing to bind itstariff at a level below the tariff thatmdashin combination with itsefcient standards choicesmdashwould be efcient and in light ofthese bindings the resulting mix of policies could not then achieveefciency And if bindings were instead set at or above the level ofefcient tariffs the implied market access commitments would beinsufcient to induce governments to select efcient trade vol-umes22 Hence this second restriction places more serious limi-

22 This last point relies on a slight further strengthening of our assump-tions We assume henceforth that W(sp( t pw )pw ) is globally concave in s withan analogous assumption for foreign-government welfare With this suppose forexample that bindings were set at the efcient tariff levels and note that in thiscase the implied market access commitments would be below the efcient levelsSuppose as well that the foreign government set the efcient foreign policies inStage 2 It may now be seen that the domestic governmentrsquos best response cannotbe the corresponding efcient domestic policies This follows with global concavity


tations on the ability to implement efcient outcomes with tariffnegotiations

With interpretations of these restrictions in hand we nowsummarize this subsection with the following proposition

PROPOSITION 3 Consider any efcient combination of tariff bind-ings and standards policies This efcient policy combinationcan be implemented under tariff negotiations if and only ifexisting standards have been set by each government in away that encourages access to its markets relative to theefcient standards policy

Figure VI illustrates a case where an efcient combination oftariff bindings and standards policies can be implemented undertariff negotiations starting from the Nash equilibrium Withstandards recorded at their initial (Nash) levels the bottom panelof Figure VI depicts t and t as the tariff bindings that wouldimplymdashin light of these initial standardsmdashthe efcient levels ofmarket access The upper panels of Figure VI then depict theadjustments that each government would subsequently make toits national standards and tariff level In the upper left-handpanel beginning from ( t sN ) the prospect of a nonviolation com-plaint induces the domestic government to operate on or belowthe iso-pw E locus and this prevents the domestic governmentfrom lowering its standard to reclaim the market access it hadgranted through tariff negotiations and thereby improve itsterms of trade Along the iso-pw E locus its preferred point isbelow the point ( t sN ) at ( tE sE ) as depicted Hence subsequentto tariff negotiations the domestic government lowers its nationalstandard to the efcient level and agrees to further reduce itstariff level as well so as to honor its negotiated market accesscommitment A similar adjustment for the foreign government isillustrated in the upper right-hand panel of Figure VI and withthese adjustments the two governments can achieve efcientpolicy combinations with tariff negotiations alone

Figure VII illustrates a case where an efcient combinationof tariff bindings and standards policies cannot be implemented

of W in s from Proposition 2 which implies that even with its tariff bound at theefcient level the domestic government would wish to alter its standards policiesfrom their efcient levels so as to restrict domestic-market access below theefcient level and it could do so in this case without violating its market accesscommitments


under tariff negotiations starting from the Nash equilibriumWith standards recorded at their initial (Nash) levels the bottompanel of Figure VII depicts t and t as the tariff bindings thatwould imply the efcient levels of market access The upperpanels of Figure VII then depict the policy adjustments that eachgovernment would subsequently need to make in order to achievethe efcient policy combination In the upper left-hand panelbeginning from ( t sN ) the prospect of a nonviolation complaintinduces the domestic government to operate on or below theiso-pw E locus and as before this prevents the domestic govern-ment from lowering its standard in an effort to reclaim the mar-

FIGURE VINegotiations over Tariffs Alone with Nonviolation Complaints When Efciency

Requires Weaker Standards


ket access that it had granted through negotiated tariff reduc-tions But along this constraint its preferred point will now beabove the point ( t sN ) at ( t E sE ) as depicted in the gure Hencesubsequent to tariff negotiations the domestic government mustraise its national standard level to achieve efciency and musttherefore be allowed to increase its tariff level as well so as topreserve its negotiated (efcient) market access commitment Asimilar adjustment for the foreign government is illustrated inthe upper right-hand panel of Figure VII The difculty is that agovernment is not permitted to make such a unilateral tariffadjustment above its bound level in the Two-Stage Tariff Nego-

FIGURE VIINegotiations over Tariffs Alone with Nonviolation Complaints When Efciency

Requires Tighter Standards


tiation Game nor is such an adjustment permitted under WTOrules Hence the efcient policies cannot be implemented withtariff negotiations in this case

It is clear from a comparison of Propositions 2 and 3 that theprospect of nonviolation complaints restricts sovereignty overdomestic policy choices in a way that in some circumstances canallow governments to reach the efciency frontier with tariffnegotiations alone In principle then the WTOrsquos existing rulescan be seen to contribute toward a solution to the problemsassociated with standards-setting while maintaining some sover-eignty over standards choices for its member governments It isalso clear from Proposition 3 however that in combination withtariff bindings the potential for nonviolation complaints does notleave governments with sufcient sovereignty over their policychoices to reach the efciency frontier in all circumstances Weexplore the possibility of modications to the WTOrsquos rules thatmight address these limitations in the penultimate section of thepaper Before doing this however we turn in the next subsectionto discuss a feature of the WTOrsquos rules not captured by theTwo-Stage Tariff Negotiation Game

C WTO Rules Reciprocity

Governments are not permitted to modify their Stage 1 tariffbindings in the Two-Stage Tariff Negotiation Game of the previ-ous subsection In combination with the prospect of nonviolationcomplaints this has the effect of ensuring that the level of marketaccess commitments implied by tariff negotiations are not erodedby subsequent changes in domestic policy However in realityWTO rules do provide governments with the right to modify theirtariff bindings Indeed as we observed in the Introduction thecentral function of Article XXIII nonviolation complaints is not toprevent governments from ever modifying their market accesscommitments but to induce them to do so explicitly by renegoti-ating their bindings under the rules of the WTO In fact GATTlegal scholars (eg Enders [1996] and Roessler [1998]) oftendescribe these renegotiation provisions as an already-availableanswer to the perceived conict between WTO market accesscommitments and strong labor and environmental standards Animportant question is therefore whether the remaining impedi-ments to efcient outcomes that we have identied above mightbe removed once the WTOrsquos renegotiation provisions are properlyaccounted for


The WTO rules of renegotiation are provided in ArticleXXVIII which sets out the procedures under which a governmentmay lawfully modify or withdraw its tariff bindings as well as therights of its trading partners in this event Under these proce-dures a government may choose unilaterally to raise a tariffbinding with the knowledge that its trading partners are thenpermitted to withdraw reciprocal concessions of their own As wehave argued elsewhere [Bagwell and Staiger 1999a] mutualchanges in tariffs that conform to reciprocitymdash equal changes inimport volumes across trading partnersmdashleave the terms of tradeunchanged23 Hence Article XXVIII provides each governmentwith the unilateral right to reduce the level of its market accesscommitments by raising the level of its tariff bindings but thereciprocal actions of its trading partners permitted under ArticleXXVIII ensure that this unilateral right does not extend to alter-ing the terms of trade In this subsection we describe an extendednegotiating structure that captures this feature of WTO rules

Formally we extend our negotiation structure to three stageswith the introduction of a ldquotariff renegotiationrdquo stage (correspond-ing to Article XXVIII) between Stages 1 and 2 of the Two-StageTariff Negotiation Game where any renegotiation satises therestriction of reciprocity as outlined above and thus results inmutual changes in tariff bindings that preserve the implied worldprice from the rst stage To ensure that the renegotiation pro-cess achieves resolution (and in line with Article XXVIII) weassume that if governments fail to agree on a renegotiated set oftariff bindings then the bindings that are implemented at theend of this stage are those that imply the greatest level of marketaccess consistent with reciprocity and the requirement that nogovernment is asked to provide greater market access than isimplied by its proposal in the renegotiation stage

Effectively this Three-Stage Tariff Negotiation Game hasgovernments determining the balance of market access commit-mentsmdashand therefore the terms of trademdashin Stage 1 tariff nego-tiations the level of market access commitments in Stage 2 tariffrenegotiations and the policy mix with which each government

23 Formally we say that the changes in trade volumes associated withthe change from one policy vector (T0T0) to another (T1T1) conform toreciprocity if pw0[Mx

1 2 Mx0] = [My

1 2 My0] Utilizing the equilibrium and trade

balance conditions at pw0 and again at p w1 allows this condition to be rewrittenas [ pw1 2 pw0]Mx

1 = 0 which implies that the terms of trade must remainunchanged


will deliver its market access commitments in the unilateraldecisions of Stage 3 We say that a policy combination can beimplemented under reciprocal tariff negotiations if given theexisting standards there exists a p w such that the stage 2ndash3subgame yields this efcient combination outcome as a subgameperfect Nash equilibrium

In our working paper [Bagwell and Staiger 1999b] we showthat an efcient policy combination cannot be implemented underreciprocal tariff negotiations if this policy combination is notpolitically optimal Intuitively any attempt to implement an ef-cient combination of policies that is not politically optimal failsunder reciprocal tariff negotiations because at such a policycombination some country desires less trade volumemdashand theconsequent increase in import-competing local pricemdashif it has theopportunity to achieve this without altering the world price Thisopportunity is provided in Stage 2 as the country can thenrenegotiate its tariff bindings subject to reciprocity We furtherestablish that reciprocal tariff negotiations cannot implement apolitically optimal combination of policies if existing standardshave been set by either government in a way that discouragesaccess to its markets relative to the efcient politically optimalstandards policy In this case the problem is that in light of theexisting standards the tariff bindings necessary to induce theefcient levels of trade volume in Stage 2 would prevent govern-ments from achieving in Stage 3 the politically optimal tarifflevels once standards were also adjusted to their politically opti-mal levels Finally with two additional conditions we establishthat when existing standards encourage market access relativeto efcient politically optimal standards politically optimal poli-cies may indeed be implemented under reciprocal tariff negotia-tions24

In summary we nd that an efcient combination of tariffbindings and standards policies can be implemented under recip-rocal tariff negotiations if and only if (a) it consists of a politicallyoptimal combination of tariffs and national standards and (b)

24 The rst condition is that at the political optimum each governmentrsquoswelfare is no less than the welfare level it could achieve in the absence ofnegotiations if it could simply commit to a unilateral increase (beyond the Nashequilibrium level) in the market access it offered to its trading partner Thisamounts to a slight additional tightening from our earlier focus on efcient pointsat which each government is below its best-response tariff The second conditionis that tariffs and domestic standards are sufciently close substitutes for meetingmarket access objectives


existing standards have been set by each government in a waythat encourages access to its markets relative to the efcientpolitically optimal standards policy Hence accounting for therenegotiation provisions of the WTO does not alter our basicconclusion the prospect of nonviolation complaints restricts sov-ereignty over domestic policy choices in a way that can allowgovernments to reach the efciency frontier with tariff negotia-tions alone but it does not leave governments with sufcientsovereignty over their policy choices to reach the efciency fron-tier in all circumstances We consider modications to the WTOrsquosexisting rules that might address this potential shortcoming inthe next section

IV TARIFF NEGOTIATIONS AND NATIONAL SOVEREIGNTY

If a government enters tariff negotiations with national stan-dards that discourage access to its markets relative to efcientstandards policy then global efciency requires this governmentto make future changes in its standards which in themselveswould increase access to its markets As we have shown in thisevent the tariff binding that would imply the efcient level ofmarket access would later impede the governmentrsquos ability todeliver this level of market access with an efcient mix of tradeand domestic standards policies This is captured by constraints(i) and (i) in (IV) and (IV) respectively and when either of theseconstraints binds the attainment of globally efcient policy out-comes through tariff negotiations is then impeded

This impediment can be removed if governments are grantedthe freedom to stabilize their implied market access commit-ments by unilaterally raising their bound tariffs when makingchanges to domestic policies that would otherwise increase accessto their markets Effectively granting this additional freedomwould amount to eliminating constraint (i) from the last stage ofthe Two-Stage Tariff Negotiation Game we have set out aboveand this would eliminate (i) and (i) from (IV) and (IV) respec-tively As a result this change would eliminate any impedimentsto achieving global efciency that were associated with the fea-tures of existing standards25

25 A limitation of this approach is the difculty inherent in measuring thetrade effects of domestic policy changes An alternative would be to allow govern-ments to record intended domestic policies at the start of tariff negotiations rather


We summarize this observation with a nal proposition

PROPOSITION 4 If governments were granted the freedom tostabilize their implied market access commitments by rais-ing their bound tariffs when making changes to domesticpolicies that would otherwise increase access to their mar-kets then any efcient combination of tariff bindings andstandards policies could be implemented under tariffnegotiations

Proposition 4 indicates that the primacy of market accessconcerns reected in current WTO rules can be harnessed toeliminate the remaining inefciencies associated with standards-setting under these rules if governments are given more sover-eignty than these rules currently provide to choose the policy mixwith which to deliver their market access commitments26

Notice that the impediment to global efciency that we haveidentied under existing WTO rules bears a resemblance to race-to-the-bottom-type fears under which the trade pressures asso-ciated with a countryrsquos WTO market access commitments couldcause it to delay the introduction or enforcement of stricter laboror environmental standards Our analysis therefore identies anelement of truth in these fears This is the case even when theexibility to renegotiate market access commitments that theWTO permits is modeled

However in light of the need for added exibility our analy-sis points to the renegotiation provisions of the WTO as a poten-tially fruitful area within which to introduce the modicationsthat could eliminate this impediment27 In this light our ndingsvalidate the WTOrsquos continued emphasis on market access con-cerns and point to renements of WTO renegotiation provisions

than existing domestic policies as our formal negotiation games assume In factthe failure to implement policy changes that were ldquopromisedrdquo at the time of around of tariff negotiations can be the basis for a nonviolation complaint underArticle XXIII (see Petersmann [1997 pp 156ndash157]) However this solution wouldintroduce an added strategic element to the choice of standards which could adda new source of inefciency (and which would be absent under the solutionproposed above)

26 We mention also that with this modication of WTO rules the restrictionthat tariffs and standards are close substitutes for meeting market access objec-tives (see note 24) is no longer needed to ensure that the politically optimalpolicies can be implemented under reciprocal tariff negotiations in the Three-Stage Tariff Negotiation Game

27 For instance the renegotiation provisions of Article XXVIII might bemodied to facilitate the possibility of offering changes in domestic standards asa ldquocompensatory adjustmentrdquo when raising the bound rate of a tariff


under which governments could better achieve globally efcienttrade and domestic policies

Also important however is a direction in which our analysisdoes not point namely the direction taken by proposals for thecreation of a WTO ldquosocial clauserdquo As we have observed abovedirect negotiations over ( t s t s) could of course allow govern-ments to move to a point on the efciency frontier But this is notwhat the social clause envisions Instead under current propos-als a set of minimum uniform international standards would rstbe negotiated and subsequently each country would be allowed todeny market access to any trading partner that did not meetthese standards These proposals reect a belief that the race tothe bottom is fueled by the policy choices of low-standards coun-tries and the trade pressures that these choices exert on theindustrialized world Under this logic the race to the bottom canbe stopped by making access to onersquos markets conditional on thestandards choices of onersquos trading partners But while we haveidentied an element of truth in the race-to-the-bottom fearsunder existing WTO rules the engines of this race are fueled bythe loss in trade competitiveness that would result from a tight-ening of onersquos own standards not by greater import competitionfrom a low-standards trading partner Consequently the inef-ciency associated with the race to the bottom cannot as a generalmatter be eliminated by modifying WTO rules to forge a directlink of the form envisioned under the social clause between onersquosmarket access commitments and the choice of standards made byonersquos trading partner28 From this perspective our analysis indi-cates that the logic of a WTO social clause is fundamentallyawed

V CONCLUSION

How should the issue of domestic standards be handled in theWTO Our analysis suggests that WTO principles are potentially

28 This can be seen formally by considering how the introduction of a socialclause would augment (IV) and (IV) and the tariff negotiating games we have setout above Under a WTO social clause as currently proposed countries would inthe rst stage negotiate as well a set of minimum international standards s andthen in the nal stage a third constraint would be added of the form s $ s to (IV)and of the form s $ s to (IV) It is now a short step from the augmented (IV) and(IV) to see that the introduction of a WTO social clause could not eliminate theinefciencies that exist under current WTO rules except in the special case whereefciency required uniform standards across countries


well equipped to address this issue and that with some modi-cation these principles could allow governments to attain globallyefcient trade and domestic policies The modication would al-low governments to increase their bound tariff rates when mak-ing changes to their domestic policies that would otherwise in-crease foreign access to their markets As WTO principleseffectively already require governments to grant compensatorytariff reductions when altering their domestic policies in waysthat would erode foreign access to their markets these modica-tions can be viewed as renements that are consistent with WTOprinciples

While in principle our results point toward a relatively sim-ple ldquoxrdquo for the contentious issue of standards in the WTO inpractice a host of important caveats must be borne in mind Firstamong these is the ldquoslippery sloperdquo argument that asks of theWTO ldquoWhy stop at labor and environmental standardsrdquo Virtu-ally all domestic policy choices of large economies such as theUnited States will affect export prices in the world economy andhence could be the subject of an analysis similar to what we haveundertaken here Where then should the WTO draw the lineAlso important is the question of how given the complexities ofthe real world the trade effects of a given change in domesticstandards could be assessed29 These and other arguments mightwell be offered up against the advisability of modifying the rulesof the WTO in the way that our formal results suggest

On the other hand the direct negotiation of a list of minimuminternational standards and the subsequent enforcement of aWTO ldquosocial clauserdquo is itself an extraordinarily complex task andnot one that is immune to the ldquoslippery sloperdquo argument At thesame time this approach crosses a boundary of national sover-eignty that has served GATT well for over 50 years Moreover aswe have observed there is a key difference between harnessingthe logic of existing WTO principles to address the issue of laborand environmental standards and negotiating a social clause We

29 The general difculty of assessing the trade effects of a given unilateralpolicy change arises as well in the context of Kemp-Wan adjustments (see alsonote 19) and in the context of customs union formation these difculties havebeen discussed by McMillan [1993] and Srinivasan [1997] Nevertheless it shouldbe pointed out that these difculties have not prevented GATT panels fromproceeding in nonviolation cases where the trade effects of domestic policychanges are precisely what is at stake Such assessments are also required whenviolation complaints result in retaliation as in the recent beef-hormone dispute[WTO 1998] Hence while accuracy is surely an issue assessing the trade effectsof domestic policy changes is already a part of GATT practice


have shown that the logic of a social clause is awed in that itrests on a fundamental misunderstanding of the nature of thepotential race to the bottom under existing WTO rules

COLUMBIA UNIVERSITY AND NATIONAL BUREAU OF ECONOMIC RESEARCH

UNIVERSITY OF WISCONSIN AT MADISON AND NATIONAL BUREAU OF ECONOMIC

RESEARCH

REFERENCES

Alesina Alberto and Romain Wacziarg ldquoIs Europe Going Too Farrdquo mimeoHarvard University March 1999

Bagwell Kyle and Robert W Staiger ldquoThe Simple Economics of Labor Standardsand the GATTrdquo in Alan V Deardorff and Robert M Stern eds SocialDimensions of U S Trade Policy (Ann Arbor The University of MichiganPress 2000) pp 195ndash231

Bagwell Kyle and Robert W Staiger ldquoAn Economic Theory of GATTrdquo AmericanEconomic Review XCIX (March 1999a) 215ndash248

Bagwell Kyle and Robert W Staiger ldquoDomestic Policies National Sovereigntyand International Economic Institutionsrdquo NBER Working Paper No 7293August 1999b

Bhagwati Jagdish ldquoThe Demands to Reduce Domestic Diversity among TradingNationsrdquo in Jagdish Bhagwati and Robert E Hudec eds Fair Trade andHarmonization Prerequisites for Free Trade Volume 2 (Legal Analysis)(Cambridge MA The MIT Press 1996) pp 10ndash 40

Bhagwati Jagdish and Robert E Hudec eds Fair Trade and HarmonizationPrerequisites for Free Trade Volume 2 (Legal Analysis) (Cambridge MA TheMIT Press 1996)

Brown Drusilla K Alan V Deardorff and Robert M Stern ldquoInternational LaborStandards and Trade A Theoretical Analysisrdquo in Jagdish Bhagwati andRobert Hudec eds Fair Trade and Harmonization Prerequisites for FreeTrade Volume 1 (Economic Analysis) (Cambridge MA The MIT Press1996)

Copeland Brian ldquoStrategic Interaction among Nations Negotiable and Non-negotiable Trade Barriersrdquo Canadian Journal of Economics XXXIII (1990)64ndash108

Ederington Josh ldquoInternational Coordination of Trade and Domestic Policiesrdquomimeo May 1999

Enders Alice ldquoThe Role of the WTO in Minimum Standardsrdquo in Pitou van Dijckand Gerrit Faber eds Challenges to the New World Trade Organization (TheHague the Netherlands Kluwer Law International 1996)

Jackson John H The World Trading System (Cambridge MA The MIT Press1989)

Kemp Murray C and Henry Wan Jr ldquoAn Elementary Proposition Concerningthe Formation of Customs Unionsrdquo Journal of International Economics VI(1976) 95ndash98

McMillan John ldquoDoes Regional Integration Foster Open Trade Economic Theoryand GATTrsquos Article XXIVrdquo in Kym Anderson and Richard Blackhurst edsRegional Integration and the Global Trading System (New York St MartinrsquosPress 1993)

Petersmann Ernst-Ulrich The GATTWTO Dispute Settlement System Interna-tional Law International Organizations and Dispute Settlement (The Haguethe Netherlands Kluwer Law International 1997)

Roessler Frieder ldquoDomestic Policy Objectives and the Multilateral Trade OrderLessons from the Pastrdquo in Anne O Krueger ed The WTO as an Interna-tional Organization (Chicago and London University of Chicago Press 1998)

Spagnolo Giancarlo ldquoIssue Linkage Delegation and International Policy Coop-erationrdquo mimeo March 1999

Srinivasan T N ldquoInternational Trade and Labour Standards from an Interna-


tional Perspectiverdquo in Pitou van Dijck and Gerrit Faber eds Challenges tothe New World Trade Organization (The Hague the Netherlands KluwerLaw International 1996)

mdashmdash ldquoThe Common External Tariff of a Customs Union Alternative ApproachesrdquoJapan and the World Economy IX (1997) 447ndash465

Winters L Alan ldquoRegionalism and the Rest of the World The Irrelevance of theKemp-Wan Theoremrdquo Oxford Economic Papers XLIX (April 1997) 228ndash234

WTO ldquoReport of the Appellate Body EC Measures Concerning Meat and MeatProducts (Hormones)rdquo Report WTDS26ABR and Report WTDS48ABRJanuary 16 1998

WTO ldquoReport of the Panel European Communities-Regime for the ImportationSale and Distribution of Bananas-Recourse to Article 215 by the EuropeanCommunitiesrdquo Report WTDS27RWEEC April 12 1999a

WTO ldquoReport of the Panel United StatesmdashImport Prohibition of Certain Shrimpand Shrimp Productsrdquo Report WTDS58R May 15 1999b


recommend necessary modications to the WTO To some degreethese labor and environmental initiatives are responsive to ldquorace-to-the-bottomrdquo concerns In the face of falling trade barriers withweak-standards countries it is feared that the labor and environ-mental standards of the industrialized world might be compro-mised in the name of international competitiveness But suchinitiatives encroach on traditional limits of national sovereigntyThey therefore raise difcult issues about the structure of inter-national economic relations among sovereign states

Motivated by the general question raised above and by therecent debate surrounding the scope of the WTO we ask here amore specic question how should the issue of domestic stan-dards be handled in the WTO We answer this question in asetting where governments choose both trade and domestic stan-dards policies and countries affect each other through their mar-ket interactions so that any externalities across countries arepurely pecuniary in nature By ruling out nonpecuniary external-ities from the start we are excluding ldquoglobal commonsrdquo issuesfrom our analysis and so countries have no reason to care abouteach othersrsquo standards choices directly We do not deny the im-portance of global commons concerns however we choose toexclude them from our analysis since the need to involve theWTO in such concerns is far from obvious1 But even in theabsence of such concerns countries may still care about eachothersrsquo standards choices indirectly because of the trade effectsthat such choices could imply Indeed it is the competitive pres-sures exerted by these trade effects that are often identied asfueling a race to the bottom And as these effects travel throughtrade they are inextricably intertwined with the business of theWTO Our paper considers the question of how labor and envi-ronmental standards should be handled in the WTO in light oftheir associated trade effects

We are of course not the rst to consider this question (seefor example the inuential volumes edited by Bhagwati andHudec [1996]) However analytical results are scarce and ofthese even fewer are concerned with the interaction betweennegotiated reductions in trade barriers and the choice of domestic

1 Except perhaps for reasons of enforcement but even here the case is notwithout qualication see for example Roessler [1998] Ederington [1999] andSpagnolo [1999] On global commons issues see Alesina and Wacziarg [1999]




























II THE BASIC MODEL







































(I) maxt s t s

W(sp( t pw)pw)



shy pw shy t D


shy pw shy t D

and

(6) (1 2 AWp)(1 2 AWp) 5 1



[dpd t ] 0 l ordm

[ shy pw shy t ]

[dpd t ] 0












(II) maxt s

W(sp( t pw)pw)


(II) maxt s

W(sp( t pw)pw)



(8) Wp 1 l Wpw 5 0


t Wp 1 Wpw G

and

(10) Wp 1 l Wpw 5 0














dWd t

1dWds


shy t 1

shy pw

shy s G


(11)dWd t

1dWds


shy t 1

shy pw

shy s G






(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES




















































II THE BASIC MODEL







































(I) maxt s t s

W(sp( t pw)pw)



shy pw shy t D


shy pw shy t D

and

(6) (1 2 AWp)(1 2 AWp) 5 1



[dpd t ] 0 l ordm

[ shy pw shy t ]

[dpd t ] 0












(II) maxt s

W(sp( t pw)pw)


(II) maxt s

W(sp( t pw)pw)



(8) Wp 1 l Wpw 5 0


t Wp 1 Wpw G

and

(10) Wp 1 l Wpw 5 0














dWd t

1dWds


shy t 1

shy pw

shy s G


(11)dWd t

1dWds


shy t 1

shy pw

shy s G






(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES



























































(I) maxt s t s

W(sp( t pw)pw)



shy pw shy t D


shy pw shy t D

and

(6) (1 2 AWp)(1 2 AWp) 5 1



[dpd t ] 0 l ordm

[ shy pw shy t ]

[dpd t ] 0












(II) maxt s

W(sp( t pw)pw)


(II) maxt s

W(sp( t pw)pw)



(8) Wp 1 l Wpw 5 0


t Wp 1 Wpw G

and

(10) Wp 1 l Wpw 5 0














dWd t

1dWds


shy t 1

shy pw

shy s G


(11)dWd t

1dWds


shy t 1

shy pw

shy s G






(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES

































(II) maxt s

W(sp( t pw)pw)


(II) maxt s

W(sp( t pw)pw)



(8) Wp 1 l Wpw 5 0


t Wp 1 Wpw G

and

(10) Wp 1 l Wpw 5 0














dWd t

1dWds


shy t 1

shy pw

shy s G


(11)dWd t

1dWds


shy t 1

shy pw

shy s G






(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES





































dWd t

1dWds


shy t 1

shy pw

shy s G


(11)dWd t

1dWds


shy t 1

shy pw

shy s G






(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES




























(12a) Ws 5 0 Wp 5 0and

























(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES














































(III) maxt s

W(sp( t pw)pw)



(III) maxt s

W(sp( t pw)pw)








QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES




























QED




















maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES











































maxt s

W(sp( t pw)pw)




maxt s

W(sp(t pw)pw)





maxt s

W(sp( t pw)pw)








E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES





























E MyE and pw E


E andM y = My














and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES


































and































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES


















































1 2 Mx0] = [My



























V CONCLUSION












RESEARCH

REFERENCES
















































V CONCLUSION












RESEARCH

REFERENCES


































RESEARCH

REFERENCES


























domestic policies, national sovereignty, and...

Documents