Moravcsik - Preferences and Power in the EU

Chapter 11

Preferences and Power in the European Community A Liberal Intergovernmentalist Approach

Andrew Moravcsik

Introduction

The European Community (EC) is the most uccessful example of i tutionalized international policy co-ordina tion in the modern ~ et there is little agreement about the proper explanation for it e

tion From the signing of the Treaty of Rome to the makin Maa tricht the EC has developed through a series of celebrat dip shygovernmental bargains e ch of which se t the agenda for an interve o ~

period of consolidation The most fundamenta l task facing a rheo~ shyical account of European integration is to explain these barg Today many would revive neo-functionalisms empha is on sui ge to

characteristics of EC institutions in particular the importance of u tended consequences of previous decisions and the capacity of supra tiona I officials to provide leadership

This article joins the debate by reas erring the self-critique advan almost two decades ago by Ernst H aas and other leading neo-funC( alists who suggested that European integra tion can only be explat ~ with reference to general theories of interna tional relations The ba claim of this article is that the Ee can be analy ed as a successful un governmental regime designed to manage economic interdepende through negotiated policy co-ordination Refinements and extensi of existing theories of foreign economic policy intergovernment negotiation and international regimes provide a plausi ble and gene shyizable explanation of its evolution Such theories rest on the as UfT shy

tion tha r state behavior reflects the rational actions of governme shyconstrain d at home by domestic societal pressures and abroad by r

Reproduced by kind permiss ion of Blackwell Publishing from Andrew Preferences and Power in the European ommunity A Li be ra l Approach JOllmal of ommoll Market Studies 1(4) 471-524 T he teX ( ha slightly as ind ica ted to fit the f rma t of thi s volume Some foo tnotes ha

loracsik ] Inre rgOenun

been oml

264

A11drew Moravcsik 265

strategic environment An understanding of the preferences and power of its Member States is a logical starting point for analysis Although the EC is a unique institution it does not require a sui gelleris tbeory

If From Pre-theory to Theory

The limitations of neo-functionalism

Despite the richness of its insigbts neo-fu nctionalism is today widely regarded as having offered an unsatisfactory account of European in teshygration (Cornett and Caporaso 1992 Haas 1975 Hansen 1969 Hoffmann 1966 Keohane and Hoffmann 1991 Keohane and Nye 1975 Taylor 1983 Webb 1983 ) The most widely-cited reason is empirical neo-functionalism appears to mispredict both the trajectory

nd the process of EC evolution Insofa r as neofunctionalism advances a clear prediction about the trajectory of the EC over time it was that the technocratic impera tive would lead to a gradual automatic and incremental progression toward deeper integration and greater suprashy

national influence (Haas 1964 70 1967 327 1976 176) Instead however the process o f Community-building has proceeded in fits and starts through a series o f intergovernmental bargains Nor has the process by which integration takes place supported the neo-funct ionalist view Integration has only intermittently spilled over into related sectors and policies and at least until recently the autonomous influence of supranational officials has increased slowly and unevenly if at all

While empirica l critiques of neo-functionalism are not without merit they should not be overstated To be sure tbe empirical evidence does not seem to confirm the stress placed by neo-functionalism on political spillover and the au tonomy of supranationa l officials But other premises particu larly the focus on economic interests may still be viable It remains plausible for example to argue thar integration is a distinctive policy response of modem welfare states to rising economic interdependence

A more incisive criticism of neo-functional ism is theoretical namely tha t it failed to generate an enduring research programme because it lacked a t11eoretical core d early enough specified to provide a sound basis for precise empirical resting and improvement Only the early variants of nco-functionalism predicted a steady development toward -eder alism Faced wi th the fail ure of Eur opean integration to advance steadily and variation in integration across issues time-periods or ountries (spillback spill-around encapsulation ) however neoshy

functionalism provided no clear direc tion for revision 1

266 Preferences and Power in the European Community

As a result further development in neo-functionalist theory seemeshy Keohan( to converge toward an increasingly complex and indeterminate idea - (largely typical description of the single case of the EC Increasing numbers 0 oping CI

epicyclical modifications and alternative causal mechanisms wert enon t1 introduced until the predictions became so indeterminate as to preshy based il clude precise testing Descriptions of alternative causal mechani llb develop prolifera ted some diametrica lly opposed to the theoryS initial fo Uo uniquer on technocratic man agement and economic planning The un e forms c developm nt of the EC in the 1960s for example was interpreted as tion in result of the influence of dramatic political actors of which de GauBe izable was the archetype - an account theoretically unrelated to H aass earlier compaJ predictions and moreover empirically unsatisfying since the malaise 1973 outlasted de Gaulles presidency By the end of the 1960s almost ao~ inducti process of decision-making among democratic states was consisten words wi th the theory (Lindberg and Scheingold 1971)

2 TtlUnderlying neo-functionalisms failure to develop predictions about variations in the evolution of the EC was its lack of grounding in The SI

underlying general theories of domestic and interna tional political schola economy In international political economy (IPE) as in other social stress phenomena it is widely accepted that prediction and explanation parshy exec ticularly over time require theories that elaborate how self-intere ted Europ actors form coalitions and alliances domestically and internationally 1993middot and how cont1icts among them are resolved Such theories must be neo-it derived independently of the matter being studied in the sense that effortgt they require a set of re trictive micro fo undations - assumptions specishy that r fying the natu re of the fundamental social actors their preferences and apprc the constraints they face In thi regard neo-functionalism is both build oddly apoliti al and lacking in any aspiration to genera lity in that it thear advances long-term pre ictions about the future of the EC without bined underlying more specific theories that identify the decisive deter mishy tant nants of politicians choices among competing alternatives While Fir stressing the domestic politics of economic policy co-ordination neoshy argUi functionalism lack an equivalent to modern theories of trade policy the F which explain government choices on the basis of models of pressure alist from predictable distributional coalitions Neo-functionalism as gene Lindberg an d Scheingold put it describes domestic processes but says denc little about basic causes of variation in national demands for integrashy uniq tion (Lindberg and Scheingold 1970 284 emphasis in original ) Neoshy erali functionalist analyses of in ternational bargaining point to the existence cent of dynamics such as log-rolling compromise and upgrading the Haa common interest through linkage and supranational mediation but bot offer no explanation - except the varia ble skill of supranational leaders inte - of how governments choose among them In

Neo-functionalisms ad hoc approach eventually detached it from sho rich currents in general theories of IPE over the past two decades (d poll

Andrew Moravcsik 267

Keohane and Nye 1975 With the exception of a few studies of (largely unsuccessful) attempts at regional in tegration among develshyoping countries the EC came to be treated as a sui generis phenomshyenon thereby impeding eHom at theoretical generalization2 This was based in large part on the a priori expectation that Europe would develop in a federa l direction which led neofunctionalists to stress the uniqueness of its institutional structure ra ther than analogies to other forms of interstate co-operation The possibility of explaining integrashytion in terms of theories of interdependence regimes or other generalshyizable phenomena was thereby lost while the potential for useful comparison and theoretical development remained limited (Pentland 1973 189- 94) For this reason neo-functionalism rema ins today an inductively derived ideal-type rather than a general theory - in the words of its creator a pre-theory of regional integration

2 The legacy of neo-functionalism

The success of the EC in recent years has fuelled efforts among scholars to resurrect neo-functional ist models in particular those that stress the unintended consequences for Member States of leadership exercised by supranational actors including Commission officials and European parliamentarians (Pederson 1992 Peterson and Bomberg 1993 Ross 1992 Sandholtz 1992) This body of work repeats many neo-functionalist themes if sometimes by other names Yet current efforts to resurrect oeo-functionalism rarely address the conclusions that neo-fun ctionalists themselves drew about the weaknesses of their approach nor do they consider the implications for current theoryshybuilding of theoretica1 developments in international re lations (IR) theory over the intervening two decades The functionalist legacy comshybined with contemporary theories of lPE suggests at least three imporshytant conclusions

First by 1975 leading neo-functionalists were nearly unanimous in arguing that regiona1 integration theory which had sought to explain the progress of the EC along the sui generis path toward a future federshyalist endpoint should be supplemented perhaps supplanted by a general theory of national policy responses to international iuterdepenshydence Rather than focusing on the fu ture aspirations that make the EC unique neo-functionalists argued that the empnasis should be on genshyeralizable aspects of the current activities of the EC Recognizing the central importance of economic management among those activities Haas carne to believe that the study of regional integration should be both included in and subordinated to the study of changing patterns of interdependence (Haas 1975)

In the language of modern th eories of IPE this implies that the EC should be treated as an international regime designed to promote policy co-ordination As Hoffmann Haas s erstwhi le critic asserted in


1982 the best way of analyzing the EEC is a an internati ad regime (Hoffmann 1982 33 ) In ternational regimes promulgate pn nil pIe norms rules and decision-making procedu res around hich al th expectations converge in given issue-areas th rough which the a a of separate individua ls or organizations - which are not in pre-exi co harmony - farel brought into c nfo rmity with one another thrau process of negotia tion often referr d to as poticy co-ordioari ur (Keohane 1984 51 Krasner 1983 1) Regime theory prov id a p ushy th sible starting point for analysis - a set of common conceptual and ( - 15 oreti al tools that can help structure comparisons with or Ol

international organizations as well as internal comparisons among 1O

ferent case of EC policy-mak ing At the same time however c nt ei

porary regime analysis requir s refinement to take account omiddot in unique institutional aspects of policy co-ordination within the EC m evidenced by the depth of its purported goals the richness of rhe ne-middot er works it ustains and above all the solidi ty of its supranational 1 ~ _ ac identity (Keohane nd Hoffmann 1991 ) va

S cond the neo-function alist legacy suggests that explanations u in tegration require stronger underlying th ories of variation in - rh sta ntive a well as inst itutional outc meso The neo-func tionali ts -

-bull

c ncerned overridingly with tracin progress toward a terminal co 3ti n called political community - the evol ution of a unique potenri

federa l political structure in Europe that would prevent war and guJ anree peaceful change (Haas 1961 Lindberg and Scbeingold 1 shy99) Accordingly the limited their defin ition of integra tion aim exclusiv Iy t instirutional characteristics of the EC - the sc pe a w in titu tional form of common decision-making T his di c ura a ttention to di tributional con fli t in the EC over issues sllch as st Ie el of external tariHs agricuJ tural prices or regulatory harmontz middot rh rion which require art mion to the substantive mea ure o f poJi) )shy 1fl

ord ination An instructive example is the creation of the CAP in 1960s While the neo-functional ts empha ize the Commis jo II

succes in crearin a policy formally lmder the control of the EC th S overlook the fact tha t it was a defeat for the Commissions orign If

substantive proposa l which fo re aw a prudently limited II

financing re i ti ely low-price reg ime3 P A broader definition of Euro pean integration might con iJer f H

dimen ions of policy co-ord inati n (1) the geographical scope f regime (2 ) th range of issues in which policies are co-ordinated the institutions of joint decision-making implementa tion and enfo - shymeor (4 ) the direction and magnitude of substantive domestic poltt adu -tm nt T hese four elements may be thought of as different dJm shyions of the same underlying variabl namely policy co-ordinau

While the first three are sil lilar to those employed by neofWJcriona shyists the fo urth - the direction and magnitude of sub tanrive


adj ustment - is based on the view that policy co-orclination is most sigshynificant where it imposes greareradjusnnent on domest ic policy Since the costs and benefits of the necessary adjustments generally vary across countries the measw-e also helps in the ana lysis of distributional confl ict

Third by the 1970s man~ neo-functionaJists had concluded that unicausal theories are unabJe to account for EC policy-making More than one theory is required (Cornett and Caporaso 1992 Pentl and 1973 189- 94 Puchala 19- 2) Modern theories of IPE suggest a number of empirical theorerical and philosophical reasons discussed in more detail in rhe ne(t section to treat the need fo r multica usal explanation as a geneT] principle Empirically robust explana tions of internationa l policy co-ordination are likely to incorporate at a minimum theories of borh national preference formation and intergovshyemmental negotiation each grounded in explicit assumptions about actor preferences constraints and choices (Moravcsik 1992b) The vagueness of neo-functionalist predictions suggest moreover that only such theories can explain rather than sim ply describe the evolution of the EC Only by meeting rhese criteria most neo-functiona lists fe lt could scholars mol~ from pre-theory to theory

3 Liberal intergovernmentalism and the rationality assumption

Rather than resurrecting neo-functionalism the approach introduced here takes serioll5)Y the self-criticisms of neo-functionalists examined above They POllt [Oward a conception of the EC more closely in line with contemporary theories of IPE Such theories suggest that the EC is best seen as an international regime for policy co-ordination the subshystantive and instirutional development of which may be explained through the sequential analysis of national preference forma tion and intergovernmental strategic interaction

This section propostS a framework within which to construct such an explanation termed liberal intergovernmentalism Liberal intershygovernmentalism buJds on an ea rlier approach intergovernmental insti tuti onalism b~ refining its theory of intersta te bargai ning and insti tutional compliance and by adding an explicit theory of national preference formation grounded in liberal theories of international interdependence ( ~lora-csik 1991 ) Various specific points seek to refine and extend the exisrjng literatu re but the result is broadly conshysistent wi th current theories of IFE in pa rticular endogenous tariff theory negotia ti on analysis and functional explanations of interna shytiona l regimes

At the core of libera l intergovernmentalism are three essentia l eleshyments the assumption of rational state behaviour a liberal theory of national preference forma t jon and an intergovemmentalist analysis of interstate negotiation Tbe assumption of rational state behaviour proshy


vides a general framework of analysis within which the costs and benshy empl efi ts of ec nomic interdependence are the primary determinants o eren na tion al preferences while the relative intensity of national prefer shy Uni ences the existence of alternative coalitions and the opportunity f r eithe issue linkages provide the basi for an intergovernmental analysis of leadi the resolution of distributional conflicts among governments Regime Mor theory is employed as a starting point for an analysis of condition_ conv under which governments will delega te powers to international institushy gave tions of tlshy

M u h contemporary IR-theory is based on the assumption of state the) rationality State action at any particular moment is assumed to be Tl minimally rational in tha t it is purposively directed toward the wor achievement of a set of consistently ordered goals or objectives- libe Governments evaluate alternative c urses of action on the basis of a tali utility function The approach taken here departs decisively however eet from those theories in IR mo t nota bly realist and neo-rear r mo approaches which treat sta tes as billiard balls or black boxes with fixed preferences for weal th security or power Instead governmems are assumed to act purpo ively in the international arena but on the III basis of goals that are defined domestically Following liberal theories DE of IR which focus on sta te-society relations the foreign policy goals of national governments are viewed as varying in response shifting presshy 1 sure from domestic social groups whose preferences are aggregat Trthrough politica l institutions National interests are therefore neither lit invariant nor un important but emerge through domestic political con shy soflict as societal groups compete for political influence national and pI transnational coalitions form and new policy alternatives are recogshynized by governments An understanding of domestic politics is a preshy

e it

condition for not a supplement to the analysis of the strategic hin teraction among sta tes (Moravcsik 1991 1992b)

[1The model of rational state behaviour on the basis of domestica ll shy a

constrained preferences implies that international conflict and co-oper shyation can be modelled as process that takes place in two successive

a t

stages governments first define a set of interests then bargain among themselves in an effort to realize those interests M etaphorically these two stages shape demand and supply functions for international co shyoperation A domestic preference fo rmation process identifies the potential benefits of policy c -ordination perceived by national governshyments (demand ) while a process of interstate stra tegic interaction defines the possible political responses of the Ee political system to pressures from those governments (supply) The interaction of demand and supply of preference and strategic oppornmities shapes the foreign polic behaviour of state 5

This conceptjon f rationality suggests that parsimonious explanashytions of international confl ict or co-operat ion an be constructed


employing two types of theory sequential ly a theory of national prefshyerence formation and a theory of interstate stra tegic interaction Unicausal explanations of European iotegration which seek to isolate either demand or supply are at best incomplete and at worst misshyleading Explaining tbe emergence in 1978-9 of the European Monetary System for example requires that we understand both the convergence of macroecononic policy preferences which led European governments to favour monetary co-ordination and the determinants of the outcomes of the tough interstare bargaining that took place over the precise terms under which it would take place

Thus liberal intergovernmeotalism integra tes within a single frameshywork two types of general lR-theory often seen as contradictory a liberal theory of national preference formation and an intergovernmenshytalist analysis of interstate bargaining and institutional creation6 In the sections that follow these sequential components are developed in more detail

III Liberalism National Preference Formation and the Demand for Integration

Liberalism and state-society relations

The theory of national preference formation set out in trus section is liberal in inspiration Liberal theories of IR focus on the effect of stateshysociety relations in shaping national preferences They assume that private individua ls and voluntary associations with autonomous intershyests interacting in ci ] society are the most fundamental actors in polshyitics State priorities and policies are determined by politicians at the head of the national government who are embedded in domestic and transnational civil society which decisively constrains their identities and purposes 7 The most fundamental influences on foreign policy are therefore the identity of important societal groups the nature of their interests and their rela tive influence on domestic policy Groups that stand to gain and lose a great deal per capita tend to be the most influential The identiry interests and influence of groups vary across time place and especially issue-area according to the net expected costs and benefi ts of potential foreign policies The factors that detershymine the identity interestS and influence of domestic groups are themshyselves both domestic and transnational In this sense second image reversed theories which assume that international constraints create patterns of societal interests that influence governments via the transshymission belt of domestic politics are characteristically liberal 8 But so are theories that stress purely domestic state-society relations due to the nature of domestic political and socia -economic institutions

- -- - - - --


Groups articulate preferences governments aggregate them For li b~ tants f(

erals the rela tionship b tween society and the government is assumed tu rn ere to be one of principal-agent societal principals delegate power to (or nalities otherwise constrain) governmental agent The primary interest of go - fits for ernments is to mai ntain themselves in offic in democratic societies jurisdict this requires the support of a coalition of domestic voters parties depend interest groups and bureaucracies whose views are transmitted direcd interdeF or indirectly through domestic institutions and practices of political Natio representation Through this process emerges the set of national intershy co-ordil ests or goals that states bring to international negotiations permjtt

T his is not to say that a ll foreign p licy proposals begin with direct This il pressu re from pluralist groups only that stare leaders must construct tive int governing coali tions out of influential groups with specific interests occur Sometimes the influence of societal groups is indi r ct In economic nation affairs for example some firms and groups pa rticu larly those with eromer fixed investments and as ets may seek to infl uence governments fl ows e

directly exercising rhe option of voice others particu larly those with domes i

more mobile invesnnents and assets may find it less expensive to shift may LO

investments to alternative activities or jurisdictions exerci ing th goal ( option of exi t (Bates and lien 1985 Hi rschman 1970 Lindblom In tIshy

1977 Magee et al 1989 13 93 102) In the liberal view even th has ~ latter c nstraint ultimately rests on the de ire of politicians to avoid policy imposing co ts on - and thereby alienating - those social groups who e pende support maintains them in office impor

Yet the intere ts of so ietal groups are not always sharply defined group W here societa l pressu re is ambiguou or divided govern ments acquire the n

a range of discretion While domest ic societal gro ups impose a basic pr m

constra int on governments the nature and tigh tn ss of this constraint resou varies with th e strength and intensity of pressures from social groups and G

At times the princi pal-agent relationship between socia l pre sures and on 10

state policies is tight at time agency slack in the rela tionship permj rs tia l fe rational governments to exercise greater discretion9 Tbl

The liberal focus on domestic interests and state-society rela tions i harm consistent with a number of piau ible motivatjons fo r govern m nt good to support (or oppose ) European integration T hese include federalist eeon (or nationalist) beliefs national security conc ens and ec nomic intershy Nati ests Elsewhere these alternative speci ficati ons of liberal theory are prod tested against one another (Moravcsik 1992a) here the foclls is on effec motiva tions that tern from economic interdependence and the ways in and which they on train governmental preferences in international n gotishy dive ations CO-I

fact 2 Interdependence externalities and cltgt-operation (C At the core of liberal theories of economic interdependence lies the claim G that increasing tran border flows of goods ervices factors or pollu- thar

I middot


rants create international policy externalities among nations which in turn create incentives for policy co-orclination Internationa l policy extershynalities arise where the pol icies of one government create cOStS and beneshyfi ts for politically significant social groups outside its national jurisdiction Where tbe achievement of domestic governmental goal depends on the policies of its foreign counterparts national policies are interdependent and policy externalities can arise (Cooper 1986 292-3 )

National governments have an incentive to co-operate where policy co-ordination increases their control over domestic policy outcomes permit ting them to achieve goals that would not otherwise be possible This si tuation arises most often where co-ordination eliminates negashytive international poLcy externali ties Negative policy externali ties

ccur where the policies of one nation imposes costs on the domestic nationals of another tbereby undermining the goals of the second govshyernments policies Examples include protectionist barriers against flows of foreign goods and capi tal competitive devaluation and lax domestic envi ronmental pollution standards Each of these pol icies may impose costs on foreign nationals thereby undermining the policy goals of foreign governments

In the modern international political economy policy co-ordination has two major p urposes each of which aims at removing a negative policy externality The fi rst is the accommodation of economic interdeshypendence through reciprocal market liberalization Restrictions on impor ts and exports are not simply of interest to domestic societa l group s but to their counterparts abroad as well The liberalization of the movement o f goods services and factors of production may promote moderniza tion and a more efficient allocation of domestic resources favouring producers in internationally competitive sectors and owners of internationally scarce fac tors of production Restrictions on imports of goods and facto rs impose policy externalities on potenshytial foreign exporters investors and immigrants

The second major purpose of economic policy co-orclination is policy harmonization in order to assure tbe continued provision of pu blic goods fo r which the srate is domestically responsible such as socioshyeconomic equality macroeconomic stabi li ty and regula tory protection National welfare provision monetary policy labour market controls product regu lation and many other domestic policies rely for tbeir effectiveness on the separation of markets for goods services factors and poll utants Where economic interdependence links jurisdictions divergent na tional policies may undermine each others effectiveness Co-ordinated (or common) policies may therefore result in greater de facto control over domestic policy outcomes than unilateral efforts (Cooper 1972 )

Contrary to the beliefs often attributed to them liberals do not argue tha t cooperation to achieve trade liberalization and the common provishy

~85

ng en Or

se 19

It

n


sion of public goods i inevitably supported by all governments The tional eJ vulnerability of governments to negative externalities may vary greatly respecti1

some are able to sustain effective policies autonomously others remain afford t(

vulnerable to negative ex ternali ties from policies abroad While the and ber latter have an incentive to support international policy co-ordination individt those that produce negative externalities or benefit from the positive cally 11 externaliri s of oth rs have an incentive to free ride on the domestic groups I

policies of their neighbours rather than cooperate (Keohane and N straint J989 12ff) Only where the policies of two or more governments ment create negative policy externalities for one another and unilateral preferel adjustment strategies are ineffective inadequate or expensive does making economic interdependence create an unambiguous incentive to co-ordishy erWI e nate policylO agreem

differer 3 The distributional consequences of policy co-ordination eluded Even where agreements are m utually beneficial governments often agricul

have different preferences concerning the distri bution of the benefits sures b leading to conflict over the precise terms of co-operation 11 The COSt At tb and benefits of policy-coordination are often unevenly distributed policie1 among and within nations rendering nearly inevitable a measure f societa international and domestic conflict between winners and losers To th Tulloe extent that it take domestic and international distributional confli r enJo into account liberal lnterdependence theory do s not as some ha positio sugge t d assume the existence of a harmony of interests or a simple desire correlation between potential tran actions and co-operation Nation cipal-a and domestic groups that are disadvantaged by policy co-ordination ments are likely to oppose it Only where governments can collectively overshy longer come such opposi middoton is co- peration possible The distribution of coaliti expected net societal costs provides a means of predicting the nature ot more J

political conflict and co-operation in the Ee both internationally and 4 Pc domestically12

Domestically governments participating in international negotiation Differ are both em powered and constra ined by important societal group benefi which calculate their interests in terms of the expected gains and losse of do from specific policies (Frieden 1991a Gourevitch 1986 Milner 198 cir UI

Odell 1982) Powerful groups disadvantaged by co-operation will se opera to obstruct government policy even where such policies generate ne basis gains for society as a whole To understand and predict the likelih 0 good~ of international co-operation in any given instance therefore requ lr _ and a more precise specification of domestic societal interests in particular

COnllissue-areas and the ways in which those interests constrain governmiddot ments At th

Societal pressure on national governments reflects not only the - in expected magnitude of gains and losses but also the unc rtaimy an poushyrisk involved The magnitude certainty and risk f domestic dimib de5i


tiona I effects of policy coordination determines oat only the goals of respective governments but the extent CO which governments can afford to be flexible in negotiation At one extreme where the net costs and benefits of alternative policies are certain significant and risky individual citizens and firms have a strong incentive to mobilize pohti cally In such circumstances unidirectional pressure from cohesive groups of producers or organized private interests imposes a strict conshystraint on government policy The prospects for international agreeshyment will depend almOSt entirely on the configuration of societal preferences in negotiations governments have li ttle flexibili ty in making concessions proposing linkages managing adjustment or othshyerwise settling au the middotlowest com mon denominator International agreement requires that the interests of dominant domestic groups in different countries converge where they diverge co-ordination is preshycluded Such conditions are approximated in EC negotiations over agricultural p rices and Ie bargaining positions are dictated by presshysures from interest groups

At the other egttrerne where the net costs and benefits of alternative policies are dipoundfuse~ ambiguous or insignificant and the risk is low the societal constraints on governments are looser (cf Buchanan and Tullock 1962 789) Under such circumstances leading politicians enjoy a wider range of de facto choice in negotiating stra tegies and positions More than one policy is likely to be consistent with the basic desire of politicians to remain in government The slack in the prinshycipal-agent relationship between society and the state permits governshyments to asswne more political risk by taking a more enlightened or longer-term view balancing winners and losers to construct broader coalitions accepting short-term losses for long-term gains or pursuing more ideologically controversial goals

4 Policy areas and national preferences in the EC

Different policy aIeas engender characteristic distributions of costs and benefits for socie[al groups from which follow varia tions in patterns of domestic political mobilization opportunities for governments to circumvent domestic opposition and motivations for in ternational coshyoperation3 EC policy areas can be divided into three categories on the basis of policy objectives the liberalization of the exchange of private goods and services the provision of socio-economic collective goods and the provision of non-economic col1ective goods

Commercial policy market access and producer interests

At the core of the EC is its Internal Market The most basic EC policies - including internal market pol icy agricultural policy competition policy industrial policy and research and development policy - are designed to liberalize or eliminate distortions in markets for private

cc


goods and services Modern theories of commercial policy begin br assuming that individ ua l and group support for li beraliza tion and proshy rn

retection reflects to a fi r t approximation the net expected co tS antI benefits of th policy change (Hillman 1989 Magee et a l 1989 ) Social w groups with an intense interest in a given policy are more likely to gr mobilize tha n those with a weak in terest si nce higher per capita gains support the costs of locating organizing monitoring and representing 0 concentra ted groups This tends to create a systematic politica l bias in f favour of producers vis-a-vis those wi th more diffuse interest sucb as fu tax-pa er and individual consumers or those with no direct acces to in the politica l pro ess uch as foreign producers (H ill man 1989 Olson er

ra1965) Following endogeno us tariff theory the approach employed here assumes that societal groups mobil ized around commercial pol iq Sl

issue are composed aim st exclusi ely of domestic producers whether 1I1

d rawn from labour or capital who organize by sector on the basis of tI calculation of net expected costs and benefits resulting from the introshyduction of new policies 14

Am ong producer the net expected cost and benefits of liberalizashy b tion reflect the fo llowing factors First the ex tent to which individu 11 d producers pro fit from commercial liberalization depends most fundashy F mentally on their competitive positions in domestic and international f markets 15 Protectionist policies not only red istribute domestic wealth fr m consumers to sheltered producers but also create negative policy externalit ies for export r excl uded from potential marke Accordingl exporters lod multinational investors tend to supp rt fre r trade which increases their profits import-competing producer tend to oppose free trad which undermines their profitabi lity Where adjustment is relatively cos tless or compensation between winners and losers can be arra nged distributional effect need oot create oppo ition to free trade W here ad justm nt and c mpen ati n are costly however a domestic prisoners dilemma among domestic veto groups - each 0

which seek to be exempted from disadvanraaeous pol icy chan e bull leading to a suboptimal outcome for society as a whole - tra lac into an international prisoners dilemma in which each governmem eeks to shelter its weakest sectors from international ma rket pre ure

Policy co-ordination helps overcome these di lemma by balancio the gains and losses of free trad wi th in and across countries thereby creshyating iable domestic coalitions in favour of li bera lj za tion 16

Second cross-cutting or balanced patterns of interests internalize lJ( costs and benefits of trade liberalization to the same sets of firm an sectors crearing a cro -cu tting set of interest that undermines oppo shytion to libera lization Most importantly intra-industry trade an investment patt rn reduce the net effects n the posi ti ons of in ruvidua pI ducers and sector 1-7 Even producers facing ubstantial importshy

Andrew Morallcsik 277

competition have an incentive to support free trade if loss of domestic marke t share is offset by exports conrrol over foreign producers or receipts from foreign investments The risk of a large loss is reduced as well Producers of finished goods also form concentrated interest groups in favour of free rrade in raw materials and intermediate inputs

T hird where the effects of policy changes are uncertain organized opposition to gouernment initiatives is diluted Uncertainty about the effects of cooperation arises where policies are stated vaguely left to future negotiation mediated by complex market processes or applied in an unpredictable way across a population Uncertain policies engender Jess opposition than those that are immediate precise and targeted Policies often become more controversial as speci fic provishysions are negotiated and the real effects become evident - as occurred in implementing EC agricultural transport and competition policy in the 1950s and 1960s

1n many cases pressure from private economic interests is enough to convince governments to libera lize Where the ne t expected costs and benefits to firms and sectors are signi ficant unam biguous and preshydictable for important segments of domestic producers pressures from producer interests will impose a relatively tight constraint on state policy Most agricultmal sectors as well as industries with chronic surplus capaci ty are characterized by inter-industry trade patterns uniform and calculable interests and high fixed irreversible investshyments and assets Net commodity exporting countries demanded libershyalization net commodity importing countries resisted it In the CAP interstate bargains have been possible only on the basis of lowest common denominator log-rolling agreements in individual sectors with the costs passed on to consumers and foreign producers Direct pressure from producer interes ts in the EC has created and maintained a system of high agricultu ra l prices and managed trade regardless of the preferences of poli ticians

In other cases the decision to liberali ze refl ects not just pressure from narrow interests but a broader calculation on the part of the governmeot When net expected costs are insignifi cant ambiguou balanced or uncerrain governments enjoy a greater autonomy from particu laristic domesTic groups that oppose co-operation which they can employ to create support for broader societal goals This they can do by negotiating international compromises and issue linkages which creates viable coalitions by balancing winners against losers By subsishydi7ing the costs of ad justment or by balancing losses of domestic market share with gains in fo reign markets they can also mu te opposishytion to Jjberalization Both agricultural trade li beralization in Germany S

5and industrial trade libera lization in France were accompanied by large 5domestic subsidies to uncompetitive producers expressly designed to

finance adj ustment The more governm ents are able to act indepenshy


dently of groups disa dvantaged by a policy thereby trading off gains SIO I

and losses over a larger constituency the more we should observe the iod compromises and upgrading the common interest predicted by neoshy am func tionalists Whereas neo-functional ism stresses the autonomy of are supranational officials liberal intergovernmentalism stresses the riol autonomy of narionalleaders iSSl

pn Ie

Socio-economic public goods provision ex1

EC policies are not limited to the co-ordination of explicit market libshy mt eralization policies but include also the coordination of domestic polishy the

fIi(cies designed to redress market fail ures or provide public goods such as those that assure macroeconomic stability social security environshy db mental protection public health and safety standards and an acceptshy co able distribution of income Rising economic interdependence often inl exacerbates the tension between uncoordinated national policies the co effectiveness of which often requires that either national markets be ffil

separated or national policies be harmonized (Cooper 1972) da Transborder inflows of air and water pollution can undermine the effectiveness of national environmental policies capital outflows can ra undermine the credibility of domestic monetary policy social P dumping can undermine the competitiveness of industry and the viashy pi bility of social compromises C

As with commercial policy an incentive for international policy coshy pi ordination exists when the configuration of domestic policies produces m negative policy externalities - domestic problems that cannot be I

resolved through dome tic regulation because of interference from cl policies pursued by fore ign governments - for more than one country P Negotiated policy co-ordination typically involves some surrender of al domestic policy autonomy in exchange for a similar surrender on the o

part of other countries Where domestic policy instruments remain II

effective governments will continue to maintain them but where govshyernments have exhausted all cost-effective domestic means of achieving domestic policy targets they have an incentive to turn to international F coordination Accordingly policy co-ordination will typically be f sought particularly by smaller governments with little control over their domestic markets and high economic interdependence and by ~ those generally with high levels of domestic public goods provision whose policies are particularly vulnerable to disruption 1

M any socio-economic pubiic goods policies have important impli ashytions for international commerce The effects of uncoordinated policies - exchange rate shif ts disparate production and product standards or divergent social welfare policies - may distort or obstruct international commerce Therefore in contrast to pure commercial liberalization the international co-ordination of such policies raises a two-dimenshy


gain ~ sional issue in that governments must strike a balance between two e the independently valued policy targets flows of economic transactions neo- and levels of public goods provision To the extent that governments

ny of are concerned about trade liberalization the incentives for internashy the tional and domestic co-operation and conflict will resemble those in

issues of pure comnlercia l policy However where governments are primarily concerned with the provision of domestic public goods the level of conflict and co-operation among governments depends on the extent to which national policy goals are compatible W hen govern shy

~[ libshy ments have divergellt macroeconomic environmental and social goals polishy then co-ordina tion is likely to be costly and difficult International conshysuch flict emerges over the division of the burden of adjustment The more 1[00- divergent national policies are to begin with the greater the costs of xeptshy co-operation Nonetheless where these costs are outweighed by the often interest in reducing negative policy externalities international policy i the co-ordi nation can help governments reach an optimal balance between tS be increased market access and the maintenance of regulatory stanshy972) dards lS

e the D ue to the tvo-dinlensional nature of the public goods issues the scan range of mobilized interests is typically broader than in commercial ociaJ policy Whereas ill pure commercial policy the public interest is bull Vla- pursued almost enti rely by national governments backed by broad

coalitions of illterested parties the public interest is represented in coshy public goods concerns by pressure from public interest groups and iuees mass publics Where ex isting domestic policy re fl ects widespread It be popular support domestic regulations are likely to be resistant to the from changes required to achieve international harmonization Alongside IDtry producer interest non-producers may either influence policy directly er of as when environmental interest groups mobilize opposition or punish n the or reward the government for the results of policy as when voters malO respond to recen t macroeconomic performance govshy As in commercial policy the level of constraint on governments ~VlDg varies depending on the intensity and eakulability of private interests ional Policies involving the direct regulation of goods and production I be processes tend to engender strong mobilization of producer groups over while the co-ordination of poljcies to provide macroeconomic public d by goods including pollution infla tion unemployment and the aggregate 51On distri bution of income generates a more diffuse pattern of societa l

interests Most producers have more ambiguous and variable interests )licltlshy in public goods provision - eg the value of the currency the level of Jjcies domestic inflation or the aggregate level of poll ution - than in issues or of pure commercial policy Where strong commercial or public intershy

ional ests are fulfil led in their deman ds for policy co-ordi nation governshytion ments will act accordingly Often however the results of negative lIlen- externalities and policy failure are more diffuse leading to a more


politi ]general economic or regulatory crisis In the la tter case governments

in teDmay act without direct pressure from interested parties ists likM acroeconomic policy provides an illustrative example While

groups do organize around the trade-related costs and benefits of mon shyetary management (Frieden 1991b) these incentives are often offset by other concerns While curren y depreciation increases the competitiveshyness of domestically-produced tradeable goods it also raises the costs

in whi qu iet

of imported intermedjate inputs and raw mate rials as well as Impoincreasing the risk of longer-term inflation Domestic moneta ry policy

Similis influenced by the autonomy of domestic moneta ry institutions and dependithe identity o f the party in power among o ther th ings Recent steps tionaliz tOward European monetary integration fo r example refl ect a set of lab le al national commitment to macroec nomic di cipline im posed by the the e~l= unsustainability of domestic policies in the face of increased internashysub rantional capiral mobil ity Only once domestic po licies had converged si n al substantially did more intensive international co-operati n become some dconceiva ble When they diverged the system once again came under funcri pressure ection

Political co-operation EC institutions and general income transfers negocia eoforeSome EC policies cannot be interpreted as direct responses to policy

The 1 externalities imposed by economic in terdependence Some sllch a a on [hecommon foreign and security policy a im to provide non-socia-ecoshypoliti II nomic co lective goods others such as general European Communi ty tion bullinstitutions and transnational (regional and structural) income transshyand unfers exist either for their own sake or to fac ilitate other poli i s ences aLiberal theory uggests that fundamenta l constraints on national prefshyto acce erences w ill re flect the costs and benefits to societa l actors where these or a E are weak uncertain or diffuse gov roment wi ll be able to pursue sup parbroader or mo re idiosyncratic goal excepnThe costs and benefi ts created by political co-operation for private strong groups are diffuse and uncertain Private producers take little int rest so yelin pol itical cooperation leaving domestic influence over the policy Nationalmost exc1u ively to partisan elites with a secondary intermitten t countri constraint impo ed by mass pu blic T he reasoning u ed to justi fy polishyerali t cies tends to be ymbolic and ideological rather than calculated and

Regi( concrete The inherent incalculabili ty of ga ins and los es in these policy

enougr areas accounts for a tro ubling neo-functional ist anomaly namely the di trib manifest importance o f ideologically moti ated heads of state (drashysibly irmatic-pol itical actors ) in matters of foreign poli y and institutional clesreform The d ifficulty of mobilizing interest groups under conditions of

general uncertai ot about peciiic winners and losers permits the posishy 5 Co tions of gOY romen particularly larger ones on questions of

T his European institution and common fore ign po licy to reflect the ideoloshyto precgies and personal comminnents of leading executive and parliamentary

bull

Andrew MoraLlcsik 281

poli ticians as well as interest-based conceptions of the national interest This may help explain the ability and willingness of nationalshyists like Charles de Gaulle and Margaret Thatcher to adopt an uncomshypromising position toward the dilution of national sovereignty as well as support by various European leaders for direct elections to the European Parliament the creation of the European Council and the quiet development of European Political Co-operation - each an issue in which the costs and benefits to organized in terest groups is near impossible to calculate

Similarly the politics of decisions about EC institutions vary widely depending on the nature of he decision-making process to be institushytionalized Where rhe consequences of institutional decisions are calcushylable and concrete Daoonal posit ions will be instrumental reflecting the expected influence of institutional reforms on the real ization of substantive interests This is for example genera lly the case with decishysions about maloril1 voting on specific economic policies Moreover some delegations of power are viewed as necessary for the effective functioning of the EC These institutions - to which we shall return in Section IV below - Delude common representation in international negotiation rh~ Commissions power of proposal under QMV and enforcement of Ee rules by the ECJ and the Commission

The more general and less predicta ble the implications of decisions on the relaon power of institutions the larger the space for leading politicians and parrisan elites to act on the basis of ideological predilecshytions National imerests would lead one to expect large self-sufficient and uncompetirie countries as well as those that hol d outl ier prefershyences on qutgtooru of public goods provision to be relatively unwillulg to accept stronger supranational institutions such as majority voting or a European Parliament British and French policy provides some support for this -ie bu t Italys consistent federalism remains an exception Suntlarly smaller countries might be expected to support strong supranattonal power The Benelux countries have indeed done so yet Danish Gr~k and Irish support has been less consistent National par1iamMlta~middot eli res appear to play an important role in countries like lcaJy Germany and the Netherlands which support fedshyeralist institutions

Regional and stmcturo11 policies - since they are neither significan t enough to prmide maior benefits to the donors nor widely enough distributed to repre5enr a policy of common interest - are most plaushysibly interpreted as side paments extended in exchange for other polishycies

5 Coneusion

This section has employed and e tended contemporary theories of IFE to predict the national preferen of EC Member States across three


types of issues commercial policy socia-economic public goods provishy among sion and other institutional pol itical or structural policies In each tage In case the magnitude distribution and certainty of net expected costs stand [ and benefits to private groups were employed to predict policy prefershy Crea ences of governments as w ell as their range of relative autonomy vis-ashy be drnJ vis those domestic groups that oppose cooperation This defines the on the demand for interna tional cooperation in the next section we turn to be ef the capacity of the international system to supply cooperation

tion shy

IV Intergovernmentalism Interstate Bargaining and the Supply of Integration

In tergovernmentalist theory seeks to analyse the EU as the result of strategies pursued by rational governments acti ng on the basis of their preferences and power The major agenda-setting decisions in the history of the EC in which common policies are crea ted or reformed are negotiated intergovernmentally but can they be consistently xplain in terms of a theory of interstate bargaining Like many intershy

national negotiations EC decisions of this kind can this be thought of as a game of co-ordination with distri butional consequences - in other words a bargaining game over the terms of co-operation (Garrett 1992b Krasner 1991 Sebenius 1991 ) The configuration of domestishycally determined national preferences defines a bargaining space of potentia lly viabI agreement each of which generates gains for one or more participants Governments if they are to pursue a common policy must collectively select one The choice between different agreeshyments often has important distributional consequences governments are therefore rarely indifferent among them Negotiation is the process of collective choice through which confl icting interests are reconciled

Bargaining games raise two analytical problems Lax and Sebenius (1986) refer to these as problems of creating and claiming value They might be thought of also as co-ordination and bargaining aspects of strat gic interaction The firs t problem concerns the efficiency of system negotiations Negotiations create value by facilitating mutually benefishy IS re cial exchanges but excessive costs of identifying negotiating and nicat enforcing bargains may obstruct co-operation Strategic behaviour may oppo lead governments to withhold information about mutually beneficial implica bargains negotiation may require costly threats enforcement may be (M ray

expensive or impossi ble International institutions can help to amelioshy bargair rate som of these problems by proposing potential agreements proshy rracted OJ

viding rules for deci ion-making and the adjudication of disputes T he numero second problem concerns rhe distributional implications of interstate menrs a bargaining The choice of a specific outcome from among many posshy thems sible ones determines the d istribution of expected costs and benefits arrange


an shy among national governments Governments barga in hard for advanshy~ach tage In order to explain bargaining outcomes it is necessary to undershyosrs stand the factors that account for the relative power der- Creating and claiming value often occur simultaneously but they can --a- be divided for analytical purposes In the following section the focus is rhe on the distributional implications Strategic interaction is assumed to

TIro be efficient the choice of agreements is restricted ro those along the Pareto-frontier and the analysis focuses on the international distribushytion of gains and losses In the following section in which the role of supranational insti tutions in assuring efficient bargaining outcomes is

d addressed these assumptions are then relaxed

Bargaining power and the intensity of preferences

It of Negotiation analysis has identified numerous factors that may influshy(heir ence the distributional outcomes of international bargaining among

tbe them the nature of the alternative policies and coalitions the level and TIed symmetry of infornlation the extent of communication the sequence ~ntly of moves the institutional setting the potential for strategic misrepreshyfltershy sentation of interests the possibility of making credible commitments It of the importance of reputation the cost-effectiveness of threats and sideshy)ther payments and the relative preferences risk-acceptance expectations rrert impatience and skill of the negotiating parties (Harsanyi 1977 Raiffa testishy 1982) In the abstract any of these factors might be important predicshye of tors of bargaining outcomes Ie or To generate precise and accurate predictions about a set of compashyma n ra ble cases such as major EC decisions detailed assumptions must be yeneeshy made about the situation in which the par ties are bargaining lents The following three assumptions about interstate bargaining offer a )cess plausible starting point for analysis of EC decision-making First intershyed governmental co-operation in the EC is voluntary in the sense that ruus neither military coercion nor economic sanctions are threatened or alue deployed to force agreement Thus fundamental decision in the EC ects can be viewed as taking place in a non-coercive unanimity voting y of system Second the environment in which EC governments bargaining nefi- is r elatively information-rich National negotiatOrs are able to commushyand nicate at low cost and possess information about the preferences and may opportunities facing their foreign counterparts as well as the technical fici al implications of policies that are of the greatest interest to them y be (Moravcsik 1993b) Third the transaction costs of intergovernmental lioshy bargaining are low Negotiations within the EC take place over a proshy

proshy tracted period of time during which member governments can extend The numerous offers and counter-offers at relatively little cost Side-payshytate ments and linkages can be made Governments can credibly commit posshy themselves to substantive policies through explicit institutional lefits arrangements Technically it is possi ble to design efficient institutions


to monitor and enforce any agreement at any desired level (The assumption of low transaction costs is rela ed in a la ter section of this article)

The asswllption of a non-coerc ive information-rich delib ra tive institutionaliz d setting may not be perfectly realized at all times during the history of the Ee but it is a rea onable firs t approximation of the context in which European governments typically negotiate O ne impli ation of these assumptions i that bargaining outcomes should be eHi ienr in the sense that confl ic t are genera lly resolved Pareto-optishymally Opportunities for useful bargains are exploited Moreover these assumptions reduce the importance of various factors that influence bargaining outcomes elsewhere such as first mover advantages trategic sequen ing traregic misrepresenta tion the use of costly coershy

cive threats and the role of unilateral precommitments EC negotiashytions can be iewed as a co-operative game in whi h the level of co-operation reflects patterns in the pref rences of national governshyments

Yet even in this relatively benign environment relative power matter Bargaining leverage stems most funda menta lly from a ymmeshytries in the relati e intensity of national preferences which reflect according to the analysjs in the previous section the relative costs and benefi ts of agreements to remove negarive externalitie In negoriating policy co-ordina tion the terms will favour those governments able to remove negative extemalitie by opening markets to which others intensely desire access modifying policies others in tensely de ire to change or di trib uting res urces others intensely desire to share The more intensely governments desire agreement the more concession and the greater effort they will expend to achieve it The greater the potential gains for a government from co-operation as compared to it best alternative policy the less risk of non-agreement it is willing to assum and therefore the weaker its bargain ing p wer over the speshycific t rms of agreement

Theorie o f barga ining and negotiation suggest three likely determishynants of inte rstate bargaining power under such circumstances (1) unishylatera l policy al ternatives (threats of non-agreement) (2) alternative oalitions ( th reats of exclusion) and (3) the potentia l fo r compromi e

and linkage

Unilateral alternatives and threats of non-agreement

A n ces ary condition fo r negotiated agreement among rational g ernm ents is that each perceive the benefits of co-operation a prefershyable to the benefits of the best al ternative ava ilable to it Wh re there exj ta p licy more desirable than co-operation a rational government will fo rgo agnement The simple but credible threat of non-agreemem - to Teect co-operation in favour of a superior alternative - provides

-----------------~


-~~ wi th high standards but off by unilateral barriers assured Far from sparking a race to tbe bottom the

_tde market under lowest common denominator barshye1res incentives for the EC to harmonize at a high leveL

lions and the threat of ex clusion

alternatives to agreement are uni lateral pol icies EC er majo r reforms can be thought of as taking place

miry voting system in which agreement requires that the ods of each country be satisfied Sometimes however ltive to agreement is not unilateral action but the forshy

n a lternative coali tion from which certain states are here alternative coalitions are possi ble a government must

alue of an agreement by comparing it not to unilateral but to its gain from alternative coalitions it would join

_ ~g it alone as [it] faces various coalitions (Raiffa 1982 rlStefl ce of opportunities to form attractive alternative

deepen existing ones) while excluding other parties --~~ -fl3 bargaining power of potential coalition members vis-ashy

~ltened with exclusion In the EC context such bargaining ult either from the threa t to co-operate with non-EC

i more common today from the possibility of forming or remative institutions within Europe while leaving some ind - a two-track or multi-speed Europe (M oravcsik coalitional dynamics tend to favour large states whose

~n is necessary for viable coalitioos and governments with -- close to the median of the EC since they are potential

more viable coalitions g negative policy externalities the formation of an alternashy

n creates an incentive for recalcitrant governments to com-Due to the much greater market power involved the threat of

om a coalition is a more powerful incentive to co-operation e states threat of non-agreement To a much greater ex tent ordinated pol icies alternative coalitions - for example an

ree trade arrangement - can create negative policy externalishy-nose left outside it By diverti ng invesunent credit trade f1nence or market confidence exclusion from an alternative l1ay impose significant costs even in the absence of military ruecoercion (Binmore and Dasgupta 1987 9 ) Under these

----uu-s a government may seek to avoid exclusion by agreeing to_

- -o-operation that leave it worse off in absolute terms than the o ante - although of course the agreement is Paretoshy

----ng in the sense that the government is better off as compared to n if the failure to reach agreement had led to the formation

-emanve coalition


A numb r of major events in the histo ry of the EC can be interpreted as responses to the threat of exclusion from an al ternative oa Lition The initial British re ponse to the formation of the Common Market in the 19505 and 1960s is an illustrative example The Bri tish government initially sought to undermine European integra tion by proposing an alternative free trade area When this failed the British sought to di lute the Common Market by negotia ting a free trade deal directly with it and subsequently formed a parallel organization the European Free Trade Association (EFTA) Only when each of these strategies had fai led did Brita in finaily apply for membership of the EC - only to find that the adjustment of other countries to the Common Market had shifted r la tive argaining power even fu rther against it Th i is a ca e in which Britajn while it would have gained from membership wouid nevertheless have preferred the status quo ante

Yet al ternative coalitions do not always create negative externalities for excluded tates and therefore pressure for geographical spillover W here a policy of exclusion has positive external ities a contrary dynamic occurs Where free t rade i assured for example governments with I w social standards often have a clear incentive to free ride rather than to compromise on common harmonized standards This h Ips explain why the threa t of exclu ion was powerless to block the British governments strikjng last minute opt-out of social policy at Maastricht Excl u i n from the so ial policy provisions of the M aastricht Treaty insofar a it had any effect at all promised to make British firms more c mpetitive on a European market from which the cannot be xcluded The adoption of high EC social protection stanshydards is thus likely to be possible only through linkage or side payshyments which play such an important role in cementing co-operation wi th the Mediterranean c uneries

The distinction between po itive and negative externalities provi des a means of predicting which policies are inherently expansive - thus resolving an am biguity in neo-functionalist theory Where policy extershynalities are nega tive non-member have an incenrive to join the organ ishyzation which will lead them to compromise on c mmon standards Where policy externalities are positive non-members have an incentive to free ride rather than compromi and agreements above the lowest common denominator are possibl only through linkages and side payshyments to which we n w turn 21 This not onJy helps to explain the dynamics of geographica l expansion in the EC but also the dynamics of current barga ining over regulatory issues

Compromise side-payments and linkage at the margin

Unilateral and coalitional policy alterna tives define a range of iable agreements which all participants prefer to the status quo Within tha t

AndreU Moravcsik 289

im at which negotiators will compromise is more - ~l1Icr particularly when more than two states are

rcal bargaining power will depend on the intensity of e margin Where uncertainty exists about the breakshy

ns or time pressure concessions tend to come disshy- om governments for which the fail ure to reach lx- least attractive - that is from those governments

the most if agreement is not reached Where such nor exist the terms of the final agreement will reflect

Ity of preferences at the margin which defines the --asible set govern ments that place a greater va lue on

bull -Me margin will gain more from negotiations22

lT3ntly for our purposes here governments often have difshyCfJce intensities across issues with marginal gains in

e lS being more important to them than to other governshymiddotIese circumstances it may be to the advantage of both an~e concessions in issue-areas about which their prefershy~ -ely weak for concessions in other areas about which T~ Even where a set of agreements taken individually

rejected by at least one national government they may i antages fo r all if adopted as a package deal

- mitation on linkage stra tegies is domestic opposition e important domestic distributional consequences They

~r mo package deals issues in which domestic groups benefit hich domestic gro ups lose Package deals tend to create losers in all countries that are party to them Where

and losses produced by linkage are only imperfectly ough compensation across issues linkage becomes a J politically risky stra tegy Since losers tend to generate aJ pressure than winners for a domestic trade-off to be tolshynnent costs to important domestic groups must be modshy[aorial compensation must be paid

rtll1ce of domestic costs and benefits suggests a number of ---~-c-~ns aboU( linkage First linkages are most likely in areas where

--F~~ces oJ domestic groups are not intense Minor issues are (0 be sacrificed to a linkage Wherever possible therefore

r srmbolic side-payments between states rather than linkshyetn substantive issues are employed The Maastricht agreeshytypical in tha t issues implicitl y linked to monetary policy 19h1y fungible resources such as increases in structural vmbolic issues such as deletion of federalist language and

vers to the European Parl iament Second package deals -e1) in the final stage of bargaining - that is at the margin

( gams and losses among issues in which all parties are close beneficiaries - rather than among issues in which nations

_90 Preferences and Power in the European Community

are la rge net winners and losers Third linkages are most likely between closely related issues - within rather than between sectors Where the costs and benefits are internalized to sect rs or firms there is more possibility for producers to adjust diversify or to balance gains and losses just as in the case of intra-ind ustry trade Sectoral organiza tions may neutrallze opposition by aggregating sectoral support and pposition into a single posicion Linkages between disshyparate sectors are mo t likely to occur where the possi bilities for intrashyissue compromise or linkage between related issues have been exhausted Fourth if linkages do impose real losses on domestic sectors they are more likely to be effective when accompanied by domestic side-payments from govemments to disadvantaged private groups In the 1960s and 1970s industrial su bsidies in France and agricultural subsidies in Germany were explicitly designed to ease adjustment to li berallzation

Linkage is thu a politically costiy second-best strategy for integrashytion Linkages that attempt too much - such as the linkage between the Common Agricul tural Policy and strong supranational institutions in the 1960s - are often unstable and are circumvented at a later stage The limitations on linkage are illustrated by the purported linkage on which the EC is said to be founded namely that between German access to French industrial marke ts and French access to German agrishycultural markets While such a linkage existed on the margin it was less central than is oft n a serted Industrialists and farmers in both countries gained French industrys objections to the Common Market were in fact relatively minor by 1959 before tariff reductions had begun in earnest they were already among the strongest supporters of acceleration Opposition to a common agricultural policy came prishymarily from economic li berals in the German government who opposed high prices and farmers who feared low prices The final agreement left farmers in every country including Germany with higher average support prices than they had en joyed previously Those elements of the CAP price structure that most disadvantaged certain fa rmers were offset by domestic compensation and adjustment assisshytance 10 the 1970s any residual loss to German farmers was more than offset by the compensation for currency movements and the subshysequent renationalisation of the CAP leaving only division f the much smaller budgetary expenditures as an outstanding issue 1n contrast to no-functionalism which viewed linkage as the core of the EC it is seen here as a strategy best pursued on the margin and of lesser imporshytance than intra-sectoral trade-offs Linkages that impose large losses on important domestic groups are unstable

So far this analysis has focused primarily on the sources of national preferences and the distributional outcomes of intergovem mental negotiations over commercial liberalization domestic public goods


~ eral political and institu tional questions We turn SIS of the distributional outcomes of intergovernshy

--6~iTinE to an analysis of its efficiency Modern regime emarional institutions as deliberate instruments to -lCncy of bargaining between states

-onal Institutions and the Efficiency of

9

nat insti tutions are often seen as the antithesis of intershyrongly so The decision to join all but the most minshy

_~ ~ uwolves some sacrifice of national autonomy which cal risk to each Member State in exchange for certain

intergovernmentalist view the unique institutional ~ is acceptable to national governments only insofar as

r than weakens their control over domestic affairs attain goals otherwise unachievable

ngthen the power of governments in two ways -TI ~t me efficiency of intersta te ba rgaining The existence

oriating forum decision-making procedures and Hance reduce the costs of identifying making and

_=aeatE ~6lers thereby making possible a greater range of coshyems This explanation reljes on the functional hich focuses on the role of regimes in reducing

- ohane 1984 H owever in order to explain the ~urion alization found in the EC this body of theory

incl ude the delegation and pooling of sovereignty YIUnl)OS strengthen the autonomy of national political

flt1cularistic social groups within their domestic ~11I1 the legitimacy and cred ibility of common polishy

rthening domestic agenda-setting power the EC ~rmiddotel game that enhances the autonomy and initiative

~_ ---oJa11 _eaders - often as noted above a prerequisite for ralizati on With a few important exceptions EC

~- ~ be explicable as the result of conscious calculashyto strike a balance between grea ter efficiency

-~ on the one hand and acceptable levels of pol itshy

stitutions and functional regime theory

al structure of the EC can be readily explained bull of regimes which argues that where transacshy identifying issues negotiating bargains cadishymonitOring and enfo rcing compliance - are


significant international in tirutions may promote greater co-operati on by provid ing information and reducing uncertain ty In the convenshytional regime-theoretica l vi w EC ins titutions serve as a pa sive strucshyture providing a contractual environment cond ucive to effi cient intergovernmenta l barga ining As compared to ad hoc negotia ti n they increase the efficiency of bargaining facili tating agreements that could no t otherwise be reached (Buchanan and Tull ck 1962 Keohane 1984 Levy et a 1992

The functional regi me theory view of international insti tutions as pas ive transaction-co t reducing sets of rules readily explains the role of EC in tirutions s a fram work for negotia ti ng major decisions from the Treaty of R me to Maastricht The acquis communautaire of the EC fun tions to stabilize a constantl y evolving set of rules and expectation which can only be altered by unanimous consent In tirutions promote international co-operation by providing a negotishyating foru m with bu reaucra tic institutions tha t disseminate informashytion and poli idea a locus for representatives of business pol itical parties national hureaucracies and inter st groups to discu s issues of omm n on ern joint decision-making procedures a common et of

underlying legal and political norms and institu tions for monitoring nd definmg national compliance Greater information and preshy

dictability reduce the cost of bargaining and the risk of unilateral oonshycomplian e Like the GAIT the G-7 and other international regimes E in tirutions provide fora in w hich to craft linkages and side-payshyment that render policy co-ordination more viab le domestica lly Package deals linking regional funds and British entry or structural lunds and [he SEA were surely ea ier to reach within a common intershynational insri t tioo Yet the large political risk inh rent in open-ended deci ion a out the future scope of EC activities means that Member

t re remain hesitant to delegate authori ty to supranational or majorishytarian in riru ions The essen e of th EC as a body for reaching major deci ion remain its transaction-cost reducing function as explicared by contemporary regime rheory

When e nun from major constitutional decision-making to the proces of everyday legislation ad ministrat ion and enfo rcement howe er the EC seems to be a fa r more unusual international institushyti n - more than a passive s t of rules codi fying previo us decision The EC differ from nearly all other international r gimes in a t least two s lient ways by pooling national sovereignty through QMV rules and by delegating sovereign powers to semi-a utonomou central instishytution These twO forms of transferring nati onal sovereignty are closely related QMV for example not only makes the formal decishysion-ma kina- of any single government more dependent on the votes of its foreign counterparts but also more dependent on agenda-setting by the Commission

llmity

-eater co-operation In the COO venshy

as a passive strucshyucive to efficient

hoc negotia tion agreements tha t

k 1962 Keohane

131 institutions as explains the role major deciSions

ommunautaire of et of ru les and

nim ous consent Oyjding a negotishyminate informashylusiness pOlitical discuss iss ues of

I common set of for monitor ing arion and preshyi unilateral nonshyational regimes s and side-payshye domestically r or str uctural

ommon intershyt in open-ended

that Member nnal or majorishyreaching major I as explicated

aking to the eniorcement

ational insti tushyious decisions Des in at least ~ QMV rules ) CfntraJ instishyIereignry are forma l decishyn the Votes of lda -settjng by


rand the conditions under which Member ~tates slOn-making un der the unanimity ru le in favour

menr to pool or delegate sovereignty contemporary t be fxrended An insightfu l starting point suggested ngast in their analysis of the ECj is to view delegashy

the problem of incomplete contracting Predicting - ~nces under which future contingencies will occur is often

i) sometimes impossible (Garrett amp Weingast 1991) governments have shared goals but are unable or

esee all future contingencies involved in the realization they may have an incentive to establish common

procedures or to empower neutral agents to propose ent interpret and enforce agreements r o f incomplete contracting per se while a useful

ption fails to explain variation in ei ther the level or the rion (or pooling) of sovereignty Delegation is after all

a nu mber of possible responses to future uncertainty [able EC decisions - including the annual determi nation and the definition of new issues under Art 235 - are

red nor pooled others - the determination of intemashyring positions and administered protection against third

- ue pooled but not delegated Elsewhere in the internashyt delegation is even rarer despite many cases of incomshy

--lcting Even withi n the EC governments often refuse to politica l risk of delegation preferring instead imperfect

-~~err and inefficient decision-making to the surrender of sovershympJete contracting appears to be neither a necessary nor a ondition for delegation en distinguishes cases of delegation or pooling from cases

unanimity voting Following pu blic choice analyses of ~ constitutional choice intergovernmentalist theory views the

o adopt QMV or delegation to common institu tions as the cost-benefit analysis of the strea m of future substantive decishy

cered to follow from alternative institutional designs For ~ember States carrying gut such a cost-benefit calculation

on to delega te or pool sovereignty signals the willingness of overnments to accept an increased political risk of being outshy

overruled on any ind ividual issue in exchange for more effishylective decision-making on the average 23 Movement beyond us voting and ad hoc negotiation for a class of decisions can

iliought of as a means of deliberately encouraging implicit linkshyvarious related issues within an itera ted game among govshy

s By facilitating linkages delegation or pooling is likely to more decisions at a lower COSt in time and energy than the

ugt negotiation of ad hoc package deals Compared to unanimity


voting delegation and pooling of sovereignty are more efficient but less controlled forms of collectiv decision-making Of the two delegashytion involves greater political risk and more efficient decision-making while pooling through QMV involves less risk but correspondingly less efficiency

Examining this trade-off more precisely the following three condishytions should en ourage national governments to support a movement from unanimity to delegated or pooled decision-making (1) The potential gains from cooperation W here time pressure previous failshyures to reach agreement the desire to implement a prior decision or a shift in national preferences requires more rapid decision-making deleshygation or pooling is more likely Ceteris paribus the less attractive the status quo and the greater the expected gains from increased co-operashytion the greater the corresponding incentive to pooL or delegate Levels of economic transactions and in particular intra-industry trade which are higher among the EC countries than among any comparable set of industrialized countries are likely to lead eventually to pressure for grea ter delegation and pooling of sovereignty Where large numbers of similar deci ions are involved the efficiency gains are correspondingly greater ( f Keohane 1983)

(2) The level of uncertainty regarding the details of specific delegated or pooled decisions Lack of precise knowledge about the form details and ou tcome of future decisions not only precludes more explicit conshytracts as nored above but also helps defuse potential opposition from those who would be disadvantaged by the impl icit linkages Where agreements can be ioreseen some governments and domestic groups would have more reason to prefer direct bargaining under unanimity as occurred in setting the initial levels of the Common External Tariff and agricultural prices in order to block policies disadvantageous to them

(3) The level of political risk for individual governments or interest groups with intense preferences Political risk can be understood as the probability of a large downside loss to a government or interest group Risk-averse governments will assent to procedures where the scope and magni tude of expected and potential losses are min imized given the goals of co-operation Governments have an incentive to delegate authority only when there is little probability that the cumulative disshytributional effects of delegated or pooled decisions will be biased in an unforeseen way against the interests of any national government or major domestic group24 The form of third-party representation agenda-setting and enforcement should involve the minimal tr ansfer of sovereignty needed to achieve desired outcomes One way to limit the scope of delegation and pooling often employed in the EC is to nest specific decisions inside a set of larger decisions already reached by unanimity thereby both diversifying and limiting political risk

but ega shymg ngly

ndi shylent The fail shyor a leleshy~ the lerashy~veJ s hicb ~t of for rs of ngJy

ated tails con shyfrom here oups n ity middotariff IS to

erest s the oup and

I the gate disshy

rl an t or lOn r of the

[lest by


the truee most important instances in which th e Treaty of ates Member State a uthority to suprana tional o fficials shy

~epresentation agenda -setting and enforcement - appears to onditions

representation Since the EC is a customs union w ith a X1ernaI tariff negotiations w ith th ird countries require a

=Mt to represent common positions In order for national govshy~5 ~ rrust the agen t it must be perceived as neutra l While this ~=-~ that an agent be delegated o nly limited independent decisio nshy

z for short periods of tim e is reqwred to carry out designated e monitoring and oversight by na tiona l governments is to be

~--1 In the com mon commercial policy of the EC for example mission represen ts the Commw1ity but tight control is mainshy

the Article 113 Committee O nly where time pressure in the gotiations forces a r apid decision and national governments

-ed can supranational officia ls advance independent initiashyl are still subject to ex post approval but may transfer some

power to the Com mission In European Po litical Co-operashyece fewer decisions are taken and a co mmon ex ternal p osition

as less in1p erative the EC is generally represented by the nis ters of its Member States

lt--lt=aJ-setting Where a wide consensus exists on a broad substan shyda it can o ften be realized m ore efficiently by gran ting a poundagenda-setting power to a supranational institu tion in this

Commission As a reliable source of independent proposals middotUmission assures tha t technical information necessary for decishyldilable More importantly as a neutra l a r biter it p rovides an ruive means of reducing the num ber of proposals to be consid shy

bullhlOrities may exist for a nu m ber of alternative proposals o n a ue with governm ents unable to red uce them to a comp romise

~-~ their vetoes This is particular ly important w here governments ght to increase the effic iency of bargaining by employing

uch circumstances agenda-setting p ower can be decisive in e which proposa l p reva ils In the EC delegating the power of ~ to the Commission provides a means of setting the agenda

voiding tinle-consunling or inconclusive cycling between difshy~ posals or an arbitrary means o f proposal selection 25

ability to select among viable proposals gran ts the 5ioD considera ble formal agenda-setting power at least in

The power is particularly decisive when the status quo is unat shy~j ~Teating general support for joint action yet there is considershy

Ueement between national governments over what should - Often a num ber of proposals might gain majority support -hich the Commissions cho ice is decisive T he most controver-

J of implementing the White Paper agenda including a number


in telecommunicarions environmenta l and social policy stand as examples (For a simila r ana lysi of pa ruarnentary power see Tsebells 1992 )

Enfor ement The possibilities for co-opera tion are en hanced when neu tral procedures exist to morn to r interpret and enforce complia nce N eutral enforcement permi t governments to extend cred ible commitshyments thus helplng to overcome the almost inevitable interstate pri shyoner s dilemma of enforcement whereby individual governments seek to evade inconveruent responsibi lities thereby undermining the integrity of the entire ysteIll By taking the definition o f compliance outside of the hands of national go ernrnents a suprana tional legal system rrengthens the credibi li ty of national commitm nts to the instishytution The COSt of such delegation whi h goes beyond the morutoring function of la iea l internationa l regimes is inc rea ed political risk Functi n of chi ryp in the EC include competi tion policy adminisshytered by the Commi sion and the interpreta tion and application of EC law arried our by the ECJ (Garrett and Weingast 1991 )

In each of the e three case there is a substantive comm itment to the a hievemenr of broad goa ls w hile the political risk is mall inso fa r as each delegated decision is relative ly insignificant Perhaps most imporshyrant che cope of delegation is explicitly limited by national governshyments

Of the three type f delegation only the enforcement power of the ECj appears to have resulted in a grant of independent initiative to supranati nat bodies beyond that w hich is minima lly necessary to perform its functions - and beyond that which appears to have been fore een by overnments The ECl has consriturionaized the Treaty of Rome built al liances with domestic courts and interest groups preshyempred national law in importanr areas and opened new venues for Commi ion in itiative as in cases hke ERTA in common commercial polic and Cassis de Dijon in technical harmonization

The expan ion of judicial power in the EC presents an anomaly for the fun tiona l explanation of delega tion as a deliberate means by national gomiddoternments of increasing the ffici ency of collective decisionshymaking While supranational delegation undoubtedly creates benefits for governments the decisions of the Court clear ly transcend what was iru tia lly foresee n and desired by most national governments (Burley an Mardi 1993) The constitutionaliza tion of the T reaty of Rome wa unexpected It is implausible moreover to argue that the current system i the one t which all national gOY rnments would currently consent as rec fit explicit limitation on the Court in the Maastrich t T reat middot demon middottrate Nor is the current institu tional form of the Court functionall y necessary Supranational dispute resolution need not take the form almost unique among international organiza tions of a semishyautonomous legal system Such a system is not a priori more approshy

AndreU MoraLlcsik 297

utes between rival interpretations and applica shy~n a dispute resolu tion panel as exist) in the il of Ministers acting under qualified majority a)

he f-lember States might simply have reserved the -arion to clarify ambigui ties Neither incomplete conshy

anal analysis can account for the precise form or hisshyi the Eel

Ie the ECJ has come to play may reflect instead a lITIi argue a number of factors idiosyncratic to the EC cgtical complexity of EC law made it difficult to foresee

~ut~ces of early Court decisions gi ving those who favoured a me leeway in drafting the trea ty (Pescatore 1981)

- - referral of cases by domestic courts and their subsequent ~_=-uwiy in the same fo rum render EC] judgments difficult to

Ii and most importantly unanimous consent of Member Ie now be required to curb its power Over the years the ursued a sophisticated strategy remaining just within a

~~ c-llT1SenSUS that protects it Any attempt to alter the current --1~~ bull ItS might be chalJenged by European federalis ts by those r-

f strong enforcement and by smaller countries which well served by a system in which QMV was employed to

e disputes (Burley and Mattl i 1993 )Ie

~he creation of common rules and procedures in functional -=-_=t ~eory al ters only the information and expectations of nationa l p --~~nts the EC goes fu rther pooling decision-making through

n ~p~ents for QMV and delegating authority over representation

genda-setting and enforcement to semi-autonomous institushy[ the delegation and pooling of authority ill the EC like the

If tion of common norms and principJes in other regimes can be a1 d by extending the centra l insight of functional regime theory that insti tutions are means of reducing the transaction costs of

if

ing negotiating and enforcing intergovernmental agreements uncer tain ty National governments strike a ba lance between

~ed decision-making efficiency and the political risk of W1conshyiUUU issue linkage The greater the potential gains the greater the

- etainr) about specific decisions and the lower the political risk re likely governments are to delegate power in these ways e

~wed in light of this trade-off independent actions by the ummission or outcomes that contravene the interests of a single

er State taken in isol ation do not constitute decisive evidence r IDst the intergovernmentalist view that the EC is grou nded fundashy

[ ally in the preferences and power of Member States Only where

actions of supranational leaders systematically bias outcomes away m [he long-term self-i nterest of Member States can we speak of ous challenge to an intergovernmenta list view While some cases of


supranational autonomy such ascertain actions of the EC] may pose such a challenge most fit comfortably within it

2 Supranational institutions and two-level games

Traditional regime theory focuses primari ly on the role of regimes in reducing the transaction costs of collective decision-making for national governments Yet EC institutions perform a second function as well namely to shift the balance of domestic initiative and influence On balance this shift has strengthened the policy auton my of national governments at the expense of particular groups (for a disshysenting view see Marks 1 91 )26 Particu larly where domestic interests are weak or divided EC institutions have been deliberately designed to assist national g vernments in overcoming domestic opposition Where institutions did not initially serve th is purpose in the Treaty of Rome new in titutions were created in order to strengthen this function - the strengthening of the Council bureaucracy in the 1960s the genesis of the European Council the 1970s and the reservation of powers over political co-operation to the Member States being prime examples

National governments employ EC institutions as part of a two-level strategy with the aim of permi tting them to overcome domestic opposishytion more successfully (d Putnam 1988 ) The EC fulfils this function in two ways by affording governmental policy initiatives greater dome tic political legitimacy and by granting them greater domestic agenda-setting power Let us briefly consider each The mantle of the EC adds legitimacy and credibility to Member State initiatives Domestic coalitions can be mobilized more easily in favour of policy co-ordination This add weight in domestic debates to both maj or reforms and everyday decisions emanating from the Ee Second the institutional structure of the EC strengthens the initiative and infl uence of national governments by insulati ng the policy process and genershyating domestic agenda-setting power for national politicians National governments are able to take initiatives and reach bargains in Council negotiations with relatively little constraint The EC provides informashytion to governments that is not generally avai lable In tergovernmental discu sions take place in secrecy national votes are not publicized D mestically parl iaments and publics generally have little legal opporshytunity to ra tify EC agreements and decisions where they do there is rarel y an opportunity to amend or revise them N ational leaders undershymine p (entia) opposition by reaching bargains in Brussels first and presenting domestic groups with an up or down choice - just as fast track procedure are employed to speed trade agreements through the US Senate (Destler 1986) Greater domestic agenda-setting power in the hands of national political leaders increases the ability of governshyments to reach agreement by strengthening the ability of governments

bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC

be es cy lor he Ice ershyla l leil la~

tal ed rshy

crshyod 1St

he tn nshyIt


~Iation for compromises or tactical issue linkages ~tS might be pressured for exemptions oversight

placed instead in the bands of the more credible lal court systems Ironically the democratic defici t

t=damental source of its success inn ing much EC decision-making has been diffishy

ept as a two-level game T he reflexive support of European federalists and those who favour the

=--=_ ecoOl1K goals of the EC grea tly ass isted the early developshycecifk EC policies In the initia l negotia tion of the T reaty of

ralizatioll of French industr ial trade o ffers a striking -e 19505 Germ any the hu b of the European trading

aging in unilateral tariff red uctions French exports e increasing r ap idly with little evidence of any proshy

non across the Rhine (Milward 1992 ) The major - France to accept the EC was not to solve an internashy

~n~rs dilemma by assur ing access to the German market em trade theory would have it Instead as French clear a t the time and de Gaulle was to reiterate even it was to employ the legitimacy of the EC to force o modernize - a goal tha t French governments had

p~tng for almost a decade without success (Institut Charles 6)

are witnessing an analogous phenomenon as the credishy ettlflS to achieve macroeconomic convergence in countries

alld Spain is bo lstered by the impression deliberately mem ber governments that the imposition of anti-inflashypline is necessary for full in volvement in Europe

i-Osed independent Eurofed offers a more recent example of Lages of insulating agenda-setting policy implementation

ric pressures - -to an extent that has been w idely criticized e-r to achieve go that would otherwise be unachievable

~ control over exchange rate policy (as well as the legitimacy Jrorean integration) has permitted German Chancellor H elmut

JJgtue a policy of moneta ry integra60n w ithout the strong either business or the Bundesbank although limits imposed

~cstic consensus have subsequently become clear In addition ~)sed European Central Bank will be do ubly insulated from

pressures in a way designed to make the common European credi ble on domestic and international markets The

~JCnr referendum in France is an exception tha t proves the ---o--2Dce of secrecy and agenda-setting power in tha t it demonshy

potential consequences when governments lose firm control DC agendas or take needless risks in ratification


VI Conclusion Beyond Liberal Intergovernmentafism

The liberal intergovern mentalist view seeks to account for major decishysions in the history of the EC by positing a two-stage approach In the fi rst stage national preferences are primarily determined by the conshystra ints and opportunitie imposed by economic interdependence In the second tage tbe outcomes of intergovernmenta l negotiations are determ ined by the relative bargaining power of governments and the functional incentives for institutional ization created by high transacshytion cost and the desire to control domestic agendas This approach i grounded in rundamenral concepts of international political economy negotiation analy i and regime theory

By bringing together theories of preferences bargaining and regimes liberal intergovernmentalism provides pIau ible accounts for many aspect of the major decisions in the history of tbe EC in a way that is barpJy dt rinct from neo-functionalism Where neo-functiona lism mpbasize domestic technocratic consensus libera l intergovernmenshy

talism looks to domestic coalitional struggles Where neo-functionshya1ism emphasizes opportuni ties to upgrade the common interest liberal inrergmernmemalism stresses the role of relative power Where neoshyfun tional ism emphasize the active r Ie of suprana tional officia ls in shaping bargaining outcomes liberal intergovernmentali m stresses in tead passive institutions and tbe autonomy of national leaders Ironically the ECs democratic deficit may be a fundamental source of its success

10reover liberal intergovernmentalism provides explanation for ome nagging anomalies inherited from neo-fun tionaJism Variation

in the tightness of domestic societal constraints is employed to ex plain the disruptive role of dramatic-political actors and the distinction between those issues where linkage or compromise is po si ble and those tn which log-rolling or lowest common denominator solu tion prevails T he distinction between posi tive and negative externa lities help explain which issues generate common solutions and spark geoshygraphical spillover and which do not The introduction of a two-level gam analysis explains why France sought industrial trade liberal izashytion with Germany in the 1950s despite the unilatera l op nness of the German economy at th time

Cri tics may challenge the approach propos d here in three ways Fir r th may dispute the basic framework arguing that sta te behavshyiour i not purposive and instrumental that preference formation does not precede the form ulation of strategies or that nationa l preference and intergovernmental bargaining are so completely manipulated by supranational officjals as to be mean ingless categories Second they may challenge rhe liberal understanding of state preferences employed

301

~jsm

r decishy In me he COD shy

nee In Ons are md the cansaeshy

aeb i nomy

gime manr

[hat is nalism nmeD shylction shyiberal e ncoshyials in tresses middotaders ouree

)5 for ia tion ICplain Jerion e and lu tion nlj tie geoshymiddotJevel aliza shyf the

ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen

ri~ of integration such

~ ded the

1 tmem that

135h between bull withou t any specification - be expected ta apply

of the EC ~--J_atry Dlay lie not in the transcendence of traditional state prefershy

d power

models of politics potentia lly applicable to

~I~ewhere

bull contemporary theories of economic inrerdepenshy-jona preferences Alternative conceptions o f eco shyertainly possible as are (liberal and non-libera l) on ideology or geopolitics T hird they may quesshy

-~mDenral theory of bargaining with its stress on barshyed in unilatera l alternatives competing coalitions

linkage and the controlled delegati on of power to rIbrirurions under conditions specified by functional t heshy

md two-level games views of domest ic polities Such

d go so far as to deny the importance of preferences Indeed a strong liberal intergovernmentalist

as a precondition for (he development of more as neo-functionalism Without

ell of sta te interests interstate bargaining and interna shyit is im possible to determine when consequences are

cammon interest is truly being upgraded o r officials are truly acting au tonomously T his vinrucates

debate between general theories of domestic na l politics is necessary Such a debate is surely p refershy

intergovemmentalist and supranationalist of the conditions under which

IOly true tha t li beral intergovemmenta lism accords supranashy-woons and officials less weight and prominence than neoshyIsm once rud Committed integrationists typically read such

ns as a disparagement of the unique achievement and futu re Yet th e real achievemen t and hope of the

but in the underlying domestic and international at have shaped nationa l preferences and power in the di rec tion

er co-operation Liberal intergovernmentalism assimil ates the a ll states thereby

ng conditions under which a similar process of in tegration may

ates

indberg and Scheingold there are five alterna tive models only one of 15 sp illover Nye expanded this to seven process mechanisms or actor stra tegies

rwo (1 an d 3) of which related to spi llover (d Lindberg a nd Schcingold 1970 134-9 md 1973 119 Schmitter 197 1 232- 64 For an overview see Nye 1971 b 64-75

_mpiricaJ exception is I-laas 1966 93- 130 Haas examines other international regimes his [beo retical explanation if inductive rather than grounded in micro-analysis of


3 Lindbergs otherwise insightful analysis in Political Dynamics largely overlooks this di shytinction See also Von der Groeben (1982)

4 Such goals are best seen not as defined across alternative policies or strategies (eg a free trade regime fixed exchange rates) but across alternative future states of the world (eg higher levels of economic transactions exchange rate stabiliry) Rational choices among policies and strategies mu t generally take into account the expected reactions of other stares and the resulting strategic interactions among them while preference across future states of the w rld do not The latter are pre-strategic preferences On this distinction more generally see Elster (1986)

5 To avoid confus ion it is important to remember that nested within the domestic definishytion of the demand function is also a national process of societal demands for and govshyernmental supply of policies T he domestic use of the metaphor of demand and supply is d rawn fr om hep Ie analyses of legislative politics (Shepsle 1992)

6 Section~ rv and V of this article deal respectively with two different aspects of interstate strategic interaction distributional bargaining and the delegation or pooling of decisionshymaking in in terna tional regimes Regime theory is treated as a theory of strategic interacshytion in the sense that the institutional measures for compliance shape the range of potential bargains The latter might also be thought of not as an element of strategic interaction but as a separate compliance stage of policy co-ordination For a model of this kind see Moravcsik 1989

7 Moravcsik 1992b from which the argument in this section is drawn 8 See Gourevitch 1978 To the extent that international factors such as economic interdeshy

pendence or external threats to national securiry influence preference formation they must pass through the domestic poliry

9 Here I do not mean to imply that all cases in which governments do not serve the intershyests of particularistic groups should be thought of as autonomous action but simply that the greater the slack in the relationship between particularistic opponents of coshyoperation the greater the possibility of pursuing a policy targeted at a larger domestic constimency or in some cases at realizing the preference of these in office

10 This diverges from the common analysis of regimes as providing public goods The instishytutional infrastructure of regimes itself may be thought of as a public good as can some common goals of regimes For the most parr however the benefits of the EC are excludshyable and to an extent rival goods cooperation stems from interdependence - the effects of national policies on the opportunities for foreigners

11 This is overlooked by Grieco (1988) but captured by Keohane and Nye (1989) and Krasner (1991)

12 The existence of such a predictive theory distinguishes liberalism from the neo- functionshyalist tradition of Haas and others (Haas 1964)

13 A more detailed model would take variations in domestic institutions into account 14 Under conditions of high domestic factor mobiliry one would expect coalitions to form

between capital and labour according to the Stolper-Samuelson theory protection will be sought by factors of production that are relatively scarce - capital in labour-abundant countries and labour in capital-abundant countries In the long term this may be valid (see Rogowski1989) In the short and medium term however many factors are unable to move between industrial sectors due in part to high fixed investments in human and physical capital Hence a specific-factors (Ricardo-Viner) approach is more appropriate in which owners of capita l (or land) and labour work together to form sectoral coalitions (Magee et al 1989)

15 This is the cornerstone of most modern empirical studies of commercial policy For empirical support see Lavergne (1983)

16 Since protectionist policies can easily be implemented unilaterally the incentive for intershynational cooperation in these areas typically stems from opportunities to co-ordinate the liberalization of market access

17 Milner (1988) stresses intra-industry trade

Uk shy

J05umiddot

~

udshyren

I~d

1011shy

ill Wt orlid wle Ilnd are OIlS

For

recshythe

Andreu Moravcsik 303

that the two are always in conflict See the examples drawn from EC ~-rmuion cired below

1991 29~1 Keeney and Raiffa 1991 Railla 1982 252- 5 Zamnan - ne and Nyc (1988) refer to this as vulnerability Hirshman develop ~ lllllployed the p hrase opportunity COSt of non-agreement and excl ushyI (1992a ) ch 1 Garrett (1992a) has applied this idea to the EC

332-4 Where a welfare reducing agreement is reached it is unlike ly to be menred Putnam (1988)

srence of alternative coalitions has been presented here as a source of power -alions over the future scope of the EC it is televant also to QMV on more In bargain ing among themselves over the precise terms of a directive or

nlnonal governments weigh the COSts of compromise which results in a rion of which they arc a member and an outcome closer to their preferred

~~_ uunst the risks of intransigence which may result in exclusion from the winning =- _-J an outcome more uncongenial to them Sash bargaining solution whereby a marginal redistribution in either d irection

acto r with concave utiliry functions would lead to an equal percentage rir utili ty This is also the equilibrium of an offer-counteroffer game in which are assumed to be equal in all respects other than their preferences See

bull M Dasgupta (1987) 5 assumes that the transaction costs of institutional creation and reform are

v but the transaction costs of individual decision-making are high The limishymore optimal international institutions stem not from transaction costs of eremiddot but from the interests of governments in reducing domestic political risk This to promote a set of decision-making rules consistent with a specific trade-off

fflCiency and risk -tirutions may be biased in a predictable manner for which a rational governshy

u1d demand compensation in negotiations manimiry voting this is less of a problem each government can compel com proshy

~ etoing any proposal that does not accommodate its views leading to a comproshynOng proposals Hence the sequence in which proposals are voted upon is less

I ugues on the basis of an analysis of the structural and regional funds that the

cata lyzing a process of diffusion whereby the decisional powers of the state are iliifted both to subnational and supranational institutions

A11drew Moravcsik 265

strategic environment An understanding of the preferences and power of its Member States is a logical starting point for analysis Although the EC is a unique institution it does not require a sui gelleris tbeory

If From Pre-theory to Theory

The limitations of neo-functionalism

Despite the richness of its insigbts neo-fu nctionalism is today widely regarded as having offered an unsatisfactory account of European in teshygration (Cornett and Caporaso 1992 Haas 1975 Hansen 1969 Hoffmann 1966 Keohane and Hoffmann 1991 Keohane and Nye 1975 Taylor 1983 Webb 1983 ) The most widely-cited reason is empirical neo-functionalism appears to mispredict both the trajectory

nd the process of EC evolution Insofa r as neofunctionalism advances a clear prediction about the trajectory of the EC over time it was that the technocratic impera tive would lead to a gradual automatic and incremental progression toward deeper integration and greater suprashy

national influence (Haas 1964 70 1967 327 1976 176) Instead however the process o f Community-building has proceeded in fits and starts through a series o f intergovernmental bargains Nor has the process by which integration takes place supported the neo-funct ionalist view Integration has only intermittently spilled over into related sectors and policies and at least until recently the autonomous influence of supranational officials has increased slowly and unevenly if at all

While empirica l critiques of neo-functionalism are not without merit they should not be overstated To be sure tbe empirical evidence does not seem to confirm the stress placed by neo-functionalism on political spillover and the au tonomy of supranationa l officials But other premises particu larly the focus on economic interests may still be viable It remains plausible for example to argue thar integration is a distinctive policy response of modem welfare states to rising economic interdependence

A more incisive criticism of neo-functional ism is theoretical namely tha t it failed to generate an enduring research programme because it lacked a t11eoretical core d early enough specified to provide a sound basis for precise empirical resting and improvement Only the early variants of nco-functionalism predicted a steady development toward -eder alism Faced wi th the fail ure of Eur opean integration to advance steadily and variation in integration across issues time-periods or ountries (spillback spill-around encapsulation ) however neoshy

functionalism provided no clear direc tion for revision 1



















-bull



















e it




a t









- -- - - - --
















I middot








~85

ng en Or

se 19

It

n




















cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe


































-bull



















e it




a t









- -- - - - --
















I middot








~85

ng en Or

se 19

It

n




















cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe


























e it




a t









- -- - - - --
















I middot








~85

ng en Or

se 19

It

n




















cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe






















- -- - - - --
















I middot








~85

ng en Or

se 19

It

n




















cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe























~85

ng en Or

se 19

It

n




















cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe



































cc





















pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe


























pn Ie























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe





























in whi qu iet









bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
















bull






5 Coneusion




tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe


















tion shy





















and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe






























and linkage



-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
















-----------------~





















-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe




































-emanve coalition

































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
















































9
























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe























llmity




k 1962 Keohane



































but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe





















but ega shymg ngly



erest s the oup and

I the gate disshy


[lest by











































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe





































if













bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe





















bull

gte

in or n

e of is-

Ire Ie

he of er

el 1shy

On fer DC


tal ed rshy

crshyod 1St

he tn nshyIt






























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
























301

~jsm



aeb i nomy

gime manr



ways ehavshydoes

renee d by rhey

oyed

~D dcorned

ether seen


~ ded the

1 tmem that



d power


~I~ewhere
















ates




















Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
































Uk shy

J05umiddot

~

udshyren

I~d

1011shy


For

recshythe
















Moravcsik - Preferences and Power in the EU

Documents

Transcript of Moravcsik - Preferences and Power in the EU