The Limits of Comparative Politics: International Relations in the European Union
Anand Menon
University of Birmingham, UK. 
a.menon@bham.ac.uk




Paper presented to the ninth biennial conference of the European Union Studies Association, 
March 31-April 2, Hyatt Regency Hotel, Austin, Texas.  


Comments welcomed. Draft only, not for citation. 


Introduction 
As political scientists have become aware of the potential gains involved in comparing the 
European Union with other political systems, more of them have come to see the United 
States as a natural comparator.  Partly, this is explicable at an empirical level, a function both 
of the many similarities between the two systems, and (as will be argued) largely misplaced 
assumptions concerning the similarity of their developmental trajectories. Because the EU 
does many things national political systems do, some observers have assumed the former 
adopted the institutions of the latter and functions as they do. They have, in other words, 
extrapolated from function to form and functioning.   Partly, too, this has been the result of a 
desire amongst some to see the methodological tools used to study American politics applied 
to the European Union.

The intellectual self-confidence of those who have applied insights derived from the study of 
the US to the EU is reflected in their growing ambitions, exemplified by claims not only that 
approaches derived from the study of American politics can help to explore the nature and 
functioning of the European Union, but also that they can do so in an intellectually more 
rigorous, more useful and more effective manner than can those formulated by scholars 
within the discipline of International Relations. 

The current paper does not dispute that approaches from Comparative Government can bring 
added value to the study of the EU. Nor is its intention to suggest that comparisons between 
the EU and nation states perform no useful function – quite the contrary. Rather, it argues 
that, in order to gain a full understanding of the European Union, observers must be conscious 
of the specific – international – nature of the politics that occurs within it. The EU contains 
elements that are irreducibly international, which makes the character of interactions within it 
fundamentally different from those within national political systems, even federal ones. 
Largely, though not exclusively, because of the unique nature of states as political actors, 
politics between them differs in numerous fundamental ways from the politics found within 
them. Comparative Government approaches are often based on insights about the nature of 
politics that are not easily transposable from the domestic to the international realm.  Their 
analyses rooted in assumptions about the texture and nature of national politics, they are 
simply unable to explain the development and functioning of the Union. 

This paper is divided into three sections. Section one discusses how and why it is that so 
many observers have been drawn to apply insights from domestic politics to the study of the 
Union and illustrates how certain of them have claimed that Comparative Government can be 
used to improve, or even supplant, International Relations as the discipline most able to 
explain the development and functioning of the Union. Section two examines three aspects of 
interstate politics in the EU that tend to undermine such claims. Section three briefly 
underlines the importance of these in understanding the Union’s nature and development over 
time. 

COMPARATIVE POLITICS AND EUROPEAN INTEGRATION
The lure of Comparative Politics
Until the early part of the 1990s, the study of the politics of European integration largely 
languished in an intellectual ghetto, divorced from examinations of other political systems, be 
they domestic or international.  Over the last decade and a half, however, scholars have come 
to recognise the debilitating effects of the ‘n=1 problem’ (ECSA 1997) and have taken steps 
to address it.  In particular, students of Comparative Government have increasingly taken both 
to using the methods of comparative politics to analyse the Union (Hix 1994; 1999; 
Jachtenfuchs 2001; Sedelmeier 2001) and, more specifically, because of its multi-tiered 
nature, to comparing the EU directly with federal political systems in general, and with the 
United States in particular  (1999; McKay 2001; Nicolaidis and Howse 2001; Kelemen 2004; 
Menon and Schain 2006). 

The tendency to compare the EU and US is easily understandable.  After all, they share 
several traits, including: potentially comparable institutional landscapes (Kreppel 2006; 
Shapiro 2006); apparently parallel developmental trajectories (Donahue and Pollack 2001) 
and strikingly similar institutional dynamics stemming from comparable federal structures 
(Kelemen 2006).  Most fundamentally,  ‘the…EU implicates and exhibits authority over its 
constituent members in ways reminiscent of a domestic polity’ (Jupille and Caporaso 1999: 
429), or, as Weiler puts it (2001: 56), '...the constitutional discipline which Europe demands 
of its constitutional actors - the Union itself, the Member States and State organs, European 
citizens and others - is in most respects indistinguishable from that which you would find in 
advanced federal states'.

Initially, it was European scholars who were the most enthusiastic proponents of a 
comparative politics approach to the EU (see notably Hix 1994), while Americans tended to 
deploy the tools of International Relations (Jupille and Caporaso 1999: 430; Jachtenfuchs 
2001: 256; Sedelmeier 2001). One possible explanation for this divergence is the fact that 
Europeans are directly exposed to the effects of European integration. Because, from a 
European perspective, the Union appears to do what states do, with similar effects to those 
exerted by state actions, it makes sense to compare the Union with a ‘normal’ polity 
(Jachtenfuchs 2001: 256). 

More fundamentally, the comparative bent stems from a growing realisation that students of 
domestic and international politics are increasingly confronted with converging empirical and 
intellectual trends, including globalisation, the salience of domestic-international linkages and 
the role of institutions both domestic and international (Jupille, Caporaso et al. 2003: 10). As 
‘domestic society has elements of anarchy (contract enforcement and  cheating are 
problematic)…the international system has substantial elements of governance and rule’ 
(Jupille and Caporaso 1999: 438), or, as Helen Milner (1998: 760) puts it, ‘within states 
anarchy threatens, whereas the institutionalization of international politics beckons’. As the 
apparent overlap of concerns between political science and International Relations has grown, 
so too has the tendency to apply theoretical approaches from domestic politics to international 
relations. 

The application to the EU of theories and approaches formulated to explore and explain 
domestic politics has yielded numerous insights, as political scientists have focused their 
attention on issues that colleagues from International Relations might well have overlooked. 
Examples include work on the role of interest groups (Streeck and Schmitter 1991; Mazey 
and Richardson 1993; Kohler-Koch 1994), public opinion (Eichenberg and Dalton 1993), 
political parties (Attina 1990), as well as analyses of the striking similarity of the judicial 
system of the Union with that of many federal states (Kelemen 2006; Shapiro 2006) , or of the 
role of veto points in shaping policy outcomes (Scharpf 1988).

The Domestic Politics Fallacy
It can hardly be denied that the insights provided by Comparative Government have proved 
instructive, nor that they have been tremendously useful in compensating for the numerous 
‘blind spots’ inherent in a discipline of International Relations often guilty of excessive 
introspection and a fascination bordering on obsession with the question of whether 
institutions matter, rather than how and in what ways they do (Martin and Simmons 1998: 
742-3).  

Increasingly, however, scholars have gone beyond merely exploring the merits of 
Comparative Government to dismissing the contribution International Relations approaches 
can make. At their least pernicious, such claims reflect a view that Comparative Politics in 
and of itself provides the tools needed to study ‘policy’ even within an international 
organisation such as the European Union, whilst International Relations approaches can be 
relegated to the study of institution building (Hix 1994; Jachtenfuchs 2001).  More broadly, 
certain scholars have claimed that IR as a distinct sub-field is losing its intellectual relevance. 
Helen Milner (1998: 759) has claimed that the separation of International Relations from 
other fields in political science has ‘limited the field’s ability to understand international 
relations even during the Cold War period, let alone since its passage.’ The remedy, it would 
appear, is for IR to turn to comparative politics for solutions in the form of a ‘steady 
infiltration of analytical concerns from comparative politics into international relations….a 
brisk import trade where the common knowledge of one field comes to be regarded as path 
breaking research in another’ (Jacobsen 1996: 95). 

Mark Pollack, in a survey of IR approaches to integration, comments that the importation of 
the new institutionalism into EU studies has succeeded in ‘enriching IR theory and reducing 
its traditional parochialism and exceptionalism’(Pollack 2001: 238). A recent survey of 
institutionalist thought concluded that empirical and intellectual developments have ‘rendered 
subfield distinctions increasingly anachronistic and potentially counterproductive. 
Institutionalism especially seemed to provide an intellectual bridge, promising, according to 
its advocates, a general theory applicable to comparative, international, and American 
politics’ (Jupille, Caporaso et al. 2003: 10; see also Milner 1998).

Such assertions raise several concerns. First, they seem based on a cursory and superficial 
understanding of the International Relations literature (Rosamund 2000: 157-186). In 
abrogating, as many students of domestic politics do, the term ‘comparativist’, they overlook 
the comparative work undertaken not only by the early neofunctionalists, but also by more 
recent IR theorists (Mattli) (though one would be justified in bemoaning the relative dearth of 
such comparative endeavours in the subfield).  

More substantively, even if one accepts the premise that the concerns of the two disciplines 
are increasingly similar, it does not follow that the consequent learning process should be 
unidirectional. In assessing two subfields that have tended to reify the distinctions between 
themselves, the question of which is exceptional, or parochial, is very much one of 
perspective. Students of domestic politics have recently come to realise the potential 
advantages inherent in the use of the kinds of non-cooperative game theory frequently 
deployed to study international politics (Martin and Simmons 1998: 739; Milner 1998). 
Moreover, if IR scholars have been guilty of, amongst other things, oversimplifying the state, 
the problem with much work on comparative politics is its failure to acknowledge a role for 
international factors in its investigations of domestic politics (Hurrell and Menon 2003). One 
illustration of this has been the paucity of attention paid by researchers on comparative 
politics to policy sectors highly sensitive to international conditions such as foreign and 
defence policy. In an era when globalization and interdependence are the themes of the 
moment, it seems somewhat bizarre to downgrade the relevance of the international for 
studies of the domestic. 

A clearer appreciation of the problems inherent in applying approaches developed for the 
study of comparative politics to the international realm can be gleaned by examining one such 
approach in more detail. 

THE LIMITS OF RATIONAL CHOICE INSTITUTIONALISM
Rational choice approaches to the study of domestic politics have been the focus of those who 
claim that comparative politics can improve upon existing International Relations scholarship. 
The forcefulness of many of the proponents of rational choice stems from their engagement in 
a methodological as much as disciplinary or theoretical crusade.  Several of those who claim 
that rational choice institutionalism represents an untapped resource in better understanding 
the world of international politics obfuscate the distinction between theory and the rationalist 
approach to politics. Thus, for Milner (1998), the adoption of rational choice institutionalism 
holds out the prospect of students of international relations being able better to comprehend 
the role both of domestic politics and of international institutions in shaping international 
affairs. 

Yet there is no necessary link between the rationalist research agenda and a substantive 
interest in either domestic politics or international institutions. Indeed, perhaps the seminal 
rationalist account of international politics studiously overlooks both (Waltz 1979). What It is 
difficult to avoid the impression that Milner is engaged simply in a thinly disguised attempt to 
provide a theoretical justification for a methodological predilection. Indeed, carried away by 
their zeal, proponents of rational choice institutionalism sometimes betray an alarming 
disdain for much of the non-rationalist literature even on domestic politics. The claim that 
American rational choice scholars ‘were the first to theorise explicitly about the control 
relationship between legislative principals and bureaucratic agents’ (Pollack 2002: 200) 
betrays, at a minimum, an extremely narrow definition of theory, one which excludes, to take 
but the most obvious example, the  work of Max Weber elucidating the inherent problems 
involved in reconciling hierarchy and expertise in bureaucratic systems (Weber 1988).

Whatever the motivation behind it, recent scholarship has placed much emphasis on so-called 
theories of delegation as a rational choice approach suited to the study of politics in the Union 
(for a critical survey, see Kassim and Menon 2003). Mark Pollack issues an explicit appeal 
that ‘students of delegation and agency outside the American context should learn from the 
experience of Americanists and draw from that literature those methods that are…transferable 
to other contexts’ (Pollack 2002). His case rests on a simple premise, notably that: 

Although not a national political system, the Union has a number of 
characteristics…which make it analytically similar to the US political system, and 
hence a promising testing ground for American-derived theories of delegation and 
agency 
Pollack (2002: 211-2)

Others have also argued that the international context of the EU should and will not detract 
from the clear benefits of applying this kind of institutional analysis (see notably Jupille and 
Caporaso 1999: 431), while Milner (1998: 780) states baldly that a ‘common language for 
thinking about institutions at both the domestic and international levels has emerged; such a 
common language is important because it allows one to make generalizations across diverse 
contexts'.  

Context, however, matters. For one thing, principal-agent approaches were deployed in the 
study of American politics primarily in relation to a specific aspect of those politics – 
implementation – qualitatively different from decision making, which has been the object of 
their application to the EU. More significantly, assuming rationality (with all that this entails) 
is, at its simplest, what rational choice is about. There is no inherent assumption here about 
the nature of the actors involved in strategic interactions (though there is an inherent tendency 
to assume all actors would act in the same way if confronted with the same situation – see 
Hay 2004). The real problem confronting the application of new institutionalist theories to the 
European Union relates to the fact that the models employed to analyse relations between 
Congress and executive agencies in the United States are grounded in assumptions derived 
from the American domestic setting. As the following sections illustrate, these assumptions 
are inherently problematic in that they cannot fully capture the nature and behaviour of the 
key actors involved in the process of delegation in the international realm. 

States and international politics
The nature of international politics and of the predominant actors within it differentiates the 
relationship between member state principals and their agents from those pertaining within 
domestic politics. This applies to the reasons for delegation as much as to its nature. The 
states that make up the European Union are motivated by factors specific to actors in the 
international realm.  The centrality of considerations of power and security in international 
politics means that institutions are often viewed as tools to manage interstate relationships, 
themselves perceived through the prism of geopolitics.  Traditional Comparative Politics 
explanations of actor strategies in policy making thus ‘cannot deal with a central motivation 
of much EU policy making – namely the management of unequal state power and the desire 
to tie certain states within a structure from which they have the option to defect.’ (Hurrell and 
Menon 1994: 392). The genesis of two of the major initiatives of recent times -  the euro and 
enlargement, cannot be understood except as a function of such power-related considerations  
(Hurrell and Menon 2003). 

In terms of the principal-agent relationship arising out of the initial act of delegation, the 
international context is again somewhat different from its domestic counterpart. Because they 
are sovereign political entities with significant resources, states do not find themselves in the 
position of mutual dependence that characterizes many if not most principal-agent 
relationships in domestic political settings. Congressional committees and executive agencies 
in the US are, to a large extent mutually dependent: they need to work together in order to 
achieve their objectives. In contrast, dependence in the EU system is unidirectional. While the 
EU institutions rely on agreement between member states in order to function effectively, EU 
membership represents only one of a number of relationships and institutional entanglements 
in which the member states are involved (Hurrell and Menon 1994). Consequently, and unlike 
domestic political actors, they posses the ability to make choices as to the most appropriate 
institutional venue for their initiatives. 

Member states, moreover, enjoy the ability to act independently of institutional settings: they 
can have a currency without relying on the ECB, and a visa policy without joining Schengen 
(see Moravcsik 2004: 346). They can also turn to alternative international institutions in 
policy sectors where European integration does not seem appropriate – hence the continued 
reliance of Europe on NATO for collective defence. The non-exclusivity of member states’ 
relationships with the Union has had important implications for the role of territory in shaping 
EU politics. Bartolini (2004) explains how the gradual replacement of territorial, in favour of 
functional, representation within European states was explicable in terms of the development 
of national boundaries:

The more closed and reinforcing the various types of boundaries, the more likely that 
territorial issues will be in the long run incorporated within broader functional cross 
local alliances. The more open the territorial boundaries and the more loosely bounded 
the polity's territories, the more likely territorial alternatives will differentiate and 
become the focus of political conflict.
(Bartolini 2004)

The control exercised by states over their borders prohibited exit strategies on the part of 
domestic actors. In contrast, member states confront few EU-imposed boundaries on their 
activities and can rely on national, European or international fora to achieve their objectives. 
Partly as a consequence, the Union enjoys only weak territoriality (Ansell 2004), particularly 
in comparison with its constituent units, whilst politics within it remains dominated by 
territorial considerations.  The contrast with the United States is instructive. There, the 
American Civil War effectively foreclosed exit options, and hence choices of institutional 
venues for the states (Fabbrini 2004), leading over time to the increasing prevalence of 
functional, as opposed to territorial, politics. This should come as no surprise to students of 
international relations. The fact that states tend to be less functionally specialised than regions 
in a federal system is a defining feature of the international realm according to the classic 
account of structural realism (Waltz 1979: 93-7). 

Moreover not only are states autonomous and well resourced, but they are also sovereign 
entities, and their sovereignty is crucial in terms of conferring upon them both authority and a 
claim to be respected. The right of sovereignty confers upon them a legal basis for their claim 
to privileged status. Internally, state power resources are reinforced by significant legitimacy. 
The survival of the nation state as the dominant form of political organisation in world 
politics is explicable to a large extent in terms of the legitimacy of nation states, which has 
proved ‘a more powerful determinant of the prevailing scale of government authority than 
either greater homogeneity of sub national communities or greater powers of supranational 
political units’ (Scharpf 1988: 240). Member states, unlike the Union are thus self-
authenticating, deriving their legitimacy directly from their peoples

This legitimacy has proved crucial in terms of shaping the relationship between states and 
international institutions, if only because populations have proved willing to accept sacrifices 
imposed by national governments, whilst their reaction to the partial impotence of 
governments in the face of domestic and international pressures has been to ‘cling to and if 
possible to reinforce the nation-state’ (Hoffmann 1982: 23). The legitimacy of states means 
that populations may be willing to accept sub-optimal outcomes as a result of autonomous 
policy choices, as long as these are national choices, whilst rejecting integrative solutions that 
are potentially more efficient. 

Thus, whilst one may or may not consider rational choice the most apposite approach to 
utilise, crucial differences exist between international and domestic contexts in terms of the 
dependence of the whole on the parts, and the nature and authority of these parts. 

States as Organisations 
The second factor differentiating states from domestic actors are their organisational 
resources. It is here that the dangers inherent in importing insights from domestic politics are 
perhaps clearest. Applications of the principal-agent model to American politics argue that 
political ‘principals’ constantly confront the danger of ‘shirking’ by executive agencies, 
because of the incomplete information at the disposal of principals concerning the possible 
actions of agents (Weingast and Moran 1983; Kiewiet and McCubbins 1991: 25).  
Information asymmetries lead to the possibility of agents exploiting `the costs of measuring 
their characteristics and performance to behave opportunistically’ (Doleys 2000: 537). Based 
on such insights, many have assumed that the EU possesses a crucial advantage over the 
member states, because `in the EU…information is largely controlled by the supranational 
Commission’ (Aspinwall and Schneider 2001: 7 see also Pollack 1997: 108 and Pierson 1998: 
40)

However, we should be wary of assuming that lessons from the study of delegation within 
states are applicable to delegation in the international realm by states.  States possess 
characteristics that render the nature of their interactions with international institutions 
qualitatively different from those pertaining between actors and institutions in domestic 
political settings. Unlike the American states which came together to form the United States, 
West European states are powerful organisations, capable of mobilizing significant resources, 
including often-sizeable ministries, to monitor developments in all aspects of public policy, 
and provide a level of expertise that the relatively small staffs of international institutions 
cannot hope to match. As Moravcsik puts it (1999: 272), why ‘should governments, with 
millions of diverse and highly trained professional employees, massive information-gathering 
capacity, and long-standing experience with international negotiations at their disposal, ever 
require the services of a handful of supranational entrepreneurs to generate and disseminate 
useful information and ideas?’  The simple fact that all the member states have embassies to 
both the European Union and the other member states serves to emphasise the organisational 
resources deployed with a view to providing rapid and reliable information on developments 
within the Union. 

Moreover, unlike in domestic political settings, the provision of information via the use of 
police-patrol oversight in a situation of multiple principals does not represent a public good 
(Pollack 1997: 111). Rather, in a context of nation states competing for influence within the 
institutional setting of the EU, information is very much a private good, to be used in securing 
comparative advantage over one’s competitors. Intelligence gleaned about the actions and 
intentions of supranational institutions is employed by states to steal a march over their rivals  
(Menon 2003). Unlike the actors in domestic politics, therefore, states have an incentive to 
ensure they are as well informed as possible, as the often complex mechanisms they employ 
to do so illustrate (Kassim, Peters et al. 2000; Kassim, Menon et al. 2001)

None of this is to argue that problems of moral hazard and adverse selection are absent in 
international institutions – and one can assume that these will be more acute for smaller 
states, whose administrations may well lack expertise even relative to that within EU 
institutions. Nor is it to deny that rational choice institutionalism could contribute much to the 
study of this relationship. Rather, member state resources imply that informational 
asymmetries will be less frequent, and less severe, than the literature on US politics implies. 
Informational problems are mitigated in the international realm, and hence assumptions 
utilised in a domestic context may need to be amended before being applied to an 
international organisation.

International Institutions
States, therefore, are unlike the kinds of actors one might find in a domestic polity.  And their 
uniqueness has implications for the structure of the principal-agent relationships into which 
they enter. For one thing, member states can attempt to design institutional arrangements that 
minimize, if not preference incompatibilities, then at least the opportunity for agents to act 
opportunistically. Several pertinent strategies are open to them. Unlike in the American 
system, where the principal (Congress) may not enjoy the power of appointment (Moe 1987: 
489), governments generally do enjoy this right over the staff of international institutions. 
Whilst adverse selection problems still bedevil them, they are nonetheless in a position to be 
able to choose candidates whose preferences are close to their own. Governments `will 
appoint people who have internalized the goals of the states rather than the organizations even 
when they are not officially there as representatives’ (Nicholson 1998: 85). Certainly, this is a 
theme addressed in the principal-agent literature (see, for example, Laver and Shepsle 1996: 
13-15), though the importance of selection has been underplayed in its applications to the 
European Union.  Almost as important, in order to avoid the danger of officials `going native’ 
in post is the power of reappointment, which can also be used to shape the behaviour of 
national officials in international posts.

More fundamentally, principal-agent approaches in economics, and the variants used to study 
American domestic politics, assume that there exist fundamental incentive incompatibilities 
between principals and agents. This assumption has been transposed by many authors to the 
relationship between member states and the EU. Thus:

EC organisations will seek to use grants of autonomy for their own purposes, and 
especially to increase their autonomy. They will try to expand the gaps in member-state 
government control, and they will use any accumulated political resources to resist 
efforts to curtail their authority. The result is an intricate, ongoing struggle. 
(Pierson 1998: 35 see also Pollack 1997: 108). 

Yet is this really credible? Member states are not only powerful in terms of their own 
resources, but also in the way they can use these to shape the actions of others. For one thing, 
they enjoy a striking ability to `introduce incentive structures into the agency relationship that 
encourage preference compatibility’ (Doleys 2000: 537).  Crucially, they can make full use of 
the control they exercise over the prospects of those who wish to return to work within the 
national administration. Whilst it is reasonable to assume that employees of a US Federal 
Agency do not aspire to become Congressmen, officials in international organizations may 
well seek to prolong their careers in national administrations, or conceivably even national 
politics.  In this case – particularly for those on short-term contracts - the home government 
enjoys a significant ability to ensure loyalty. After all, where one stands depends not only on 
where one sits (Allison and Zelikow 1999: 307), but also on where one wants to recline 
later.  

That  those working for the EU institutions may well feel the temptation to aspire to senior 
positions back ‘home’ was clearly illustrated under the last European Commission, as 
Commission President Prodi was widely seen to have neglected his duties in Brussels to focus 
on manoeuvring for a return to Italian politics. It is hard indeed to imagine an analogous 
situation in the United States, involving a President intent on securing a Governorship. And 
whilst one could imagine a situation in Europe where the structural equivalent of State 
Governors – Heads of State and Government – focussed on the ambition of high office at the 
European level, it is imaginable only in the case of smaller member states. The point, again, is 
not to claim that similarities do not exist between federal systems and the European Union, 
but that the assumptions generally used in studies of the former, especially in terms of actor 
incentives, may be called into question in the case of the latter.      

The ability of member states to influence the officials who staff international institutions 
reinforces the arguments made above concerning the resources of the member states.  The 
presence of nationals within the institution concerned, nationals who may find it in their 
interests to cooperate closely with their home nation, serves further to diminish the problem 
of informational asymmetries confronting the member states.  The nature of international 
politics, as well as the nature, resources and organisation of states makes for a different 
texture of politics between, as opposed to within, states. 

International Relations and the development of the European Union
Those who compare the US and EU systems are drawn, perhaps inevitably, from assertions 
regarding the comparability of the two institutional systems, to arguments implying the 
probable similarity of their developmental trajectories. Thus, the claim that the Union shares 
with federal systems like that of the US ‘a constitutionally defined…separation of powers 
between a central government and individual Member States’ (Donahue and Pollack 2001: 
95) leads to the assertion that the US and EU: 

display similar alternations of centripetal and centrifugal impulses….the centralising 
impulse is likely to reassert itself in both polities in the century’s early years. The next 
period of predominant centralisation, moreover, should be somewhat sharper and 
shorter in Europe than in the US, as the asynchronous rhythms of the two polities - one 
a mature federation, the other not - move out of phase. 
(Donahue and Pollack 2001: 116)

Less explicitly, much of the most influential work applying principal-agent models to the 
Union draws on a similar logic, applying, somewhat uncritically, the literature on the United 
States to emphasise the problems member states face in controlling supranational agents, and 
the ‘gaps’ in their control that subsequently emerge (Pollack 1997; Pierson 1998).  

Turning first to the question of structure, it is profoundly misleading to talk in terms of some 
kind of ‘separation of powers’ between the Union and the member states.  The nature of 
politics between states has, from the first, been highly influential in shaping the institutional 
structure of European integration. Particularly in comparison to that of federal systems, this 
reflected the desire on the part of the member states to safeguard their authority over their 
creation.  Consequently, and in stark contrast with the US, the member states designed a 
system in which their institutional interests are represented via the Council of Ministers 
(Sbragia 1991: 2, 5; see also McKay 2001: 130-1). The Senate never functioned as a ‘Council 
of the states’, even under the short-lived, and ultimately unsuccessful ‘doctrine of 
instructions’ (McKay 2001: 133). Rather, Senators represent their constituents. In contrast, in 
systems where state governments participate directly in central decision making, it is the 
interests of these governments that are voiced. Even here we can distinguish between systems 
of state representation at the centre, such as Germany, where, the central government enjoys 
political identity, resources and strategic and tactical capabilities of its own, and the EU, 
where the central institutions enjoy no such advantages over the constituent units.  As Sbragia 
puts it: 

Community politics and national politics are institutionally intertwined rather than 
insulated from each other….one can discuss American politics and policymaking 
without mentioning a single governor...the privileged status enjoyed by member 
governments in the Community's political system has no analogue in the US system. 
American state officials do not participate personally in national policymaking.  For 
their part, American national politicians represent voters who happen to reside in states; 
they do not represent governors or state legislators. Thus, in the United States, the 
representation of the institutional interests of state governments in national 
policymaking is not constitutionally entrenched.
(Sbragia 1991: 5)

Member state representation, moreover, is more than symbolic. As argued above, their 
administrative capacities give them the ability to exploit to the full their privileged 
institutional position and to represent and defend their interests within the Union (Sbragia 
1991: 2).

The institutional system of the Community as a whole was designed in such a way as to 
facilitate the stolid defence of perceived national interests should the desire for enhanced 
cooperation falter. Scholars who take their cues from a domestic analogy tend to emphasise 
the importance of the Commission’s monopoly over the right of legislative initiative as a 
cornerstone of its influence (Pollack 1997). An alternative reading, however, would be that 
member states ensured ultimate control for themselves by ensuring that they could react 
against any Commission proposals. Thus, Taylor comments, with reference to the 
‘eurosclerosis’ of the 1970s, that:  

the institutions of the Communities made it easy for national governments to adopt a 
defensive stand if they so wished; they were designed to allow a relatively undramatic, 
stonewalling approach if it happened that in practice the government's expectations of 
future compatibilities were unrealised
(Taylor 1975: 339). 

Agenda setting power is crucial in any political system, but the influence of the agenda setter 
is dramatically curtailed in a situation when the core legislators are not interested in working 
together. 

Member states form the core of the EU decision-making system. Certainly their role is not 
exclusive, and the European Parliament, most notably, has seen its legislative powers 
enhanced considerable during the course of successive treaty revisions. Yet whilst the EP, 
acting in concert with coalitions of member states in the Council, represents a more effective 
legislative body than many, if not most, national parliaments, it is, ultimately, the Council 
which is the crucial legislator in the EU system. And it is the member states qua member 
states that sit in the Council. In this sense, even in matters of core legislative activity, the EU 
is in no real sense separate from its constituent member state parts. 

Moreover, and far from incidentally, it is the member states, via the European Council that 
control the size of the EU budget. As a consequence, this is limited to around 1% of the 
Union’s GDP, thereby severely limiting the ability of ‘Brussels’ to buy either loyalty or 
compliance, and ensuring the absence of a tool employed to powerful effect by many federal 
systems, notably the US, where ‘“[c]oercive cooperation” is underpinned by the cash which 
Washington dispenses to state capitals’(Sbragia 2006). 

Structural differences between the EU and federal systems help account for differences in 
their developmental trajectory. Here, it is important to differentiate between what Taylor 
(1975) refers to as the ‘scope’ of integration and its ‘level’. On the one hand, it is clear that 
the attribution of tasks by member states to the European level has continued, with ever more 
sensitive areas of public policy – immigration, cooperation on counter-terrorism, and defence 
policy amongst them – being affected by the Union. Yet EU involvement or activity in a 
policy sector does not imply a shift in power from the parts to the centre, given that the 
member states themselves wield such tremendous influence over what occurs in ‘Brussels’. 
Contrast this with an American system in which ‘the exercise of federal government functions 
is formally independent of the governments of the American states, and those functions that 
have been taken over by the federal government are effectively nationalised’ (Scharpf 1988: 
242)

And if the 1990s witnessed a steady accretion of new competences at the European level, this 
was accompanied by persistent and consistent measures adopted to ensure member state 
control. Crucially, the international character of the Union empowered the member states in 
their attempts to claw back authority. Because the Union is based on a treaty rather than a 
constitution, member states enjoy levels of control over its development that are unparalleled 
in national federal systems. For one thing, the treaty base uniquely empowers states as 
institutions to intervene between citizens and the Union in treaty negotiations (Chopin 2002: 
43-8).  Moreover, not only is no formal role given to the EU institutions in negotiations over 
treaty change, but such negotiations uniquely empower the negotiators. Whilst:

amendments to a constitution resemble a continuing dialogue with previous political 
and constitutional developments, the formulation of new treaties can differentiate 
among whatever institutional innovations were made in previous treaties. Treaties 
allow for much greater discontinuity in institutional development, a disjuncture that 
permits national governments to control the timing and shape of institution building 
relatively closely.
(Sbragia 1991: 273)

Thus the treaty base provides member states with a large amount of latitude in shaping the nature 
of the system, for instance by isolating some areas of policy from the ambit of the EC and 
limiting the power of the supranational institutions over them, as they did with the creation of the 
pillared structure in the treaty on European Union. Finally, member states ensure a large degree 
of control over the development of the Union by insisting that substantive changes to its nature 
and functioning be decided on via treaty amendment. In order that they be bound in new areas, a 
Treaty revision must be unanimously agreed – in contrast to the profound changes ushered in in 
the United States by the New Deal – without the need for constitutional amendment (Sbragia 
2006). 

The trend towards increasing member state assertiveness first manifested itself at Maastricht, 
with the introduction of the principle of subsidiarity, explicitly intended to put an end to 
‘competence creep’ (Majone 1998), and the creation of the pillared structure (possible 
because of the treaty rather than constitutional base), which distanced the ‘supranational’ 
institutions from the sensitive areas of foreign and security policy and justice and home 
affairs. As the decade progressed, member states tightened their hold on developments within 
the Union in numerous ways, including via the increased prevalence of Intergovernmental 
Conferences, with progressively more detailed agendas, the growing dominance of the 
European Council, and the introduction of new forms of decision making, such as the Open 
Method of Coordination, or the purely intergovernmental method adopted for the emergent 
European Security and Defence Policy (Kassim and Menon 2005). The still to be ratified 
constitutional treaty merely reaffirms and reinforces such trends, bolstering subsidiarity with 
the provision for member state parliaments to vote on draft EU legislation. It also underlines 
the degree to which member states remain anxious to preserve their position as ultimate 
drivers of the integration process via the creation of a President of the European Council.


In sum, if analysts were able, in the early 1990s, to assert that the Governments of the twelve 
exerted immeasurably more influence on Community policies than American state 
governments exert on federal policies within the United States (Sbragia 1991: 2), this is all 
the more true after the events of the last decade. These have served to highlight the 
instrumentality of a Union system created and maintained to serve the interests of its 
constituent member states and continually recast in order to reflect changing state 
preoccupations and purposes. 

Conclusions
The purpose of this paper has not been to claim that International Relations necessarily 
provides all the answers for those interested in studying the European Union.  Indeed, it is not 
difficult to find serious shortcomings in the IR literature on European integration, including 
the failure (ironically) to recognise the key role of geopolitics (Moravcsik 1998), and the 
obsession, referred to above, with the debate about whether institutions matter or not. Nor is 
its intention to claim that Comparative Government specialists should not engage in 
comparisons of the Union with federal states for, although ‘the Community is unique, analysis 
is more likely to suffer from studying it in isolation from other systems than from using the 
comparative method in less than ideal circumstances.' (Sbragia 1991: 268).

There is, however, a danger in the over-zealousness of some in assuming both that the EU and 
US are obviously comparable cases (thereby overlooking the possibility that other federal 
political systems - whether Germany, or Switzerland, or Canada, or India - might represent 
more apposite comparators), and that comparisons with domestic political systems are the 
appropriate methodology to deploy.  A fixation with the United States is particularly 
inappropriate given the fact that the US itself arguably represents something of a sui generis 
system of highly centralised multilevel governance and one which, for some, is not consistent 
with the federal principle at all (Chopin 2002). 

Much of this zealotry stems not from a loyalty to the sub-field of comparative politics per se 
but, rather, from a desire to emulate those rational choice theorists of American politics 
whose work has become so influential in the discipline in recent years.  Claims concerning 
the applicability of rational choice institutionalism as formulated for the American domestic 
context are based more often than not on a belief in the superiority of rational choice as a 
method. Rooted in the belief that ‘the ability to generalize across contexts while incorporating 
local specificity is a signal advantage of institutional analysis’ (Jupille and Caporaso 1999: 
438) proponents of such an approach tend to overlook the unique features of international 
politics, and of states as actors in international politics that require, at the very least, a 
rethinking of certain of the theoretical assumptions underlying new institutionalist 
assessments of US politics.

Herein lies the dual paradox of a rational choice institutionalism derived from the study of 
American politics. On the one hand, it is too specific, in that 

it seems clear that both the theory and the research designs of American studies can be 
applied beyond the US context only with considerable adaptation. Precisely because 
they were developed by Americanists with the aim of understanding the specific 
institutional features of the American political system - that is, congressional 
committees and regulatory agencies - the theories analysed above have been developed 
at a relatively low level of abstraction, which makes it difficult to transfer their insights 
to the very different, and primarily parliamentary, political systems of Western Europe
(Pollack 2002: 210)

On the other, it is too general. Despite the contention that they hold the promise of providing 
a more sophisticated understanding of the nature of the state (Milner 1998), Principal-Agent 
approaches tend to assume that states can be treated like any other principal in any other 
relationship. Whilst, in the 1960s and 1970s, scholars rooted in study of domestic politics 
were not deceived by the optimistic neofunctionalist prediction that power would 
progressively be transferred from states to a new European centre (Wallace 1982: 64), the 
newer brand of comparativists, in their quest for parsimony and methodological 
sophistication, risk failing to appreciate the unique nature of the institution most central to 
their substantive concerns. 






Bibliography

(1999). Consumer row hampers sales. London: 2.
Ansell, C. K. (2004). 'Territorial Authority, and Democracy' in C. K. Ansell and G. D. 
Palma Restructuring Territoriality: Europe and the United States Compared. 
Cambridge, Cambridge University Press.
Aspinwall, M. and G. Schneider (2001). 'Institutional Research on the European 
Union: Mapping the Field' in G. Schneider and M. Aspinwall The Rules of 
Integration: Institutionalist Approaches to the Study of Europe. Manchester, 
Manchester University Press.
Attina, F. (1990). "The voting behaviour of European parliament members and the 
problem of Euro parties." European Journal of Political Research  18: 557-
579.
Bartolini, S. (2004). 'Old and New Peripheries in the Processes of European 
Territorial Integration' in C. K. Ansell and G. d. Palma Restructuring 
Territoriality. Cambridge, Cambridge University Press.
Cappelletti, M., M. Seccombe, et al., Eds. (1986). Integration Through Law: Europe 
and the American Federal Experience. Berlin and New York, Walter de 
Gruyter.
Chopin, T. (2002). L'Héritage du Fédéralisme? Etats-Unis/Europe. Paris, Fondation 
Robert Schuman.
Doleys, T. J. (2000). "Member states and the European Commission: theoretical 
insights from the new economics of organization." Journal of European Public 
Policy  7(4): 532-553.
Donahue, J. D. and M. A. Pollack (2001). 'Centralisation and its discontents: the 
rhythms of federalism in the US and the EU  ' in K. Nicolaidis and R. Howse 
The Federal Vision: Legitimacy And Levels Of Governance In The US And 
The EU. Oxford, OUP.
ECSA (1997). "Does the European Union Represent an n of 1?" ECSA Review  X(3): 
1-5.
Eichenberg, R. C. and R. J. Dalton (1993). "Europeans and the European Community: 
The dynamics of public support for European integration." International 
Organization  47: 507-34.
Elazar, D. J. (2001). 'The US and the EU: Models for their Epochs ' in K. Nicolaidis 
and R. Howse The Federal Vision: Legitimacy and Levels of Governance in 
the US and the EU. Oxford, OUP.
Fabbrini, S. (2004). 'The European Union in American Perspective: The 
Transformation of Territorial Soverneignty in Europe and the United States' in 
C. K. Ansell and G. d. Palma Restructuring Territoriality: Europe and the 
United States Compared. Cambridge, Cambridge University Press.
Hendrickson, D. C. (2003). Peace Pact: The Lost World of the American Founding. 
Lawrence, University of Kansas Press.
Hix, S. (1994). "The Study of the European Union: The Chaloenge to Comparative 
Politics." West European Politics  17(1): 1-30.
Hix, S. (1999). The Political Sytem of the European Union. London, Macmillan.
Hoffmann, S. (1982). "Reflections on the Nation State in Western Europe Today." 
Journal of Common Market Studies  21: 21-37.
Hurrell, A. and A. Menon (2003). 'International Relations, International Institutions, 
and the European State' in J. Hayward and A. Menon Governing Europe. 
Oxford, OUP.
Jachtenfuchs, M. (2001). "The Governance Approach to European Integration." Journal 
of Common Market Studies  39(2): 245-264.
Jacobsen, J. K. (1996). "Are all Politics Domestic? Perspectives on the Integration of 
Comparative Politics and International Relations Theories." Comparative 
Politics  29(1): 93-115.
Jupille, J. and J. A. Caporaso (1999). "Institutionalism and the European Union: 
Beyond International Relations and Comparative Politics." Annual Review of 
Political Science (2): 429-444.
Jupille, J., J. A. Caporaso, et al. (2003). "Introduction: Integrating Institutions 
Rationalism, Constructivism, And The Study Of The European Union." 
Comparative Political Studies  36(1/2): 7-40.
Kassim, H. and A. Menon (2003). "The Principal-Agent Approach and the Study Of 
The European Union: Promise Unfulfilled?" Journal of European Public 
Policy  10(1).
Kassim, H., A. Menon, et al., Eds. (2001). The National Co-ordination of EU Policy 
Making: The EU Level. Oxford, Oxford University Press.
Kassim, H., B. G. Peters, et al., Eds. (2000). The National Co-ordination of EU Policy: The 
Domestic Level. Oxford, Oxford University Press.
Kelemen, R. D. (2004). The Rules of Federalism: Institutions and Regulatory Politics 
in the EU and Beyond. Cambridge, Ma., Harvard University Press.
Kelemen, R. D. (2006). 'Federalism and Democratization: The US and EU in 
Comparative Perspective' in A. Menon and M. Schain Comparative 
Federalism: The European Union and the United States. Oxford, OUP.
Kiewiet, D. R. and M. D. McCubbins (1991). The Logic of Delegation; Congressional 
Parties and the Appropriation Process. Chicago, University of Chicago Press.
Kohler-Koch, B. (1994). "Changing patterns of interest intermediation in the 
European Union." Government and Opposition  29: 166-180.
Kreppel, A. (2006). 'Understanding the European Parliament from a Federalist 
Perspective: The Legislatures of the USA and EU Compared' in A. Menon and 
M. Schain Comparative Federalism: The European Union and the United 
States. Oxford, OUP.
Majone, G. (1998). "Europe's 'Democratic Deficit': The Question of Standards." 
European Law Journal  4(1): 5-28.
Martin, L. L. and B. A. Simmons (1998). "Theories and Empirical Studies of International 
Institutions." International Organization 52(4).
Mazey, S. and J. Richardson, Eds. (1993). Lobbying in the European Community. 
Oxford, OUP.
McKay, D. (2001). Designing Europe: Comparative Lessons from the Federal 
Experience. Oxford, Oxford University Press.
Menon, A. (2003). "Member States and International Institutions: 
Institutionalizing Intergovernmentalism in the European 
Union." Comparative European Politics  1(2).
Menon, A. and M. Schain (2006). Comparative Federalism: The European Union and 
the United States. Oxford, Oxford University Press.
Milner, H. V. (1998). "Rationalizing Politics: The Emerging Synthesis Of International, 
American, And Comparative Politics." International Organization  52(4): 759-786.
Moe, T. M. (1987). "As assessment of the Positive Theory of `Congressional 
Dominance'." Legislative Studies Quarterly  12(4, November): 475-520.
Moravcsik, A. (1999). "A New Statecraft? Supranational Entrepreneurs and 
International Cooperation." International Organisation  53(2): 267-306.
Moravcsik, A. (2004). "Is there a 'Democratic Deficit' in World Politics? A 
Framework for Analysis." Government and Opposition  39(2): 337-363.
Nicholson, M. (1998). 'A rational Choice analysis of International organizations: How the 
UNEP helped to bring about the Mediterranean Action Plan' in B. R. a. B. Verbeek 
Autonomous Policy Making by International Organizations. London, Routledge.
Nicolaidis, K. and R. Howse, Eds. (2001). The Federal Vision: Legitimacy and Levels 
of Government in the US and the EU. Oxford, Oxford University Press.
Pierson, P. (1998). 'The Path to European Intgegration: A Historical-Institutioanlist 
Analysis' in W. Sanholtz, and Stone Sweet, Alec European Integration and 
Supranational Governance. Oxford, OUP: 27-58.
Pollack, M. (2001). "International Relations Theory and European Integration." 
Journal of Common Market Studies  39(2): 221-44.
Pollack, M. (2002). "Learning from the Americanists (Again): Theory and Method in 
the Study of Delegation." West European Politics  25(1): 200-219.
Pollack, M., A. (1997). "Delegation, Agency, and agenda setting in the European 
Community." International Organisation  51(1, winter): 99-134.
Riker, W. H. (1964). Federalism: Origin, Operation, Maintenance. Boston, Little 
Brown.
Rosamund, B. (2000). Theories of European Integration. London, Macmillan.
Sbragia, A. M. (1991). 'Introduction' in A. M. Sbragia Euro-Politics: Institutions and 
Policymaking in the 'New' European Community. Washington, The Brookings 
Institution.
Sbragia, A. M. (1991). 'Thinking about the European Future: The Uses of Comparison
' in A. M. Sbragia Euro-Politics: Institutions and Policymaking in the 'New' European 
Community. Washington D.C., The Brookings Institution.
Sbragia, A. M. (2006). 'The US and the EU: Overcoming the Challenge of Comparing 
Two “Sui Generis” Systems' in A. Menon and M. Schain Comparative 
Federalism: The European Union and the United States. Oxford, OUP.
Scharpf, F. W. (1988). "The Joint-Decision Trap: LEssons from German Federalism 
and European Integration." Public Administration  66: 239-278.
Sedelmeier, U. (2001). "Comparative Politics, Policy Analysis and Governance - A 
European Contribution to the Study of the European Union?" West European 
Politics  24(3): 173-182.
Shapiro, M. (1991). 'The European Court of Justice' in A. M. Sbragia Euro-Politics: 
Institutions and Policymaking in the 'New' European Community. Washington 
D.C., The Brookings Institution.
Shapiro, M. (2006). 'The U.S. Supreme Court and the European Court of Justice Compared' 
in A. Menon and M. Schain Comparative Federalism: The European Union 
and the United States. Oxford, OUP.
Stein, E. (2000). Thoughts from a Bridge: A Retropsective of Writings on New 
Europe and American Federalism. Ann Arbor, University of Michigan Press.
Streeck, W. and P. C. Schmitter (1991). "From national corporatism to transnational 
pluralism: Organized interests in the single European market." Politics & 
Society  19: 133-159.
Taylor, P. (1975). "The Politics of the European Communities: The Confederal 
Phase." World Politics  27(3): 336-360.
Wallace, W. (1982). "Europe as a confederation: the Community and the nation 
state." Journal of Common Market Studies  21: 57-68.
Waltz, K. N. (1979). Theory of International Politics. New York, Random House.
Weber, M. (1988). Wirtschaft und Gesellschaft. Tübingen, Mohr Siebeck.
Weiler, J. H. H. (2001). 'Federalism Without Constitutionalism: Europe's Sonderweg ' 
in K. Nicolaidis and R. Howse The Federal Vision: Legitimacy And Levels Of 
Governance In The US And The EU. Oxford, OUP.
Weingast and Moran (1983). "Bureaucratic Discretion or Congressional Control? 
Regulatory Policy-Making by the Federal Trade Commission." Journal of 
Political Economy.




  The author would like to thank Dionyssis Dimitrakopoulos, Hussein Kassim, Ed Page, Martin Schain, 
Elizabeth Edgington and particularly Dan Kelemen for comments and suggestions on earlier drafts. In 
fairness to Dan in particular, I should add that certain suggestions were ignored. 
  In contrast, lawyers had been relatively quick to approach the study of the EC from a comparative 
perspective, see notably Stein (2000) and Cappelletti, Seccombe et al.(1986).
  The notion of federalism was itself, in its original usage, a term applied to relations between 
sovereign states rather than one used to classify a certain genus of state (Hendrickson 2003: 22-4).  
  Decontextualisation is not, of course, a problem unique to comparative politics. 
  Note Sbragia’s comment (1991: 274): 'The legacy of the European sovereign state is that national 
boundaries are extraordinarily important shapers of most aspects of life. The territorial claims that 
national governments represent, therefore, are exceedingly strong….National identity, political party 
organisations, party systems, partisan ideology, interest groups, taxing and spending arrangements,  
educational systems, electoral constituencies, the internal organisation of the state, executive-legislative 
relations, the appointment of commissioners and European Court justices, the role of the judiciary, 
legal systems, and administrative apparatuses are all defined by national territory.' 
  Waltz points to two structural differences between the international system and domestic political 
systems. The first is the absence of hierarchy amongst states, in contradistinction to domestic systems, 
which are ‘centralised and hierarchic’ (p. 88). The second is the fact that ‘states…are not formally 
differentiated by the functions they perform’ (p.93). That is to say that states ‘try to perform tasks, most 
of which are common to all of them’ (p. 96), whereas national politics ‘consists of differentiated units 
performing specific functions’ (p. 97).
  Care should be taken even here. It is only recently that Congressional committees have enjoyed 
particularly the staffing resources necessary to hold agencies to account, and this situation was 
exacerbated by the cuts implemented by Newt Gingrich in the 1990s. I am grateful to Dionyssis 
Dimitrakopoulos for pointing this out to me. 
  Clearly, this is a question of degree.  Larger states enjoy larger resources, and the European Union, 
since the enlargement of 2005, now contains a significant number of small states 
  Certainly, promises of future advancement, especially when these are political rather than purely 
administrative, might be dependent upon relative stability in the partisan composition of government. 
On the other hand, this is not the case in the context of apolitical administrative systems. Moreover, the 
partisan nature of government has had relatively little impact on what have been markedly stable 
member state preferences about European integration over time.
  The phrase was suggested to me by Dan Kelemen.
  The contrast has not always been so clear cut. The United States of the eighteenth century was one in 
which the relative attractions of state and federal levels were perceived to be more equal. Justice John 
Rutledge resigned from the Supreme Court in 1791 to become Chief Justice of South Carolina. John 
Jay, Chief Justice of the Supreme Court, was elected  governor of New York in 1795, declining to 
return to the Court when asked to do so by President John Adams in 1800. My thanks to Dan Kelemen 
for informing me of this. Moreover, the European Court of Justice is arguably highly similar to its 
American counterpart in terms of the prestige implied by appointment to the bench.  Yet while the 
argument presented above applies most clearly to the European Commission, the ECJ also has many 
features that reflect the entrenched power of member states within the EU system: judges are selected 
from national judges nominated by governments; all member states are represented on the court, and 
national courts are ‘coordinate’ rather than inferior courts (Ansell 2004). Moreover, unlike the Supreme 
Court, the ECJ cannot invalidate a member state statue or act, but has to rely on the state itself to do so, 
and there have been instances where this has not occurred (Shapiro 1991: 125).
  In addition to interest-based motivations, it is interesting to speculate as to whether notions of loyalty 
or identity condition the behaviour of member state nationals in the EU institutions. 
  Similarly William Riker assumes (1964: 6) that either a centralising or peripheralising tendency will 
win out in two level systems. For him, the structure of the party system represented the crucial variable 
which would determine the ultimate outcome (1964: 129-36)
  This is not to say that Pollack slavishly follows the American literature. He explicitly makes the 
point that he is wary of  the extremes of either the Congressional abdication hypothesis or the 
Congressional dominance school, stating that `EC institutions…neither run amok nor blindly follow 
the wishes of member governments but rather pursue their own preferences within the 
confines of member state control mechanisms whose efficacy and credibility vary from issue 
to issue and over time’ (Pollack 1997: 110). Yet his analysis of the constraints upon member 
state ability effectively to monitor agents and their root causes is fully consistent with the 
assumptions of students of American politics. 
  The notion of separation of powers of course is generally used to refer to relations between branches 
of government rather than to the vertical distribution of power in federal states. I would like to thank 
Dionyssis Dimitrakopoulos for making this point to me. 
  Elazar makes a similar point, noting that the ‘EU’s constitution and common institutions have 
developed in such a way as to minimise the threat to the Member States’ (Elazar 2001: 38)
  Note the wording of the UK referendum question on the constitutional treaty, which asks the voting 
population to decide if the United Kingdom should sign the Treaty, rather than whether they approve it. 
 
 
 
 


23