Each of these approaches has its limitations. Westminster’s formal-legalapproach ignores larger debates both in the study of comparative politics and in
political science. It can revel in archaism, taking its stance from Bagehot (Bogdanor
1999 , 175 ). The sheer scale of Blondel’s modernist-empiricism is impressive but it is
the scale that poses problems. Dogan and Pelassy ( 1990 , 116 ) comment that
such comparative studies disappoint because ‘‘comparability is very low.’’ Citing
Blondel’s ( 1980 ) analysis of all ‘‘heads of government in the post-war period,’’ they
ask: ‘‘what sense is there in comparing the ‘regular ministerial career’ in the Middle
East and in the Atlantic and communist worlds? Aren’t we here misled simply by
verbal similarities?’’ Ignoring the general criticisms of rational choice (see Green
and Shapiro 1994 ;Hay 2004 ), there are the speciWc limitations of principal–agent
theory when applied to the study of executives. For example, the assumption of
hierarchy does not hold. Ministers are embedded in webs of vertical and horizontal
dependencies and only the former can be conceptualized as principal–agent chains.
Webs or networks are conspicuous by their absence despite their centrality to both
delegation and accountability.
3 Debates and Challenges
.........................................................................................................................................................................................
This section will cover the growth of prime ministerial power, referred to as the
presidentialization thesis; executive coordination; policy advice and policy capacity;
accountability; the eVects of institutional diVerences; and comparative analysis.
- 1 Presidentialization of Prime Ministers
The conventional cliche ́is that the prime minister is no longer ‘‘Wrst among equals’’
in the government but the elected ‘‘Wrst magistrate’’ (Crossman 1963 , 22 – 3 ; Mack-
intosh 1968 , 627 ). There is a corresponding decline in cabinet government. It is
diYcult to overstate the scale of this debate in the academic literature. It is the
deWning debate of the Westminster approach and refers to three main claims: there
has been a centralization of coordination, a pluralization of advice, and the
personalization of party leadership and elections. 1 The broad argument is common
1 On the several deWnitions of the presidentialization thesis see: Foley 1993 , ch. 1 ;Pryce 1997 , 37 , 67 ;
Mughan 2000 , 9 – 10 ; and Poguntke and Webb 2005 a, 5 , 8 – 11. For the key articles on prime ministerial
power see Dunleavy and Rhodes 1995 ; King 1969 , 1985.
executives in parliamentary government 327