Ministerial Resignations 1945–97
| DOI | http://doi.org/10.1111/1467-9299.00109 |
| Author | Won‐Taek Kang,Keith Dowding |
| Date | 01 September 1998 |
| Published date | 01 September 1998 |
MINISTERIAL RESIGNATIONS 1945–97
KEITH DOWDING AND WON-TAEK KANG
This article
1
reports on data collected on ministerial resignations and non-resig-
nations 1945–1997. It analyses the reasons why ministers resign and patterns that
emerge in terms of the types of issues that are more likely to lead to resignation,
and variances between different Prime Ministers, parties and over time. It provides
the first fully quantified analysis of ministerial resignations in Britain in the post-
war period to enhance the impressionistic analyses which have been offered before.
I INTRODUCTION: GENERAL FIGURES
A great deal has been written on the role of ministers, their individual and
collective responsibilities, their relationships with colleagues, civil servants,
their party and the Prime Minister. It is naturally, and correctly, assumed
the most important relationships ministers have are with their own back-
benchers, their cabinet colleagues and with the Prime Minister. There are
many ways in which we can examine ministerial relationships; one route
is via the issue of resignations. Much has been written on why ministers
resign, are sacked or reshuffled, and much also on the conditions under
which they should resign. However, no quantitative analysis of the pattern
of ministerial turnover has ever been undertaken. This article is a first
attempt to analyse data collected on ministerial turnover in the 52 years
from the beginning of the Attlee government to the end of the Major admin-
istration.
The first part of this article charts the general patterns of ministerial turn-
over, the second part concentrates upon individual ministerial resignations,
leaving aside the data on reshuffles.
2
There have been few systematic
attempts to examine the causes of ministerial resignations. Most of the
literature discusses recent resignations and compares them with previous
cases (Fry 1969–70; Baker 1972; Ganz 1980; Pyper 1983; Hennessy 1986; Oli-
ver and Austin 1987; Doig 1989, 1993; Phythian and Little 1993) but offers
no systematic account of resignations. In all of this literature there is sur-
prisingly little empirical evidence concerning the particular causal reasons
leading to the resignation. In discussing individual resignations authors
tend to assume the minister ‘had to go’, and compare and contrast the case
Keith Dowding is in the Department of Government at the London School of Economics and Won-
Taek Kang is at the Institute for Far Eastern Studies, Kyungnam University, Korea.
Public Administration Vol. 76 Autumn 1998 (411–429)
Blackwell Publishers Ltd. 1998, 108 Cowley Road, Oxford OX4 IJF, UK and 350 Main Street,
Malden, MA 02148, USA.
412 KEITH DOWDING AND WON-TAEK KANG
with the almost mythical Crichel Down Affair to see how current practice
shapes up to the so-called constitutional convention. The articles tend to be
normative as much as empirical, as they are concerned with constitutional
convention: those rules ‘accepted as obligatory by those concerned in the
working of the constitution’ (Wheare 1966, p. 179). But constitutional con-
vention is created through customary behaviour and not defined by one or
two cases. To understand when ministers should resign through consti-
tutional convention one needs to discover when they do resign, through
systematic analysis of past cases. Finer’s (1956) classic article remains per-
haps the only attempt to try to do so historically, though more recently
Woodhouse (1994) has looked at cases in the 1980s. Even these consider
only the famous resignation cases and do not look systematically or quanti-
tatively at all instances. Previous literature, with the exception of Wood-
house (1994) and Dowding (1995), also suffers a major methodological flaw.
One cannot examine the causes of resignation by considering only cases
where ministers resign: one also needs to consider the cases where they do
not resign. One cannot claim that the causes of ministerial resignation are
such-and-such if one examines only cases where ministers have resigned
and those conditions obtain, for there may be a significant number of cases
where such-and-such occurred yet the ministers did not resign. This article
is thus the first systematically to study ministerial resignations over a long
period of time to try to discover the true causes of ministerial turnover
and thus to generate empirical propositions about the nature and changing
conditions of the conventions and expectations surrounding ministerial res-
ignations.
3
Ignoring turnover caused by changes of government following a general
election, there have been 368 resignations.
4
Of these cases, 295 (80 per cent)
occurred either through cabinet reshuffles or retirement without undue
controversy (including those who went on to other jobs). The remaining 71
resignations, an average of 1.4 per year, have been coded into a set of eight
categories.
5
The analysis conducted allowed for each resignation to be
coded into one of the eight categories as the proximate cause of the resig-
nation, and also for each to be coded for other or background reasons.
6
‘Non-resignations’ were similarly coded. A non-resignation is defined as
any case where the press, a non-political organization or MPs in the House
have suggested the minister should resign (not including ritual catcalls).
Non-resignations are just as important as resignations if there is to be any
attempt to understand the causal process of resignation.
7
For far too long
academics, journalists and self-proclaimed constitutional experts have dis-
cussed the conventions of ministerial resignation, dredging up past cases
of resignation, without fully acknowledging the extent of non-resignation:
134 cases (including 8 threatened resignations) or 2.6 per year. These cases
are also important for understanding the degree of controversy faced by
different cabinets over the years. It should be understood, however, that
given the basis of the data collection, differences over time reflect changes
Blackwell Publishers Ltd. 1998
To continue reading
Request your trial