The length of ministerial tenure in the United Kingdom, 1945-97

British Journal of Political Science

Volumn 37, Issue 2, 2007, Pages 245-262

  Berlinski, Samuel ^a   Dewan, Torun ^b   Dowding, Keith ^b  

^a UNIVERSITY COLLEGE LONDON   (United Kingdom)

^b LONDON SCHOOL OF ECONOMICS   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33947619547     PISSN: 00071234     EISSN: 14692112     Source Type: Journal    
DOI: 10.1017/S0007123407000129     Document Type: Review

References (53)

1. James E. Alt, 'Continuity, Turnover and Experience in the British Cabinet, 1868-1970', in Valentine Herman and James E. Alt, eds, Cabinet Studies: A Reader (London: Macmillan, 1975), pp. 33-54, at p. 33.
2. Jean Blondel, Government and Ministers in the Contemporary World (London: Sage, 1985), p. 8.
3. For example, Peter Hennessy, Cabinet (Oxford: Basil Blackwell, 1986);
4. Richard Heffernan, 'Prime Ministerial Predominance? Core Executive Politics in the UK', British Journal of Politics and International Relations, 5 (2003), 347-72.
5. For example, G. P. Thomas, Prime Minister and Cabinet Today (Manchester: Manchester University Press, 1998);
6. Peter Hennessy, The Prime Minister: The Office and Its Holders Since 1945 (London: Penguin, 2001);
7. Dick Leonard, A Century of Premiers: Salisbury to Blair (Houndmills, Hants.: Palgrave Macmillan, 2005).
8. For example, Sir R. Scott, 'Ministerial Accountability', Public Law (Autumn, 1996), 410-26;
9. Diana Woodhouse, 'The Reconstruction of Constitutional Accountability', Public Law (Spring, 2002), 73-90.
10. Keith Dowding, The Civil Service (London: Routledge, 1995), chap. 8;
11. Keith Dowding and Won-Taek Kang, 'Ministerial Resignations 1945-97', Public Administration, 76 (1998), 411-29;
12. Torun Dewan and Keith Dowding, 'The Corrective Effect of Ministerial Resignations on Government Popularity', American Journal of Political Science, 49 (2005), 46-56.
13. Notably Gerald Kaufman, How to be a Minister (London: Faber and Faber, 1997).
14. Bruce Headey, British Cabinet Ministers (London: Allen and Unwin, 1974).
15. Following the standard use in formal principal-agent modelling we use the female pronoun for prime ministers (principal) and male pronoun for ministers (agents) - unless we refer to actual people, where we use the appropriate pronoun.
16. Valentine Herman, 'Comparative Perspectives on Ministerial Stability in Britain', in Herman and Alt, eds, Cabinet Studies, pp. 55-76.
17. Alt, 'Continuity, Turnover and Experience'.
18. Putting constitutional niceties aside we refer to everyone in our dataset as a 'minister'. In fact, it includes full cabinet ministers, ministers of cabinet rank, junior ministers and government whips. All positions are within the patronage of the prime minister. It does not include parliamentary private secretaries, as these, whilst confirmed by the prime minister are usually chosen by the minister.
19. Gary King, James Alt, Nancy E. Burns and Michael Laver, 'A Unified Model of Cabinet Dissolution in Democracies', American Journal of Political Science, 41 (1990), 846-71.
20. See Headey, British Cabinet Ministers,
21. or Bruce Headey, 'The Role Skills of Cabinet Minsters: A Cross-National Review', Political Studies, 22 (1974), 66-85, for discussion of what makes able ministers.
22. David Walter, The Oxford Union: Playground of Fewer (London: MacDonald, 1984),p. 11,
23. cited by Jeremy Paxman, The Political Animal (Harmondsworth, Midx.: Penguin, 2002), p. 53.
24. In fact, we only code for majority at the start of the government's term of office. Its majority at any point thereafter is highly correlated with its majority at the start of its term. To code each time an MP dies, switches parties or a by-election is held would require a new entry for each minister on each occasion.
25. It could be further claimed that our analysis is biased because we include ministers who entered after the government started. However, our results are robust to looking only at ministers who entered at the start of government.
26. Janet M. Box-Steffensmeier and Brad Jones, 'Time Is of the Essence: Event History Models in Political Science', American Journal of Political Science, 41 (1997), 1414-61.
27. These days the post-election shuffle rarely takes more than a couple of days, but in the past shuffling the cabinet was a more leisurely affair.
28. We chose this rule since David Butler and Gareth Butler in Twentieth Century British Political Facts 1900-2000 (Houndmills: Macmillan, 2000), from which we get most of our data, report different end dates for ministers following a new government forming with its usual reshuffle. Some are given as ending just before the new government forms, some with the reshuffle. Some whips resign between the calling of an election and the election itself. However, all these are on a par. We believe there are no cases of genuine resignation other than as part of the general clear out and reshuffle during the last two weeks of government. Might this censoring rule effect the estimates of the hazard rate? In this respect the Cox proportional model assumes that the duration function may take any form, so that the beginning or end of government is unlikely to bias our analysis as long as the proportionality assumption is correct.
29. If they appear in Butler and Butler, British Political Facts, then virtually all ministers as we define them are included in our sample. Reasons for exclusion include lack of information on age or inconsistencies in Butler and Butler that we were not able to rectify from other sources.
30. The numbers in the latter four categories of observations of ministers is larger than the number of people since many people have had positions at various levels.
31. Prime minister, chancellor of the exchequer, home secretary and foreign secretary.
32. Again these numbers do not add up to total people or total 'observations', since some people are counted as one observation of a 'ministerial career' having held office both as an elected MP and a peer.
33. From now on, prime ministers are excluded from the sample.
34. It is also worth noting that the distinctions are maintained at full cabinet level. In terms of the full cabinet, whereas only half of Attlee's initial twenty ministers had attended university (five going to Oxbridge), twenty of Major's initial twenty-two went to university (seventeen to Oxbridge). It is also worth noting that it was not unusual for all the cabinet ministers to have been to public schools in some of the early Conservative administrations, indeed ten of Eden's eighteen cabinet ministers had been to Eton, and even six of Thatcher's first cabinet had attended that school (and two of Attlee's had also been to Eton). See Butler and Butler, British Political Facts, p. 71.
35. Major's second administration drops to 3.88 years which is still on the high side overall.
36. We should note that whilst we have 2,235 spells, we do not have that number of separate people serving as ministers - since many people have served at several levels and during several administrations.
37. This picture of ministerial stability prevails if each government is analysed separately, though we do not report these results here.
38. Logrank tests confirm that the differences in the survivor functions of Oxbridge and experienced ministers are statistically significant. However, at conventional levels of statistical significance, we cannot reject the null of equality of survivor functions between female and male minsters.
39. A logrank test confirms that the difference in the survivor functions of ministers serving in the first and second term of a prime minister is statistically significant.
40. We have also included dummy variables for highest level of education attained which proved to be statistically insignificant.
41. We included a squared term in age which was not statistically significant.
42. We coded for experience in two ways. First, an experience variable measuring the years of ministerial experience in previous governments at the start of a spell. Secondly, a variable (some experience) coded 0 on first appointment and 1 for subsequent appointments in any later government term. The 'experience' variable (plus a squared term on 'experience') was insignificant when the 'some experience' variable was included. Thus, we have left only the categorical variable and when we refer to experienced ministers we mean those who have served a spell in a previous government. Newly appointed ministers are those appointed in the current term of government.
43. Mattei Dogan and Peter Campbell, 'Le personnel ministriel en France et en Grande-Bretagne', Revue Française de Science Politique (1957), 313-45;
44. Philip Maynard Williams, Crisis and Compromise: Politics in the Fourth French Republic (Hamden, Conn.: Archone Books, 1964), p. 206;
45. Herman, 'Comparative Perspectives'; John D. Huber and Cecilia Martinez-Gallardo, 'Cabinet Instability and the Accumulation of Experience: The French Fourth and Fifth Republics in Comparative Perspective', British Journal of Political Science, 34 (2004), 27-48.
46. Huber and Martinez-Gallardo, 'Cabinet Instability', pp. 46-7.
47. Dewan and Dowding, 'The Corrective Effect'.
48. We find no statistically significant partisan or majority effects. Ministers who are serving in the second term of a prime minister have 46 percentage points higher hazard rates relative to those serving in the first term, whereas those in a third term have a hazard rate 55 points higher.
49. Chief executives may have similar hazards. Someone might work their way up through the management team of a company, but once at the top have a high hazard rating for that company, choosing to move to another such post sooner rather than later, perhaps whilst their company is doing well rather than when it fails.
50. We might note the finding from Table 2 that although Attlee's 1945 administration was the least experienced, most commentators see it as an efficient and enterprising government; and though Major's first administration was the most experienced (and his second highly experienced), they were seen as tired and inefficient.
51. For example H. Best and M. Cotta, eds, Parliamentary Representatives in Europe 1848-2000: Legislative Recruitment and Careers in Eleven European Countries (Oxford: Oxford University Press, 2000).
52. Huber and Martinez-Gallardo, 'Cabinet Instability and the Accumulation of Experience'.
53. Dewan and Dowding, 'The Corrective Effect'; Dowding and Kang, 'Ministerial Resignations 1945-97'; Patrick Dumont, Leon De Winter and Regis Dandoy, 'Demissions gouvernmentales et performances electorales des majorities sortantes (1946-1999)', Courrier hebdomadaire du CRISP, No. 1722 (2001).