Hostname: page-component-745bb68f8f-kw2vx Total loading time: 0 Render date: 2025-02-06T20:42:12.621Z Has data issue: false hasContentIssue false

Heir Apparent Prime Ministers in Westminster Democracies: Promise and Performance

Published online by Cambridge University Press:  13 September 2018

Ludger Helms*
Affiliation:
Ludger Helms, School of Social and Political Sciences, University of Innsbruck, Austria
*
*Corresponding author. Email: ludger.helms@uibk.ac.at
Rights & Permissions [Opens in a new window]

Abstract

While the grand narratives of political leaders and leadership in parliamentary democracies tend to centre on victorious campaigners, prime ministers ‘inheriting’ the office from their predecessor between two parliamentary elections are a widespread occurrence in constitutional practice. Focusing on four Westminster democracies (Britain, Canada, Australia and New Zealand), this article inquires how such heirs apparent have fared in terms of prime ministerial performance. Although in light of their experience, expertise and public recognition, heir apparent prime ministers can be, and have been, considered to be particularly well placed to succeed, when eventually securing the most powerful political office, most of them have actually been conspicuous under-performers. The single most important and strongly counter-intuitive finding of an empirical investigation of different prime ministers is that extensive experience in government, both in terms of duration and diversity of ministerial offices held, seems to correlate more with failed rather than particularly successful premierships.

Type
Articles
Copyright
Copyright © The Author 2018. Published by Government and Opposition Limited and Cambridge University Press

The very notion of ‘heir apparent’ does not sit comfortably with the concept of democracy. In fact, structures and incidents of inheritance in the recruitment of holders of senior political office are the defining feature of hereditary monarchies, observed across different ages and regions of the world (see Finer Reference Finer1997). Hereditary monarchies apart, the great majority of clear-cut examples of hereditary succession in political office and/or of incumbents picking their own successors, relate to regimes which would fail most of the tests for an entirely consolidated democracy or even be considered outright autocratic regimes. Mexico’s infamous tradition of the dedazo provides a particularly prominent example (see Langston Reference Langston2006), but hereditary successions of one kind or another are a much more general feature of leadership selection in republic-style dictatorships.

Where power-holders in autocratic regimes are able effectively to choose their successors freely, more specific patterns can be identified. Comparative political inquiries suggest that ‘rulers … prefer sons over alternative figures more inclined to hasten the succession through assassination or coup attempts’ (Brownlee Reference Brownlee2007: 605), and sons are most likely to succeed fathers whose power pre-dates the party’s (Brownlee Reference Brownlee2007: 628). Reminiscent of the primogeniture in ancient pre-democratic regimes, full-blown autocratic regimes have been generally marked by forms of male inheritance. However, particularly in hybrid regimes, the daughters or wives of power-holders have featured no less prominently in the succession game. As Farida Jalalzai (Reference Jalalzai2008: 212) has pointed out, ‘due to prevailing gender norms, women are more appropriate “heirs” than male relatives since they are deemed natural representatives of men, uncontaminated by their own political ambitions’.

For obvious reasons, there is no clear-cut equivalent of hereditary succession in democratic regimes. That said, heirs apparent of one kind or another are by no means completely alien to, or absent from, contemporary established democracies. Indeed, the phenomenon has been discussed for different types of democratic regime. In the US, vice-presidents in particular have been studied with regard to their possible qualities as heirs apparent (Kincade Reference Kincade2000); there have also been more encompassing conceptualizations of heirs apparent which extend to, inter alia, former cabinet secretaries (Jones Reference Jones2008). The French experience suggests that the notion of heirs apparent is by no means confined to democratic regimes with an established office of ‘vice’ or ‘deputy’ political chief executive. Indeed, individual political figures, such as Georges Pompidou or Valéry Giscard d’Estaing, have occasionally been referred to as heirs apparent (see Bernstein and Rioux Reference Bernstein and Rioux2000: 86; Cordell Reference Cordell2005). However, it is parliamentary democracies with their more flexible and fluent recruitment regimes that have proved particularly fruitful grounds for scholarly and public debate about heirs apparent for senior political office, and where this article seeks to break new ground.

The next section sets out to develop a reasonable conceptualization of ‘heir apparent prime ministers’, which will then be applied to a sample of prime ministers from four Westminster democracies: Britain, Canada, Australia and New Zealand (1945–2016). There is an established tradition of studying prime ministers and core executive politics in these Westminster democracies from a comparative perspective (see, for example, the classic study by Weller Reference Weller1985; and, more recently, Bennister Reference Bennister2012; Esselment et al. Reference Esselment, Lees-Marshment and Marland2014; Rhodes et al. Reference Rhodes, Wanna and Weller2009; Shaw and Eichbaum Reference Shaw and Eichbaum2014) that forms the background and basis of this more specific investigation.Footnote 1

The inquiry, however, extends beyond developing a conceptual focus on the phenomenon of heirs apparent in the office of prime minister: it seeks to assess the impact that this pattern of political advancement has tended to have on prime ministerial performance. That the politics of succession in leadership positions matters for how voters weigh the past executive’s performance and hold leaders to account has been demonstrated recently (De Ferrari Reference De Ferrari2015), but we still know very little about the difference that track records of candidates make in terms of the perceived performance of an incumbent political leader, and in particular of prime ministers. Even the otherwise most valuable and empirically rich volume by Paul ‘t Hart and John Uhr (Reference ‘t Hart and Uhr2011) on ‘transition and succession in government’ has conspicuously little to say about this important aspect. By contrast, the politics of succession has figured prominently in recent research on political parties (see, for example, Bynander and ‘t Hart Reference Bynander and ‘t Hart2008; Cross and Blais Reference Cross and Blais2012; Cross and Pilet Reference Cross and Pilet2016; Horiuchi et al. Reference Horiuchi, Laing and ‘t Hart2015), and some authors have specifically looked into the ‘predecessor effects’ on an incoming leader. Long-term party leaders have been identified as being particularly ‘hard acts to follow’ (Horiuchi et al. Reference Horiuchi, Laing and ‘t Hart2015) – a finding to be remembered when assessing the performance of prime ministers from our sample (see below).

The closing section, in addition to providing a brief summary of the main findings, looks at the implications of the suggested causal relationships and addresses some of the key items on the agenda for future research.

Conceptualizing heir apparent prime ministers

Generally, the position of heir apparent tends to be firmly associated with a shared party affiliation of an incoming and an outgoing political leader. Just succeeding an incumbent from the same party in office is, however, certainly not sufficient to meaningfully classify a candidate (in retrospect) as a political heir apparent. This is true even if one were to add another criterion – the successor’s lack of his or her own electoral mandate – and to include only those candidates who won office between two parliamentary elections. Otherwise, the world of parliamentary democracies would be full of heirs apparent. Indeed, even in the UK, historically the beacon of adversary politics and sweeping alternations in power, nearly half of all prime ministers appointed between 1945 and 2016 initially gained office by means of an intra-party succession, rather than as successful opposition challengers, and there are few, if any, countries that have not had any similar experience. Such prime ministers are best characterized as ‘takeover prime ministers’ (Worthy Reference Worthy2016), rather than genuine heir apparent prime ministers. Thus, more specific criteria are needed to substantiate the latter category of political leaders.

Still, at least in party-centred parliamentary democracies, following an incumbent from the same party marks a necessary, in fact the single most important, feature of a candidate’s profile as heir apparent. For parliamentary democracies, which are in effect party democracies, with competitive and power-seeking parties being more or less central to virtually every stage of the public policymaking process, the notion of an heir apparent from a different party would come close to a contradiction in terms.

Another key criterion should be seen in a candidate’s membership of the previous government. Other patterns of leadership recruitment are possible, but they do not really conform to the basic idea associated with an ‘heir apparency’ in political terms. Whenever a governing party reaches out for a candidate located beyond the established political executive elite at the national level to replace an incumbent head of government (which may bring in holders of electoral office from the subnational level as well as technocrats), this is likely to indicate the determination deliberately to prevent any impression of installing anything like a political heir to the outgoing leader.

Third, it seems crucial that a candidate is widely seen in the media and the public to be in the position of heir apparent. While party approval is the most obvious necessary condition to establish an heir apparency, it is public recognition of a candidate as the ‘natural successor’ that turns him or her into an heir apparent to the incumbent prime minister. The public acknowledgement of a candidate as an heir apparent in relevant sections of the mass media is arguably the closest equivalent to an institutionalized entitlement to succeed an incumbent in pre-democratic regimes. This publicly assigned status, which can be expected to build up a candidate’s authority even before eventually assuming the premiership, distinguishes genuine heir apparent prime ministers from mere takeover prime ministers. By contrast, a candidate’s explicit endorsement by the outgoing prime minister is not essential for obtaining the status of political heir apparent.Footnote 2

Finally, I suggest that for a candidate to be an heir apparent he or she must not actively challenge and oust the incumbent prime minister. Candidates chasing and toppling the sitting prime minister (which, in the Westminster democracies, implies a successful formal challenge for the party leadership) are, after all, much better characterized as ‘usurpers’ rather than ‘heirs’. This final distinction may possibly seem to be more about semantics than politics; however, it is important to devise conceptual distinctions that would seem likely to stand the test of terminological plausibility. Also, if the way in which candidates win office is considered likely to shape their standing and performance as prime minister (see Weller Reference Weller2014: 494), usurpers and heirs, as distinguished above, would in fact seem to make for two different categories of prime minister. All else being equal, and apart from other issues of perceived legitimacy, the former can be expected to be much more committed to breaking away from his or her predecessor’s legacy than the latter, which is likely to become an independent factor shaping the political fate of a given prime minister.Footnote 3

These observations underscore the need to put the debate on heirs apparent in democratic politics, and in executive politics more specifically, on a more elaborate conceptual basis. By way of concluding the discussion above, I suggest the following criteria for identifying heir apparent prime ministers in parliamentary democracies. A candidate:

  1. 1 must come from the same party as the outgoing prime minister,

  2. 2 must have been a minister in the previous government,

  3. 3 must have been perceived as the most likely successor to the sitting prime minister by the media and the public for at least several weeks before the change in the prime ministerial office takes place, and

  4. 4 must not have directly challenged and ousted the incumbent prime minister.

Needless to say, additional criteria – such as the public policies that outgoing and incoming prime ministers stand for – are possible, but their inclusion would make comparative empirical classifications of individual candidates significantly more difficult and vulnerable, as the assessment of continuity and change at the level of public policies lies much more in the eye of the beholder than the criteria suggested above.

Heir apparent prime ministers and leadership performance

The first thing to note about the number of possible heir apparent prime ministers from the four countries under consideration is that the number, and shares, of candidates meeting the basic criterion for an heir apparent prime ministership – namely the same party affiliation as the outgoing prime minister – varies considerably between the countries. For the period 1945–2016, the score ranges from 35.7% (Canada) to 47.1% (Australia), with New Zealand (44.4%) and the UK (46.7%) falling in between.Footnote 4 A more detailed picture emerges if the total period is divided into two equal halves (1945–80 and 1981–2016): the UK then stands out as the only country which has experienced a slight decrease in the number of takeover prime ministers since the early 1980s (three instead of four). However, this is to a large extent due to the fact that there have been fewer British prime ministers in the second period distinguished (six instead of nine), which has been dominated by the long-term premierships of Margaret Thatcher and Tony Blair. On a different count, it is worth noting that all more recent British prime ministers winning the office in the first place in the wake of an election victory (Thatcher, Blair and David Cameron) were succeeded by takeover prime ministers (John Major, Gordon Brown and Theresa May). Thus, their percentage share actually went up from 44.4% to 50%, making the UK the country with the largest percentage share of takeover prime ministers in the period 1981–2016. While the figures were stable for New Zealand (four), both Canada (three, up from two) and Australia (four, up from three) have had a greater number of prime ministers belonging to that category since the 1980s. Accounting for the percentage share, the figures remained unchanged for New Zealand (44.4%), while they went up both for Canada (from 33.3% to 37.5%) and Australia (from 37.5% to 44.4%). For all four countries combined, the share of takeover prime ministers increased from 39.9% to 44.1%.

Career patterns and profiles

Table 1 lists not only those post-1945 prime ministers who succeeded a departing prime minister from their own party, but also indicates which of those candidates were not classified here as genuine heirs apparent, and why. Those candidates who failed to qualify as heirs apparent either held no ministerial office in the outgoing government or enjoyed no status as a likely successor to the sitting prime minister in the media and among the public, and/or actively challenged and toppled the incumbent prime minister.

Table 1. Intra-Party Successors to the Prime Minister and Heirs Apparent in Britain, Canada, Australia and New Zealand (1945–2016)

Sources: https://en.wikipedia.org; https://www.gov.uk/government/people; http://www.thecanadianencyclopedia.ca; http://primeministers.naa.gov.au.

Notes: All data compiled by the author. a=not a minister in the outgoing government; b=not perceived in the media and the public as a long-standing successor; c=direct challenger of the incumbent prime minister.

As to the second criterion – holding ministerial office in the outgoing government – there is conspicuously little variation between our four countries. With just one exception (John Turner, Canada), all candidates held ministerial office in the previous government, though some of them resigned from office shortly before making their bid for the premiership (such as Paul Keating and Malcolm Turnbull in Australia). Even Turner was a special case: in the early 1970s, when he became minister of finance under Pierre Trudeau, ‘he was clearly the heir apparent if anything should happen to the Prime Minister’ (Gordon Reference Gordon1977: 327); but then, a few years on, Turner left politics in favour of a legal and business career, and when he returned in the mid-1980s in order to seek the premiership, he did so as an outsider.

While the picture is remarkably clear-cut for our sample of countries, this criterion would still appear to possess a substantive discriminatory power for comparative research. Indeed, other findings strongly suggest that membership of the previous government marks a particular feature of Westminster-style democracy rather than a more general feature of parliamentary democracy. For example, an empirical investigation conducted by the author reveals that in Austria, Germany and the Netherlands (three countries from the consociationalist end of the spectrum of parliamentary government types), more than half of all takeover prime ministers/chancellors between 1945/9 and 2016 succeeded someone from their own party without having been a minister in the outgoing government (though some of them had earlier experience in government).

The third criterion – being publicly perceived by the media as a natural successor to the sitting prime minister – tends to be somewhat more elusive, and contested, in political reality. For example, in the UK both James Callaghan and John Major, while holding very senior political offices in the previous government(s), emerged from a quickly arranged open party contest rather than being offered the party leadership and the prime ministership on the basis of a widespread perception as a ‘natural successor’ to the prime minister. In Major’s case, Geoffrey Howe, Thatcher’s longest-serving minister and deputy prime minister until his resignation on 1 November 1990 (just three weeks ahead of Thatcher’s resignation), was in many respects more of a long-expected heir apparent to the prime minister than Major himself. Callaghan’s winning the premiership may have been somewhat less of a surprise. He was Prime Minister Wilson’s preferred successor, but in the first round of the contest for the Labour Party leadership he was defeated by Michael Foot, and in any case was ‘not the anointed heir apparent’ (Rubinstein Reference Rubinstein2003: 313). To be sure, with regard to this aspect, Callaghan and Major were still closer to the status of heir apparent than Harold Macmillan and Alec Douglas-Home who both, alongside several other hopeful contenders, faced a powerful deputy prime minister (Rab Butler, who could meaningfully be characterized as the established heir apparent to both Eden in 1956/7 and Macmillan in 1963). In contrast, Theresa May’s case was clear-cut: while being a member of the Cameron–Clegg government, she was clearly an outsider in terms of a would-be prime ministerial candidate. Indeed, for the best part of Cameron’s premiership, perceptions of a possible heir apparent in the media, and the Conservative Party, strongly focused on George Osborne, the chancellor of the exchequer.

Failing to be perceived as the natural successor, however, is no peculiarity of some British candidates: New Zealand’s 30th prime minister, Bill Rowling, also could not really claim to hold the status of a natural successor to Prime Minister Norman Kirk. It was Hugh Watt, not Rowling, who served as deputy prime minister under Kirk, and who also became interim prime minister when Kirk resigned. Watt was also Labour’s long-standing deputy party leader. He lost the party leadership contest on 6 September 1974 to Rowling mainly because the parliamentary party felt that, at 61, he was too old and Labour and the government needed a younger leader. A similar case was John Gorton, Australia’s 19th prime minister, who followed Harold Holt after the prime minister’s mysterious sudden disappearance without having been seen as a particularly likely successor by many.

The fourth criterion – the absence of a direct and successful attempt to oust the prime minister – bears empirical relevance within our sample only with regard to some candidates from Oceania’s two largest parliamentary democracies. What was a rare exception in New Zealand’s post-war political history – the toppling of Prime Minister Jim Bolger by Jenny Shipley in late December 1997 – would be a much more established pattern in Australia. In fact, four out of five Australian takeover prime ministers failed to qualify as heir apparent prime ministers because they usurped rather than inherited the office. Major contextual differences apart, this applies to William McMahon, Paul Keating, Julia Gillard and Kevin Rudd as well as Malcolm Turnbull.

Overall, 13 out of 27 post-war takeover prime ministers from the four countries under consideration can be classified as heir apparent prime ministers. In the UK, both Anthony Eden and Gordon Brown were truly exemplary cases of heirs apparent. In Eden’s case this status was also formally underscored by his position as deputy prime minister and as deputy leader of the Conservative Party. While Brown was neither deputy prime minister nor deputy leader of the Labour Party – positions held by John Prescott throughout the Blair era – there was never any doubt about Labour’s pecking order with regard to No. 10.

In the same period, 1945–2016, Canada experienced no fewer than four heir apparent prime ministers: Louis Saint-Laurent’s winning the premiership may not have been a foregone conclusion; there were other contenders from within the Liberal Party. But Saint-Laurent was Prime Minister King’s personal choice, and the latter campaigned hard for his preferred successor to win the Liberal Party leadership, which Saint-Laurent did on the first ballot. Neither was Pierre Trudeau the safe successor to the outgoing prime minister and won the party leadership only on the fourth ballot, yet he was still the frontrunner and the clear favourite with the Canadian public. Kim Campbell, while also not being the only candidate seeking to win the party leadership and premiership from the hands of Brian Mulroney, effectively ‘became his heir apparent and a media darling’ (Campbell Reference Campbell2001: 95) long before Mulroney announced his resignation. Among Canadian heir apparent prime ministers, Paul Martin arguably marks the most clear-cut case, even though he spent many years in conflict with the prime minister whose office he eventually inherited (see Sands Reference Sands2002: 4).

There were two clear-cut cases of heirs apparent in Australia: Harold Holt, who held senior ministerial posts under his predecessor and was, years before winning the premiership, touted in the press as a certain successor to Menzies and a potential prime minister. In his book on Australian prime ministers, Brian Carroll (Reference Carroll2004: 201–206) therefore correctly labels Holt as ‘the heir apparent’. The only other Australian post-war prime minister coming very close to Holt’s record is Ben Chifley. Chifley, who became prime minister in July 1945, was a member of the previous governments and arguably the politically most senior member of the Curtin government (though formally not deputy prime minister). When John Curtin died on 5 July 1945, Frank Forde, the deputy prime minister, was appointed prime minister. Yet he held the office for just over a week, and on 13 July 1945, after having won the leadership contest in the Australian Labor Party, Chifley assumed the office of prime minister. That said, Chifley very much deserves to be classified as a genuine heir apparent as he did not actively seek the role of challenger and defeater (see Hughes Reference Hughes1976: 134).

In the same period, New Zealand witnessed no fewer than five heir apparent prime ministers who met all four criteria suggested here. While being remembered in particular for his second, more extended term as prime minister (1960–72), Keith Holyoake was a classic case of an heir apparent prime minister when he first won the office in the late 1950s. He was the first person ever to be formally appointed to the post of deputy prime minister and advanced to the premiership from this position when Prime Minister Sidney Holland retired due to ill health in September 1957. The story of New Zealand’s second post-1945 heir apparent prime minister, Jack Marshall, is strongly reminiscent of Holyoake’s initial coming to power: being deputy prime minister in the Holyoake administration, he joined the leadership contest within his party, and won, after the prime minister had declared his resignation from office. Geoffrey Palmer, the third in line, provides another example of a deputy prime minister turning prime minister. Mike Moore, the country’s fourth heir apparent prime minister of the post-war period, was a less clear-cut case of an heir apparent than Holyoake, Marshall or Palmer. Rather than being a long-standing ‘crown prince’ to the prime minister in the ailing Palmer administration, he was simply the best candidate to lead Labour in the 1999 general election, which was widely being predicted as a certain defeat for Labour. His only challenger in the party leadership race, Richard Northey, an MP, was a real surprise candidate and would have been a much less classic heir apparent than Moore himself. Finally, with Bill English, appointed prime minister in December 2016, New Zealand experienced another textbook case of an heir apparent prime minister who possessed all the defining features suggested above, plus the explicit endorsement of the departing prime minister.

As these assessments demonstrate, the share of takeover prime ministers who can also be characterized as heir apparent prime ministers (according to the criteria suggested above) differs greatly between the individual countries covered. The proportion of candidates falling into this category is largest in Canada and New Zealand, where between 71% (New Zealand) and 80% (Canada) of all candidates following in the steps of a prime minister from the ranks of their party possessed all the defining features of an heir apparent. The share was 25% for Australia and 28.5% for the UK.

Reflecting the multidimensional features of the concept suggested here, there are several reasons that might account for this pattern. Not having been considered by the media as a likely successor, and having formally challenged and ousted the incumbent prime minister are responsible for five clear-cut disqualifications each, with one other case relating to not having held a post in the previous government. The second reason for not qualifying is by far the most intriguing one. As Patrick Weller (Reference Weller2012) has argued, the likelihood that a party leader, and prime minister, is ousted by an intra-party challenger, is essentially the result of different sets of rules which shape the choice of strategies adopted by potential challengers. In a more recent study, Mark Bennister and Tim Heppell (Reference Bennister and Heppell2016) have contended that the reasons for party leader and prime ministerial survival are more complex and include different party cultures and political circumstances. However, both approaches help explain why the number of ‘usurpers’ has been incomparably higher in Australia than in the other three countries.

That the UK, notwithstanding the more incumbent-friendly party environment, hosts a very similar share to Australia’s of takeover prime ministers falling short of being genuine heir apparent prime ministers points to a different factor: the size of the pool of candidates considered eligible for the office of prime minister (which is, again, not independent of the rules for intra-party succession). Several British takeover prime ministers arrived at No. 10 rather unexpectedly, thereby failing to qualify as an established and long-standing heir; in this regard Theresa May is just the latest case in point. By contrast, the considerable share of heir apparent prime ministers among the total of takeover prime ministers in New Zealand may to some extent be attributed to the notable centrality of the office of deputy prime minister as the natural recruiting base for mid-term successors, for which there is no equivalent in the other three countries. That said, the Canadian case suggests that a greater pluralism with regard to the office held by respective candidates immediately before advancing to the premiership between two elections (indeed, all five moved in from different places) does not necessarily undermine the prospect of having a very sizeable share of genuine heirs apparent. Overall, then, while institutional differences clearly matter even within what is widely considered to constitute an established sample of ‘most similar cases’, there is a striking amount of contingency shaping the politics of prime ministerial succession in those four countries.

Prime ministerial performance

How have heir apparent prime ministers fared as political leaders? Any substantive inquiry into matters of prime ministerial performance faces considerable theoretical as well as methodological challenges, which can be addressed only very briefly within the scope of this article (see, however, Helms Forthcoming). Further, to avoid a purely exploratory approach, it would also seem useful to revisit some earlier assessments of the difference that long apprenticeships and heir apparency are believed to make in terms of prime ministerial performance, before assessing the cases from our sample.

Theoretical and empirical preliminaries

Longevity and re-election track records of incumbents have been widely considered to constitute important indicators of prime ministerial performance (see, for example, Rhodes and ‘t Hart Reference ‘t Hart2014: 13). Yet, even in the age of the ‘permanent campaign’, governing and campaigning have remained two fundamentally different activities, and it is easy to confuse longevity with prime ministerial excellence. Indeed, in parliamentary democracies with complex coalition governments, heads of government effectively cannot be ousted, or at least not by the voter. In this type of democratic regime, longevity of a prime minister is shaped primarily by the dominant governing party’s (and its candidate’s) coalition-building potential. Other things being equal, junior coalition parties may even be more willing to join a coalition if the prime minister representing a different party is weak, and likely to be seen as a figure with limited political stature.

Thus, compared to longevity, which is strongly context-dependent and tends to be no more than a fairly vulnerable proxy-variable for any particular leadership performance even under the most favourable circumstances, job approval ratings of an incumbent provide a much more direct and valid indicator of perceived successful prime ministerial performance. However, in the end only more sophisticated surveys specifically seeking to capture different dimensions of performance can provide a reasonable basis for comparative performance-related assessments of different incumbents. Therefore, the assessments below draw on a series of expert surveys on rating and ranking prime ministers, in particular those reported in the empirically rich case studies gathered by Paul Strangio et al. (Reference Strangio, ‘t Hart and Walter2013) and several more recent sources. They include a wealth of different dimensions, from personality-related features (such as trustworthiness) and political dimensions of a candidate’s performance (such as party and coalition management) through concrete policy achievements.Footnote 5 While no methodological approach is free of inherent weaknesses,Footnote 6 expert surveys would, after all, still appear to be perhaps the least deficient approach to assessing the performance of political leaders in a given country (see Bose and Landis Reference Bose and Landis2003; ‘t Hart and Schelfhout Reference ‘t Hart and Schelfhout2016: 153–156).

The status of heir apparent – succeeding a prime minister as a broadly acknowledged natural intra-party successor and, usually, after a distinguished and extended ministerial career – has drawn little, if any, particular attention in the literature. This notwithstanding, long and distinguished political careers have been widely considered to provide favourable conditions for performing successfully in the office of prime minister. After all, this experience-focused philosophy is at the very heart of established recruitment regimes in parliamentary democracies, under which it normally takes more than 20 years for a candidate to climb the greasy pole (with extended stretches both in parliament and as a government minister). The complexity of contemporary politics, and the increasing demands on those who take charge of directing the public policymaking process, is widely understood to require, and justify, an advanced degree of political professionalization and careerism (see Borchert Reference Borchert2003). This idea has also figured prominently in recent comparative work on political chief executives. For example, in their empirical assessment of prime ministers in Central Eastern Europe, Florian Grotz and Ferdinand Müller-Rommel (Reference Grotz and Müller-Rommel2015) suggest that the more previous political experience a prime minister has, the more likely he or she is to be a powerful and successful prime minister. More specifically, various scholars, while usually eschewing more particular conceptual issues, have highlighted the greatness of precisely those prime ministers they characterized as heirs apparent (see, for example, Pakulski and Tranter Reference Pakulski and Tranter2015: 79). Thus, the central hypothesis emerging from this work is that heir apparent prime ministers possess special resources (in terms of experience, expertise and public recognition), and thus would appear set to perform successfully in office.Footnote 7

Before proceeding to the empirical assessment of different prime ministers’ performance, one possible caveat is to be addressed briefly here. In the eyes of many party scholars, both takeover and heir apparent prime ministers are conceived of as the inevitable victims of ‘party exhaustion’: voters are believed to get tired of the same parties the longer they govern, largely irrespective of the faces they may present, and those arriving in office late in the day are doomed to fail. Governing parties indeed generally tend to incur electoral costs of governing, and thus prime ministers are potentially at risk of getting dragged into their party’s struggle against the downturn (see Green and Jennings Reference Green and Jennings2017: chs 5 and 6). However, the relationship between prime ministerial performance and the electoral performance of his or her party is much more complex. For one thing, prime ministerial performance cannot be meaningfully reduced to aspects of electoral performance. Even when prime ministerial performance is defined in terms of the development of his or her party’s electoral support, the prime minister is not merely a passive element in a purely party-driven game. Indeed, in their major recent study on the costs of governing, Jane Green and Will Jennings (2017: 194) specifically acknowledge that ‘a change of leadership in the midst of a party’s period in office may offer a chance to alleviate … some governing costs under certain conditions’. A closer look reveals several more particular things: first, prime ministers do not simply become less and less successful the longer they govern. Long-term prime ministers are not always considered to have had their strongest performance during their first term; for example, Thatcher’s second term was judged to have been considerably more successful than her first and third (Theakston Reference Theakston2013: 235). Second, it is not always the case that a later incumbent in a series of takeover prime ministers is widely seen as the worst prime ministerial performer. For example, Harold Macmillan is unequivocally considered a better prime minister than his predecessor, Anthony Eden.Footnote 8 Finally, some early popularity assessments of takeover prime ministers – for example, after a few months in office, John Major was hailed as ‘one of the most popular British leaders of the last 30 years’ (Whitney Reference Whitney1991), and Theresa May was even found to be the most capable British leader in more than 40 years (Osborne Reference Osborne2017) – strongly suggest that being a takeover prime minister is by no means synonymous with fighting a losing battle from the start.

Empirical patterns of prime ministerial performance

The picture that emerges from an inquiry into prime ministerial performance in the four Westminster democracies is clearly dominated by assessments that see the great majority of heir apparent prime ministers as under-performers. This is certainly true for British Prime Ministers Eden and Brown, who featured in the 2010 expert survey organized by Kevin Theakston and Mark Gill among the three worst performing post-1945 prime ministers: Brown, was ranked number 10 out of 12, with an average score of 3.9, and Eden was number 12 with an average score of 2.3 (Theakston and Gill Reference Theakston and Gill2011). In particular, Eden’s status as one of the worst British prime ministers after World War II, and arguably the worst of all, is documented in many different popular opinion and expert surveys. Indeed, he came in with the lowest score in all but one of the surveys and assessments reported by Theakston (Reference Theakston2013), the only exception being the Angus Reid poll published in 2010, in which Eden achieved one percentage point more in the category of good prime minister, and one percentage point less in the category of bad prime minister, than Alec Douglas-Home (Theakston Reference Theakston2013: 222).

Among Canadian heir apparent prime ministers, Louis Saint-Laurent and Pierre Trudeau fared comparatively well. In a recent survey by Norman Hillmer and Stephen Azzi (Reference Hillmer and Azzi2011), they came in as number 5 (Trudeau) and number 7 (Saint-Laurent) out of a total of 22 prime ministers ranked. In an earlier survey conducted by Norman Hillmer and Jack Granatstein (Reference Hillmer and Granatstein1997) Saint-Laurent even featured in the second highest category (‘near great’) out of six categories (running from ‘great’ to ‘failure’). The assessments of Trudeau are to be considered carefully in our context, as Trudeau was appointed prime minister twice, in 1968 (serving to 1979) and in 1980 (serving to 1984), and he was not an heir apparent prime minister on returning to office in the 1980s. Also, some of his achievements shaping his overall performance and assessment score relate to his second tenure. Saint-Laurent’s and Trudeau’s ratings contrast starkly with the significantly less favourable evaluations of the performance of Paul Martin, ranked 15, and Kim Campbell who was ranked number 22 and thus at the very bottom of the Canadian prime ministerial league. Campbell also received the lowest score of all 23 prime ministers evaluated in a more recent expert survey (Azzi and Hillmer Reference Azzi and Hillmer2016).

Among Australian prime ministers, Harold Holt tends to be considered less of a failure than Eden and Brown in Britain. However, he features prominently as a candidate in the contest for the ‘all-time worst prime minister’ (where he was ranked number 4, ex aequo with Joseph Cook in a small expert survey conducted in 2001 by the Australian Financial Review), and was not mentioned among the 10 best prime ministers identified (see Strangio Reference Strangio2013: 272). In another expert survey, carried out in 2004, Holt received the second lowest score of all 11 post-war prime ministers to that date (Strangio Reference Strangio2013: 274). While his overall ranking in another expert survey directed by researchers from Monash University is less bad, he was placed in the lowest quarter of 23 prime ministers from Edmond Barton to Kevin Rudd (Strangio Reference Strangio2013: 278). For reasons of fairness it should be added, though, that Holt’s career as a prime minister was short and ended abruptly by what seems to have been a fatal accident. Very much unlike Holt, Ben Chifley is regularly ranked in the best 25% of all Australian prime ministers. In the 2010 Monash survey he came in as third out of 23 (‘overall rating on a scale from outstanding to failure’) and fourth respectively (on the basis of cumulative ratings on 10 performance indices; see Strangio Reference Strangio2013: 278–279). In an older survey, the Canberra Times poll conducted in 1992, Chifley even led the field of then 15 prime ministers, with a total score more than 15% higher than that of the second best rated prime minister (Strangio 2013: 269).

Among New Zealand’s heir apparent prime ministers, Keith Holyoake received by far the highest scores in various expert rankings. In a scholarly ranking carried out in 2011, Holyoake secured the fifth highest overall score of a total of 32 prime ministers (Johansson and Levine Reference Johansson and Levine2013: 305). However, like his Canadian counterpart Trudeau, Holyoake had two tenures as prime minister, and was an heir apparent only when he won the office first in 1957. Moreover, while Trudeau’s first term was considerably longer than his second, Holyoake’s first term ended after just three months. The ‘real’ premiership of Holyoake, which won the praise of many observers, started only in late 1960 when he won the office from the ranks of the opposition and held it for more than 11 years. In the same ranking, Palmer, Marshall and Moore came in as numbers 15, 22 and 27. In another ranking list, which includes only those prime ministers appointed after 1945, and serving until 2011, Palmer, Marshall and Moore are numbers 7, 9 and 11 out of 12. The premiership of Bill English, New Zealand’s most recent heir apparent prime minister (2016–17), has not received any substantive scholarly assessment yet, which is why his premiership is not considered in more detail in the analysis below. Yet in light of the various assessments available, it would seem reasonable to expect him to receive future rankings somewhere towards the lower end.

The picture that emerges from these assessments is clear: the large majority of heir apparent prime ministers identified have tended to disappoint, with their performance being perceived as average at best, and more often significantly below average. The closest contenders for the status of heir apparent prime minister from our sample of countries are probably Callaghan and Major from the UK, and Rowling from New Zealand. However, if one were to add their names and evaluations to the performance record of heirs apparent in the four countries under consideration, it would not change the overall picture at all; in fact, all three were judged as slightly or, more often, clearly below average (see Johansson and Levine Reference Johansson and Levine2013; Theakston Reference Theakston2013).

It is important to note, though, that, while many heir apparent prime ministers have been judged as under-performers, a disappointing performance is by no means a foregone conclusion. There can be no doubt that heir apparent prime ministers may perform successfully. The few but major exceptions from our sample of countries are: Saint-Laurent and Trudeau (Canada), Chifley (Australia) and Holyoake (New Zealand). However, as mentioned above, in particular the extended premiership of Holyoake – split into a three-month period (1957), when Holyoake was indeed an heir apparent, and a second incomparably longer term (1960–72) which developed out of the electoral victory of the National Party in late 1960 – cannot be counted as that of an heir apparent. Further, given their particular track records, Saint-Laurent, Trudeau and Chifley were to a lesser degree textbook cases of long-standing heirs apparent than, for example, Eden, Brown or Holt.

While this rather small-n sample of successful heir apparent prime ministers may make such an inquiry vulnerable, it is still tempting to look for shared features that mark the political rise and the tenures of the most successful heirs apparent from our sample (Saint-Laurent, Trudeau and Chifley, who are widely seen to belong to the most accomplished 50% of post-war prime ministers in their respective countries), and for possible contrasts in the profiles of the worst-ranked heir apparent prime ministers from each country.Footnote 9 The poorest (or at least worst ranked) heir apparent prime ministers from our four countries were: Eden and Brown (UK), Campbell and Martin (Canada), Holt (the only other Australian heir apparent prime minister apart from Chifley) and Marshall and Moore (New Zealand).

The comparative assessment below focuses on four aspects: the parliamentary experience (overall, i.e. until being appointed prime minister; and net, i.e. excluding years as government minister while holding a seat in parliament), the number of years as government minister, the number of different ministerial offices held, and the length of the term of the previous prime minister (see Table 2). The length of term of the heir apparent prime ministers themselves is not being considered here because durability in office is often taken as a proxy for a successful performance (see above), which creates an obvious methodological problem. Also, the length of experience as party leader before assuming prime ministerial office has not been considered here, as (unlike candidates winning the prime ministership from the position of opposition leader) heir apparent prime ministers, at least in Westminster contexts, usually become party leader only a few days before (or even after) being appointed prime minister.

Table 2. Best- and Worst-Ranked Heir Apparent Prime Ministers, and Top-Ranked Prime Ministers Compared

Sources: https://en.wikipedia.org; https://www.gov.uk/government/people; http://www.thecanadianencyclopedia.ca; http://primeministers.naa.gov.au.

Notes: All data compiled by the author. * Including only those heir apparent prime ministers who have widely been considered to belong to the best 50% of all post-war prime ministers in their respective country (see note 9). ** Chifley’s immediate predecessor, Frank Forde, held the office for just one week, yet Chifley was effectively much more an heir apparent to John Curtin than to Forde (see above), which is why figures for both tenures have been calculated. *** Eden’s intra-party predecessor, Winston Churchill, held the office of prime minister twice (1940–5 and 1951–5), which is why two figures have been calculated.

As Table 2 indicates, there are major differences to be found within the family of heir apparent prime ministers: the three best-performing heir apparent prime ministers from our sample of countries had significantly less extended parliamentary and ministerial careers and also held fewer different ministerial offices before becoming head of government.

As to the first two aspects – the length of parliamentary and ministerial experience – the findings do not support the widespread assumption that a more extended career as a professional politician benefits a candidate when eventually winning the premiership. It seems that candidates are, ceteris paribus, better off if they advance to the top after just a few years in parliament and/or government, before they possibly start running out of energy and ideas or adopt entrenched departmental views and habits. In particular, exceptionally long parliamentary careers do not seem to benefit prime ministerial performance.

The second finding – that those who performed better held, on average, slightly fewer different ministerial offices than the most notable under-performers – is perhaps even more remarkable. It would, again, appear plausible to argue that serving in a number of different departments before eventually becoming prime minister provides candidates with a most useful set of political and policy-related experiences when it comes to providing effective cross-departmental leadership. However, this is not what our cases and findings suggest. More than two-thirds (five out of seven) of the worst-ranked heir apparent prime ministers gathered experience in four or more different departments before advancing to the premiership.

The third finding – that the worst-ranked heir apparent prime ministers, with few exceptions, tended to follow particularly long-serving prime ministers – is certainly more in line with conventional wisdom, yet nevertheless well worth noting. More than half of the worst-ranked heir apparent prime ministers from our sample took over from an outgoing prime minister who had been in power for more than 10 years.Footnote 10 Thus, what Yusaku Horiuchi et al. (Reference Horiuchi, Laing and ‘t Hart2015) found for party leaders also applies for prime ministers. Other things being equal, the longer the term of an incoming prime minister’s predecessor, the longer his or her shadow, and the heavier the legacy to come to terms with. This would appear to be true not only in terms of public policy but also, and perhaps in particular, with regard to established public expectations of public leadership, including leadership rhetoric and media management.

The importance of the patterns highlighted above is further underscored by a comparative analysis of the top-ranked prime ministers assuming office after 1945 from the countries under consideration (Clement Attlee, UK; Lester Pearson, Canada; Bob Hawke, Australia; and Helen Clark, New Zealand). Overall, the profiles of these top performers are much more similar to the profiles of the most successful heir apparent prime ministers than to those of the strongly under-performing heir apparent prime ministers. Three of the four top performers had previous ministerial experience in just one department (Pearson and Clark) or no ministerial experience at all (Hawke); and their average length of experience as government minister before becoming head of government was just five and a half years. The picture looks different for the parliamentary experience of prime ministers: with 14.4 years on average, the overall parliamentary experience of the top performers was much more similar to that of the worst-ranked heir apparent prime ministers than to that of the more highly ranked heirs apparent (20.35 and 5.33 years respectively). With 9.67 years, the average score concerning net parliamentary experience (‘MP only, without simultaneously holding a ministerial office’) was also significantly closer to that for the under-performing heir apparent prime ministers (8.67 years) than to that for the best-performing heirs apparent (2.15 years). Thus, unlike the two executive-related factors distinguished, the length of parliamentary experience does not seem to be a strong determinant of prime ministerial performance – which, at least in the context of Westminster democracies, can still be considered a finding notable in its own right. Finally, the importance of the fourth factor distinguished (the length of the tenure of the previous prime minister) is also reflected at the level of top-ranked performers: with an average figure of less than five years, the prime ministers belonging to this group had a lighter legacy to deal with than either group of heir apparent prime ministers.

In light of these figures, it can be concluded that, everything else being equal – and very much in contrast to conventional wisdom – a modest ministerial track record in terms of length and in particular diversity tends to benefit the performance of a prime minister. Such effects relating to a candidate’s own track record may combine with a short or moderate tenure by the outgoing prime minister, which also tends to have a favourable effect on the fortunes of an incoming prime minister. The difference in the perceived performance of relatively successful heir apparent prime ministers on the one hand and top performers on the other suggests that the status of heir apparent somewhat limits the beneficial effects of these features. Put differently, winning the office as a challenger from the opposition would appear to mark another important prerequisite for a truly great prime ministerial performance. Indeed, all top-ranked post-war prime ministers from all four countries won the premiership in the aftermath of an electoral victory by their respective parties.

Conclusions

As the analysis above suggests, patterns of advancement matter in terms of prime ministerial performance. Overall, heir apparent prime ministers are significantly more likely to fail and disappoint, even though their extended experience as well as public prominence and recognition on assuming office would appear to provide good reason to expect otherwise. Moreover, the analysis has identified more specific elements in the track record of individual incumbents which provide important clues as to why some prime ministers were more successful than others.

While this article has consciously focused on a particular set of factors, it is important to note that this obviously implies a strong ceteris paribus clause. A first major factor not accounted for here is the personality traits of different candidates: in fact, some people may bring all the talents required to perform exceptionally well as second-tier players, while being unable to maintain that performance when placed at the top level, whatever their professional experience and expertise may be (on studying leaders’ personality traits, see e.g. Hermann Reference Hermann2003). The other factor, or rather set of factors, consciously ignored in this study is political context broadly defined. In reality, the political and economic state of affairs, the presence or absence of crisis situations, as well as the strengths or weaknesses of potential challengers constitute potentially powerful factors likely to shape the political fortunes of a prime minister (see ‘t Hart Reference ‘t Hart2014), which would all have to be on the agenda of a more extensive comparative assessment of prime ministerial performance. Indeed, studying context is indispensable for a deeper understanding, and perhaps fairer assessments, of leaders and their performance; yet, ultimately, it is not context that makes or breaks leaders. Comparative research on leaders’ styles and performance in context suggests that low-opportunity leaders by no means necessarily perform less successfully than high-opportunity leaders, nor are they necessarily perceived as less effective than their better situated counterparts (see Genovese et al. Reference Genovese, Belt and Lammers2014; Theakston Reference Theakston2002).Footnote 11

In addition to studying more closely the personality traits of prime ministers and the particular context of their premierships, it would be intriguing to learn more about the psychology of heirs apparent (see also Bynander and ‘t Hart Reference Bynander and ‘t Hart2006). Does the status of heir apparent create a latent ‘office fatigue’ among incoming candidates right from the start, or possibly even before the start? Is there perhaps a certain amount of mental exhaustion at work that falls upon an incumbent who may not only have been tired of waiting but also tired from waiting? Is the political attention of heir apparent prime ministers distracted, and their judgement impaired, as they suffer from a fixation to do things differently from their immediate predecessor at virtually any cost? There is a more general inclination of incoming prime ministers to go out of their way to break free from their immediate predecessor’s legacy and style (Helms Reference Helms2016: 296–297), but this may create particularly difficult challenges and delicate choices for heirs apparent in the office of prime minister.

Further, it would be important to know more about the difference that major strategic choices of incoming prime ministers make in terms of their perceived performance. Related research on incoming party leaders suggests that – notwithstanding the actual facts of a succession – there is considerable room for ‘framing’ a succession with much leeway in terms of image-building and public promotion of a candidate in the competitive arena (Bynander and ‘t Hart Reference Bynander and ‘t Hart2008: 402). What seems clear, though, is that heirs apparent in the office of prime minister are more likely than ‘normal’ prime ministers to find themselves confronted with deeply conflicting public expectations about continuity and change, even among their supporters. This may all too easily turn into a major source of disappointment and disenchantment, which may indeed be a crucial part of the story behind the strange political failure of many heir apparent prime ministers. After all, there can be little doubt that perceptions of prime ministerial performance tend to be strongly shaped by expectations (see Helms Reference Helms2017: 12).

Finally, there would appear to be several important implications of the key findings of this article. To the extent that the suspected causal relationship between career patterns and performance in office can be substantiated by future comparative research, such insights could become the basis for revisiting, and revising, the written and unwritten rules of political recruitment to top leadership positions. For example, to the extent that there is a proven causal relationship between extended tenures as a departmental minister and later disappointment and failure in the office of prime minister, it might be worthwhile to consider the likely effects of term limits for possibly ‘over-experienced’ and ‘exhausted’ contenders for the premiership. Moreover, if extremely extended premierships are actually problematic not only for an incumbent and his or her government, but also for his or her successor, and the country more generally, as the findings above suggest, then measures addressing this problem would seem well worth considering, too.Footnote 12

Acknowledgements

The author would like to thank the three anonymous referees of this journal for their generous and most valuable suggestions. A previous draft of this paper has been presented at the third PUPOL international conference, ‘Leadership in a Changing Environment’, held at the Swedish Defence University in Stockholm, 19–20 April 2018. I am grateful to all participants in the workshop on ‘Leaders and their Context – How Parties and their Selection Processes Shape Leadership’, in particular to my discussant, Paul Webb. Any remaining errors are mine.

Footnotes

1 A brief note on the Westminster-type of parliamentary democracy is in order. In the wider comparative politics literature, such as and in particular with Arend Lijphart (Reference Lijphart1984, Reference Lijphart2012), ‘Westminster democracies’ are prominently presented as a synonym to ‘majoritarian democracies’, a distinct set of institutions and agency-related patterns, conceptualized as the direct opposite to ‘consociational democracies’ (for a recent reappraisal, see Bogaards Reference Bogaards2017). To be sure, in real-world politics no less so than in Comparative Politics as an academic discipline, notions of Westminster democracy have tended to be diffuse and contested. In a rigorous comparative assessment, Graham Wilson (Reference Wilson1994: 193) identified ‘the unity between the legislature and executive secured through a disciplined political party’ as the only basic feature of the Westminster model to be found across the family of Westminster democracies. Notwithstanding countless minor and several major institutional differences (in particular at the level of territorial politics), the four Westminster democracies studied here are marked by a set of institutions that have given rise to more or less powerful executives (towards parliament and beyond), prime ministers that tend to be dominant players (while operating within a system of collective ministerial responsibility), and a notably prominent role of political parties at all stages of the governing process. Indeed, it is the inseparable link between party leadership and premiership that continues to distinguish Westminster democracies even from several other parliamentary democracies with a powerful chief executive, such as Germany, where winning the chancellorship is not dependent on holding the party leadership, and some chancellors were strikingly aloof from their own party (see Helms Reference Helms2002).

2 Often, the nomination and promotion of a candidate by the prime minister and his or her particular status in the media go hand-in-hand. However, explicit prime ministerial endorsement of a candidate should be considered no more than a secondary indicator, and whenever the incumbent’s personal preferences and the party’s and media’s assessments of a candidate are at odds, preference should be given to the latter when it comes to classifying that candidate as an heir apparent.

3 Of course, in political reality the fates of heirs and usurpers may be similar, as a closer inspection of Australian prime ministers through 2010 (see Strangio Reference Strangio, ‘t Hart and Walter2013) and the perceived performance of several more recent usurpers (Julia Gillard, Kevin Rudd and Malcom Turnbull) would appear to suggest. Moreover, there are admittedly borderline cases. For example, Hawke and Keating, Australia’s 23rd and 24th prime ministers, obviously had an understanding that Keating was to be the next leader, until Hawke decided he did not want to leave the prime ministership. This notwithstanding, it would seem reasonable to keep conceptual and empirical issues apart.

4 The calculations above include all prime ministers that have been appointed or reappointed since 1945. On that count, the earliest prime ministers to be included are Attlee (UK), King (CA), Forde (AUS) and Fraser (NZ). All returning prime ministers – Holyoake (NZ), Rudd (AUS), Trudeau (CA) and Wilson (UK) – were counted twice.

5 It seems important to note that the status of incumbents – elected, takeover or heir apparent prime ministers – has conspicuously been ignored in those expert surveys, which makes revisiting the findings all the more intriguing.

6 For example, even the most carefully arranged survey may suffer from ‘presentism’, uneven levels of historical knowledge as well as the mediating and filtering role of opinion-makers and the mass media regarding the performance and achievements of individual leaders (see Theakston Reference Theakston2013: 223).

7 It is to be noted that occasional work looking more specifically into the performance of the larger group of intra-party successors in the office of British prime minister has arrived at considerably less favourable assessments (see Worthy Reference Worthy2016); however, apart from major conceptual differences, this research has focused purely on the UK.

8 Similar evidence can be found beyond Westminster: for instance, Austria’s Chancellor Franz Vranitzky (1986–97) is widely seen as an incomparably more successful political leader than his immediate predecessor, Fred Sinowatz (1983–6), himself a Social Democratic takeover and heir apparent chancellor.

9 As no heir apparent prime minister from the UK and New Zealand has been considered to belong to the best 50% of the holders of their country’s post-war premiership, the sample of best-ranked performers presented here includes only Canadian and Australian prime ministers.

10 If one accounts for Winston Churchill’s war-time years as prime minister, and Eden’s unchallenged status as his ‘natural successor’ to be firmly established even before Churchill’s return to No. 10 in 1951, Eden could also be classified as following in the steps of a long-term prime minister, which would further underscore the importance of this nexus.

11 This would appear to mark a more general phenomenon cutting across different types of government, including presidential systems: for example, George W. Bush’s status as high-opportunity president did not save him from being considered by many as one of the worst presidents in recent American political history. The opposite is true, for example, for Bill Clinton, a low-opportunity president who continues to receive average and above-average assessments in many expert surveys (see Genovese et al. Reference Genovese, Belt and Lammers2014).

12 Indeed, few observers have considered the final term of a long-term prime minister to be particularly impressive, and usually the gradual decline of a prime minister is felt strongly, especially when he or she has once been perceived as exceptionally capable and successful.

References

Azzi, S and Hillmer, N (2016) Ranking Canada’s Best and Worst Prime Ministers. Maclean’s, 7 October 2016, www.macleans.ca/politics/ottawa/ranking-canadas-best-and-worst-prime-ministers.Google Scholar
Bennister, M (2012) Prime Ministers in Power: Political Leadership in Britain and Australia. Basingstoke: Palgrave.10.1057/9780230378445CrossRefGoogle Scholar
Bennister, M and Heppell, T (2016) Comparing the Dynamics of Party Leadership Survival in Britain and Australia: Brown, Rudd and Gillard. Government and Opposition: An International Journal of Comparative Politics 51(1), 134159.Google Scholar
Bernstein, S and Rioux, J-P (2000) The Pompidou Years, 1969–74. Cambridge: Cambridge University Press.Google Scholar
Bogaards, M (2017) ‘Comparative Political Regimes: Consensus and Majoritarian Democracy. In Thompson W (ed.), The Oxford Encyclopedia of Politics, published online 1 March 2017, doi: 10.1093/acrefore/9780190228637.013.6, pp. 131.Google Scholar
Borchert, J (2003) Professional Politicians: Towards a Comparative Perspective. In Borchert J and Zeiss J (eds), The Political Class in Advanced Democracies. Oxford: Oxford University Press, pp. 125.10.1093/0199260362.001.0001CrossRefGoogle Scholar
Bose, M and Landis, M (eds) (2003) The Uses and Abuses of Presidential Ratings. New York: Nova Science Publishers.Google Scholar
Brownlee, J (2007) Hereditary Succession in Modern Autocracies. World Politics 59(4), 595628.10.1353/wp.2008.0002CrossRefGoogle Scholar
Bynander, F and ‘t Hart, P (2006) When Power Changes Hands: The Political Psychology of Leadership Succession in Democracies. Political Psychology 27(5), 707730.10.1111/j.1467-9221.2006.00529.xCrossRefGoogle Scholar
Bynander, F and ‘t Hart, P (2008) The Art of Handing Over: (Mis)Managing Party Leadership Successions. Government and Opposition: An International Journal of Comparative Politics 43(3), 385404.10.1111/j.1477-7053.2008.00259.xCrossRefGoogle Scholar
Campbell, C (2001) The Vicious Circle of the Constricting State as Viewed through the Failure of Executive Leadership in Anglo-American Democracies. In Flynn P and Hodgkinson VA (eds), Measuring the Impact of the Nonprofit Sector. New York: Springer, pp. 81101.Google Scholar
Carroll, B (2004) Australia’s Prime Ministers: From Barton to Howard, Kenthurst: Rosenberg.Google Scholar
Cordell, J (2005) Unity and Plurality in the French Right. South European Society and Politics 10(2), 191206.10.1080/13608740500134861CrossRefGoogle Scholar
Cross, WP and Blais, A (2012) Politics and the Centre: The Selection and Removal of Party Leaders in the Anglo Parliamentary Democracies. Oxford: Oxford University Press.10.1093/acprof:oso/9780199596720.001.0001CrossRefGoogle Scholar
Cross, WP and Pilet, J-B (eds) (2016)The Politics of Party Leadership: A Cross-National Perspective. Oxford: Oxford University Press.Google Scholar
De Ferrari, I (2015) The Successor Factor: Electoral Accountability in Presidential Democracies. Comparative Political Studies 48(2), 193230.10.1177/0010414014546442CrossRefGoogle Scholar
Esselment, AL, Lees-Marshment, J and Marland, A (2014) The Nature of Political Advising to Prime Ministers in Australia, Canada, New Zealand and the UK. Commonwealth and Comparative Politics 52(3), 358375.10.1080/14662043.2014.919731CrossRefGoogle Scholar
Finer, SE (1997)The History of Government from the Earliest Times, 3 vols. Oxford: Oxford University Press.Google Scholar
Genovese, MA, Belt, TL and Lammers, WW (2014) The Presidency and Domestic Policy: Comparing Leadership Styles, FDR to Obama, 2nd edn.New York: Routledge.Google Scholar
Gordon, WL (1977) A Political Memoir. Toronto: McClelland and Stewart.Google Scholar
Green, J and Jennings, W (2017) The Politics of Competence: Parties, Public Opinion and Voters. Cambridge: Cambridge University Press.10.1017/9781316662557CrossRefGoogle Scholar
Grotz, F and Müller-Rommel, F (2015) Schwache Regierungschefs? Politische Erfahrung und Amtsdauer von Premierministern in Mittel- und Osteuropa. Zeitschrift für Parlamentsfragen 46(2), 310327.10.5771/0340-1758-2015-2-310CrossRefGoogle Scholar
Helms, L (2002) ‘Chief Executives’ and their Parties: The Case of Germany. German Politics 11, 145164.10.1080/714001286CrossRefGoogle Scholar
Helms, L (2016) The Politics of Leadership Capital in Compound Democracies: Inferences from the German Case. European Political Science Review 8, 285310.10.1017/S1755773915000016CrossRefGoogle Scholar
Helms, L (2017) When Less is More: ‘Negative Resources’ and the Performance of Presidents and Prime Ministers. Politics, published early online, November, https://doi.org/10.1177/0263395717738964.CrossRefGoogle Scholar
Helms, L (Forthcoming) Performance and Evaluation. In Andeweg R et al. (eds), The Oxford Handbook of Political Executives. Oxford: Oxford University Press.Google Scholar
Hermann, MG (2003) Assessing Leadership Style: Trait Analysis. In Post JM (ed.), The Psychological Assessment of Political Leaders. Ann Arbor: University of Michigan Press, pp. 178212.Google Scholar
Hillmer, N and Azzi, S (2011) Canada’s Best Prime Ministers. Maclean’s, 10 June 2011, www.macleans.ca/news/canada/canadas-best-prime-ministers.Google Scholar
Hillmer, N and Granatstein, JL (1997) Historians Rank the Best and the Worst Canadian Prime Ministers. Maclean’s, 21 April, pp. 3439.Google Scholar
Horiuchi, Y, Laing, M and ‘t Hart, P (2015) Hard Acts to Follow: Predecessor Effects on Party Leader Survival. Party Politics 21(3), 357366.10.1177/1354068812472577CrossRefGoogle Scholar
Hughes, CA (1976) Mr Prime Minister: Australian Prime Ministers, 1901–1972. Melbourne: Oxford University Press.Google Scholar
Jalalzai, F (2008) Women Rule: Shattering the Executive Glass Ceiling. Politics and Gender 4, 205231.Google Scholar
Johansson, J and Levine, S (2013) Evaluating Prime-Ministerial Performance: The New Zealand Experience. In Strangio P, ‘t Hart P and Walter J (eds), Understanding Prime-Ministerial Performance: Comparative Perspectives. Oxford: Oxford University Press, pp. 291317.10.1093/acprof:oso/9780199666423.003.0013CrossRefGoogle Scholar
Jones, CO (2008) Vice Presidents and Other Heirs Apparent: The Historical Experience of Experience. Presidential Studies Quarterly 38(3), 422432.10.1111/j.1741-5705.2008.02653.xCrossRefGoogle Scholar
Kincade, VR (2000) Heirs Apparent: Solving the Vice Presidential Dilemma. Westport, CT: Praeger.Google Scholar
Langston, J (2006) The Birth and Transformation of the Dedazo in Mexico. In Helmke G and Levitsky S (eds), Informal Institutions and Democracy. Baltimore: Johns Hopkins University Press, pp. 143159.Google Scholar
Lijphart, A (1984) Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries. New Haven: Yale University Press.10.2307/j.ctt1ww3w2tCrossRefGoogle Scholar
Lijphart, A (2012) Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries, 2nd edn.New Haven: Yale University Press.Google Scholar
Osborne, S (2017) Theresa May’s Leadership More Popular than Margaret Thatcher or Tony Blair, Poll Reveals, Independent, 26 April, www.independent.co.uk/news/uk/politics/theresa-may-more-popular-margaret-thatcher-tony-blair-leadership-uk-prime-minister-polls-latest-a7703591.html.Google Scholar
Pakulski, J and Tranter, B (2015) The Decline of Political Leadership in Australia? Changing Recruitment and Careers of Federal Politicians. Basingstoke: Palgrave Macmillan.Google Scholar
Rhodes, RAW and ‘t Hart, P (eds) (2014) Puzzles of Political Leadership. In ‘t Hart, P and Rhodes, RAW (eds), The Oxford Handbook of Political Leadership. Oxford: Oxford University Press, pp. 121.Google Scholar
Rhodes, RAW, Wanna, J and Weller, P (2009) Comparing Westminster. Oxford: Oxford University Press.CrossRefGoogle Scholar
Rubinstein, WD (2003) Twentieth-Century Britain: A Political History. Basingstoke: Palgrave.CrossRefGoogle Scholar
Sands, C (2002) Escaping the Leadership Muskeg. Canada Focus 3(2), 19.Google Scholar
Shaw, R and Eichbaum, C (2014) Ministers, Minders and the Core Executive: Why Ministers Appoint Political Advisers in Westminster Contexts. Parliamentary Affairs 67(3), 584616.CrossRefGoogle Scholar
Strangio, P (2013) Evaluating Prime-Ministerial Performance: The Australian Experience. In Strangio P, ‘t Hart P and Walter J (eds), Understanding Prime-Ministerial Performance: Comparative Perspectives. Oxford: Oxford University Press, pp. 264290.CrossRefGoogle Scholar
Strangio, P, ‘t Hart, P and Walter, J (eds) (2013) Understanding Prime-Ministerial Performance: Comparative Perspectives, Oxford: Oxford University Press.CrossRefGoogle Scholar
Theakston, K (2002) Political Skills and Context in Prime Ministerial Leadership in Britain. Politics and Policy 30, 283323.CrossRefGoogle Scholar
Theakston, K (2013) Evaluating Prime-Ministerial Performance: The British Experience. In Strangio P, ‘t Hart P and Walter J (eds), Understanding Prime-Ministerial Performance: Comparative Perspectives. Oxford: Oxford University Press, pp. 221241.Google Scholar
Theakston, K and Gill, B (2011) The Post-War Premiership League. Political Quarterly 82(1), 6780.CrossRefGoogle Scholar
‘t Hart, P (2014) Contextual Analysis. In ‘t Hart P and Rhodes RAW (eds), The Oxford Handbook of Political Leadership. Oxford: Oxford University Press, pp. 210224.Google Scholar
‘t Hart, P and Uhr, J (eds) (2011) How Power Changes Hands: Transition and Succession in Government. Basingstoke: Palgrave.CrossRefGoogle Scholar
‘t Hart, P and Schelfhout, D (2016) Assessing Prime-Ministerial Performance in a Multiparty Democracy: The Dutch Case. Acta Politica 51(2), 153172.CrossRefGoogle Scholar
Weller, P (1985) First Among Equals: Prime Ministers in Westminster Systems. London: Allen & Unwin.Google Scholar
Weller, P (2012) Fight, Flee, or Fulminate: Prime Ministerial Challengers, Strategic Choices, Rites of Succession. Political Quarterly 83(1), 152164.CrossRefGoogle Scholar
Weller, P (2014) The Variability of Prime Ministers. In Rhodes RAW and ‘t Hart P (eds), The Oxford Handbook of Political Leadership. Oxford: Oxford University Press, pp. 489502.Google Scholar
Wilson, G (1994) The Westminster Model in Comparative Perspective. In Budge I and Kay DM (eds), Developing Democracy: Comparative Research in Honour of J.F.P. Blondel. London: Routledge, pp. 189201.Google Scholar
Whitney, CR (1991) Gray Flannels or Not, John Major Is Riding the Crest of Popularity in Britain, New York Times, 28 February, www.nytimes.com/1991/02/28/world/gray-flannels-or-not-john-major-is-riding-the-crest-of-popularity-in-britain.html.Google Scholar
Worthy, B (2016) Ending in Failure? The Performance of ‘Takeover’ Prime Ministers 1916–2016. Political Quarterly 87(4), 509517.CrossRefGoogle Scholar
Figure 0

Table 1. Intra-Party Successors to the Prime Minister and Heirs Apparent in Britain, Canada, Australia and New Zealand (1945–2016)

Figure 1

Table 2. Best- and Worst-Ranked Heir Apparent Prime Ministers, and Top-Ranked Prime Ministers Compared