Keeping Party Leaders Honest

Canadians do it differently from most — and quite possibly not as well.

Not a leader? Who looks like a leader? Where is the leadership today? Who do you know in the leader’s office?

Today leadership is what Canadian political culture amounts to; the rest is window dressing. So what could be more useful than a tough, data-driven analysis of how Canadian political leadership systems stand up to international benchmarks? Politics at the Centre: The Selection and Removal of Party Leaders in the Anglo Parliamentary World looks like the right book at the right time.

Its authors, both Canadians, are veteran scholars with impeccable credentials in political science: William Cross at Carleton University, André Blais at the Université de Montréal. As we might expect, Cross and Blais deliver lots of data. They compare leadership procedures in Australia, Canada, Ireland, New Zealand and the United Kingdom—a selection that may evoke the old idea of the “white commonwealth,” as if parliamentary democracy only functions in states that evolved under the rule of British peoples. They do not include data on Commonwealth states such as India or Jamaica, or countries such as Japan, Spain or some of the new parliamentary democracies of Eastern Europe.

Still, even with their “anglo” sample, they have taken on no small challenge. For just five countries between 1965 and 2008, they had to chart and interpret more than 200 leadership changes in 22 different political parties. For every leadership career ended, for instance, Cross and Blais carefully parse the differences between “resigned voluntarily,” “resigned under pressure” and, most ominously, “removed.” If you seek to compare the rise and fall of our own Dions, Campbells, Clarks and Martins against the fates of Bob Hawke of Australia, Jenny Shipley of New Zealand, Margaret Thatcher of Britain or Ireland’s Bertie Ahern (among hundreds of others), the materials are here at hand.

They have the information. Can they also enlighten us as to what it means?

Here Cross and Blais run into what seems an existential problem of political science. Political scientists have developed very sophisticated expertise in gathering, handling and displaying quantitative data on political phenomena. To my non-expert eye, Cross and Blais are exemplars of this exacting discipline. But the other side of the commitment to data often seems to be ambivalence about political thought. Anyone can think about politics, anyone can have a political opinion and political judgement is not easily reduced to quantifiable data. Perhaps as a result, quantitative political science sometimes seems to shy away from subtleties of political thought. To the rest of us, seeking meaning more than data, texts in political science sometimes read like Dylan Thomas’s Christmas gift books, the ones that told him everything about the wasp, except why.

I find that wasp problem acute in Cross and Blais’s study of political leadership. But before taking up the problem of what the data mean, let’s swim a little in Cross and Blais’s river of data. They present leadership events and situations that to a Canadian observer seem not merely eyebrow raising but head-snappingly strange.

Canadians will be familiar with the account here of leadership selection in this country. In Canada leaders have long been chosen by the party at large, either in delegated conventions or by mass membership voting. At the end, a leader is chosen by an electorate that instantly dissolves, leaving the new boss formally accountable to no one. Until the next slow, expensive race develops, probably years in the future, party and leader are effectively fused.

When they turn to the other countries in their international sample, Cross and Blais’s data sets open windows on a different world. Many Canadians know how Britain’s onetime prime minister Margaret Thatcher, at the height of her power, was replaced in a matter of days by her own members of Parliament. Some Canadians may have noted the recent adventures of Julia Gillard, who became Australia’s prime minister in the summer of 2010 by persuading her Labour Party caucus colleagues to remove the incumbent prime minister just before a general election—and then, early in 2012, had to face down her resurgent predecessor, who was still in caucus and cabinet. In these situations, there are no crippling campaign debts, no vote-selling competitions, no months of a leadership vacuum. The verdict is almost instant, the cost practically zero and the accountability constant.

In Canadian discourse, processes like these are invariably dubbed “coups” or “mutinies” and framed as violent, extra legal and lacking in political legitimacy. Describing Gillard’s rise to power, the dean of Canadian political columnists, Jeffrey Simpson, declared that Australia’s system was “coups and knifing” and “elitism,” which would be acceptable only to those Canadians who wanted “intrigue and blood-on-the-floor politics.”

Cross and Blais, like most Canadian commentators, share Simpson’s disapproval. But their data sets demonstrate that the process Canadians take as an unquestioned norm (ours) has long been alien and bizarre in the rest of the parliamentary world. Every time Cross and Blais compare leadership details across their five countries, they are obliged to distinguish the Canadian way: “everywhere, except Canada,” “Canadian parties were the clear outliers,” “the norm everywhere except in Canada,” “the Canadian … parties are the exception,” “with the exception of the Canadian parties,” “the exceptions here are the Canadian campaigns,” “the situation has been dramatically different in Canada,” and so on. Their data sets reveal the Thatcher-Gillard rules as having been the norm around the world: citizens elect MPs, and their MPs not only influence policies but also hire and fire leaders to implement them. What Canadians call coups, parliamentary democracies around the world have long endorsed as a vital chain of political accountability.

Cross and Blais’s data sets make possible a thought experiment that compares the strengths and weaknesses of two systems that have coexisted in the parliamentary world. In the four other countries under review, party leaders have generally been accountable to MPs who can hire or fire them. MPs are accountable to citizens who can deny them re-election. So when the British Conservative MPs removed Thatcher and when Australian Labour MPs first chose Gillard and then reviewed their choice, they were surely calculating how best to save their own seats. But they were thereby aligning themselves with the will of the citizenry—something the Canadian system hardly allows. Those Australian government MPs have been under daily pressure to make the correct strategic choice about who should lead the country. Nobody calls them nobodies.

Exploring in Cross and Blais’s data sets, I found myself marvelling over what rich material their comparative study provides for fresh thinking about our Canadian situation. We struggle with the “democratic deficit” incarnated in the “friendly dictator,” the leader who forms a government and does as he pleases for years at a time, formally accountable to no one and gleefully firing our elected representatives should they provide anything but fawning deference to him. (I could write “to him or her,” but I have one of Cross and Blais’s head-snapping observations in mind. In their period of study, “not a single full-member [i.e., Canadian-style] vote process has resulted in the selection of a female leader.”1)

Could a different model of accountability give Canadians fresh ways to conceive of political leadership in this country? Sadly, Cross and Blais do not think like that. They are firm adherents of Simpson’s view that any leadership model other than the Canadian one is all coups and knifing. Where elected representatives hold leaders to account, Cross and Blais denounce them as “a small unrepresentative group of elites.” They hold that parliamentary systems in which the people’s elected representatives make leadership choices are remnants of an outdated elitism.

Structuring their data by this dictum, Cross and Blais find a simple progression: from elitism to democracy, from wrong to right. At the start of their study period in 1965, leadership selection in the five states they study rested with MPs (except in Canada, to be sure!). By 2008, they show, there had been movement in some parties in some countries toward the Canadian model, with at least some of them empowering party membership holders over elected MPs. Cross and Blais’s theory, then, is less a comparison of two systems than a simple linear change of which they warmly approve. “While the expansion of the leadership selectorate is far from universal, it is widespread and unidirectional,” they write. “There is no going back to more elite-controlled processes.”

For all their scrupulous care, this is one place where their use of data wobbles a little. Widespread? In their sample of just five countries, they must disregard as “laggards” two (Australia and New Zealand) that have rejected the trend and a third (Ireland) where it has had little purchase in major parties. Unidirectional? Their own tables show that in Britain a mass-party selectorate chose Tony Blair as Labour leader in 1994, but the caucus alone made Gordon Brown leader (and prime minister) in 2007. Cross and Blais see only last stands of elitism, but around the world the debate on principles is alive.

Although their study is international and comparative, Cross and Blais are thoroughly Canadian in their preference for mass party members as the crux of accountability. They are instinctively hostile to any parliamentary system that has not yet moved to the Canadian model. Leadership selections made by the citizens’ elected representatives, they declare, are not just disqualified by elitism, but “plagued with Machiavellian-like tactics.” They reflect “ugly politics,” have “the aura of palace politics” and are “marked by deception and often clandestine-like intrigue” leading to “coronations.” They find no fruitful comparisons in their data, only the good Canadian way and the benighted but gradually reforming habits of quasi-democratic foreigners.

In a single paragraph about the case for MPs’ control of leadership, Cross and Blais acknowledge (and dismiss as elitist) one argument: that MPs, as professional politicians, are “best situated to judge” who will become an effective party leader. Considering Canadian experiences with such mass party choices as Kim Campbell, Stockwell Day, Stéphane Dion and Michael Ignatieff, it is easy to see the merit in that claim. Could MPs possibly do worse? (Actually, Cross and Blais have data on this too; parliamentary leaders have about the same life expectancy no matter what system produces them.)

But despite Cross and Blais, expertise is not the only, or even the principal, reason advanced by thoughtful advocates of making or keeping leaders accountable to MPs who are themselves accountable to citizens.

I claim nothing like the breadth of information about leadership selection that professors Cross and Blais have compiled and organized, but over the years I have interviewed a then-future leader of the British Conservatives, a then-recently deposed New Zealand prime minister and a kingmaker in Irish leadership politics, among others. And in my recollection, an argument to which Cross and Blais seem to have been entirely deaf came up regularly.

Cross and Blais’s tables show that William Hague was chosen as leader of the British Conservative party by the party caucus in 1997 and resigned from the leadership when caucus was about to fire him in 2001. (He accepted his demotion, went back into caucus and is today Britain’s foreign minister.) When I interviewed him, he was still a backbencher, fresh from the excitement of Margaret Thatcher’s removal. His argument then in favour of caucus selection still strikes me as persuasive.

Having the leader elected in parliament strengthens parliament itself. Without the power to change the leader, to elect the prime minister, backbench MPs would have less influence, would have less power over their party leader. All of us who are constituency MPs, trying to represent our constituents and our interests in different parts of the country, know that we are strengthened by having this colossal power at our disposal. In other systems, where party conventions do the choosing of the leader, individual members of Parliament have less influence throughout most of the life of the parliament. And it can very clearly be argued that democracy suffers as a result because the ability of members of Parliament to bring influence to bear is fundamental to democracy.2

To be sure, this is not an idea that polls well in Canada. Cross and Blais stand squarely in the Canadian mainstream in refusing to consider that leaders should be accountable to representatives accountable to citizens. They are entitled to their opinion. But it is discouraging to see such gifted and dedicated students of politics rejecting as inconceivable what seems to have worked quite well in much of the parliamentary world. Why commit so much labour to the gathering of vivid comparisons and then shrink from comparing the real and serious ideas that underpin them? It is as if they measure every parliamentary wasp to the millimetre and then dismiss as deformed all the subspecies that fly differently from our local variety.

Finally, is it at all plausible that accountability as practised in other parliaments could come to Canadian politics?

The old adage that power corrupts and absolute power is even better suggests that any such reform would be ferociously opposed by the parties and by potential leaders (and consultants) enticed by the untrammelled authority the Canadian system promises. Given the still sacred status of unaccountable leadership in Canadian public opinion, it is hard to predict where the change might start or gain traction.

But fundamentally, Canadian party caucuses have always been free to assert authority over their leaders. No significant structural difference separates Canadian practice from that of other parliamentary democracies, just habits and convictions. The constitutional accountability of executive to legislature and leaders to elected representatives is a power long unused by Canadian MPs, but it remains as potent as ever. At bottom MPs need only to grasp they have the authority any time they are prepared to wield it. But who is going to give them the hint?


  1. Kim Campbell? Audrey McLaughlin? Cross and Blais are here distinguishing all-members-may-vote systems from delegated conventions as well as caucus selection, but in general the rule holds. Thatcher, Gillard and New Zealand’s Shipley and Clark: so far women prime ministers who stick have all been caucus choices.  

  2. William Hague, MP, interviewed by Christopher Moore for CBC Radio Ideas, “Leadership Conventions,” broadcast February 1, 1993 (from CBC Ideas Transcripts). The quotation here is slightly condensed from the spoken original.