Ideas in Democracy: Co-Leadership, Multiple Heads or Multiple Headaches?
Around the world, several political parties use a co-leadership model to govern themselves. The practice goes back thousands of years, but today it faces old and new challenges alike.
This is the first piece in an open-ended series about ideas that shape democracy.
I hadn’t thought about co-leadership much when I learned Canadian Green Party leadership hopefuls Anna Keenan and Chad Walcott running to share the job. I was aware that the model was practiced here and there. I knew the Romans had done it but I couldn’t have told you how. In fact, when thinking about a power-split in history, the Roman Empire came to mind, with all its decayed glory and the structural incompetence and malevolence that accompanies decline. I didn’t think about the much more successful experience of the Roman Republic. Perhaps because we are conditioned to think of leadership in politics as a solo affair—particularly for those at the top—I dismissed the idea. Then I spoke with Keenan and Walcott and asked them about how the co-leadership model works. After that, I wanted to learn more.
When Canada’s Green Party launched its leadership contest, it was coming off a disastrous election, years of decline, and a leader, Annamie Paul, who’d been run out of the party. Beset by internal fighting, resignations, incoherent messaging, and a suite of problems too long to catalogue here, the Greens’ search for a new leader risked becoming an utter catastrophe. In some ways, it has been. In September, a struggle over the race erupted as the party’s federal council discussed suspending the contest while the Greens grappled with the misgendering of interim leader Amita Kuttner during a Zoom event launching the leadership race.
As Kuttner wrote in a statement, “this incident is reflective of a larger pattern of behaviours that a few in the party are perpetuating. Over the years the party has documented reports which indicate a systemic issue disproportionately affecting Black, Indigenous, and racialized people and 2SLGBTQIA+ people…” By the end of the month, and after a handful of resignations from the leadership organizing committee, the party scrapped a second round of online voting and moved to a single round over the course of a week in November.
Beyond the episodic and structural issues of the party, however, the leadership race is introducing people to a fascinating idea in democracy—and one with a long history. Anna Keenan and Chad Walcott are running for the leadership as co-leaders. So too are former party leader Elizabeth May and co-candidate Jonathan Pedneault. And even though the notion seems strange to some observers—How could that possibly work?—reading beyond the headline and setting aside one’s initial scepticism is worth the work.
What is co-leadership?
Co-leadership isn’t common around the world, but it’s probably more common than you think. In 2020, Politico reported that parties were adopting co-leadership models that divided power along gender lines. Zei Weise noted that while it’s long been a favoured practice for Green parties globally—in New Zealand, Germany, Belgium, for instance—the German Social Democrats adopted the model as have other parties in the Bundestag and the European Parliament.
The co-leadership model divides authority between two equal partners. As mentioned, it’s not a new practice. In “Co-Leadership: Lessons from Republican Rome,” Behavioural economist David Sally traces the shared power structures of the ancient Roman polity. In the Republic, Sally notes, “the pre-eminent social conflict occurred between the orders—the patricians (nobles) and the plebians (common people).” In short, Rome grappled with class conflict. Over time co-leadership evolved to allow (wealthy) plebians a chance to govern and introduced a kind of balance. Co-leadership existed across offices in the republic, also including dual censors, praetors, aediles, and questors, meaning the polity sought a broader political balance than you might find in single-holder offices.
Sally outlines a “Roman way” of co-leadership that includes “a number of rules, norms, structures, and behaviors.” One rule was co-leaders had to serve together at the same time—entering and leaving office in tandem and maintaining symmetry such that if one died, no new co-consul was appointed, so as to avoid an imbalance in experience. Moreover, no co-consul was permitted to permanently serve as the solo leader.
The co-leadership of the consuls was also one office. While two individuals held authority, the office bound them together as one, sharing an emblem, with each having a formal veto power over any and all decisions. In public, the co-consuls were to maintain solidarity at all times to preserve the integrity of the office and the model.
Sally notes that co-leadership was “part of a system of general power sharing” spread across the republic. “The downfall of the Republic,” Sally writes, “occurred as inordinate power accrued to the consuls and to dictators.” The broader system in which leadership is shared was, and is, important.
A co-leadership model in contemporary politics centres around two political party leaders who share power, cooperating and making decisions together based on consensus. Each is likely to hold veto power, thus checking the power of the other and of the party apparatus. Like republican Rome, the animating idea behind co-leadership is that consensus, checks, balances, and veto points produce better outcomes and limit the vulnerabilities, aporia, egos, and deficiencies that come with the concentration of power into the hands of one person. In some instances, as in the case of certain Green parties and provincial party Québec Solidaire, the party leaders serve as co-spokespersons rather than traditional top-down leaders.
Are two heads better than one?
Today, co-leaders of political parties yield substantive and strategic benefits to their side as they challenge the ego-driven attempts at frictionless government from the centre. In her study of co-leaders and the Green Party of New Zealand, Eva Hartshorn-Sanders found that being in two places at once allowed Green co-leaders Rod Donald and Jeanette Fitzsimmons to cover more ground and, at least during campaigns, draw greater media coverage. The party’s model also generated more capacity with two leaders, particularly insofar as the two held complementary skillsets. As in republican Rome, the dual model produced checks and balances and a “frank” and equal check-in between leaders who could help keep one another from veering off course. The equality aspect is important, as Hartshorn-Sanders emphasizes. Co-leaders aren’t staffers. They aren’t advisors. They are equal participants in governance and their interactions are conditioned by the expectation of equality.
Alongside its merits, a co-leadership model has in-built challenges. Veto points and friction can produce better decisions, slowing down processes to encourage reflection, but that also produces…veto points and friction. This approach can be slower and can stall decisions. It can also produce unclear or incoherent outcomes if the co-leaders don’t work and communicate well together, internally and externally.
The co-leadership model assumes, or rather requires, a doubling of high-level talent: excellent communicators who can work extraordinarily well in a team. Such people can be hard to find. At the same time, as Hartshorn-Sanders finds, there’s a risk that staff will try to play co-leaders off one another, potentially creating a split between the two. And while two leaders can bring double sets of networks, capacities, areas of expertise, and skills, they can also duplicate those same goods.
Perhaps the biggest challenge to co-leadership is structural. Parties and legislatures often aren’t institutionally designed for the setup and, indeed, may not recognize two heads. This is a problem noted time and time again by proponents and opponents of the model. Making it work requires either institutional amendment (or a constitutional change) or workarounds—such as appointing one person a deputy leader. But such kluge arrangement may undermine the equality required for co-leadership to work. Nonetheless, in the case of Canadian Green Party hopefuls Keenan and Walcott, each has pledged to name the other as a co-spokesperson (the Green’s bottom-up iteration of leader) if they win.
The broader culture in which we operate is also a problem for co-leadership models. Our politics tend not to rest on consensus. Ego often drives political life. A 24-hour news cycle and the hunt for wedges, gaffes, and gotcha moments by some in the media risks sundering even the closest co-leaders. Substantive debate, disagreement, and humility can be seen as a weakness and weaponized against those who want to take their time and work out together what ought to be done. Changing the political culture in which we operate is therefore key to making co-leadership work—and that is a tremendous challenge. That said, having co-leaders model more productive leaderships might be just what we need to instigate that shift.
Season 6, Episode Two: “The Meeting.”
The American sitcom The Office featured a story line in which Michael Scott and Jim Halpert become co-managers. In the episode “The Meeting,” the show skewers co-leadership as Michael resents sharing power and tries to “micro co-manage” Jim. Oscar, one of the office’s accountants, assesses the situation, saying “Look, it doesn’t take a genius to know that any organization thrives when it has two leaders. Go ahead and name a country that doesn’t have two presidents. A boat that sets sail without two captains. Where would Catholicism be without the popes?” (Indeed, the Catholic Church does have a history of dual—and dueling popes—but that’s not quite the same thing.) But the joke is as much about the dysfunction of the office, the state of contemporary business practices, and the peccadilloes of Michael Scott as it is about co-leadership. The episode hammers home the point that structure and personalities matter. But structures change. Personalities come and go.
While a co-leadership model is full of challenges and easy to mock or dismiss at first glance, it shouldn’t be ignored. One’s knee-jerk reaction to the approach may be driven by life in a toxic political culture that begs for structural change, slower politics, greater consensus, more checkpoints, and a shift to a more engaged, thoughtful, and productive working out of disagreement. Sharing power isn’t easy. But it’s been done—successfully. It might be done once more.
I wanted to get a bit more into that but I didn't quite have the time. And thank you!
I guess delving into Quebec Solidaire would have derailed this point. I thought they had co-leaders but looking it, they have co-spokespeople with leadership being more fuzzy.
As always, I like the examination of structures in a fun way that's a hallmark of your writing.