Boards buffer organizations during crises, but only when they maintain leadership continuity.
By Tim Glowa
In September 2023, Dr. Michael Strong’s term as president of the Canadian Institutes of Health Research came to an end. His office stayed empty for fifteen months. Two interim leaders cycled through. A successor was announced in November 2024 and did not start until January 2025.
Between October and December 2024, while the chair at the top was still effectively empty, federal employees filled out the Public Service Employee Survey. The CIHR results were striking. Ratings of immediate supervisors held nearly flat, within a point or two of where they had been two years earlier. Confidence in day-to-day teammates and team operations: unchanged. But ratings of senior leadership collapsed by 17.8 points on a 100-point scale, across all six leadership questions in the survey. The drop was nearly three times the size of the broader organizational decline.
That asymmetry, where the leadership questions move dramatically while the supervisor questions stay flat, is the empirical signature of a leadership vacuum. And it is the most vivid finding in a four-year analysis that suggests boards do something specific when the weather turns.
WHAT THE DATA SHOW
The analysis covers 13 federal organizations across four cycles of Canada’s Public Service Employee Survey, 2019 through 2024. Six are governed by boards or governing councils that sit between the executive and the responsible minister. Seven are minister-direct departments matched on mandate and workforce type.
Before COVID, the two groups produced identical results. In 2019 and 2020, employees in board-governed organizations rated senior leadership the same as employees in minister-direct departments. The pandemic hit, and a four-point gap opened. Minister-direct departments lost about 12 points between 2020 and 2024. Board-governed organizations lost about 8. The gap appeared in 2022 and held in 2024, across all six leadership items, and survived every robustness test the analysis could throw at it.
Board governance buffered employee confidence in senior leadership when stress arrived. The buffer was invisible in calm conditions and decisive under pressure.
WHY THE BUFFER HELD IN SOME PLACES AND COLLAPSED IN OTHERS
Within the board-governed group, the variation tells the rest of the story. The Natural Sciences and Engineering Research Council and the Social Sciences and Humanities Research Council kept their executives in place through the pandemic. Both produced the strongest resilience numbers in the sample. CIHR, with the same governance architecture, the same statutory independence, the same insulation from direct ministerial control, lost its president and produced the worst.
Same form. Same insulation. Different outcome. The variable that moved was leadership continuity.
The CIHR vacancy was not a surprise departure. The outgoing president retired at the end of a defined term, which means the governance system had months of advance notice and likely had years. Despite that runway, the position sat empty for fifteen months. This is the harder version of the lesson. Surprise vacancies happen, and even disciplined boards face transition gaps when a CEO leaves unexpectedly. Planned retirements are different. When a board knows the end date and the chair still sits empty for over a year, that is not process failure under stress. That is the absence of process during exactly the period when succession planning was supposed to happen.
WHY CORPORATE BOARDS SHOULD CARE
The federal public service is not a corporation. The principal is the public, not shareholders. There is no stock price to discipline the board, no exit market to consult, no quarterly earnings call. But the structural anatomy is familiar. Boards approve budgets, oversee senior leadership, and absorb risk between the operating organization and the political principal above. Corporate boards do the same work, with capital markets and activist investors playing the role that ministers and Parliament play in Ottawa.
The corporate version of the buffering test arrives through scandals, regulatory shocks, supply-chain breakdowns, hostile takeover attempts, activist campaigns, and CEO transitions that go sideways. The structure that buffers a federal research council during a pandemic is the structure that buffers a Fortune 500 company during an FDA recall, an SEC investigation, or an unplanned CEO departure. The mechanism in both settings is the same: the board’s capacity to maintain leadership continuity through the stress.
That is a higher bar than most boards meet. It is also testable in advance. Three practices distinguish the boards positioned to buffer their organizations from the boards positioned to fail when the next crisis arrives.
THREE THINGS BOARDS CAN DO BEFORE THE NEXT CRISIS
1. Build succession depth. Every board should be able to name three internal candidates ready to step into the CEO role on ninety days notice. Not in theory. In practice, with the bench you have, the development program you have funded, and the willingness to act on either. If the answer is not three names, the work to develop them starts at the next board meeting. Boards that wait for a vacancy to begin succession work discover they cannot manufacture a bench during the vacancy.
2. Engineer institutional memory. Term limits and director refreshment policies should stagger departures so institutional knowledge survives any single transition. A board that turns over alongside the CEO has lost its capacity to stabilize the organization at exactly the moment that capacity is most needed. Stagger the rotation so that experienced directors anchor the room when new ones arrive.
3. Build relationships beyond the CEO. The board’s read on the organization cannot run only through the chief executive. Structured exposure to the executive layer below the CEO is how directors learn who can step up, where the gaps are, and how to maintain momentum through a transition. This is not a courtesy meeting at the annual offsite. It is the mechanism through which the buffering function actually operates.
THE BOTTOM LINE FOR BOARDS
The next crisis will not look like the last one. Pandemic, regulatory upheaval, technology disruption, geopolitical shock, financial distress, leadership malfeasance. The categories rotate. The pressure does not.
Board governance buffered employee confidence in senior leadership through the most significant external stress event in a generation. The buffer held across thirteen organizations, six leadership questions, and two post-pandemic survey cycles. Where boards maintained leadership continuity, the buffer held. Where boards failed to maintain it, the buffer collapsed alongside the empty chair.
The work is invisible in calm conditions and decisive under pressure. The boards that do the work before stress arrives have a buffer. The boards that wait discover they cannot build the capability during the storm. The empty chair at CIHR is the cautionary case. The 17-point collapse is the price of letting succession fail. The buffer is real. It does not maintain itself.
Based on a working paper by the author: Glowa, T. (2026). “Invisible Until It Rains: Board Governance and Leadership Resilience in the Canadian Federal Public Service.” Working paper, Haskayne School of Business, University of Calgary.
Tim Glowa is a DBA student at the Haskayne School of Business, University of Calgary, and the author of “Smart Board Governance for the AI Revolution.” He is the founder of CleverTrout consulting and Smart Board Governance.