The Canadian government has an ambitious agenda for change. The translation of bold intent into meaningful impact will depend critically on successful implementation of policies and programs. In our experience, recent governments in Canada have struggled with the mechanics of implementation. Too often, they have failed to achieve their stated policy goals, resulting in delays, increased costs, inefficiencies and public controversy.
We focus on two factors that are responsible for this outcome. First, the federal government has failed to sufficiently build implementation considerations into the original policy decision. Second, current internal processes create unnecessary complexity, delay and friction.
To be clear, this is not due to a lack of effort. Public servants are pursuing implementation with diligence and pace, often driven by considerable impatience from cabinet.
Nor is the question whether the public service can deliver on very challenging implementation issues. The track record of speedy and successful implementation in crisis situations such as the COVID-19 pandemic and the 2008 global financial crisis shows that timely delivery is possible when implementation issues are brought to the fore (Breton et al., 2024; Jean & Jahn, 2021; Lindquist, 2022). In the case of the COVID-19 pandemic experience, some innovations and process reductions were made permanent, some were retained with modifications and others were codified to be triggered for the next crisis.
The issue is that public servants are operating within an unnecessarily complex and risk-averse rule set, and political actors do not have sufficient visibility on implementation issues at the front end of the decision-making process. These factors can result in the selection of options with considerably higher implementation risks.
Lessons on improving implementation outcomes can be drawn from other Westminster countries, provincial governments and the federal government’s experience during crises.Our core perspective is simple: implementation considerations need to be at the heart of all policy decision-making.
Process improvements will, of course, never be a panacea or guarantee success. For political actors, communications considerations often drive the intent, timing and structure of policy decisions. The key is to get alignment around the notion that the announcement is only the beginning of the process and that successful delivery is ultimately what the government will be measured by. The friction between political decision-making and implementation realities imposes substantial direct and opportunity costs, while fuelling frustration among ministers and public servants. Together, these costs make the status quo difficult to defend.
Many policy ideas are shaped outside of internal government processes. Federal election campaigns increasingly feature detailed platform commitments. The commitments often speak not only to what the government proposes to do, but also how it will do so, with what resources and by when. The political actors developing the platform generally have little experience in implementation. They are also not directly involved in the implementation of their proposals. No CEO would announce the launch of a major new product without consulting their engineers and operations managers on how it might best be built and by when, but this is increasingly what is happening with campaign commitments.
In practical terms, policy intent can be thwarted or delayed by factors that the platforms’ authors cannot foresee from their external vantage point. Examples include the limitations of existing computer and information management systems, and legal limitations such as constitutional constraints and binding trade commitments.
The current approach creates considerable frustration for a government when it comes to power and its commitments meet the realities of actual systems. It is not uncommon to hear the phrase, “Had we known that, we would have committed to a different approach.” But governments are often reluctant to backtrack on their public commitments. This means implementation is often achieved at higher cost and greater risk than would have been the case had they had the relevant information in advance.
Canadian political parties currently don’t have the option of consulting with the public service on their draft campaign commitments. While the Parliamentary Budget Office has a mandate that allows political parties to indirectly consult the public service on costing issues, it is not permitted to assess the feasibility of proposals. By contrast, Australia has had guidelines in place since 1976 that allow confidential consultations with the public service on implementation issues associated with potential policies (Coombs, 1976). Their Guidance on Caretaker Conventions, first passed in 1987, allows political parties to filter out non-feasible proposals or change their design before announcing them, so there is a better prospect for successful implementation (Department of the Prime Minister and Cabinet, 2024).
The intent is not to involve public servants in partisan or political matters, but simply to afford political actors an early opportunity to learn about key factors which might influence successful implementation. The Canadian government should adopt a similar approach as that taken by Australia’s government.
Policy decisions involve both individual departments and central agency oversight. Any initiative requiring a policy change or regulatory change approved by cabinet is first subject to review by internal departmental personnel. The Department of Finance, Treasury Board Secretariat and Privy Council Office also exercise oversight.
In normal practice, policies are approved through cabinet or budgetary decision-making. While key financial factors around an initiative are disclosed at the cabinet stage, the decision is generally based on overarching strategic, policy and communications goals. This policy stage is then followed by a more detailed review by the cabinet committee of ministers called the Treasury Board. The department that supports the Treasury Board, the Treasury Board Secretariat (TBS), contributes to the review with materials relating to human, financial and technological resources. There is a subsequent exchange of questions and answers between Treasury Board ministers and TBS officials. The review also includes submissions of information from the department that proposed the initiative in question.
In our experience, Treasury Board ministers can be surprised by the powerful constraints revealed during the more detailed review. In particular, the time and detailed planning required for information technology delivery, procurement and successful negotiations with provincial, municipal and private-sector delivery partners can result in unmet ministerial and public expectations.
Ontario approaches this same decision-making process with different sequencing. Policy proposals first go to a cabinet sub-committee for initial or preliminary endorsement. The Treasury Board of cabinet then undertakes a more detailed review of fiscal and resource implications. Final approval occurs at the cabinet level, with ministers able to review both cabinet sub-committee and Treasury Board of cabinet reports.
Australia builds program design and implementation assurance into the cabinet and budget submission process. The cabinet secretary is authorized to reject submissions that do not contain all the information the cabinet requires to assess the implementation challenges of a proposal. There is no later Treasury Board process.
Both models provide cabinet decision-makers with an integrated picture of both policy intent and the time, resource and partnership requirements for successful implementation. The integration of cabinet and Treasury Board reviews imposes discipline on both the timing and volume of information needed for decision-making, encouraging consolidation and clarity for ministers.
The federal government should adopt one of these models to ensure operational and implementation issues are considered at the front end of the cabinet decision-making process.
Not all decisions proceed through the cabinet process. A prime minister may choose to make decisions personally on key files — out of preference, or out of necessity during a crisis, such as the COVID-19 pandemic. This model can increase implementation risks by limiting opportunities for the responsible minister or departmental officials with operational expertise to raise concerns. In these cases, it is particularly critical that operational considerations feature prominently in the advice provided to the prime minister. Ideally, the responsible minister and department officials would brief the prime minister directly on these issues.
Wherever the policy decision is made at the political level, our core advice is that those decisions need to be implementation-informed. This includes advice on the implementation pathway, timing, cost and risk mitigation associated with each option considered.
Treasury Board and its Secretariat are guardian institutions that oversee federal spending, set administrative policies and ensure accountability for public funds. Along with its sister institution, the Department of Finance, TBS has generally fulfilled this role by exercising a “challenge” function over departmental requests. Much of the TBS review is focused on detailed verification of new departmental requests. TBS experts do a deep dive into costing, human resource requirements, program terms and conditions, procurement needs and other factors that enable policies to be put into action. Risk aversion is not a bug in the system but rather a design feature.
The Department of Finance and TBS are also guardians of the fiscal framework. Their systems are designed to challenge new spending requests. They usually suggest final allocations that are lower than what the department identified as necessary to achieve implementation. Spending avoidance is a core design feature of the system. Meanwhile, accountability for implementation rests with departments who must live with those decisions.
Risk aversion and spending avoidance are meant to ensure responsible stewardship of public spending. But when government decisions require an accelerated operational tempo, these features can become serious impediments to successful delivery.
When funding levels are set too low, teams underinvest in delivery capacity or scale down the project’s scope. Operational contingencies — funds and strategies to deal with the inevitable bumps that occur in project implementation — are generally removed from budget decisions. Including these contingencies is a best practice in the public and private sectors. Cutting them creates time delays and an impression of mismanagement if departments return for new funding when, inevitably, those bumps emerge. Government priorities should be resourced to maximize the chance of success, not just minimize the initial fiscal impact. Building appropriate contingencies into the original funding decision is essential.
When departments don’t have the ability to move project money between fiscal years, it prevents them from taking advantage of opportunities to accelerate. It also creates administrative burden and delay as they await to see if their request to move money to the next fiscal year has been approved. Large projects should be given multi-year flexibility in allocating their spending.
Another costing issue that creates challenges is the pressure to make early announcements on a project when the cost estimate is preliminary and intended only as a broad approximation. This is particularly true of defence procurements. As with all major projects in their early stages, confidence in costing estimates gets much tighter when the project details get closer to being finalized. Costing announcements made too early in the process often end up underestimating the true project cost and contribute to a perception that a project has been mismanaged and gone over budget. The federal government should either include a wider range of potential costs in the early announcement or delay announcing cost estimates until the project planning is further advanced.
Even the most carefully constructed implementation strategy will require adjustments, usually requiring a visit to the Minister of Finance or the Treasury Board to request permissions. Central agencies need to be aware that, in some cases, their own review processes may become a barrier to successful implementation. Central agency approval processes operate on specific cycles that can be slower than the pace required for real-world decision-making on projects with significant technological, partnership or inter-governmental elements.
As a result, senior officials in charge of major projects often end up devoting their attention to securing internal approvals at the expense of overall partner and project management. The savings from strong oversight and control can be vastly overwhelmed by the actual and opportunity costs imposed by process requirements. For complex projects, time delays related to approvals or additional funding can materially increase overall costs and delay delivery.
There is scope to further calibrate access to Treasury Board and streamline conditions imposed on projects. One of the measures retained after the COVID-19 pandemic was a much more streamlined submission and approval process for lower-risk initiatives. This should be expanded. Departments with high implementation capacity and lower program complexity should be given greater freedom to implement without further oversight.
Beyond approval processes, central agencies should also reconsider how they allocate oversight effort. Particularly complex projects need more ongoing engagement with central agencies, but less in the form of “challenge” and more in the form of assistance to help move the project forward. Treasury Board should delegate more ongoing oversight and approval responsibilities to central agency leaders to ensure timely decision-making. The significance of the program and project would dictate the seniority level of the TBS official tasked with the control. Access to contingency funding should be delegated to officials at the Department of Finance. The key point is the need to delegate outside of the political level to allow for faster decision-making.
Treasury Board and its Secretariat are generally focused on assessing and constraining the risk of new proposals. “New” is treated with caution and diligence and subject to greater scrutiny. This is a disincentive to innovation, despite the fact that innovation is a critical ingredient to advancing projects more efficiently. The focus on “new” comes at the expense of assessing existing programs and the risks of maintaining the status quo. Treasury Board and TBS’s work plans should ensure that the risks of change are not overweighted. They should focus greater effort on reviewing base spending that may not be meeting performance objectives.
Overall, in a world of rapid change where timely delivery is critical, we believe that central agencies need to shift their balance of effort from the “challenge” function, focused on reducing risk in departmental proposals, toward the “lift” function, focused on enabling departments to get proposals implemented quickly.
Adding complexity to a program design almost always increases the cost of delivery and makes tasks more challenging to automate. This is why successful companies spend significant time at the design stage minimizing project complexity.
For example, Canada’s employment insurance system is one of the most complex income support systems in the world. It is the result of dozens of decisions across different governments which have continually added to its complexity. This complexity increases program delivery costs, makes it almost impossible to scale in a crisis, and greatly increases the risks and costs of moving the system to a modern digital platform.
As part of an internal red tape reduction exercise, Treasury Board should require all departments to review existing programs for simplification opportunities. These could include everything from changes to core program rules to administrative and process changes. Simplification also reduces the risks associated with moving programs to modern digital platforms and makes it easier to automate tasks on those platforms. Savings from the exercise could be dedicated to moving more programs to digital platforms — generating, in turn, even greater savings through automation efficiencies. For all new programs, the full costs and risks of complex policy options should be identified and weighed against simpler alternatives.
Taken together, these changes involve a significant shift in how the federal government operates. They alter the division of roles, responsibilities and accountabilities. As with all changes, they come with risks.
Some of the changes are drawn from best practices of other Westminster jurisdictions and could be adopted at relatively low risk. Some are extensions of how the government has operated in past crisis contexts where risks of the new approach are relatively well understood.
Other elements of our proposal introduce new risks, reflecting a deliberate decision to prioritize timely implementation over eliminating all potential vulnerabilities. We believe that inaction itself is a bigger risk than moving forward with some uncertainty. Given the current context facing the country, we believe that this bias toward action is justified and necessary.
Breton, C., Han, J., McLaughlin, D., & Woodward, C. (2024). Resilient institutions: Lessons from Canada’s pandemic response. IRPP Report (March). Montreal: Institute for Research on Public Policy. https://centre.irpp.org/research-studies/resilient-institutions-learning-from-canadas-covid-19-pandemic/
Coombs, H. C. (1976). Report of the Royal Commission on Australian Government Administration. Australian Government Publishing Service. https://apo.org.au/node/34221
Department of the Prime Minister and Cabinet. (2024, December 9). Guidance on caretaker conventions. Australian Government. https://www.pmc.gov.au/resources/guidance-caretaker-conventions
Jean, D., & Jahn, A. (2021). Public service innovation and leadership during COVID-19: What can we learn? Public Policy Forum. https://ppforum.ca/publications/public-service-innovation-and-leadership-during-covid-19-what-can-we-learn/
Lindquist, E. (2022). Canada’s response to the global financial crisis: Pivoting to the Economic Action Plan. In E. Lindquist, M. Howlett, G. Skogstad, G. Tellier, & P. t’ Hart (Eds.), Policy success in Canada: Cases, lessons, challenges (pp. 457-477). Oxford University Press.
https://doi.org/10.1093/oso/9780192897046.003.0023
This memo was commissioned as part of Capacity for Change: Designing a Public Service Built to Deliver, an IRPP research project that aims to provide solutions to barriers to state capacity through research-backed recommendations from leading experts. The publication was developed under the direction of IRPP President and CEO Jennifer Ditchburn and Professor Jennifer Robson, Director of Carleton University’s School of Political Management. It was copy-edited by Maya Lach-Aidelbaum, proofread by Zofia Laubitz, editorial co-ordination was by Étienne Tremblay and production was by Chantal Létourneau.
Graham Flack and Peter Wallace are former Secretaries to the Treasury Board of Canada. Graham was a Deputy Minister in five federal departments. Peter also served as Toronto City Manager, Secretary to Cabinet and Head of the Ontario Public Service, and Ontario Deputy Minister of Finance and Secretary to the Treasury Board.
This project was made possible in part thanks to support from the Max Bell Foundation and the Metcalf Foundation. The IRPP is an independent think tank, and retains control over the scope, methodology, conclusions and recommendations of our work.
A French translation of this text is available under the title Les intentions face à la réalité :
Intégrer la mise en œuvre dans le processus décisionnel gouvernemental.
To cite this document:
Flack, G. & Wallace, P. (2026). Where intent meets reality: Embedding implementation into government decision-making. Institute for Research on Public Policy. https://doi.org/10.26070/5y1f-2k67
