Herd Nation: How Canadians Are Managed, Nudged, and Trained to Comply Alexandra Kitty

Instalment 2: Think What We Need You to Think: The Federal Behavioural Insights Apparatus

There is a unit inside the Privy Council Office, the nerve centre of the Canadian federal government, the office that serves the Prime Minister directly, whose stated purpose is to change the way Canadians make decisions. It is called the Impact and Innovation Unit (IIU). It has over 500 members embedded across the federal public service. It recruits “top-tier external talent” through a fellowship programme with expertise in “sludge audits,” “outcomes-based financing,” and “advanced data and behavioural science analytics”. It runs experiments on Canadians using randomised control trials, the gold standard of scientific testing, to determine which psychological interventions are most effective at producing compliance.

It does not require your consent to do any of this.

What “Nudge” Actually Means

The word “nudge” was popularised by American legal scholars Richard Thaler and Cass Sunstein in their 2008 book of the same name, and it carries a deliberately gentle connotation. A nudge, in their formulation, is “any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives”. Putting fruit at eye level in the school cafeteria is a nudge. Banning junk food is not.

The philosophy behind nudging rests on a body of research in behavioural economics and cognitive psychology showing that human decision-making is systematically irrational in predictable ways. People avoid effort. They follow social norms. They give disproportionate weight to losses over gains. They are heavily influenced by how choices are framed. They defer to defaults. These are not occasional lapses, they are structural features of human cognition, replicated across thousands of studies. The nudge theorists argue that since these biases operate whether governments intervene or not, responsible government should design the choice environment to steer people toward outcomes that are good for them.

The EAST framework, Make it Easy, Make it Attractive, Make it Social, Make it Timely, is the operational tool through which these principles are applied. Make it easy: remove friction from the desired behaviour, add friction to the undesired one. Make it attractive: frame benefits to emphasise their appeal; use scarcity, personalisation, and optimism. Make it social: deploy peer comparisons, social norms, and “messenger selection” to use the influence of trusted figures. Make it timely: present the desired behaviour at the moment when the person is most psychologically receptive.

The Canada School of Public Service teaches this framework to federal public servants as a tool for “improving policy-making and service design”. Policy Horizons Canada, a federal foresight body, has published briefings on its application. The OECD endorses it globally. Deloitte sells it to the Canadian public sector as consulting advice. By the time a citizen encounters a redesigned government form, a reworded email, or a reconfigured benefit enrolment flow, the nudge has already been tested, iterated, and optimized for compliance, and the citizen has not been asked whether they consented to being the subject of a psychological experiment.

The Architecture Inside the Government

The IIU was established in 2015 under the Trudeau government as a vehicle for “innovative policy and programme approaches”. It sits within the Privy Council Office, which means it has direct access to the highest levels of federal decision-making and can embed behavioural interventions across all departments simultaneously. It founded the Behavioural Insights Community of Practice in 2016; by 2026, that community counted over 500 government members connecting regularly across departments.

The IIU’s methodology is explicit and systematic:

Identify: Consult departmental leadership to determine “priority target outcomes” and the “key behaviours to target.”
Understand: Conduct literature scans, stakeholder interviews, observation, surveys, and administrative data analysis to map the “drivers and barriers” to those behaviours.
Design: Develop behavioural interventions referencing the scientific literature; prototype and iterate based on partner feedback.
Test: Deploy randomised or quasi-experimental trials to measure which interventions produce the greatest behaviour change.
Scale: Roll out successful interventions across departments.

This is, in structure, identical to the methodology used by the private sector to design advertising, product interfaces, and social media feeds, all of which have been widely criticised for exploiting psychological vulnerabilities to extract engagement, purchases, or data. In the government context, the “product” being optimised is compliance.

The Canada Revenue Agency has used nudges within specific segments of the small business file to monitor and “optimise” compliance behaviour. Employment Insurance has been a target for behavioural redesign, with interventions aimed at steering claimants toward faster re-employment. Veterans Affairs, health promotion, tax filing, benefit enrolment, and public health communications have all been documented targets of IIU and departmental behavioural science interventions.

A parallel unit, the Public Health Agency of Canada’s Behavioural Science Office (BeSciO), was established in 2021 specifically to apply behavioural science to public health. In its first two years, BeSciO listed as priority areas: “vaccine confidence, public health measures, combatting mis/disinformation, borders measures”. The OECD’s Observatory of Public Sector Innovation records BeSciO as having a team of 20 behavioural scientists analysing “data about individual behaviour to look for patterns and predict behaviours”.

What Happened During COVID

The clearest window into how the nudge apparatus operates under pressure came during the COVID-19 pandemic, and what the documentary record shows is considerably more aggressive than anything the government acknowledged publicly.

In March 2020, Impact Canada launched a “program of applied research grounded in Behavioural Science to help support the Government’s response effort”. The stated objective was to “integrate evidence-based, behaviourally-informed insights and recommendations to public communication materials, policy and programmatic considerations, and whole-of-government decision-making”. In practice, according to the Justice Centre for Constitutional Freedoms’ November 2025 report Manufacturing Consent: Government Behavioural Engineering of Canadians, this included:

Emotional framing: Deploying fear, reassurance, and urgency to increase compliance with lockdowns, mask mandates, and vaccine requirements, with specific emotional triggers tested experimentally against different audience segments.
Message manipulation: Emphasising or omitting details to shape how Canadians interpreted adverse events following COVID vaccination, making such events “appear less serious”.
Fictitious news testing: Exposing thousands of Canadians to fabricated news reports to assess which emotional triggers most effectively neutralised public anxiety about vaccine side effects.
Targeted vaccination uptake campaigns: Designing communications explicitly to achieve the government’s 70% vaccination uptake target, defined internally as reaching “herd immunity”.

The choice of words in that last point is worth pausing on. The government’s internal documents used the phrase “herd immunity.” The intervention was designed to move the herd. The metaphor was operational, not rhetorical.

Academic review of pandemic nudging has found that nudge-based interventions during COVID produced mixed results, with effectiveness heavily dependent on situational context, and a tendency to suffer from “habituation” after repeated exposure or to “backfire due to inappropriate use”. But effectiveness is not the only question. The prior question, whether the government had any right to run covert psychological experiments on its population without disclosure or consent, received no serious parliamentary debate before, during, or after the intervention.

The Sludge Audit: When Nudging Goes Both Ways

The IIU’s work is not limited to increasing compliance with government programs. It also runs “sludge audits”, structured reviews of unnecessary frictions in government processes that prevent citizens from accessing services or benefits they are entitled to. Sludge, in the behavioural science lexicon, means “excessive or unjustified frictions or barriers in a process that make it harder for people to follow through on their intentions”.

This sounds benign, and in isolation it often is: reducing the complexity of a disability benefit form, simplifying a tax filing instruction, removing redundant approval steps from a government service. But sludge audits exist alongside the nudge infrastructure, not instead of it. The same government that removes friction from benefit access can add friction to non-compliance. The same architecture that makes it easy to sign up for a vaccination programme also makes it easy to design a system where opting out requires effort, knowledge, and deliberate resistance.

The Institut économique de Montréal, a market-oriented think tank, published a detailed warning in 2023 arguing that Canada’s nudge apparatus lacks any meaningful democratic oversight. There is no requirement for public disclosure of which interventions are being deployed on which populations. There is no parliamentary committee with jurisdiction over the IIU’s experimental programme. There is no mechanism by which a Canadian citizen can determine whether their behaviour is currently being actively managed by a government-designed psychological intervention, or opt out if they object.

The IIU’s own guidance to public servants includes a list of ethical principles: be mindful of identifying potential ethical concerns, obtain informed consent “where required,” be transparent about the purpose of collected data. The qualifier “where required” is doing significant work in that sentence. The IIU does not define the threshold at which consent is required. In practice, population-level interventions, redesigned forms, reworded letters, changed default options, do not require individual consent under any current federal law or policy.

The Consent Nobody Gave

The philosophical defence of nudging has always rested on a distinction between coercion and architecture. Nudges do not force people to do anything, the argument goes; they merely make the preferred option more visible, more accessible, or more socially normal. The individual retains the freedom to choose differently, they just have to work slightly harder to exercise it. This is called “libertarian paternalism”: the state acts in your best interest while technically preserving your autonomy.

The criticism that this defence cannot ultimately answer is simple: who decided what your best interest is, and when did you agree to have the state experiment on your decision-making to produce it?

Nudge theory was built on the premise that human cognition is imperfect and that institutional architecture can correct for it. But the institutions deploying the architecture have their own interests, their own cognitive limitations, and their own political pressures. The vaccination compliance target of 70% was not derived from a value-neutral epidemiological consensus, it was a political goal pursued through psychological engineering. The decision to test fictitious news articles on Canadians to shape their perceptions of vaccine adverse events was not a disinterested public health intervention, it was a government shaping how its citizens thought about the government’s own decisions.

The British Behavioural Insights Team, from which the IIU draws its model and after which it patterns its methodology, has been formally criticised by the UK Parliament’s Science and Technology Committee for operating with insufficient transparency and inadequate ethical oversight. The Canadian equivalent has received no comparable parliamentary scrutiny.

The 500 members of the Behavioural Insights Community of Practice embedded across the federal public service are not a conspiracy. They are, in their own framing, professionals improving service delivery and evidence-based policy. But they are also a standing apparatus for the systematic redesign of the decision environment in which 40 million Canadians live and make choices, without those Canadians being told the environment is being redesigned, without their consent, and without any mechanism to find out whether it is being done to them right now.

The census asks you to disclose yourself to the state. The nudge unit designs the world you disclose yourself into.

Next in Herd Nation, Instalment 3: Read What We Pay For: The Media Subsidy Regime and Narrative Dependency

Additional research by Perplexity