HEW-NODE-0133 · How Education Works · regulatory impact assessment, policy appraisal, education regulation, problem definition, alternatives, cost-benefit analysis, administrative burden, equity impact, consultation, implementation feasibility, monitoring, review clauses and regulatory quality
A new education rule can be perfectly well intentioned and still make the system worse.
A reporting requirement can absorb thousands of teacher-hours. A licensing rule can improve quality and also create a shortage. A compulsory programme can protect learners while imposing costs some schools cannot absorb. A new accountability indicator can clarify performance and simultaneously encourage narrow behaviour.
Regulatory impact assessment exists for the moment before the rule hardens into obligation.
Regulatory impact assessment asks whether a proposed rule solves the right problem, whether a less costly route exists, who will bear the burden, and how the system will know if the rule worked after it becomes real.
This node sits beside the How Education Works hub, Education Policy, Education Law, Education Law & Regulatory Hierarchy, Education Policy Pilots & Scaling, Education Sector Planning & Annual Operationalisation, Education Complaints, Appeals & Redress and Education Internal Controls & Fraud Risk Management.
Those pages keep their jobs. Education Policy owns policy goals and choices broadly. Education Law owns legal authority. Regulatory Hierarchy owns how legal instruments relate. Policy Pilots owns testing reforms in live settings before scale. Sector Planning owns implementation programmes. This node owns the ex-ante appraisal layer: how a proposed education law, regulation, mandatory standard or administrative rule is examined before adoption for necessity, options, cost, equity, burden, implementation feasibility, unintended effects and future review.
The 60-Second Read
- RIA begins with the problem, not with the preferred regulation.
- A rule should identify the market, institutional or rights failure it is trying to repair.
- The “do nothing” baseline is necessary because regulation should be compared with what happens without it.
- Alternative options can include information, funding, standards, guidance, incentives, licensing, inspection or direct prohibition.
- Costs include compliance time, training, systems, reporting, enforcement and opportunity cost.
- Benefits can include learning, safety, access, trust, reduced fraud, better information and stronger rights.
- Distribution matters: the same rule can benefit the average learner and harm a small high-need group.
- Schools of different size and capacity can face very different compliance burden.
- Teacher time is a real regulatory cost.
- Digital systems can lower burden and create new data, privacy and vendor risks.
- Consultation should test assumptions, not merely collect support.
- Stakeholders closest to implementation can reveal hidden costs.
- Regulators should ask whether the intended authority actually has capacity to enforce the rule.
- Unenforced regulation can weaken trust and create unequal compliance.
- Transition time and sequencing affect feasibility.
- Sunset or review clauses can prevent obsolete rules from persisting indefinitely.
- Monitoring indicators should be defined before implementation where possible.
- Post-implementation review closes the loop and tests whether the ex-ante assumptions were correct.
- RIA is not a substitute for political judgement; it makes trade-offs more explicit.
- The objective is not fewer rules by default. It is better justified rules with visible consequences.
One-Sentence Definition
Education regulatory impact assessment is the structured appraisal of a proposed law, regulation or mandatory policy rule to determine whether intervention is necessary, compare alternatives, estimate benefits and burdens, test equity and implementation feasibility, and define how the rule will later be reviewed.
The First Distinction: Problem Definition Is Not Rule Justification
“Schools need to submit monthly data reports” is a proposed rule, not a diagnosis. The problem might be that government lacks timely attendance data. The correct response could be monthly reporting, automated extraction, sampling, termly reporting or improvement of an existing data system.
Starting from the preferred rule narrows the analysis before it begins.
The Second Distinction: Lawful Is Not Necessarily Proportionate
A ministry may have legal authority to impose a requirement and still need to ask whether the burden is proportionate to the risk being addressed.
Regulatory quality lives between “can we do this?” and “should we do it this way?”
The Third Distinction: Consultation Is Not Impact Assessment
Consultation supplies evidence and perspectives. It does not replace analysis. A rule can be popular and still expensive, or unpopular and still necessary to protect rights or safety.
The regulator remains responsible for weighing evidence and explaining the decision.
OECD Treats RIA as a Core Regulatory-Control Tool
OECD’s Government at a Glance 2025 describes regulatory impact assessment as a way to examine the likely positive and negative effects of proposed regulation, compare possible pathways, make trade-offs explicit and support transparent decision-making. Its 2024 iREG evidence also shows that methodology has generally developed faster than oversight and quality control.
That pattern matters in education. Writing an impact-assessment template is easier than building institutional capacity to challenge weak assessments.
Start With the Failure the Rule Is Intended to Correct
- information failure;
- unsafe practice;
- quality variation;
- rights violation;
- coordination failure;
- misaligned incentives;
- market power;
- fraud or misrepresentation;
- external costs imposed on others;
- inconsistent public-service standards;
- missing accountability;
- administrative ambiguity.
Different failures justify different tools. Licensing can address minimum competence. Disclosure can address information asymmetry. Funding can address affordability. Inspection can address hidden non-compliance. A ban is not the natural answer to every problem.
Define the Baseline
What happens if government does not introduce the rule? The relevant baseline is not “nothing changes” unless that is realistic. Existing technology, demographic trends, court decisions, market behaviour or voluntary standards may already be changing the system.
Every claimed regulatory benefit is measured against a baseline, whether stated or hidden.
Specify the Policy Objective Before the Instrument
“Reduce school transport incidents by 40 per cent” is an objective. “Require a new annual certification form” is an instrument. Keeping the two separate creates room for alternatives.
Generate Real Alternatives
- no new intervention;
- better guidance;
- public information or disclosure;
- professional standards;
- funding incentives;
- contract conditions;
- self-regulation with oversight;
- licensing;
- inspection;
- mandatory reporting;
- minimum service standards;
- prohibition or restriction;
- technology-based compliance;
- pilot before full obligation.
A meaningful RIA compares alternatives that could plausibly solve the problem, not one preferred rule against obviously inferior straw men.
Administrative Burden Is Educational Burden
A new reporting form may require only twenty minutes. Multiply by every class, every month, every school and every teacher, and the system can consume hundreds of thousands of hours that would otherwise support instruction, planning or student care.
Time should be costed when regulation requires professional labour.
Count Setup Cost and Recurring Cost Separately
A new safeguarding standard may require one-off training, software configuration and policy rewriting, followed by modest annual maintenance. A reporting regime may have low setup cost and high recurring burden.
The distinction changes affordability and transition planning.
Small Schools Face Different Fixed Costs
One compliance officer can serve 2,000 students or 200. Fixed regulatory requirements therefore weigh more heavily on small schools and providers unless scaled appropriately.
Proportionality can include simplified reporting, thresholds or shared services without lowering the underlying protection standard.
Benefits Need the Same Discipline as Costs
Regulatory analysis can become asymmetrical: costs are quantified carefully while benefits are described vaguely as “improved quality.”
Benefits should be specified: fewer injuries, fewer fraudulent credentials, more timely intervention, reduced dropout, more reliable assessment, lower administrative disputes, stronger accessibility or improved information.
Not Every Benefit Can Be Monetised
Child safety, dignity, procedural fairness and equal access cannot always be reduced sensibly to currency. Cost-benefit analysis is useful but should not force every education value into an artificial price.
Multi-criteria appraisal can preserve non-monetary values explicitly.
Equity Is Not an Appendix
A rule can lower average cost and increase inequality. For example, digital-only compliance can reduce paperwork for most families while excluding those with poor connectivity or accessibility needs.
RIA should examine who gains, who pays and who may lose access because of the rule.
Rights Constraints Can Override Efficiency
A cheaper option is not acceptable if it unlawfully discriminates, violates privacy or removes due process. Legal and rights review should constrain the feasible option set before optimisation.
Implementation Capacity Is Part of the Impact
A regulation requiring every school to employ a specialist may look strong on paper while the country lacks enough specialists. The immediate effect can be widespread technical non-compliance rather than better service.
RIA should test whether the workforce, systems, budget and enforcement capacity exist on the required timeline.
Enforcement Cost Belongs in the Calculation
If compliance cannot be observed, inspected or audited, the rule may rely on self-reporting. If government intends serious enforcement, inspectorate capacity and appeal handling must be funded.
A rule without an enforcement model can create unequal compliance: conscientious organisations bear the burden while non-compliant organisations gain advantage.
Digital Regulation Can Reduce Friction and Increase Dependency
Automated reporting can eliminate repetitive forms. It can also require interoperability, identity systems, vendor contracts, cyber controls and data-protection safeguards.
The impact assessment should include the infrastructure required to make the “simpler” digital rule work.
Data Collection Should Pay Rent
If government requires a new field from every school, who will use it and for what decision? Data collected because it “might be useful” can accumulate into permanent burden.
Each mandatory data item should have an owner, purpose, retention rule and review date.
Consultation Should Search for Disconfirming Evidence
Do not ask stakeholders only whether they support the proposal. Ask what the rule assumes incorrectly, what burden is hidden, which cases do not fit, what they would do to comply and how behaviour might change strategically.
Good consultation pressure-tests the model.
Include the Quiet Implementers
Principals, registrars, finance staff, counsellors, special-needs coordinators, IT administrators and local officials often understand operational cost better than senior policy teams.
Implementation knowledge is distributed through the system.
Consult Affected Learners and Families Where the Rule Reaches Them
A technically clean admissions rule can create confusing family behaviour. A reporting requirement can expose privacy risks. A transport regulation can increase journey time for disabled learners.
Those effects can remain invisible if consultation occurs only among institutions.
Behavioural Response Can Defeat the Rule
When incentives change, regulated actors adapt. If a funding formula penalises high absence, schools may change attendance coding. If a licensing rule raises entry costs, providers may leave the market. If a ranking metric becomes public, institutions may optimise for the metric.
Impact assessment should model behaviour, not only formal compliance.
Thresholds Create Edge Effects
If schools with more than 500 students face a new reporting regime, organisations may resist expansion around the threshold. If providers need a licence only above a revenue level, firms may split operations.
Thresholds simplify regulation and can also distort behaviour.
Transition Design Is Part of Regulation
A good rule with an impossible deadline is still bad implementation. Systems may need phased entry, grandfathering, transitional funding, provisional licences or staged reporting.
Grandfathering Solves One Problem and Creates Another
Allowing existing providers to continue under old rules can prevent disruption but may create two quality standards. Transition exemptions should therefore have an endpoint or review path.
Sunset Clauses Force Reconsideration
A temporary rule can expire automatically unless renewed. This is useful when evidence is weak, technology changes rapidly or emergency regulation should not become permanent by inertia.
Review Clauses Are Often More Practical Than Sunset
Some education rules need continuity. A mandatory review after three or five years can require government to examine outcomes and burden without creating legal uncertainty from automatic expiry.
Define the Monitoring Plan Before Adoption
- implementation rate;
- compliance cost;
- enforcement activity;
- learner outcomes;
- access effects;
- provider entry and exit;
- complaints;
- appeals;
- data incidents;
- subgroup effects;
- unexpected behaviour.
If no data will exist to review the rule later, the ex-ante assessment should say so.
Post-Implementation Review Tests the Original Story
Did the problem shrink? Were costs close to forecast? Did small schools face larger burden than expected? Did the rule create workarounds? Did technology reduce or increase compliance time?
The purpose is not to defend the original decision. It is to update the regulatory system using observed evidence.
Current OECD Practice Emphasises Oversight
OECD’s 2025 regulatory-policy evidence reports that countries have improved the systematic use of RIA, while oversight and quality control remain weaker than methodology in many systems. Its October 2025 report Applying Regulatory Impact Assessment at Regulatory Authorities provides a practical methodology for assessing economic, social and environmental impacts.
The education lesson is direct: a ministry needs not only analysts who fill in impact assessments, but reviewers who can reject weak ones.
RIA and Policy Piloting Work Together
RIA can identify uncertainty before adoption. A pilot can generate evidence about those uncertainties in a bounded setting. The separate Education Policy Pilots & Scaling node owns live implementation testing.
Appraisal asks what we expect; piloting asks what actually happens under controlled early implementation.
RIA and Cost-Effectiveness Are Related but Different
Cost-effectiveness compares ways to achieve an outcome. RIA asks a wider regulatory question: whether intervention is necessary, whether regulation is the right tool, what burdens and rights effects occur, and how implementation will work.
Case Study: The Monthly Reporting Rule
Invented example: government proposes monthly manual reporting of student attendance by every school. The problem is slow detection of disengagement. RIA estimates 420,000 staff-hours annually and finds most schools already record attendance digitally.
The final design uses automated weekly transfer from existing systems, with manual reporting only for schools lacking compatible systems. The objective becomes stronger while the burden falls.
Case Study: The New Specialist Requirement
Invented example: every school is required to employ one specialist counsellor within twelve months. Workforce data show there are only enough qualified professionals to cover half the required posts.
The regulation is phased, training capacity expands and regional shared-service models are permitted during transition. The quality standard remains; the implementation path changes.
Case Study: The Digital-Only Application Rule
Invented example: a scholarship programme moves to digital-only application to reduce administrative cost. Most users benefit, but consultation identifies rural connectivity gaps and accessibility barriers.
The regulation keeps digital as the default while requiring assisted application routes and accessibility standards. Average efficiency no longer erases edge cases.
Case Study: The Ranking Rule That Changed Behaviour
Invented example: institutions must publish one headline employment-rate metric. Providers begin excluding certain graduates from the denominator and reorient programmes toward easily placed occupations.
Post-implementation review replaces the single metric with a transparent indicator set and standard denominator rules.
Failure Modes and Repairs
- Rule-first appraisal: repair by defining the problem before the instrument.
- Do-nothing baseline omitted: repair by modelling the realistic counterfactual.
- Only one option assessed: repair with genuine alternatives.
- Teacher time treated as free: repair by costing professional compliance burden.
- Average effects only: repair through subgroup and distribution analysis.
- Implementation capacity ignored: repair by testing workforce, systems and enforcement feasibility.
- Consultation as popularity poll: repair by asking stakeholders to challenge assumptions.
- Digital simplification myth: repair by including infrastructure, privacy and vendor dependencies.
- No review mechanism: repair with sunset or formal post-implementation review.
- RIA as paperwork: repair by giving an independent oversight function authority to challenge weak assessments.
The Education RIA Operating Chain
- Define the education problem.
- Identify legal authority and rights constraints.
- Define the baseline.
- Set the policy objective.
- Identify affected learners, families, schools and providers.
- Generate regulatory and non-regulatory options.
- Estimate setup cost.
- Estimate recurring compliance cost.
- Estimate government enforcement cost.
- Estimate expected benefits.
- Identify non-monetary benefits and rights effects.
- Assess equity and distribution.
- Assess small-provider and small-school burden.
- Test data and technology dependencies.
- Test workforce feasibility.
- Model behavioural response.
- Identify threshold effects.
- Consult implementers and affected users.
- Revise assumptions.
- Compare options transparently.
- Design transition arrangements.
- Define enforcement.
- Define monitoring indicators.
- Add review or sunset clauses.
- Subject the RIA to quality oversight.
- Publish the rationale where appropriate.
- Adopt the chosen instrument.
- Monitor actual burden and outcomes.
- Conduct post-implementation review.
- Amend, simplify, retain or repeal based on evidence.
A Regulatory Impact Dashboard
- problem definition;
- legal authority;
- baseline;
- options considered;
- estimated compliance hours;
- estimated fiscal cost;
- estimated benefits;
- groups gaining;
- groups bearing cost;
- small-school impact;
- workforce constraint;
- digital dependency;
- privacy impact;
- consultation responses;
- major assumptions challenged;
- transition period;
- enforcement capacity;
- review date;
- actual burden after implementation;
- actual outcome against forecast.
Canonical Owner Boundaries
- Education Policy owns policy goals and broad policy design.
- Education Law owns legal rights, duties and authority.
- Education Law & Regulatory Hierarchy owns relationships among statutes, regulations, subordinate instruments and guidance.
- Education Policy Pilots & Scaling owns live pilot design, evidence and expansion.
- Education Sector Planning & Annual Operationalisation owns implementation planning after policy choice.
This node owns the pre-adoption appraisal of mandatory policy instruments: problem definition, alternatives, costs, benefits, burden, equity, implementation feasibility, consultation, monitoring design and later review.
The Return Path
Return to the draft regulation before anyone is legally required to follow it.
This is the cheapest moment to discover that the reporting burden is enormous, that the workforce does not exist, that the threshold creates avoidance, that a simpler instrument would work, or that one group will lose access.
After adoption, those discoveries become implementation problems, political disputes and sunk cost.
The purpose of regulatory impact assessment is not to stop education systems from acting. It is to make them pay attention before everyone else has to pay the cost of a badly designed rule.
Return to the How Education Works hub.