How should a research organization translate findings into model legislation and operational policy without overstating certainty, hiding value choices, or neglecting delivery?
Publication
Practice Note
Published
Pillar
Public Policy Development
Evidence base
8 primary sources
Review status
Foundation synthesis; not peer reviewed
Executive synthesis
Evidence can establish the scale of a problem, test an intervention, and reveal tradeoffs. It cannot by itself choose whose interests should prevail, create legal authority, appropriate funds, or make an institution capable of delivery. A credible research-to-law process therefore makes the handoffs visible: from observation to explanation, from explanation to policy choice, from choice to legal instruction, and from a legal text to a funded service that can be evaluated.
This practice note proposes an operating method for True Tomorrow’s Policy Lab. It sets out a proposed repeatable method for developing the Foundation’s model legislation; it does not claim legislative sponsorship, enactment, or implementation. The proposed outputs are a decision brief, a claim-and-evidence ledger, an options appraisal, a legal-instructions packet, a model text reviewed in the target jurisdiction, an implementation plan, and an evaluation protocol. Each artifact should have an owner, version, unresolved questions, and a public status.
The objective is a reviewable proposal that survives contact with legal drafters, frontline staff, affected people, budget officers, and evaluators. The process should preserve disagreement and uncertainty rather than smoothing them away for a coalition message. A model bill is a starting point for democratic deliberation, not evidence that the public supports the proposal or that it will deliver its intended effect.
Findings
What the evidence supports.
Source markers link findings to the evidence below.
01
Evidence use is a process, not a single citation
WHO’s evidence-informed decision-making guide brings together methods for identifying, appraising, and mobilizing diverse evidence in health policy. It describes an evidence ecosystem involving producers and users rather than a simple transfer of a finished paper. This is methodological guidance; transferring it to non-health legislation requires adaptation. The proposed Policy Lab should therefore commission a decision question before commissioning a model bill. The question should name the population, problem, feasible choices, desired outcomes, and decision date. That structure prevents a collection of relevant publications from being mistaken for an answer to a specific legislative problem. [1]
02
Research communication can influence adoption in a bounded setting
In randomized evaluations with Brazilian municipal policymakers, researchers found that inviting mayors to a session on tax-reminder evidence increased adoption of the practice. The authors emphasize that the interventions were relevant, familiar, and relatively inexpensive, and caution about extrapolation to more complex policy. This is causal evidence about the invitation in that setting, not proof that reports produce legislation or that the adopted policy improved every outcome. The implication is to test the fit between evidence, official authority, operational cost, and the decision at hand. Accessible communication matters, but political agreement and delivery capacity remain separate questions. [2]
03
Regulatory analysis should begin before the preferred instrument is fixed
OECD’s regulatory-impact principles call for problem definition, proportionate analysis, plausible alternatives including non-regulatory choices, stakeholder engagement, and attention to implementation and enforcement. They also warn against equating every assessment with a full quantitative cost-benefit exercise. This is normative guidance drawing on regulatory practice. True Tomorrow’s proposed options table should compare existing practice, administrative improvement, spending or service changes, and legislation where feasible. If the chosen remedy requires a bill, the reason should be explicit: for example, creating a duty or power that an agency cannot lawfully create through guidance alone. [3]
04
A benefit-cost ratio cannot decide the policy alone
The UK’s 2026 Green Book distinguishes appraisal from policy choice and asks decision-makers to consider monetized and unmonetized impacts, distribution, risk, and delivery. Its mandates apply within its stated UK institutional setting. The practical lesson is to show the fiscal cost separately from social benefit and to identify groups bearing transition burdens. A proposal can have an attractive aggregate estimate while leaving rights, feasibility, or distribution unresolved. Our options table would therefore include a separate column for legal constraints and non-negotiable protections, alongside financial and empirical judgments. [4]
05
Evaluation design belongs in the initial proposal
The 2026 Magenta Book emphasizes early evaluation thinking, explicit theories of change, and methods appropriate to the question and context. Counterfactual methods help distinguish intervention effects from other change; theory-based approaches can investigate mechanisms where a credible comparison is unavailable. These are methodological standards, not guarantees that an evaluation will produce a decisive answer. The Policy Lab should specify what evidence could revise its recommendation and identify baseline data before rollout. An increase in service use after a law is adopted should remain a descriptive result until a defensible design addresses alternative explanations. [5]
06
Legislative drafting requires accurate operational instructions
The UK Office of the Parliamentary Counsel’s drafting guidance emphasizes understandable, accurate, and effective bills while recognizing its own institutional scope. It is not a universal drafting manual. Our proposed handoff gives local counsel the policy decisions and unresolved questions before asking for legal clauses: who owes what duty, to whom, under which conditions, with which powers, remedies, exceptions, and start date. Counsel should identify interactions with existing law and the limits of delegated powers. A polished model clause cannot substitute for this analysis or establish that the proposal fits another constitutional system. [6]
07
Participation and review need a route back into decisions
OECD participation guidance calls for meaningful room to influence the decision, appropriate inclusion, and feedback on how input was used. Its regulatory-review principles support systematic, proportionate assessment after adoption and modification when outcomes disappoint. Both are governance guidance rather than causal proof of effectiveness. Together they motivate a public response table and a scheduled review duty in the proposed workflow. Consultation should show which provisions changed and which objections remain. Review should examine whether the objective is still justified, whether the instrument worked, and what replacement or adjustment is feasible. [7][8]
Institutional Implications
What this changes in practice.
These are design implications drawn from the synthesis, not claims that one institutional model will fit every jurisdiction.
01
Maintain a claim-and-evidence ledger
Original proposal: assign an identifier to every material factual claim, recording its source, study design, population, date, limitations, and the wording it supports. Label a causal estimate, descriptive statistic, legal rule, forecast, and value judgment differently. A quantitative claim should retain its denominator, comparison, uncertainty, and time period. The ledger should also show contradictions and evidence that would weaken the recommendation.
02
Make each clause answer an implementation question
Original proposal: connect each proposed duty or power to the problem it addresses, the responsible body, the action required, necessary resources, and a reviewable outcome. Use examples of ordinary, difficult, and exceptional cases. If a clause cannot be explained through an operational scenario, its meaning or feasibility needs further work before endorsement.
03
Publish the value choices separately
Original proposal: state how the proposal weighs equity, liberty, security, cost, and urgency, and which body has democratic authority to decide. Invite challenges to those judgments without treating disagreement as ignorance of evidence. Explain distributional effects and tradeoffs in the coalition brief; do not remove caveats simply to make the message easier to circulate.
04
Develop the budget and delivery plan with the text
Original proposal: cost staffing, training, information systems, enforcement, assistance, communications, and evaluation over a realistic implementation period. Distinguish recurring from transition costs and identify who would authorize funding. Record dependencies on local institutions and secondary rules. A proposed right that cannot be exercised because the review body has no staff is an incomplete design.
05
Use independent challenge before public advocacy
Original proposal: commission review by someone tasked with finding a credible failure mechanism, including displacement, gaming, administrative burden, unequal access, or legal conflict. Respond with evidence or revise the proposal. Publish substantive disagreements and conflicts of interest. A successful internal review means the proposal is ready for deliberation, not that it has been validated in practice.
04 / Policy options
Choices and their tradeoffs.
Foundation analysis of the options. Evidence informs the choices; it does not remove their costs or uncertainty.
1
An administrative pilot where authority already exists
Original option: test a service or procedural improvement within existing powers and funding. This can expose delivery problems before wider reform, but must not become an informal way to bypass statutory entitlements or required consultation. Define the permitted scope and the decision the pilot will inform. [3]
2
A model legislative package
Original option: where a new legal duty or power is necessary, produce jurisdiction-specific drafting instructions, reviewed model clauses, explanatory notes, a fiscal and delivery assessment, and an evaluation plan. Publish unresolved constitutional or institutional choices instead of disguising them through generic wording. [6]
3
A staged reform with a review obligation
Original option: pair a durable protection with phased implementation and a scheduled public review. Use a sunset only where the consequences of expiry are acceptable; automatic expiry of an essential safeguard can create avoidable harm. Review should have a responsible body, evidence requirements, resources, and a response deadline. [8]
05 / Implementation
From a proposal to a working system.
A proposed sequence for local adaptation. These stages are not a claim of funded or completed Foundation activity.
1
Commission and scope
Name the decision owner and target jurisdiction; define the problem, public objective, affected groups, and feasible instruments. Publish the research question, inclusion criteria, and a timetable proportionate to the decision.
2
Synthesize and appraise
Build the evidence ledger, assess quality and transferability, map the proposed causal mechanism, and compare options. Record costs, distribution, legal constraints, uncertainty, and reasons for rejecting credible alternatives.
3
Draft and test the service
Give local counsel complete instructions. Develop the implementation plan alongside the draft. Test concrete cases with frontline staff and affected people, then publish a response table and an independently reviewed version.
4
Track the public process accurately
Maintain separate labels for proposed, consulted, introduced, amended, adopted, commenced, and evaluated. Link each claim to the responsible institution’s record. Adoption and commencement are different events, and neither is proof of impact.
5
Evaluate and revise
Collect the agreed baseline and implementation data, protect evaluation independence, and report findings including null or adverse results. Publish whether the evidence supports continuation, modification, expansion, or withdrawal, with the accountable body’s response.
06 / Measurement
What progress would mean.
Proposed measures for evaluation. Baselines, thresholds and reporting responsibilities must be agreed before implementation.
Claim traceability
Share of material factual claims linked to an inspected primary source and an explicit evidence classification.
Interpretation: A fully cited document can still select evidence unfairly; audit omitted or conflicting findings.
Implementation readiness
Each proposed obligation has a legally competent owner, resource estimate, process, commencement dependency, and review mechanism.
Interpretation: Readiness is a design assessment, not evidence that funding or political approval exists.
Consultation influence
Documented substantive changes, reasoned rejections, and unresolved objections following participation.
Interpretation: Counting submissions does not establish representativeness or legitimacy.
Outcome and attribution
Prespecified public outcomes assessed against baseline and a justified evaluation design, with distributional results.
Interpretation: Legislation passed, meetings held, and reports downloaded are outputs rather than demonstrated social impact.
Research Programme
What the foundation should test next.
Apply the complete proposed workflow to one bounded policy question and ask an independent reviewer to identify every unsupported leap from evidence to clause.
Give frontline staff and affected participants anonymized scenarios and proposed instructions; compare their interpretations and document operational contradictions.
Compare comprehension of a short coalition brief with the full evidence note, checking whether readers retain the main limitations and distinguish proposal from law.
Pre-register an evaluation plan for a legally authorized pilot, including outcome definitions, comparison strategy where feasible, missing-data treatment, and a rule for publishing unfavorable results.
Method & Limits
How to read this brief.
Method
Targeted synthesis completed 23 September 2026. Eight primary sources were inspected: WHO methodological guidance, three OECD governance publications, current UK appraisal and evaluation guidance, official legislative-drafting guidance, and a study-author account of randomized evaluations in Brazil. The randomized evidence supports a narrow claim about policy adoption; the remaining material provides methods and institutional principles. The workflow, artifacts, measures, and options are original proposals, not a description of established Foundation operations or a systematic review.
Limitations
No original legislation, local legal opinion, fiscal model, stakeholder consultation, or impact evaluation was commissioned for this note. UK guidance is used for its explicit methods rather than presumed global legal applicability. A research ledger cannot remove publication bias, political disagreement, or uncertainty about transferability. Implementation and evaluation plans need a real institutional sponsor, data access, funding decisions, and participation from the people affected before any proposal can be treated as ready for adoption.
Unresolved evidence
The effectiveness and cost of the complete proposed Policy Lab workflow have not been tested.
Evidence about inexpensive municipal adoption does not establish how legislatures respond to costly, contested, or constitutionally sensitive proposals.
Cross-country policy transfer requires local evidence about legal authority, institutional capacity, public priorities, and implementation conditions.
Source Record
Read the official sources.
[1]
World Health Organization · 8 July 2022 · Accessed 2026-09-23
