2.1 Foresight in the Policy Cycle

Foresight in the Policy Cycle

Foresight creates value when it changes a decision. Beautiful scenarios that remain detached from agendas, budgets, regulation, implementation, and evaluation are intellectually interesting but institutionally weak. This session shows how futures methods can be connected to the policy cycle while respecting an important reality: policymaking is rarely a neat linear sequence. Problems, interests, evidence, authority, and political opportunity interact continuously.

1. Where Foresight Enters Policy

At the agenda-setting stage, horizon scanning can identify emerging issues before they become crises. During problem definition, systems mapping reveals relationships, feedback loops, and groups affected by the current framing. In option development, scenarios test whether proposed interventions remain useful under different conditions. During implementation, indicators track changes in the operating environment. In evaluation, foresight asks not only whether a programme achieved its original objective, but whether that objective and its assumptions remain appropriate.

2. Horizon Scanning

Horizon scanning is the systematic search for signals of change. Sources can include scientific research, patents, regulation, investment, social movements, cultural shifts, environmental observation, and frontline experience. A signal is not important merely because it is new. It becomes strategically relevant when it may alter a system, challenge an assumption, create a new capability, or change the distribution of risk.

A disciplined scan records the source, date, domain, affected actors, degree of uncertainty, and possible implications. Learners should separate evidence from interpretation. One article does not establish a trend, but a weak signal can justify further investigation. Diversity among scanners is essential because professional training and institutional position influence what people notice.

3. Drivers and Critical Uncertainties

Drivers are forces that shape the focal issue over time. They may be social, technological, economic, environmental, political, legal, or ethical. Some drivers are relatively predictable; others are deeply uncertain. Scenario work often identifies two uncertainties that are both highly impactful and genuinely variable. Crossing them produces contrasting conditions in which a policy can be tested.

The goal is not to reduce the world to two axes. It is to create a manageable structure for strategic conversation. A strong scenario contains a coherent causal logic, identifies winners and losers, and avoids both utopian optimism and catastrophic theatre.

4. Backcasting and Robust Options

Forecasting moves from the present toward possible outcomes. Backcasting begins with a desirable future condition and asks what would have to happen, in reverse sequence, to reach it. This is useful when incremental improvement cannot meet a long-term target. Backcasting should be paired with political and institutional realism: who has authority, what resources are required, which interests may resist, and what early steps build capability?

Policy options can then be classified as robust, contingent, or experimental. Robust actions perform reasonably well across several scenarios. Contingent actions become appropriate when specified indicators appear. Experiments generate evidence at limited scale. This portfolio approach is often wiser than placing all resources behind one assumed future.

5. Wind-Tunnelling Policy

Wind-tunnelling tests an existing or proposed policy against multiple scenarios. For each scenario, ask whether the policy remains feasible, effective, legitimate, and equitable. Identify failure points, unintended consequences, and dependencies. Then modify the policy with safeguards, triggers, review clauses, modular components, or alternative pathways. The result is not a perfect future-proof policy; it is a more adaptive design with visible assumptions.

6. Evidence, Uncertainty, and Communication

Foresight must not blur empirical evidence and imaginative exploration. A useful policy product labels established facts, contested interpretations, assumptions, and speculative possibilities. Probability language should be used carefully. Decision-makers need to understand both what is known and why uncertainty matters. Visual simplicity is valuable, but it must not erase disagreement or imply precision that the evidence cannot support.

Prototype Exercise: Policy Foresight Brief

Prepare a two-page prototype for the challenge selected in Module 1:

Write a precise focal question with a ten-year horizon.
Document six signals from at least four credible source types.
Group the signals into four drivers of change.
Select two critical uncertainties and construct four short scenario conditions.
Test one current policy or organisational strategy in all four conditions.
Recommend one robust action, two conditional actions, and one bounded experiment.
Define five indicators that would trigger review.

In the AI Laboratory, an enrolled learner may use the linked foresight prototype to cluster signals or generate challenges to a scenario. The learner must verify sources, remove invented claims, and explain any significant AI contribution. AI is a critical-thinking partner, not an authority.

Quality Checklist

Is the focal question decision-relevant and time-bounded?
Are sources traceable and sufficiently diverse?
Do scenarios differ in causal structure rather than cosmetic detail?
Have affected groups and distributional consequences been considered?
Are recommendations linked to institutional authority and resources?
Are assumptions and review triggers explicit?

Key Takeaway

Foresight strengthens policy when it is embedded at specific decision points. The essential movement is from signals to interpretation, from interpretation to alternatives, and from alternatives to actions that can be monitored and revised. The quality of the process depends on evidence, plurality, transparency, and a clear connection to institutional responsibility.

Slide deck

A ten-slide HTML presentation summarising this lesson. It prints, translates and reads on any device.

A ten-slide presentation of this lesson. Every slide is plain HTML, so it prints, translates and reads on any device.

Slide 1 of 10