RIPPLE-4 Policy Thinker’s Field Guide

RIPPLE-4 rapid assessment for Saint Lucia

Research by Kevin L. Michel

This section is deliberately self-contained. Copy it into a Cabinet-paper template, programme concept note, regulatory impact assessment or AI policy-analysis prompt.

The 30-minute protocol

Minutes 0-5 - Reference the decision

Write the decision being requested in one sentence.

What measurable public outcome is sought, for whom and by when?
What happens under the no-policy counterfactual?
What constraint is binding: money, skills, infrastructure, law, trust, time or foreign exchange?
Which rights, fiscal limits, environmental assets or public obligations are non-negotiable?
What result would constitute failure?

If the decision, counterfactual or success measure cannot be stated clearly, stop and redefine the proposal.

Minutes 5-10 - Identify immediate incidence

Record who receives the immediate benefit and who carries the immediate cost. Check three forms:

Cash: revenue, expenditure, prices, income and debt service.
Real resources: labour, land, water, energy, equipment and administrative time.
Balance sheets: government liabilities, utility revenue, bank exposure, household debt and foreign-exchange demand.

Ask who is excluded and which cost has merely moved off-budget, into the future or onto another institution.

Minutes 10-15 - Predict adaptation

Assume actors respond to the policy, not merely receive it. For households, firms, public officers, financiers, political groups and international partners, ask:

What behaviour becomes more rewarding?
What rule will be avoided, gamed or reinterpreted?
What investment will be accelerated, postponed or abandoned?
Who gains an incentive to preserve the policy after its original purpose expires?
Which external actor can rewrite the payoff?

Minutes 15-20 - Propagate the cascade

Place the instrument at the centre of four rings labelled O1 to O4. For every material arrow record direction, delay, precondition, actor, probability range and evidence confidence.

Trace the path through seven ledgers: households; firms; government; external account; finance; institutions and trust; natural and infrastructure assets. Circle any path that returns to the Consolidated Fund, a utility, bank, port, airport, tourism market, scarce technical team or foreign-exchange pool.

Minutes 20-25 - Load uncertainty

Assign low, central and high conditional probabilities. For material effects score magnitude, reach, persistence, irreversibility, coupling and control deficit.

Use two screens:

Expected System Value = sum[p x magnitude x reach x persistence x time factor]

Critical Tail Exposure = sqrt(p upper) x severity x irreversibility x coupling x control deficit

The first compares expected effects. The second prevents a low-probability catastrophe from disappearing inside an average. Neither replaces legal, ethical, environmental or rights-based limits.

Minutes 25-30 - Engineer learning and exit

Choose one posture:

Approve: effects are understood and controls adequate.
Pilot: uncertainty is material, but learning is reversible.
Sequence: enabling infrastructure, law or capability must come first.
Redesign: benefits remain plausible, but the cascade is unsafe.
Hedge: proceed with insurance, reserves, caps, redundancy or contingency finance.
Pause or reject: exposure is irreversible, uncontrollable or incompatible with public obligations.

Every approval must name an owner, data source, review date, threshold, automatic response and stop rule.

Ten dangerous blind spots

Blind spot	Diagnostic question
Headline fixation	Are we treating the announced output as the public outcome?
Policy-chain neglect	Have we stopped before adaptation and feedback?
Static-actor bias	What changes because incentives, information or fairness perceptions changed?
Fiscal illusion	Which tax expenditure, guarantee, utility loss, maintenance cost or future obligation is missing?
Average-case blindness	Which adverse effects share a cause or critical node?
Present/electoral bias	What falls beyond this Budget or term of office?
Capacity optimism	Which existing function loses staff, attention or procurement bandwidth?
Goodhart’s trap	How will the target be gamed, relabelled or substituted?
External-sovereignty illusion	Which foreign visa, banking, airline, trade or regulatory rule can change?
Sunk-cost escalation	What evidence would make us stop despite money already spent?

Feedback-loop typology

Loop	Pattern	Example
Accumulation	Surplus funds capability, producing further surplus	Mauritius used rents, education and infrastructure to build export capability
Cost spiral	Falling scale raises unit costs, inducing further exit	Caribbean banana volume and shipping economics
Fiscal ratchet	Temporary revenue creates sticky expenditure and political dependence	A windfall becomes wages or open-ended commitments
Credibility loop	Weak safeguards provoke external response, reducing programme value	Citizenship concerns, visa action and weaker demand
Capacity congestion	Policy demand overloads scarce people or infrastructure	Simultaneous hotel projects raise construction cost and water demand
Scarcity balance	Growth raises the price of a binding input and slows itself	Lewis’s industry-food-wage-profit mechanism
Rebound/substitution	Efficiency or subsidy changes use elsewhere	Lower energy cost increases use or shifts peak demand
Learning/trust	Visible delivery improves adoption and legitimacy - or failure reverses it	Water tariff, capital works and willingness to pay

Systemic adjustment or critical tail?

Ask four questions:

Does an accounting identity, physical constraint or legal rule make the effect difficult to avoid?
Does an affected actor have both a clear incentive and the capability to respond?
Has the adjustment appeared in a relevant reference class?
Does it occur under optimistic, central and adverse scenarios?

If at least three answers are yes, put the effect in the baseline as a systemic adjustment, not a risk footnote.

A critical tail has a low central estimate but a meaningful upper probability and combines severe harm with irreversibility, concentrated exposure, correlated losses, weak early warning or low government control. Prevent, cap, insure, stage or preserve an exit option. A severe systemic adjustment and a critical tail can coexist.

Probability structure	Consequence	Default response
High lower bound	Moderate	Include in baseline and budget
High lower bound	Severe	Redesign unless benefits clearly dominate
Low centre, meaningful upper bound	Moderate	Monitor or insure
Low centre, meaningful upper bound	Catastrophic or irreversible	Prevent, stage, hedge or do not proceed

Five questions to ask aloud

What will people do because this policy exists?
Where will the cost migrate?
Through which shared institution or bottleneck will the effect return?
Which external actor can rewrite the payoff?
What observable evidence would make us change course?

RIPPLE-4 effect card

Decision:
Responsible minister and implementing owner:
Decision date / evidence vintage / review date:
Intended outcome and deadline:
No-policy counterfactual:
Binding constraint:
Non-negotiable safeguards:

Order	Effect and mechanism	Actor / ledger	Sign	Delay	Probability L/C/H	Magnitude range	Evidence confidence
O1 direct
O2 adaptation
O3 propagation
O4 inherited state

Critical return paths: Which effects return to the same budget, utility, bank, infrastructure system, sector, foreign-exchange pool or external gatekeeper?
Reinforcing and balancing loops:
Thresholds and lock-in:
Distribution: Who gains, who loses and who can organise?
External dependencies: Which foreign government, market, lender, insurer or regulator can alter the outcome?
Expected System Value:
Critical Tail Exposure:

Leading indicator	Data source / frequency	Trigger threshold	Automatic response	Accountable owner

Posture: Approve / Pilot / Sequence / Redesign / Hedge / Pause
Reason and pathway that changed the decision:
Next review:
Revision or stop rule: