Method
The instrument behind the number
A range is only worth what the method behind it is worth. This page is that method, in full: how a market is scored, what is watched between assessments, and the rules the number has to survive before it reaches a decision.
- Version
- GRS v3.0
- Weighted pillars
- 8
- Sub-risks
- 29
- Scale
- 0–10, anchored, non-linear
The instrument
How a market is scored
Every market we cover carries one number, produced the same way each time: 8 weighted pillars, 29 sub-risks, on an anchored 0–10 scale. GRS v3.0 is internal working method. It is not a product, and we do not publish country scores.
Seven bands, unequal on purpose
The top of the scale is compressed. The distance between elevated and high is a matter of degree; the distance between severe and extreme is a matter of kind, and the scale should not flatter it. Each band is drawn at its true width, and each one says what it means for a company already operating there.
- Low0.1–0.7Ordinary commercial risk. Geopolitics is not the binding constraint.
- Moderate0.8–1.5Watchable. Nothing that should change a plan on its own.
- Elevated1.6–2.3Worth a named owner and a defined trigger.
- High2.4–3.1Belongs in the plan with a stated contingency.
- Very high3.2–4.3Capital decisions should carry an explicit range.
- Severe4.4–6.4Continuity, not optimisation, is the question.
- Extreme6.5–10.0Presence itself is the decision under review.
What the number is made of
Eight pillars, each with a question it exists to answer and a weight that is published. A weighting that cannot be shown is a weighting that cannot be argued with — and a client who cannot argue with the weighting is being asked to take the number on trust.
- PSC
Political Stability & State Capacity
Can this state make a decision and carry it out?
- Government stability
- Policy continuity
- Institutions
- State capacity
0.16 - DSS
Domestic Security & Social Stability
Is the domestic order holding?
- Civil conflict
- Protest and unrest
- Terrorism and armed groups
- Organised criminal violence
0.14 - ISC
Interstate & Strategic Conflict
Is this state in, or heading into, a fight with another?
- Interstate war
- Territorial disputes
- Proxies and military escalation
0.14 - MFV
Macro-Financial Vulnerability
Can money get in, and can it get out?
- FX and inflation
- Reserves
- Convertibility and capital controls
- Sovereign stress
0.14 - ECF
Economic Coercion & Fragmentation
Is this market being used as an instrument by someone else?
- Sanctions
- Tariffs and restrictions
- Export controls
- Bloc fragmentation
0.12 - CDF
Critical Dependencies & Flows
What has to keep moving for this to work?
- Energy and chokepoints
- Logistics
- Critical minerals
- Supply chains
0.12 - SRI
Sovereign & Regulatory Intervention
Can the state reach into the business itself?
- Expropriation
- Contracts and licences
- Controls and taxation
0.10 - ESS
External & Systemic Spillover
What reaches this market from outside it?
- Regional contagion
- Third-party alliances and dependencies
- Systemic and cyber escalation against critical infrastructure
0.08
8 pillars · 29 sub-risks · weights sum to
1.00Two layers, one cap
A score that only moves slowly is a reference work. A score that moves with the news is a feed. The instrument carries both, and keeps them apart.
- Structural
- Reviewed on a standing cycle
- The standing condition of the market: institutions, dependencies, macro-financial footing, exposure to coercion. It moves when the underlying situation moves, not when the coverage does.
- Dynamic
- Moves with live signals
- What is happening now, expressed as a deviation from the structural score and capped at plus or minus one point. A cap that cannot be overridden is what keeps a scoring system from becoming a news ticker.
±1.0
The dynamic layer can move a country by no more than one point against its structural score. The cap is the honest part: it is what stops a scoring system from becoming a news ticker, and it cannot be overridden.
What is watched
The chain that runs whether or not you asked
Five steps, in the order they actually run. Nothing moves an assessment without passing through all of them, and the last one has a name on it.
01Sources
Every source carries a grade. A ministry statement, a shipping manifest and a well-sourced report are not the same evidence and are never recorded as if they were.
02Signals
A signal is a specific observed change, with a date, an author and an attached piece of evidence — including evidence that argues against it.
03Situations
Related signals resolve into a situation with named actors and a defined escalation ladder, so that 'it got worse' has a step number.
04Triggers
Observable conditions with thresholds set in advance. A trigger firing does not silently change an assessment — it opens a review with an analyst's name on it.
05Review
A human decides whether the assessment moves, and records why. The reason is mandatory. There is no path through the system that skips it.
One risk, end to end
From the disruption to the decision
The chain run once for real: the channels it travels down, the markets it reaches, the range it puts on the table and the trigger that would change the answer.
Insurance moves before freight does
Roughly a fifth of seaborne oil moves through the Strait of Hormuz. When the corridor comes under pressure, underwriters reprice cover within days — continuously, and on judgement. Charter rates follow on schedule and on volume. A risk function watching freight is reading the confirmation, weeks after the window to act on it opened.
War-risk premium, Gulf transits
Share of hull value, indexed to plan assumption = 100
Above this, downside assumptions apply
Four structures, one family
Baseline, downside, severe and tail are variants of the same situation, each with the assumptions that define it and a probability range rather than a single number. Building them together keeps the probabilities coherent and makes the comparison meaningful.
The chain from the strait to a purchase decision
Insurance moves first, then freight, then energy, then headline inflation in importing markets, then currency and policy, then the discretionary spend of the customer. Naming each step is what later makes the number arguable instead of asserted.
War-risk insurance
Hull and cargo premia for Gulf transits; availability of cover by flag and operator.
Freight & routing
Charter rates, rerouting distance, transit time and schedule reliability.
Energy
Crude and refined product prices, regional gas, and the cost of everything indexed to them.
Inflation
Headline and imported inflation in energy- and food-import-dependent markets.
Currency & policy
Import cover, currency pressure, policy tightening and, at the extreme, capital controls.
Demand & purchasing power
Discretionary spend, subscription retention, advertiser budgets and price sensitivity.
Two assessments, held apart
Vulnerability belongs to the market: import cover, currency regime, fiscal room. Exposure belongs to your business: revenue, cost base, logistics, payment routes. We assess them separately and combine them in the open, because a reader who cannot pull the number apart cannot argue with it.
Türkiye
High energy import dependence, currency highly sensitive to external shocks
Price pass-through: Full
Subscription revenue, local pricing, hardware landed cost
Egypt
Thin import cover, subsidised fuel, managed currency regime
Price pass-through: Administered
Payment collection reliability, advertiser budgets
Pakistan
Constrained external position, import compression already in use
Price pass-through: Partial
Discretionary spend, subscription retention
India
Large absolute import bill, fiscal room to absorb the first round
Price pass-through: Partial
Scale of the growth plan, hardware supply chain
The same shock, four different outcomes. India carries the largest absolute exposure and the lowest transmission intensity; Türkiye is the reverse. A single regional assumption would be wrong in both directions.
What it costs, as an interval
Landed cost of goods, freight and energy-linked inputs, indexed against the plan. Each variant produces a band, not a point. A range without a visible assumption register is not defensible, so the two are delivered together.
- Baselinep 55–65%101–103
Crude +5% to +12% · Freight +10% to +20% · Headline inflation, exposed importers +0.3 to +0.8 pp
- Downsidep 20–30%104–109
Crude +15% to +30% · Freight +35% to +60% · Headline inflation, exposed importers +0.9 to +1.8 pp
- Severep 6–12%110–122
Crude +35% to +60% · Freight +80% to +140% · Headline inflation, exposed importers +2.0 to +3.5 pp
- Tailp 1–4%123–148
Crude +60% to +120% · Freight +150% to +300% · Headline inflation, exposed importers +3.5 to +7.0 pp
Indexed landed cost of goods, freight and energy-linked inputs. Plan = 100.
What would change our mind
Each assumption that can move becomes an observable with a threshold set in advance and a decision attached. A crossing convenes a review; it never rewrites the assessment on its own.
The tick sits at the same point on every track — the distance to it is the comparison
War-risk premium, Gulf transits
Above 0.7% of hull value, sustained for two weeks
Approaching
On crossing — Move to downside assumptions. Analyst review convened within five working days.
Carrier transit suspensions
Two or more major carriers suspend Gulf transits for more than seven days
Within tolerance
On crossing — Escalate to the severe branch. Activate the pre-agreed freight contingency.
Regional benchmark crude
Sustained above +25% versus the plan assumption for ten trading days
Within tolerance
On crossing — Re-run the pass-through model. Bring the pricing decision forward.
- Pass-through policy
- Agree, in advance, how much of a freight and energy surcharge is absorbed and how much is repriced — by market and by channel, before the pressure arrives.
- Hedge tenor
- Extend cover on energy-linked input costs across the window in which the downside and severe branches remain live.
- Launch sequencing
- Re-sequence hardware-dependent launches in the four exposed markets rather than absorbing landed-cost volatility at launch.
- Provisioning band and review date
- Set the provision against the probability-weighted range, with the date and the evidence that would cause it to be revisited.
Illustrative example — constructed to show the method. Every parameter is a stated assumption, and no figure here comes from client work.
Governance of the number
What separates an instrument from an opinion with decimals
Five rules. Each one is a constraint the system enforces on us, and each one costs us something — which is the only reason to believe it is real.
- 01
Scores are append-only
A score is never edited. A new one is recorded, with its author, its date and its reason, and the previous value stays readable. You can always ask what we thought last quarter, and why we changed our mind.
- 02
Confidence is not evidence
How sure we are and how good the evidence is are two different things, recorded separately and never averaged into one comforting number.
- 03
Contrary evidence is recorded
Evidence that undermines an assessment is attached to that assessment, not left out of it. An analysis with nothing against it has not been tested.
- 04
Published figures stay tied to their source
Every number in a document points back to the object that produced it. If that object moves after publication, the document says so rather than quietly ageing.
- 05
We keep score of ourselves
Probability statements are logged and scored against what actually happened. Calibration is a property we measure, not a claim we make.
Each of these is enforced in the system itself. None of them is a policy an analyst is asked to remember.
None of this is the deliverable.
The scale, the pillars, the chain and the governance exist for one reason: so that a short document, written in your own figures, can be put in front of people who have to decide — and hold when they push on it.