Geopolitics has no base rate

Two days on the lake

Lausanne, 4 November. IMD’s Geopolitics for Boards programme: a room of board members from more than ten countries, two days, one subject: how a board governs geopolitical exposure the way it already governs credit, cyber and succession. Simon Evenett, IMD’s professor of geopolitics and strategy, ran it. James Davis, professor of international politics at St. Gallen and a former dean there, taught the sense-making. Christian Dussey, until this spring director of the Swiss Federal Intelligence Service and before that ambassador in Tehran and head of the foreign ministry’s crisis centre - the desk that ran the case when two Swiss citizens were held by the Taliban in Pakistan for eight months - taught leverage and cyber from the inside.

Geopolitics is not new on a board agenda. What is new is that it is being used as an instrument: in January the World Economic Forum put “geoeconomic confrontation” - sanctions, tariffs, export controls, the plumbing of trade and finance used as leverage - at the top of its global risk ranking for the first time, above armed conflict. The two days were about what a board does with that.

Why traditional risk management cannot hold it

The programme opened with a question to the neighbour: a factory fire, a defective product, a lost key customer - your company knows how to handle those. What is different about a geopolitical event?

Start with what makes the fire manageable. It has happened often enough that there is a base rate. The base rate is why an insurer will price it, why a procedure exists for it, and why it sits comfortably in a heat map with a likelihood on one axis. Take the base rate away and all three go with it. An export ban, a frozen account, a licence that is not renewed: too rare to learn from, too linked to trace - one government’s move triggers another’s - and too contested for the room to agree what it means. You can’t predict it, you can’t insure it, and nothing you do stops it happening. You can only manage your exposure to it.

Then the harder part, which is not about the event but about us. The toolkit assumes the rules hold. In geopolitics, states change the rules. So the assumptions a board carries without ever writing them down - we were not the target; the damage stays local; normal comes back; the business model survives; we can stay neutral - are exactly the ones most likely to be wrong. The challenge is not only the shock outside. It is the map inside our heads, formed in thirty years when the rules held.

Possibilities, not probabilities

What replaces the register is a different question. Under deep uncertainty - the fog in which you do not yet know which questions to ask - probability is not available even in principle, so the first question is not “how likely?” but “what is going on here?” Work with possibilities, not probabilities: write the scenarios down before they happen, test the company against each, and keep more than one alive, because the same facts usually fit several and a board that settles on one too early has made its biggest decision without noticing.

The map

The tool for that is a map. Not a dashboard: a drawn causal picture of the forces acting on the company - rivalry, resource scrambles, fiscal strain, technology controls - linked to the markets that matter and then to the choices the board actually controls. The programme worked through one for a global food company. It fitted on a page: about sixteen boxes in four layers, from trade policy and currencies at the top down to cost of living, sales and margins, with every arrow marked as making things worse or holding them back. A map makes you write your assumptions down and name the signal that would break each one.

Without a map, data is noise.

The map has to contain the plumbing. Supply chains, payment rails and the data centres a company’s software runs in were built for efficiency; the hubs they pass through can now see everything and switch anyone off. That is what “leverage” means in practice, and it is why the intelligence lens Dussey taught - what does the state want, what leverage does it have, what would make it use it, what signal would tell us we are wrong - belongs on a board’s page and not only in a ministry.

The proof

And then the proof that the map is not enough on its own. Two exercises, two days apart. On day one, a red team: each table first attacked a fictional company - find the three dependencies you would squeeze - and every table found them in minutes, because they were in the company’s own description of itself. Then the same tables switched sides and defended it, with the attack plan on the table in front of them, and the second half was the harder one. Knowing where you will be hit is not the same as knowing what to do about it. On day two, a board simulation: a decision on acquiring a company with dual-use products next to a conflict zone - yes, no, or yes on conditions - with the scenario getting harder each round as new facts arrived. Different exercise, same lesson, this time under the clock.

Success is crisis organisation times practice. Anything times zero is zero.

For boards

The job is not to have a view on the world. It is to make sure three things exist and get used. A map, owned by management, challenged by the board, assumption by assumption, with the signals that would break each one named in advance. A stress test that uses it - not the kind with a probability attached, the kind that says this route closes, this region switches off, this bank holds our receipts for thirty days. And a drill with a geopolitical scenario in it, dated; if the date is “never”, that is the agenda item.

For start-ups

You did not choose your exposure; it came with the defaults. The same three things, in your units. A one-page map of what you depend on. Six lines of stress test: cloud region, single source, payment rail, a rule change in your main market, cap table - which investor, or which of their backers, could get you flagged by a bank’s sanctions screening - and days of payroll with money on hold. And one afternoon a year in which the founders rehearse what they would do on day one.

What I still don’t know

One thing is clear to me: a board that cannot name the three dependencies an adversary would squeeze first is not ready, however good its risk register. What I still don’t know is the next step - when that justifies a team of its own, with a budget line, and how a board would tell a year later whether it was worth it. I came out of Lausanne with a map and a drill. Not with that answer.

Geopolitics has no base rate. Which is why it cannot be priced, and why it has to be rehearsed.

Notes. The frameworks are IMD’s, from Geopolitics for Boards, November 2026. Uncertainty: John Kay and Mervyn King, Radical Uncertainty (2020). Networks as leverage: Henry Farrell and Abraham Newman, Underground Empire (2023). Board pillars: Didier Cossin, High Performance Boards (2020). Risk ranking: World Economic Forum, Global Risks Report 2026.