How Long Geopolitics Actually Stays in the Price
 |
When a conflict escalates, the market moves and the explanations arrive within the hour. Everyone can tell you why it fell. Almost nobody tells you the more useful thing, which is how long it stayed fallen.
That second question has an answer, and the answer is shorter than most people expect.
The Record
Research covering decades of geopolitical shocks — coups, invasions, terror attacks, escalations — finds a pattern that repeats with unusual consistency.
The average total drawdown across those events is about 4.7%. Narrow it to wars and major military operations and the average deepens to roughly 7%.
The average time to reach the bottom is about 19 days. The average time to recover the entire loss is about 42 days.
Across the major military conflicts since the Second World War, 19 of 20 saw the index back to where it started within roughly a month.
Read those numbers together and a shape appears. The typical geopolitical shock is a 7% move that resolves inside two months. Not nothing. But considerably smaller and considerably shorter than the volume of commentary around it would suggest.
Reading That Number Correctly
There is a second statistic that gets quoted alongside those, and it is the one most often misused. Looking at a few dozen geopolitical events across eighty-odd years, the index has averaged something like 7.8% over the following twelve months, and finished higher about 68% of the time.
That gets reported as stocks go up after wars. It is not what the number says.
Compare it to what the index does over an ordinary twelve months — any twelve months, chosen at random, with no event attached — and the two are hard to tell apart. The post-event year is not a boom. It is an average year.
That is the actual finding. Twelve months later, the event has mostly stopped being visible in the data. It did not cause a rally. It stopped mattering.
Which is a far more useful thing to know than a direction, because it tells you what kind of information you are holding when a headline arrives: something that will move prices this week and be difficult to locate in the chart by next spring.
Why the Effect Is Usually Short
The mechanism is simpler than it looks, and it is worth stating plainly because it also tells you when the rule breaks.
A market does not price events. It prices cash flows and the risk attached to them. An event only matters to a share price to the extent that it changes what a company will earn, or what someone will pay for those earnings.
Most geopolitical events have a narrow path to that. A conflict in a region that supplies little to the global economy changes almost no company's revenue. The initial fall is not a re-pricing of earnings — it is a re-pricing of uncertainty, and uncertainty is the one input that resolves on its own simply by time passing. Two months later, whatever happened is known rather than feared, and the discount for not knowing comes back off.
The commentary, meanwhile, describes the event. The market is pricing the path from the event to a cash flow. Those two things are only occasionally the same conversation.
When It Is Not Short
The same research carries a caveat that matters more than the averages: the deepest and longest drawdowns come from energy supply disruptions.
That is the exception that proves the mechanism. An oil shock is not a distant event with a narrow path — it is an input cost for nearly every business on earth, and a direct hit to every household budget. The path from event to cash flow is short, wide, and unavoidable.
The other durable category is structural: when a conflict changes the plumbing rather than the mood. When roughly $300 billion of one country's reserves stopped being spendable in 2022, that did not resolve by time passing. It changed how the institutions that hold the world's savings think about where those savings sit — and they are still acting on that conclusion years later.
So the working test is not how serious does this sound. It is: does this change a cost, a contract, or a settlement system — or does it only change how confident people feel this week? The first kind stays. The second kind has a half-life of about six weeks.
The Average Is Made of Two Different Things
There is a trap in that 4.7% figure, and it is worth stepping into deliberately.
An average is not a typical case. That number is the midpoint of a distribution containing a great many events that did essentially nothing and a small number that did something enormous. Almost no individual event has ever produced a 4.7% drawdown. They cluster near zero, with a long tail.
The tail is where the useful information is.
In late 1973 an oil embargo took the price of crude up several times over within months. What followed was not a 7% dip resolving in six weeks. It was one of the deepest and longest bear markets of the post-war era, and the recovery was measured in years rather than days.
Nothing about that event was more dramatic than other conflicts of the period. What was different was the path. Energy is an input to almost every business and every household simultaneously. There was no waiting for uncertainty to resolve, because the cost had already changed and everyone could see it on an invoice.
Now hold that next to the ordinary case. An escalation in a region that exports little, sits on no major shipping lane, and settles nothing systemically important changes essentially no company's costs. The market still falls, because a market prices uncertainty and uncertainty went up. Then the uncertainty resolves — not necessarily well, just resolved — and the discount comes off.
Same headline volume. Completely different arithmetic underneath.
And here is the uncomfortable part: in the first week, the two look identical. Prices fall, coverage escalates, and confident people appear on both sides. The distinction between a shock and a change is not visible in the price action. It is visible in the path — and the path takes a while to trace.
Which means this is not a trading rule. Nobody can use "most events resolve in six weeks" to decide what to do on day three, because the events that do not resolve in six weeks look exactly the same on day three. What it can do is set your default. If most of these are noise and you cannot tell which is which early, then the position that survives both is the one sized to sit through either.
What This Looks Like From Korea
There is a version of this that is obvious from here and invisible from most places.
Korea has lived with a geopolitical headline as permanent background for seventy years. Every so often something escalates on the peninsula and the international coverage treats it as a market event.
The local market's reaction is almost always smaller than outside observers expect, and shorter. Not because anyone here is indifferent, but because a risk that has been present continuously for decades is already inside the price. It is not news. It is a standing condition.
That distinction — between a risk that arrives and a risk that has always been there — is the whole thing. What people call the discount on this market is not a reaction to events. It is a permanent subtraction that events barely move, in either direction. A headline can only reprice a risk that was not already priced.
Why Posts Like This Keep Getting Written
It is worth asking why geopolitical market commentary is so abundant when its predictive record is so thin.
The honest answer is that it is unusually easy to write and unusually hard to check.
A geopolitical story has a complete narrative — actors, motives, a beginning and an anticipated end. It explains a price move in a way that feels satisfying. And by the time anyone could evaluate whether the call was right, the situation has changed enough that nobody bothers.
Contrast that with writing about, say, the arithmetic of fees. There is no narrative, no villain, and the claim can be checked with a calculator by anyone who cares. One of those is much more comfortable to produce.
This page previously held a piece of exactly the first kind. It predicted a ceasefire, called a market boom "inevitable," and assigned motives to named political figures I had no way of knowing. It was confident, it was tidy, and none of it was checkable. That is the genre working as designed, and it is why the genre is worth distrusting — including when I am the one writing it.
What to Do With It
Very little, actively. Quite a lot, in terms of what you decline to do.
Match your reaction to the half-life. If the typical shock resolves in about six weeks and your money is for a decade from now, the correct response to a headline is usually nothing at all. Selling into it means paying a real cost to avoid a temporary one — and the days you miss by stepping out tend to cluster exactly in the recovery.
Compare the half-life to your own horizon. A six-week shock is not equally dangerous to everyone holding through it. If the money is needed in six months, a six-week drawdown is a real and immediate problem, and the right response was to not have it there in the first place. If the money is for a decade from now, the same six weeks is a rounding error you happened to watch in real time. Same event, same percentage, completely different exposure — and the variable that decides it is a fact about you rather than a fact about the news.
Check the path, not the drama. Before assuming an event matters to what you own, ask which cost, contract, or settlement system it touches. If you cannot name one, you are probably looking at a six-week story.
And notice who has already sold. A market that has fallen for a week on a headline has usually already removed the people who could not hold it, which is a separate and more useful thing to look at than the headline itself — the subject of what actually moves a market on an ordinary day.
If the honest answer is that you cannot sit through a 7% drawdown without acting, the problem is not the geopolitics. It is the size of the position, and that is decided before the headline arrives, not during it.
A Closing Note
None of this says the events do not matter. People are living inside them, and the human cost of a conflict is not measured in index points and never should be.
It says something narrower and, for an investor, more useful: the market is a poor instrument for measuring how much a thing matters. It measures one specific thing — the path from an event to a cash flow — and it measures that path quickly and then goes quiet.
The headline will be the loudest thing in the room for about six weeks. Then it will be a small notch on a chart that nobody points at, and the money that stayed put will have been the money that noticed the difference between a shock and a change.
Data as of 2026-08-08. Sources: LPL Research studies of S&P 500 behaviour around geopolitical events and post-war military conflicts. This post is observation, not investment advice.
Related Posts:
The Cost of Stepping Out — Why Markets Punish the Absent
Central Banks Are Buying Gold Like It's 1971
The News Was Ordinary. The Move Was Not.
Nobody Pays Off a National Debt
Visuals on this post are AI-generated. The author works with AI as a research and drafting assistant; topics, judgments, and final edits are the author's own. This post is observation, not investment advice. See full Disclaimer for details.
