Key takeaways:
- We looked at nearly 40 conflicts going back to WWII and found equities sell off initially but tend to recover, unless it's a world war, a lasting energy shock, or there’s something else going on. Historically, stocks care a lot more about a credit crisis than even a major war
- Gold has been a steady hedge, but not this time; oil is fast-acting but fades quickly. Meanwhile, yields tend to grind higher as inflation filters through, and the dollar tends to drift up as lingering war angst finds a home in the safe-haven currency
- A dedicated ‘war bucket’ could help reduce portfolio risk. One idea: a diversified basket spanning oil futures, precious metals, defence and fossil-fuel stocks, quality, momentum, trend following, and S&P puts
My mother had high-minded ideals about not letting me play with violent toys. It was at the point where I chewed my toast into the shape of a gun in the desperate pursuit of weaponry, that she caved to the inevitable. I know I am not the only investor who said their role model was Warren Buffet but really dreamed of being James Bond all along. And in times like these, that suppressed vocation finds some pale release in the form of a transformation, from portfolio manager to armchair military strategist.
No, enough, you’ve read too many of those notes, so in an object lesson in self-denial, I’m going to go light on discussion of asymmetric warfare and materiel (with an ‘e’) and instead focus on conflict price action, and what the ‘war bucket’ in your portfolio might look like, and indeed whether it’s worth having one at all. Two caveats. First, war is always a human tragedy, compared to which the investment ramifications are a distant, distant second. And secondly, events move fast, and in both directions. By the time you read this, the current Middle East War may be at an end. But as The Book says, until He returns, “you will hear of wars and rumours of wars, but see to it that you are not alarmed”.1 Hopefully this ‘cut out and keep’ piece stands the test of time, such that you might not be alarmed when the next conflict arises – from a portfolio perspective, at least.
Part one: what happens?
What happens from a market perspective in a geostrategic event? And what counts? We could go back and forth on this through the entire note, so in the interests of keeping momentum (kind of a thing, for us), let’s agree on the list shown in the appendix. A combination of Claude, the investment banks and some of our own nous. Crucially, we have selected these dates before looking at performance data.
In this section, I’ll take a brief tour through five asset allocation primary colours, for different subsequent periods following the kick-off date. I suspect you can look at a picture as well as I can, but herewith some observations for the words folks.
The buildup
As with the Spanish Inquisition, the greatest weapon of war is surprise. If we look at the 20 days prior to the outbreak of each geopolitical event in our sample, the positive hit rate for the S&P 500 is 51%. For the dollar? 54%. Gold: 50%.2 You get the idea (not shown graphically because it would be too boring). As much as it pains the aforementioned armchair chin-strokers, no one has a clue. If your portfolio has a war-hedge component, it’s a permanent component: don’t bother trying to time troop movements.
Equities
Stocks are risk-on assets. War is a risk-off event. It is natural that as the world grapples with loud, hard-to-explain noises, equity investors tend to sell first and ask questions later. On average, per Figure 1, the bourses are negative in the week and month following the event. But the market folklore is “buy to the sound of cannons, sell to the sound of trumpets”.3 The data bears this out to some extent, with returns swinging back into positive territory on the longer look-forwards, and in the green in more than 70% of instances six months after the action begins (hover over bars to see hit rates and best and worst events). Moreover, you almost always get a sharp relief rally at some stage during the 12 months, even if it ends up as a bull trap. The median largest 20-day equity move in the 12-month period following 35 completed events is 10%.4 I shrink at the sense of war profiteering, but from a pure investment perspective, periods of conflict may warrant at least being alert to pockets of market resilience.
Figure 1: Forward S&P 500 price returns following war dates (defined in appendix)
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
It is worth referring to the times when the Rothschild heuristic fails. Germany’s invasion of France and the Low Countries in World War II (10 May 1940), the Yom Kippur War (6 October 1973) and 9/11 (11 September 2001) saw stocks down 22%, 38% and 16%, respectively, in the first year. For me, this neatly encapsulates three scenarios where the theory comes unstuck.
- It’s the big one (WWII)
- It has a large and persistent effect on energy markets (Yom Kippur War)
- Something else is going on (9/11 – the initial market recovery from the attack on the World Trade Center was drowned out by a second leg in the dot-com crash)
Where does this leave us today? I’ll take a pass on existential risk. Lasting disruption to fuel prices? Not beyond the realms of possibility. Other stuff we’ve temporarily forgotten? Private credit, the ‘SaaSpocalypse’, the return on investment (ROI) on AI capex, fiscal sustainability… No shortage of candidates.
Gold
The most reliable war hedge there is, at least on historic precedent examined here. ‘Fiat’ means ‘let it be done’. Fiat currencies are a kind of alchemy, whereby a government pronouncement imbues value to paper. There’s nothing like a bit of wartime turbulence to have people question the pronouncer, and revert to old-school money. That big red deficit-spend button can look mighty tempting when you’re in a fight. Per Figure 2, gold is positive 85% of the time in the five post-event days, stays above 60% on virtually every look-forward, and runs well above the unconditional mean on all.
The drawdown we have seen in the yellow metal this time around appears to be unprecedented. Around -16%, from 27 February through 23 March 2026, close to close. If we look at the 17 trading days following prior episodes listed in Appendix 1, the next worst result on the same timeframe is the -3% that came following the US invasion of Grenada on 25 October 1983. At the end of last year, we wrote about gold’s strange behaviour on the way up, and so far it looks like that weirdness is extending to the way down.5
Figure 2: Forward gold price returns following war dates (defined in appendix)
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
Duration and the US dollar
I group these together, because I think the message is similar. And that is, while the heat of war price action (usually) cools quickly, it can leave an underlying dis-ease, manifesting in lingering concern on inflation and the unpredictability of events.
Wars almost always have some impact on inflation, even if ‘transitory’ (*winces*), and therefore your base case for yields is higher. This is evidenced in Figure 3, and particularly on a longer horizon. In the 12 months following our event dates, yields are higher 62% of the time, and by an average of 28 basis points (bps), miles above the unconditional. Perhaps this is reflective of the percolative effect of inflation information into the bond market. It takes time before the market properly twigs that supply-chain disruption has a cost, and it’s usually the consumer who is presented with the bill.
Is this effect present today? Quite possibly. I have seen estimates ranging from 20 to 50 bps on the Consumer Price Index (CPI) per 10% move in the oil price. If we take the average 2024 barrel (US$69), year-end inflation (2.7%), and assume we get some partial settling to US$85, that would imply 2026 could print somewhere between 3.2% and 3.9%. No heart attack, but well above target, and indeed, the completion of six years without a one-handle. Pre-COVID inflation expectations were around two. Post-COVID, this has settled at around 2.3%. Is 3% the new 2%? If you miss your target that consistently on the upside, there’s got to be some point where in practice you’ve got a new one. Yields are probably up as inflation expectations adjust, in this telling.
Figure 3: Forward change in 10-year US Treasury yield following war dates (defined in appendix)
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
I wonder whether a similar effect of dawning realisation is present in the dollar moves, per Figure 4. The greenback is a traditional ‘flight-to-safety’ asset. War is a flight stimulus, both because of crystallisation of fears around what inflation might do to economic activity, and because of a general fear of the unknown. Furthermore, war is a test of strength, and US military spending is formidable: close to US$1 trillion a year. You add up the next five (China, Russia, Germany, India and the UK), and you’re still over US$250 billion shy of that.6 While the dollar struggles to make up its mind in the early aftermath of conflict, as time passes it’s intuitive that, as we reflect on the risks this violent and unpredictable world presents, the unit of account for the biggest kid in the playground rises.
Figure 4: Forward trade-weighted USD returns following war dates (defined in appendix)
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
Oil
In the very early days of a major conflict, oil has historically risen sharply. The 60% hit rate on the one-week periodicity in Figure 5 undercooks it, as many of the negative readings are low magnitude: Operation True Promise (-3.5%), fall of the Berlin Wall (-1.8%), collapse of the Soviet Union (-1.2%) and the 2014 Syrian intervention (-1.2%). These were also arguably of small military bearing, even if historically consequential.7 The average return is 19 times the unconditional. While gold might be the most stable war hedge historically, black gold might as well be the fastest acting and most powerful.
Such excitement is hard to sustain, however, and on a six-month view, the hit rate falls below a coin toss, and the average return is a fifth of the unconditional. So while oil might be first to jump with the jitters of war, it is also one of the first to feel the geopolitical risk premium unwind, when the headlines calm down.
Figure 5: Forward price return to the first oil futures contract following war dates (defined in appendix)
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. Source: Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
Part two: the war hedge allocation
Given our discussion around the short-lived nature of war impact on risk assets, you would be forgiven for having stopped reading. And indeed, I would accept as a defensible position, something to the effect of, “I’ll wear the near-term geostrategic drawdown pain and look for opportunities in the hole, given I know the precedent is for it to quickly move to rear view.” Easier said than done of course: pain will make short-termists of us all. But even if you do have the requisite psychological fortitude to weather the volatility, as we have seen, there is an inflationary impact which risks harming more in the portfolio than just stocks. What is more, every so often it is something more, perhaps much more, and who wants to be Neville Chamberlain?
So, assuming you have at least some interest in introducing portfolio ballast against war risk, read on. My first stab: Figure 6.
Figure 6: Harry Neville’s war bucket
Problems loading this infographic? - Please click here
Elegant qualitative logic, though I say so myself. “But what about the data?”, the pernickety numbers, people cry out. It’s hard to run sims for some of these assets, but give me Claude Code and a Bloomberg terminal and I assure you I’ll come up with something. Figure 7: the valiant attempt, profit and loss shown alongside equities, for reference.
Figure 7: A simulation of figure 6 portfolio performance
Problems loading this infographic? - Please click here
Simulated and/or hypothetical. Past performance is not indicative of future results. This chart contains live historical returns and applies hypothetical allocations. Source: Bloomberg. Date range: 31 December 1926 - 25 March 2026. Portfolio comprised of the 8 elements listed in Figure 6, included as they appear historically. Oil futures is the return on the first contract for Brent / WTI. Precious metals is physical return to gold, and to silver where gold doesn’t have a daily return prior to the gold standard. Fossil fuel producers is the average of the oil and coal segments of Kenneth French’s 49 sector split, outperformance versus the market. Defence is the same, but for the ‘Guns’ sector. Quality is AQR’s QMJ factor. Momentum is the Kenneth French portfolio. Trend is simulated by Man AHL: The result of applying a simple trend-following rule to markets as they appear through time, starting in 1928. S&P puts are 10% OTM, on a 3-month roll. Puts are given a static 10% weight, everything else is equal risk-weighted.
All the usual caveats on these long-term simulation exercises apply, but it’s a good starting point. The top panel of Figure 8 shows portfolio performance specific to war start dates, in the same manner we went through in part one. The shape of returns is as we would hope, i.e. higher in war aftermath than in general, and with high hit rates. The lower panel (and perhaps this is a contortion too far) shows the 12-month period performance of the war hedge portfolio and equities following each of our start dates, back-to-back. Stocks outperform in pure performance terms; as we have seen, once the initial volleys are passed, risk appetite returns quickly. But the more than halving in risk, both on a standard deviation and max drawdown basis for the war hedge portfolio , feels to me like it shouldn’t be sniffed at. Perhaps I’m just gun-shy.
Figure 8 : Forward return to the war hedge portfolio following war dates defined in appendix (top panel) and P&L limited to 12 months following war dates (lower panel)
Problems loading this infographic? - Please click here
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This chart contains live historical returns and applies hypothetical allocations. Source: As per Figure 7.
A reckoning
War is the unknown. It is chaos. It is disruption: both of orderly price discovery and of economic activity. But it is also brittle and non-linear, often starting and ending unpredictably. For investors, I would humbly posit the following lessons. First, don’t try to time war. Second, consider opportunities, once you’ve been surprised by war. Third, have some segment of your portfolio which, at least in simulation, seeks to provide resilience during periods of conflict. Thus, you freak out less, thus you are better able to do (2). You’ve seen my suggestion. Fourth, don’t forget the big things you cared about before the war; they might come back. Fifth, watch energy markets like a bald eagle (apparently the best eyesight of any bird, who knew). They tend to go up in the near term, but if they don’t meaningfully retrace over the next 12 months, you should worry more broadly. For investors, whether they end up scenting victory in the morning like Colonel Kilgore, or going mad in the jungle like Colonel Kurtz, is likely in that balance.
1. Matthew. 24: 6 (NIV)
2. Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
3. Attributed to financier, Nathan Mayer Rothschild (1777-1836)
4. Bloomberg and Man Group, based on qualitative judgement from Man Group based on similar lists from Deutsche Bank and Morgan Stanley.
5. See: https://www.man.com/insights/road-ahead-golden-fears
6. SIPRI Military Expenditure Database
7. To be fair, Operation Midnight Hammer (21-22 June 2025) is perhaps the exception that proves the rule, given oil fell 12% in the five business days following. We do not include given lack of one year forward data. But we would point out that, in this instance, a lot of the oil upside had come in anticipation of the strike, with the commodity rising 23% from the start of June to 19 June. The speed and contained aftermath of the strikes served as a relief unwind of geopolitical tension in markets.
Appendix 1: List of selected geopolitical events from WWII to present
Problems loading this infographic? - Please click here
You are now leaving Man Group’s website
You are leaving Man Group’s website and entering a third-party website that is not controlled, maintained, or monitored by Man Group. Man Group is not responsible for the content or availability of the third-party website. By leaving Man Group’s website, you will be subject to the third-party website’s terms, policies and/or notices, including those related to privacy and security, as applicable.