A deep dive into FX hedging [Members]
For MOGULS by The Investor
on July 30, 2026
All investors with holdings in foreign assets are doing macro investing – but very few have decided which macro trade they are actually running. So argues long-time Monevator reader and commenter Ho Simpson in this special guest Moguls post on FX hedging for retail investors.
There’s a popular myth in personal finance: remove FX volatility from your portfolio – that is, the ups and downs of currency swings – and you’ll sleep better at night.
This article can be read by selected Monevator members. Please see our membership plans and consider joining! Already a member? Sign in here.
Comments on this entry are closed.
I’ve got nothing to say other than I really enjoyed this and am looking forward to part 2!
This is a great article – from someone who really knows the space. Of course I’m biased because they agree with my (substantially less technical) https://compass-signal.live/dont-currency-hedge-your-equity-portfolio/%3C/a%3E%3C/p%3E
I’m slightly biased to be more sympathetic to hedging if it provides positive carry (even though this is a “layered” bet – I am indeed making a side bet on carry!).
Pretty much the only thing I actually might hedge are short-term bonds held in the accounts that I manage for older members of the family – because they have much higher volatility aversion in GBP than the rest of us.
Thanks. I really enjoyed it and look forward to part 2 too.
Thanks for this article @TI. it has expanded my knowledge of the hedge greatly.
I have but one fund that has (overt) hedging which I bought many years ago and took advice from a seasoned investor friend to buy the hedged version – IJPH.L
It is a full half a percent fatter on the fees than the nearest similar exposure to the same area, and now a very small part of my portfolio (1/2% ish) but looking at the timescale I have held (13Y+), it seems to have paid off. Moving forward, who knows. As with all things, a little luck might have had something to do with it.
I look forward to part 2.
JimJim
This area feels very much above my pay grade but I really enjoyed it. Especially the deep dive on gold and hedging. Thus found this paper from Vanguard on global bonds and currency management helpful when I was working out my approach.
https://www.vanguard.co.uk/content/dam/intl/europe/documents/en/going-global-with-bonds-unlocking-diversification-through-global-fixed-income-eu-en-pro.pdf
Looking forward to pt2 too
@all — Thanks for the feedback, really appreciated since as most of you will have heard me bemoan before the quieter comments on Moguls is a bit frustrating. Our membership continues to grow, but I suppose many Moguls are too busy running their vast empires et cetera. 😉
Seriously, part two is great also, and would love to have more from Ho Simpson on Moguls in the future. He’s probably more Mogul-y than I am!
Very accessible article and I now feel more confident about explaining why I intuitively forego the hedged version of funds!
I disagree with the fundamental idea that “Currency hedging is a trade”. I’d say it’s the exact opposite. The passive view is to be currency hedged. You are not taking a view on any currency bilateral. Functionally, each position is being funded locally. Taking GBP and buying US equities, FX hedged is equivalent two discrete trades: owning a GBP deposits and owning US equities funded by borrowing USD. No mixing of views.
Being currency unhedged is the trade. It’s an explicit active view on a currency bilateral. Buy US equities currency unhedged, you are funding your purchase by selling GBP depos. If you believe that taking currency risk will lower portfolio volatility or increase diversification, that’s taking a view on correlation. If you think having an open FX exposure is a tail hedge against GBP devaluation, that’s another view. It’s a view I may well agree with but it’s still a view. Whether it works or not is irrelevant.
I think being currency unhedged is a 100% active view and has no place in a true passive portfolio. It’s the same illogical thought process that drives so called passive investors to avoid long-duration bonds, or commodities or whatever. They are assuming they know better, whilst telling people it’s impossible to know better! Total cognitive dissonance.
AndyJ — thanks for sharing the Vanguard article. I think there are some good points in favour of a more global bond allocation, e.g. reducing concentration risk, diversifying across different sovereign yield curves and economic cycles, and not relying entirely on one country’s fiscal and term-premium dynamics.
Where I get a bit more sceptical is when “global bonds” becomes shorthand for “buy the Global Agg” (the first footnote in the Vanguard paper).
DM government bonds, EM government bonds, corporate bonds and securitised products are all labelled “fixed income”, but they do rather different jobs in a portfolio. A DM sovereign bond provides ballast in a vanilla demand-side recession, equity crash or other risk-off event (less so during stagflation). Corporate bonds and EM sovereign debt have a much larger risk-on component. Going global+hedged can provide issuer diversification and income, but they are not the same thing as crisis ballast.
I’m also not entirely convinced that the FX hedging argument applies exactly as Vanguard are suggesting. Unless I’m missing something on page 5, they seem to argue that one of the benefits of global bonds is exposure to multiple interest rate and inflation environments, but then later argue that foreign government bonds should generally be hedged back into the investor’s domestic currency. Well, hedging a foreign government bond back into sterling removes the currency exposure and, with it, a significant part of the foreign ongoing monetary/inflation regime exposure. What remains is diversification through differences in yield curves, term premia, fiscal positions and economic cycles. That is a valid benefit, but it is a rather different argument from owning foreign interest rate regimes or inflation environments.
For corporate bonds, the credit cycle may matter more than the interest rate cycle anyway. A hedged corporate bond fund can provide issuer diversification and income, but it is still not the same thing as defensive ballast. People have different views about whether to hedge risk-on assets, and one of the great joys of not being a bank or insurer is that nobody makes you write a policy document justifying your opinion.
As an aside, the Bloomberg Global Aggregate includes a meaningful allocation to Chinese bonds (c. 6–7%). I’m not suggesting Chinese bonds are uninvestable, but if someone tells me their “defensive” bond allocation includes meaningful exposure to a market with capital controls, questionable data transparency and a rather different legal and political framework, my eyebrow does tend to rise. It also has c. 11–13% in securitised debt. I’m not anti-securitisation, it absolutely has its place, but it is not the same asset as vanilla corporate debt. I know “see: tobacco bonds” is a cheap straw man — and they were never in the Global Agg anyway — but… see tobacco bonds.
So I think there is a strong case for a globally diversified portfolio of high-quality sovereign debt, which might or might not be hedged back to sterling. In the hedged scenario, the diversification comes from different economic and fiscal cycles, yield curves and term-premium behaviour. What I’m less convinced by is the leap from that perfectly sensible idea to “therefore buy the Global Agg”. The market portfolio answers the question, “What debt has been issued?” It does not necessarily answer the question, “What assets do I want to own when equities are down 40%?” Fixed income is one area where passive investing warrants a bit more curiosity about what’s in the tin, and the”bond fund” label tells you rather less than you might hope.
(I’ve ranted about bond indexing elsewhere, if you ever struggle with insomnia: https://3652daysblog.wordpress.com/2025/12/09/dont-panic-part-iii-the-planet-sized-brain-error-of-treating-bonds-like-stocks/)
One final wrinkle on the FX point: have you ever come across the argument that unhedged US Treasuries can sometimes provide a better portfolio diversifier against global equities than UK gilts for UK investors? The reason is partly currency. A global equity index already has a very large USD exposure, and in a risk-off event sterling is a pro-cyclical currency. So if you leave the equity currency exposure unhedged but hedge the DM sovereign bond allocation back into sterling, you can sometimes end up removing one of the channels through which bonds can diversify the portfolio. Not saying this is universally true as currencies have a habit of making fools of people, but it does make the “hedge bonds” rule not quite straightforward when you think about the whole portfolio rather than asset classes in isolation. It all comes down to what your bond allocation is for. Look, Vanguard make some good points, I’m just prejudiced against bond indexing in general. The only bond ETF I own tracks BBG Global Treasury AAA-AA Capped (unhedged) and that is currently underwater, which I’m ok with for the time being as it seems to be behaving as expected in my portfolio 🙂
Well, when I say “the only bond ETF”, I mean the only longer-duration bond ETF. I also have some short-term stuff, but those are part of my cash allocation rather than what I’d consider my bond allocation.
@ZXSpectrum48k — interesting framework, but it only holds if you define currency exposure purely at the instrument level, as the currency in which the price is reported.
Consider a FTSE 100 tracker. The shares are priced in GBP, so by your definition buying it unhedged involves no currency trade as you’re funding a GBP instrument with GBP deposits. Except roughly 70-75% of FTSE 100 revenues originate abroad. Shell, AstraZeneca, Unilever — these are global businesses reporting in GBP but earning in dollars, euros and yen, which are then translated to sterling purely for accounting purposes. The GBP share price is in the reporting currency, not the economic currency. The foreign income streams are already embedded in the instrument at the spot rate before you buy it.
So, following your logic, in order to be genuinely currency neutral in any economically meaningful sense, you’d need to hedge each underlying company’s foreign earnings back to GBP, which is impractical, never done, slightly absurd and arguably impossible given the complexity of multinational revenue streams across dozens of currencies and jurisdictions. And what would you do with poor dual-listed CCL? Pay to hedge the NYSE-listed shares back to sterling? Which would be, like, the opposite of arbitrage. I have a bridge I’m willing to let go for a reasonable price if you’re interested.
The same logic applies to an S&P 500 tracker hedged to GBP. You’ve hedged the USD share price, but you still own Apple’s yuan revenues, Microsoft’s euro revenues, and everything else that gets consolidated into USD before your hedge even touches it.
So “passive is hedged” describes currency neutrality in terms of the instrument’s reporting currency; a reporting convention rather than a genuine statement about economic currency exposure. True neutrality, as you define it (exchanging your GBP deposit for a pure GBP income stream) is essentially unachievable for any investor in equities.
And that’s before we even consider what FX hedging instruments actually are. There’s no such thing as a pure FX derivative. Each one has two or three (or six) additional exposures embedded in it. Take an FX forward: you get (1) the spot FX component — the bit you think you have — plus (2) the interest rate differential, equivalent to an FRA pair or funding trade, giving you domestic and foreign curve exposure, plus (3) the XCCY basis, the basis swap component, because CIP doesn’t hold in practice, plus (4) credit, funding, and convexity effects (CVA/DVA for OTC forwards, convexity adjustment for FX futures). Overlay all of that on your Apple shares and call it passive. Cool.
I think the argument about companies with foreign revenues is increasingly a red herring. The vast majority of companies with market caps in the tens of billions or greater do not leave their currency exposure unhedged. Plus, they often fund directly in major foreign markets or swap funding into local currency.
If we take S&P500, it’s companies have about 30% of their net revenues in foreign currencies. The number of US companies FX hedging has only risen over the past decade and their hedge ratios have also only gone up. For US companies with market caps over $1bn (much smaller than a typical S&P constituent), over 90% now FX hedge. The hedge ratio was over 60% as of 2Q26. So the total exposure to foreign currency flows is in the region of 10% and probably far less once cap-weighted for S&P500 companies.
As for items such as xccy basis or convexity impacts, I think we can ignore those for the purposes of short-dated FX hedging in G10 currencies. At least in non-stress scenarios.
The expense of FX hedging is low. In G10, both spot and fx swaps trade basically choice in anything less than a few hundred million. Right now, I can trade GBPUSD spot in £100mm 1 pip wide (1.34324/1.34334). If I leave an iceberg order at mid, I’ll get filled within seconds. The spot vs. 3m, fx swap is 0.08 pips inverted in £100m. The PB will charge a £600 fixed fee on those three legs in £100mm each. Where are these large transaction costs? If FX hedging costs, it’s Vanguard using it as an excuse to gouge index investors.
I’d be fine if someone said neither FX hedged or FX unhedged is passive. What I don’t accept is that “currency-hedged passive investors often end up running complex macro exposures in the least informed, most expensive, and least compensated way possible.” I’d say the same about currency-unhedged passive investors. How are they more informed or compensated by not FX hedging? Every complexity due to being hedged is the same/worse being unhedged.
I always start my portfolio construction FX hedged. Assume everything is locally funded. It’s make it far easier to discretise your risks from each other. True, I might well decide there are cross-asset correlations including FX-asset correlations. At least, however, I’m then explicitly trying to quantify them.
FWIW I can endorse what ZX says r.e. companies hedging (some/most) of their foreign currency exposure. I look at a lot of multi-national group accounts (…) and I can’t think of one recently where that hasn’t been the case. Worth highlighting a point that ZX also said, a lot of these groups borrow/fund operations in local currency as well. To give an example that was in the news, Google issued (amongst others) an 100-year GBP bond.
An aside, AZ’s functional currency is USD and it utilises a lot of hedging (don’t ask how I know…)
I think there is a case for switching to £ hedged versions of our index trackers when the £ falls towards parity with the dollar, as it did in 2022 and somewhat in 2020. Our non-hedged assets rose usefully in value compared to the hedged versions in 2022 and I didn’t want to give up the gains, using IWDG, EQGB and XDPG for a while. There’s tricky decisions to be made as to when to get in and when to get out. My trips were $1.10 and $1.40 although hung onto EQGB (Nasdaq 100) until this Monevator prompt prompted me to press the button on unhedged EQQQ. Well worth the price of entry for this piece alone.
@ZXSpectrum48k — a few thoughts on your response.
On “passive is hedged” — I note you now concede you’d be “fine if someone said neither hedged nor unhedged is passive”. That’s a bit of a reversal of your initial view, since you came in with a binary “passive is hedged, unhedged is the active trade” and are now going with the middle ground that the article was arguing from the start. Fine, but worth acknowledging.
You ask how unhedged investors are “more informed or compensated” and suggest every complexity of being hedged is “the same or worse” as being unhedged. The complexities are not symmetric. Hedging introduces carry drag, counterparty risk, margin calls, XCCY basis, and roll costs that don’t exist in the unhedged position. An unhedged investor’s “complexity” is accepting FX volatility, which is passive, requires no instruments, costs nothing in transaction terms, and introduces no counterparty exposure. Treating these as equivalent kinds of complexity is … not a neutral observation.
On XCCY basis — you suggest we can ignore it for short-dated G10 hedging in non-stress scenarios. But the article is explicitly about retail investors managing long-term portfolios over decades. A 40+ years of a hedged portfolio lifecycle results in c. 160 quarterly derivative rolls across the full distribution of market conditions, including every stress event where basis widens and bid-ask blows out. The “non-stress scenarios” and “G10” caveat is doing a lot of heavy lifting here. Stress scenarios are exactly when basis costs matter. Also, XCCY basis is a tradable derivative with its own term structure and volatility so if it were really ignorable, the cross-currency basis swap market wouldn’t exist.
On corporate hedging — the hedge ratio argument mixes up two different things. Most corporate treasury hedging covers near-term transactional exposures. When you own an equity investment, you own a claim on all future earnings in perpetuity at a multiple of current earnings. The vast majority of that present value consists of future earnings that no treasury hedging programme can touch. So your 60% hedge ratio has essentially no bearing on the FX sensitivity of a 25x earnings multiple applied to profits that don’t exist yet and can’t be hedged by any treasury team. (I think the current stat of S&P500 P/E ratio is something around 25 – 26). True economic exposure (ie the long-horizon sensitivity of a business’s pricing power and cost structure to FX moves) is the largest and most important component of equity FX risk and it’s unhedgeable, hence “participating in FX hedging” and having your equity value insulated against currency moves are not the same thing.
On foreign currency borrowing as a natural hedge — yes, companies routinely issue debt in foreign currencies or use swaps, that’s standard operating procedure, but it’s sized to investment requirements in foreign operations, not to the foreign operation’s contribution to the market cap. Where companies do use currency borrowing as a net investment hedge, the size is capped at (but usually lower than) the sub’s net assets — not the NPV of the ongoing income stream that drives the valuation. So we’re back to the earnings multiple.
“I always start my portfolio construction FX hedged” — Ok, starting from a hedged baseline and then explicitly deciding which FX exposures to add is a legitimate approach — for an institutional fixed income investor. But then you apply your institutional fixed income portfolio construction logic to a retail equity investor context, and when the variables don’t fit you’re narrowing them rather than questioning the framework’s applicability to the asset class. I think in this case the cognitive dissonance charge sits more comfortably in the direction it came from than the direction it was aimed. But if you prefer — agree to disagree.
NB – I’ve accepted your hedging stats without challenge, but I have a strong inkling that the “90% of companies hedge at 60% ratios” comes from that MillTech survey of 200 “senior decision makers” (translation: middle managers with time on their hands for answering surveys) at mid-sized North American companies, which then got cited by a few finance journos, then got extrapolated from “companies that answered the survey” to “corporate America” to “global equity markets” and has since achieved the status of an inalienable fact through sheer repetition. Just sayin’
@The Details Man — I’m not sure if your AstraZeneca USD functional currency example was offered as a rebuttal to my point about the GBP-denominated FTSE 100 tracker not being a GBP asset economically, but it actually illustrates my point quite perfectly instead of contradicting it. Think about it.
Also, I have no prior knowledge of or interest in AZN (no offence, I find all non-financial corporates rather dull), but a cursory glance at a couple of pages of their 2025 accounts tells me the following:
1. The functional currency at group level is indeed USD, but that’s not the case for all operating subs, a meaningful number of which have different functional currencies.
2. AZN does almost no meaningful net investment hedging. What little exists is probably for JVs, associates, something acquired on the fly for patents, or whatever the hell Global Diabetes Alliance is. No idea and don’t care. The point is, they’re sitting with $1.6bn of unrealised FX paper loss on the subs with different functional currencies, down from $4.1bn last year. The swing suggests the subs in question are probably in the UK, Sweden and/or the Eurozone.
3. Most of their hedges are cash flow hedges, which in corporate treasury usually means swaps hedging debt.
4. They hedge transactional exposures, which is standard operating procedure for every multinational treasury everywhere and proves nothing about economic FX neutrality at the group level.
5. Oh, but here we go: (A) “Approximately 59% of Group Total Revenue in 2025 was denominated in currencies other than the US dollar, while a significant proportion of manufacturing and R&D costs were denominated in pound sterling and Swedish krona”; (B) “Surplus cash generated by business units is substantially converted to, and held centrally, in US dollars” (total cash is c. 5% of the balance sheet and “surplus cash” will be a fraction of that, i.e. peanuts); and (C) “The impact of movements in exchange rates is mitigated significantly by the correlations which exist between the major currencies to which the Group is exposed and the US dollar”. Which, taken together, is the treasury equivalent of a shrug. They borrow in (or swap borrowings into) the currencies in which investments are made and costs are incurred as a natural hedge (if they need to borrow), but otherwise aren’t particularly bothered.
There’s probably more in there but I can’t be arsed reading it. Because: boring.
Anyway, this brings me to the problem with corporate hedging stats generally, “utilising a lot of hedging” doesn’t always tell you very much about the net economic FX exposure.
@hosimpon. My numbers were very conservative. Based on what our systems collect from major custodians such as JPM then hedging is roughly as follows: contractual cashflows 95%+. Forecast cashflows for the next 12m 75%+. Cashflows 12m+ 60%+. Hence my use of 60%+ as a conservative estimate.
Now, FX hedging of translation (balance sheet) exposure is only around 30% because that is not hedging any actual cashflows but instead mainly concerned with the potential impact of FX moves on future competitiveness. I would note, however, that when I started 30 years ago, translation hedging was basically zero. Yes, for US corps, it’s jumped substantially in the last 18-months due to the impact of Trump trade policy. The trend though is clear: all forms of FX hedging are increasing because it only ever gets easier and cheaper for corps.
As for you comments on xccy basis, then as a interest derivatives trader/portfolio manager with 30 years experience I’m perfectly aware these are tradeable markets. Either that or I’ve been hallucinating for a very long time. I trade fx swaps, xccy swaps, FRASs and IRS every day and that’s mostly in EM. Even then for retail types, this is not something they have to be hugely concerned about. It will tend to average out over the long-term.
Moreover, the carry and roll costs you seem to be bothered about are just assumptions. It’s an active view to assume you will earn the carry or roll-down. It’s would be same as arguing that the return on an asset is it’s dividend or coupon. Historically, FX carry does work (over the long term and with some shocking drawdowns) but it’s clearly not a passive strategy.
Passive investing is about hedging a the future liability structure as best you can. It’s about reducing risk. Not about taking views on spot FX or carry. FX hedging is, by construction, locking in the forward and the zero P&L trajectory. I cannot see how it can be any more passive.
I will leave it here. Yes, we will disagree.
@hosimpson, it wasn’t intended as a rebuttal or criticism, so I don’t know why you’ve been snippy with me. I don’t see the need for the tone of your reply to me (or ZX for that matter). For what it’s worth, I’m more on your ‘side of the fence’ in your debate with ZX.
That said, I don’t agree with some of your analysis or editorialising of the accounts. However, I appreciate TI will not want us to go down a tangential rabbit hole so I’ll leave it there.
@ZXSpectrum48k and @HoSimpson – this has been a really interesting exchange, thank you both for your comments.
Just picking up on something that ZX said:
“Passive investing is about hedging a future liability structure as best you can.”
I fully accept that most of my future liabilities are going to be priced in GBP, but from a relatively uninformed (though very interested) retail point of view:
– Despite being priced in GBP, I’m personally of the view that most of the economic exposure of my future liabilities is likely not GBP. We import a lot of food, energy, raw materials, etc., so even things that appear to GBP liabilities (for example home maintenance) could have complex underlying foreign-currency drivers.
– I think this is particularly true for affluent households who spend more of their income on imported luxury goods, holidays etc.
– The other thing I’m quite concerned about is effectively being ‘left behind’ on a global scale by being so exposed to GBP, and I personally view holding a lot of USD as a way to help mitigate that.
– My house, my future income, my state pension etc are all priced in GBP, I’ve got a huge amount of GBP exposure, and I’m not really sure why I’d want to take on even more exposure in my investments. It seems to be quite risky to concentrate all my wealth in the currency of a small, relatively insignificant island, just because that’s where I happen to live.
If either of you has any comments, I’d be glad to hear them; it’s always good to have my biases challenged.