To Hedge, or Not to Hedge, That is the Question

[DM94]

We all know that when we invest in an ETF the value of our positions move up and down due to the price of the underlying stocks changing throughout the day but I feel that people tend to gloss over the power of currency hedging.

As a (predominantly) boring index investor myself, a recent personal example comes from being invested S&P 500 in the US. I buy my ETF shares in GBP but that gets converted to USD as the underlying stocks are in USD, giving me potentially unwanted currency exposure.

In November 2020 I was invested in a GBP hedged S&P 500 ETF from iShares (GSPX) as I was concerned about USD continuing to decline post-Covid crash. It did continue to fall and this meant that by August 2021 my hedged position had produced a return of +23.9%. If I was unhedged during this time I would've 'only' had a return of +20.4%.

In September 2021 I was started thinking about missing out on the opposite as USD seemed poised to start fighting back against GBP. Luckily for me it did and I switched to one of the most popular unhedged S&P 500 ETFs, VUSA by Vanguard. Since then I’ve enjoyed the protection that being unhedged brings during market downturns and as a result my investment has returned +5.6% instead of -6% that remaining hedged would've given me.

Seemingly for any non-US investors who’re invested in the US, being unhedged will suppress returns but cushion losses (although this might not always be the case I’m sure).

From what I've seen online, most fund managers recommend being unhedged as over the long term investors reap similar returns either way, and the additional cost of hedging typically outweighs the benefits. But since you can now invest in hedged ETFs with virtually the same fees as their unhedged counterparts, is there now an argument to be made for switching from hedged to unhedged, and vice versa, over the long term? Or would this simply fall into the 'trying to time the market' category?

Interested to hear your thoughts/experiences.

Cheers!

DM

[Lootman]

I never bother with currency hedging. There is always a cost to hedging and the FX market is probably the hardest one to make any money in, so I don't try and second guess currency movements.

That said the pound sterling has pretty much being going down my whole life, minus a few blips. So I am always willing to take FX "risk".

At the end of the day if you choose to invest in, say, US shares then you are effectively betting on Uncle Sam and that includes his currency. There is no real way around that, and which currency your ETF is actually denominated in is irrelevant. You are investing in USD-denominated assets.

[Hariseldon58]

Clearly if you hedge and things go your way then you win, however the hedging costs money and its imperfect.

When the Brexit referendum looming in early 2016 I pivoted to more US dollar exposure, which I have kept. ( both equities and bonds) taking positions on FX rates is clearly not a great idea generally, however I am with Lootman, the long term trend is for GBP to weaken against US$.

I think that move was a good one and has earned me a decent 6 figure sum in excess returns.

Looking forward, no one knows but I still retain a US$ heavy exposure, including a large holdings in intermediate and long TIPS.

[1nvest]

Firms often opt to hedge their FX exposure to better stabilise earnings in the currency they are domiciled to. So determining how much a particular index of stocks should be hedged by is a daunting calculation.

When you buy a stock you're buying into a number of factors, such as FX, short bonds ...etc. For instance the average stock borrows around half of its book-value (sells corporate bonds/is short bonds), and share prices are around twice book-value, so every £1 of stock purchase might also be considered as having 25p of short bond exposure as part of its total return.

As with bonds, that broadly are zero sum (sometimes borrowers win, other times lenders win, overall neutral), so also does FX tend to be zero sum, such that hedging is just a additional cost/drag.

Simplest way is to count US stocks as both stock and £/$ volatility exposure, sometimes they correlate, sometimes they inversely correlate.

Or like with stocks, you could go 80/20 stock/bond - to negate the 25% short bond exposure that stocks on average include, but then you're in effect lending 20% of your capital to yourself, for broadly overall 0% nominal return, worse after inflation is factored in.

The better way is to diversify FX risk. Maybe 50/50 UK/US stocks. Without opting for the currency hedged version of US stock index tracker. Some take that further, and hedge fiat currency exposure (such as US$ and/or GB£) with non-fiat commodity currency holdings - such as gold and/or silver. As investors we seek to minimise risks/maximise rewards, and concentration risk is major risk factor that otherwise is easily diversified away. If you're all-in on the US$ or other single currency, then your over-concentration risk is high. So you're right to think about such matters, but hedging FX generally isn't the way to go.

[1nvest]

I should add that broadly you might expect the US$ to decline relative to gold over the mid/longer term, a natural function of fiat-currencies. Equally you might expect the £ to decline relative to the US$, again just standard characteristic. But not consistently/reliably, at times there is high volatility.

Longer term however, 50/50 gold and hard US$ currency (dollar bills) went a considerable way to offset UK inflation. Made the loss of purchase power that more gradual. When US$ were instead deposited/invested then the tendency was to entirely offset UK inflation, or swing the line around to being positively sloped.

For US investors, 50/50 US stock/Gold since 1972 has near-as yielded the same total return as 100% stock, but done so with less volatility - which is commonly considered as being a better risk adjusted reward.

Consider that stock price only and gold might broadly be expected to pace inflation, but with high volatility. For a compound 0% real when yearly averages exhibit a 20% standard deviation (as is common for both stock and gold) the assets yearly average gain has to be 2%. 50/50 however reduces the volatility by half - without reduction of the yearly average and a 2% yearly average with a 10% standard deviation yields a 1.5% annualised benefit, which is a real return as we're talking after inflation values here. Attribute that 1.5% portfolio gain to the 50% gold holdings and if stocks pay a 3% dividend you end up with a portfolio total return of 3% real. 50/50 stock/gold compared to 100% stock total returns (with dividends included), but did so with lower portfolio volatility, higher Sharpe Ratio (better risk adjusted reward).

[AWOL]

I should add that broadly you might expect the US$ to decline relative to gold over the mid/longer term, a natural function of fiat-currencies. Equally you might expect the £ to decline relative to the US$, again just standard characteristic. But not consistently/reliably, at times there is high volatility.

Longer term however, 50/50 gold and hard US$ currency (dollar bills) went a considerable way to offset UK inflation. Made the loss of purchase power that more gradual. When US$ were instead deposited/invested then the tendency was to entirely offset UK inflation, or swing the line around to being positively sloped.

For US investors, 50/50 US stock/Gold since 1972 has near-as yielded the same total return as 100% stock, but done so with less volatility - which is commonly considered as being a better risk adjusted reward.

Consider that stock price only and gold might broadly be expected to pace inflation, but with high volatility. For a compound 0% real when yearly averages exhibit a 20% standard deviation (as is common for both stock and gold) the assets yearly average gain has to be 2%. 50/50 however reduces the volatility by half - without reduction of the yearly average and a 2% yearly average with a 10% standard deviation yields a 1.5% annualised benefit, which is a real return as we're talking after inflation values here. Attribute that 1.5% portfolio gain to the 50% gold holdings and if stocks pay a 3% dividend you end up with a portfolio total return of 3% real. 50/50 stock/gold compared to 100% stock total returns (with dividends included), but did so with lower portfolio volatility, higher Sharpe Ratio (better risk adjusted reward).

I love the historic profile of these Equity/Gold portfolios but the problem is that for 41 years up to 1974 US investors were prohibited from buying gold and then from 1974 onwards they could buy the forbidden fruit. I have no way of knowing how much this and Bretton-Woods prior to this distorted the gold price. So it looks great in models but in the future, who knows?

Regarding FX, I wouldn't currency hedge equities as equity volatility is greater than currency volatility (US investors buying "International" may feel currency risk is too high for them but I am UK), I would consider hedging FX for global bonds as FX volatility is greater than bond volatility.

Having said that my US bond ETF is unhedged as I am taking a punt.

[1nvest]

I love the historic profile of these Equity/Gold portfolios but the problem is that for 41 years up to 1974 US investors were prohibited from buying gold and then from 1974 onwards they could buy the forbidden fruit.

Americans were only prohibited from holding investment gold within the US pre 1970's. US based physical gold had to be submitted to the state in the early 1930's at a fixed price. Americans could have held physical gold in London/wherever vaults outside of the US. Or fallen back onto poor-mans gold (silver) as a replacement.

Prior to the 1930's (or rather WW1 when it started to break down), money and gold were the same, money was backed by physical gold. Being finite that meant broad low/no inflation. Lending to the state (swap gold into money and buy Treasuries) was paramount to the state paying you for it to securely store your gold, and the rates were pretty good, 4% type yields. As such most investors simply held 'bonds', and had reasonable security to plan around the real yields they paid.

The US ended that when money/gold were decoupled. Money became just a number, permitting the US to print/spend and devalue all other notes in circulation. Inducing just inflation and where bonds were more inclined to broadly yield 0% real (sometimes borrowers won, other times lenders won).

Of the two, the fiat/depository US style choice rather than the custodial/non-fiat (gold backed) British Empire choice, and there's greater stability under the latter. Under the former even with multiples more of 'enough to retire' there's still uncertainties, under the latter you could better plan with much less having been accumulated.

Banks shifting over to being depository (money deposited becomes the banks money) rather than custodial (safe keeping of your deposits) is more inclined to see depository based banking taking high risk/reward bets, heads they win, tails they don't pay back your 'loan' (deposit).