This article can be read by selected Monevator members. Please see our membership plans and consider joining! Already a member? Sign in here.
Reducing lifetime portfolio risk with leveraged ETFs [Members]
Next post: Weekend reading: you can bank on this
Previous post: Monzo investments: what’s on offer, is it any good?
Comments on this entry are closed.
Thanks for another really brilliant article @Finumus.
No Amelia’s here but Mrs TLI and I are of the FS (and CARES) DB pension cohort that you mention.
So, whilst we each run with all/nearly all Global Equities in our ISAs and SIPPs, our total, DB pensions inclusive, risk on asset allocations are, in effect, rather less.
Following on from your excellent and eye opening LETF for the Long Run piece, and inspired also by your similarly quite brilliant cheapest ETF world tracker portfolio article, I came up with a sort of version of latter using the lower spread CSP1 and EQQQ (for the US non-leveraged ETFs), and using 3LUS and LQQ3 for leveraged US exposure; and then switching between the pairs (from leveraged to unleveraged) based upon the SP500’s & NASDAQ 100’s 3 month absolute momentum; but then scaling down the cap weight allocation to those indices by a factor of 3 each (from 57% to 19% overall) both to better reflect the use of leverage (when leveraged) and a choice to be underweight (the rather expensive) US markets (when not leveraged).
This, in effect, would ‘create’ another 38% in the equity allocation for likes of PE exposure and multifactor ETFs etc.
Your article today has made me think that I could try instead and diversity away SOR risk over time by:
– For the first half of the period between now and when I eventually anticipate needing to access to my ISA (for care home fees etc) in 25-30 years time (when I’m in my 70s-80s), i e. for the next 12 to 15 years of those 25-30 years: switching between the leveraged, risk on 3LUS and LQQ3 and either a risk off, low volatility, leveraged, complimentary asset like 3TYL or a low spread, unleveraged one like IBTL, IBTM, CU71, BBLL, BBM3, IGLH or FEDG (again using 3 months momentum both to choose between them and to signal the switch back to risk on); and,
– For the second half of that period (from my early to mid 60s onwards): switching between unleveraged, risk on CSP1 and EQQQ to one of the above listed risk off assets, again according to their past 3 months’ momentum.
This scales down risk exposure (via risk on, leveraged assets) as I enter the second half of the period.
Rather than a cliff edge change, I could instead of course taper into unleveraged only, risk on US equity exposure over the second half of the 25-30 year period.
In any event, your very thorough and extremely well expressed piece today gives me much food for thought.
Again, thank you for the excellent article today.
> Beyond short-term cash needs, then, it should be odd for such a person to invest in a portfolio which had a lot of cash or bonds in it, because they already have their retirement savings and income sorted. Yet you see it all the time.
I came to this conclusion many years ago. I didn’t hold any bonds at all, though I did hold about three years cash through the gap between quitting work and drawing the DB pension, to forestall becoming a forced seller in a down market.
It worked for me, I did eventually manage to accumulate enough ISA to beat out the notional capital standing behind the DB pension of 16x annual income.
I am not sure I would have had the brass nuts to tackle a leveraged approach, because the gap (between retiring early) and being able to draw the pension is another hazardous point where your risk tolerance is lowered. It’s all very well having a large notional pension but if you spend your early fifties living under the railway arches because you can’t get your hands on that or you happen to eat a tough SOR on your leveraged assets then it’s not so good. If you are of TI’s predilection that you will always be working then of course that midlife limitation does not apply.
I am surprised that the cost of carry of leveraged ETFs is low enough that this is workable for timescales measured in years!
I think the point about time diversifcation is really important. There is too much focus on diversification thought of only in terms of holding a few asset classes over a few countries. It often creates very little diversification at all. Meanwhile, time diversification is basically ignored.
What I don’t see is why you use LETFs rather than futures (or options). LETFs are essentially products that compound daily changes. This a long term strategy that really doesn’t utilize that part of their feature set. The only reason I can see to use LETFs is in the specific situation of ISAs since they cannot do derivatives. For everything else, I find F&O cheaper than LETFs to run these sort of long delta overlays.
@All: dynamical risk aversion is perhaps something to consider here. Scenarios and people vary of course, but I’d guess that it’s often the case that if you’re young and with comparatively little starting wealth then just a small absolute loss can be crushing, even if it would seem quite trivial when older and with more assets. Yet, as Finumus’ piece shows, rationally risk aversion should be higher later in life because of the impact and importance of SOR risk. But it may not necessarily seem that way when you’re young. Stick £9k into a 3x Nasdaq LETF and see it fall 45% to 5k (rather than by a rather more palatable amount in a normal ETF) and you can be racked by guilt. But 40 years on with, say, upto £1mm in ISA, SIPP and GIA, with the mortgage paid off, and with some cash savings to boot etc; then going and putting £100k into an LETF and then seeing a paper loss of 60% might neither matter much, nor affect you. Yet, speaking theoretically, you should be taking bigger risks when young with compounding time and human capital on your side.
BTW: 1. really enjoyed @Finumus’ Twitter (X??) threads, which are linked to at end of piece; and 2. the idea at #1 above is just that, a WIP paper possibility/plan, which I haven’t (at least yet) implemented.
@ZXSpectrum48k – Curious to hear more about your implementation of (long term) leverage using options. Are you using deep ITM LEAPs?
Further to the discussion in the comments on @TI’s recent”A Question of Trust” piece discussing the present unavailability in the UK of US listed non-UCITS Capital Efficient ETFs (i.e. moderately leveraged ETFs with largely negatively correlated asset class pairs – i e. 100 Global Shares / 100 Intermediate Duration US Treasury Bonds), “the 7circles” have covered both this type of ETF and also life cycle investment using more heavily levered LETFs (which is not for the faint hearted). The link covering all of the relevant articles there is below:
https://the7circles.uk/category/active/what-works/leverage/
Any thoughts on using LETFs as part of a bar bell strategy of putting 10% into maximally risky assets (say the ‘Max Pain’ 3x leveraged ETF rotation system from Dual Momentum Systems, previewed in 7circles, as per the link below); with the remaining 90% going into minimally risky – but inflation protected – assets (i.e. TIPS and/or ILGs held to maturity)?
No rebalancing between the 10% max risk and the 90% min risk parts in order to try to increase overall expected returns. Worst possible loss is 10% of total initial portfolio size – i.e. 100% of the 10% in maximally risky assets.
https://the7circles.uk/dual-momentum-systems/
@Finumus: following the tag line for your piece that “dumb leverage blows you up” this March 2024 paper shows that LETFs cannot work beyond a certain level of volatility and also shows: the sensitivity of LETF performance to rising uncertainty about future volatility, a discontinuous trend of increasing volatility, and the sensitivity of LETFs to the inflation regime:
https://arxiv.org/abs/2403.01088
My provisional thoughts are rotate to LETFs when:
– realised volatility low (e.g. VIX below 25); and,
– expected volatility low (i.e., from Gayed’s Leverage for the Long Run paper, when the index is above its 200 Day Moving Average); and,
– inflation low (below 1.5% p.a. from chart in linked paper above); and,
– rates low (low financing cost of leverage, i.e. low cost of carry, see latter comments in thread on your earlier Moguls’ leverage piece covering this point).
So a low rates, low inflation regime with low realised volatility and an uptrending market should hugely increase the odds of LETF outperformance; and the opposite for high inflation, high rates, high realised volatility and below trend line markets.