This article can be read by selected Monevator members. Please see our membership plans and consider joining! Already a member? Sign in here.
The natural yield model portfolio wheels are turning [Members]
Next post: Weekend reading: National scandal
Previous post: Commodities are working
Comments on this entry are closed.
> portfolio value rose steadily, before falling precipitously
Look on the bright side. That income’s gotten a bit cheaper, albeit a bit more fragile 😉 Indeed I recall reading should you swap your shares for an investment trust on a discount in those desperate post GFC days and think I ain’t got any shares but I’ll take the IT on a discount. Those were the days my friend and all that. And indeed I’ve just done a little bit of IT topping up in the GIA, getting on for 20 years on, what with the lower tax on dividends etc
Thanks for the update. Would just question the use of Vanguard 80% as the comparative. Wouldn’t the 60% or even 40% be more comparable for a retiree in drawdown?
I’m going to be that person who makes the obvious point you flag.
Re: The “starting yield on the Vanguard fund is only about 1.6%. So even in this Inc fund form you’d need a bigger pot and/or cash buffer and a plan to sell down capital to live off it”:
And therein lies the rub.
Cakeism usually doesn’t (easily) work. Chose income or capital. But it’s quite hard, but not impossible, to get a growing income and growing capital.
Buffett receives over $93,000 per hour from Coca Cola dividends, $816 million annually on a 30 year old investment that cost just $1.3 billion. His annual yield on cost now exceeds 62%, meaning he recovers his entire original investment every 18 months in dividends alone. In 1994, when Berkshire completed its $1.3 billion Coca Cola purchase (~400M shares bought 1988–1994 for ~$1.3B), the initial dividend was just $75 million, a respectable but unremarkable ~5.8% yield on cost for 1994.
But you have to take a risk with single names. Quality/wide moat with some dividends and with buybacks when accretive (costs below WACC), with high value factor loading (i.e. a relatively low P/E and/or low P/S and/or P/B to own history) and coming off a steep draw down (with a judgement call as to no fundamental impairment), plus some momentum on the rebound.
Otherwise you’re just trading income for capital growth, e.g. Schwab US Dividend ETF 3.9% yield, up 28% over 3 years on price compared with SPDR S&P 500 ETF Trust (aka the SPDR/Spider) on 1.2% yield up 73% over 3 years (price only, both for 36 months upto 28 February 2026, when the war kicked off).
Investment Trusts probably have a better chance of optimising for income and capital than dividend ETFs (Aristocats, HY, HY+Low Volatility etc, there’s no shortage of passive choices). But, given the remit, I suspect they’ll deliver essentially sustainable income in exchange for less growth and probably less total returns than say VWRP (Acc)/VWRL (Dist) ETFs.
You’d really need an IT here to be explicitly targeting growing dividends even at the ‘cost’ of a low starting yield.
Chevron (2.8%), Morgan Stanley (2.4%), Merck (2.7%) and Coca Cola (2.5%) all yield double the SPDR but are on puny starting yields to your Natural Yield portfolio @TI. However, I suspect that, along with JPMorgan Chase (2%) and Exxon (also 2%), they may end up after 30 years on a higher yield on cost (again, all yields quoted are as at 28/2/2026).
@ermine — Indeed, although the discount narrowing has been a long time coming for a lot of trusts this time around! Sure lower prices are nice for a new investor buying into a new income portfolio. But remember this portfolio is now a sunk cost, give or take any future rebalancing. I’ll only reinvest any income if the cash buffer gets unwieldy. So the best result here for the model portfolio owner would be soaring share prices for the trusts *and* dividends rising nicely!
@ChuckieB — A 60/40 or 40/60 might make more sense in terms of how people potentially invest in drawdown, but I’m not sure it’d be a fairer comparison. If you look at the model portfolio only 20% is in bonds. Some 50% is in equities and the rest is in assets that @TA would be quick to remind us should come out of the ‘kinda like equities’ bucket. Maybe a 70/30 would be closer, accordingly, but Vanguard doesn’t offer that. For now it doesn’t matter. When I’m having some big debate in a decade no doubt it will… 😉
@Delta Hedge — Well yes we’re going to be able to say and write that with every update, and it’s much covered in this post and the two previous ones. This portfolio isn’t about getting something for nothing. I explicitly state I don’t expect it to beat the market for total return. It is about (probably) paying some cost in growth to exchange capital volatility for a smoother income, as stated several times. 🙂 Finally the low-yielding dividend growers you cite may well be on higher yields-on-cost in 30 years, but by then even a fairly early FIRE-ee is getting towards the end of their innings, and in practice many such portfolio owners might have have handed over the passwords and shuffled off to that great platform in the sky. Compromises have been made to get a starting yield of 5%, and I have been clear about them. 🙂
Cheers for thoughts everyone!
There’s a decent argument to say that Investment Trusts should always trade at a discount, reflective of the NPV of the future management fees of the trust. Though, I’ve never bothered to actually calculate what level of discount that would imply for various trusts.
Did you get any inspiration @TI for future tweaks or reinforcements to the Natural Yield Portfolio from the latest AIC 16 March list of 20 Investment Trust dividend heroes (those consistently increasing their dividends for at least 20 years in a row, half of which have increased for 50 or more consecutive years) and/or from its subsequent (23 March) list of 30 future dividend heroes (the 30 ITs that have increased their dividends for 10 or more consecutive years, but less than 20)?:
https://www.theaic.co.uk/aic/news/press-releases/30-investment-trusts-make-up-next-generation-of-dividend-heroes-0
My Coca Cola example is an extremity of course. $1.3B capital in 1988-1994 turns into $29B in February 2026, with distribution yield falling from 5.8% in 1994 to 2.8% in 2026, even as the dividends rise from $75 million to $816 million, and yield on cost hits 62%. There’s few of them about, that’s for sure.
@Delta Hedge — Thanks for the link, I included a nod to the same (very much in stealth mode) in the article, to a pointer from This Is Money I think. I’m afraid that there won’t be chopping or changing unless (until?) anything goes wrong, which is why I really only want to be reviewing this portfolio once a year. After all, it’s supposed to be outsourcing income gathering to the trust managers.
Of course it’d be more fun to make changes along the way but I don’t think it’d meet the mission statement of the TLIY portfolio. 🙂
@SkinnyJames — It’s an interesting point, and I’ve seen people attempt to do various calculations for some high-profile trusts, though I’m not inclined to do it myself. My observation is that discounts wax and wane over time, so whatever the theoretical validity of one existing, often they do not, when market conditions are right.
The exception perhaps are somewhat controlled trusts like Hansa, Caledonia, or perhaps Pershing Square (which I hold) the way things are going.
If you were to calculate it then I’d suggest you’d also perhaps have to include offsetting factors, such as a trust being able to gear much more effectively/safely than an individual, which must be worth something, and having an independent board of directors overseeing the manager, which too often demonstrably isn’t! 😉
@The Investor plus any alpha that you think the manager might have (positive or negative!), and adjust for any differences between what you think the true NAV of any private assets are vs. the reported NAV.
I think this all becomes too difficult and I just wouldn’t bother.