HYP1 is 19 - thread discussing income and capital diversification

[Itsallaguess]

With the recent 19th Birthday post from pyad regarding the fantastic HYP1 income-portfolio, I've updated my spreadsheet that keeps a track of the lumpiness of both income and capital delivered by HYP1.

I've started a separate thread to discuss this lack of income and capital diversification in HYP1's ongoing make-up, and have deliberately opened a separate thread this year on the subject with a hope that any discussions regarding this aspect can avoid the possibility of disrupting pyad's main thread on the subject, which can be found here -

https://www.lemonfool.co.uk/viewtopic.php?f=15&t=20380

My hope is that by starting a separate thread on this particular subject, that it can be ignored by anyone completely uninterested in this aspect of HYP1 analysis, as I know some people are...

HYP1 is being run as a completely 'non-tinker' income-portfolio, and over the years there has been a clear trend towards a relatively small section of the HYP1 portfolio delivering a very high percentage of the dividend-income derived from it.

For the income HYP1 received this year, that trend has continued to rise, as we can see in the table below where we see that the percentage of overall dividend income derived from the largest 5 income-components has now risen from last year's 75.65% to a new all-time high of 77.86% -



The underlying historical HYP1 income and capital data for the above table can be found here - https://i.imgur.com/onwguKf.png

In addition to the above increased concentration of HYP1 income on the top 5 dividend-payers, there has also been a rise in the percentage of income paid from the two largest income-providers. Last year Persimmon and BA Tobacco together paid out 46.43% of total HYP1 income, and the crown this year goes to Rio Tinto and Persimmon, who between them paid out a huge 47.65% of overall HYP1 income -



I think HYP1 is a great experiment in how a completely 'hands-off' income-portfolio might develop, but given the above analysis, I continue to believe that the developing concentration-risks to income and capital of taking such an approach was never fully highlighted in the original HYP concept.

When we look at Pyad's original explanation regarding HYP, he had this to say -

Whatever the money available, even very large sums, no more than about 15 shares are necessary to strip out the excessive risk of too few shares. Stick to FTSE 100 companies and spread the holdings around sectors.

https://web.archive.org/web/20140219210446/http://news.fool.co.uk/news/foolseyeview/2000/fev001106c.htm

I'm not sure if Doris, if she were to wake from her multi-decade 'investment-slumber' and wished to take a look under the HYP1 bonnet, might wonder if that initial 'stripped out excessive risk' of too-few shares had lasted as long as she might have hoped, or that the income-delivering holdings that were initially 'spread around sectors' were still diversified as much as the HYP label originally seemed to reach for...

There's no doubt that HYP1 has managed to keep delivering on the promise of a rising portfolio income, and of course Pyad should be congratulated on both coming up with the original HYP concept and also maintaining this ongoing 'non-tinker' income-portfolio experiment, but I hope we're allowed to perhaps sometimes ask 'at what cost, Doris?' when we're given an opportunity to take a peek under the HYP1 bonnet ourselves....

Cheers,

Itsallaguess

[moorfield]

and the crown this year goes to Rio Tinto and Persimmon, who between them paid out a huge 47.65% of overall HYP1 income -

But remember Rio's contribution has been grossly skewed by the special dividends this year. According to my own records Rio has paid 492.65p per share this year, of which 233.37p were specials. Stripping out those adjusts pyad's income to £1,340.47 or 14.3% of £9,350.76 overall.

On my own portfolio reports I have always stripped out specials when computing income weightings. I don't assume the same specials will be paid again in a subsequent year.

[vrdiver]

On my own portfolio reports I have always stripped out specials when computing income weightings. I don't assume the same specials will be paid again in a subsequent year.

The specials will make for a lumpy payment profile, but if you strip them out, I think you've double-removed them.

My logic is that if the company hadn't paid a special, it would have had that money to reduce debt, invest in the business or increase the regular dividend by a smaller amount, spread over multiple years, thus improving your future returns one way or another. By removing the special from your receipts record you have lost both the current cash receipt and been penalised by losing the alternative future use of that money.

One solution, should you wish to strip out specials, is to reinvest them in the company that paid them (attempting to mimic no-tinker by putting the cash back into the company that paid it out.) Then you would get the future benefits that the special has in effect pulled forward as a cash alternative.

My own method is to treat specials that have no share consolidation as dividends, but where there is a consolidation, to treat them as a forced return of capital*.

E.g.
100 shares at £1 each
Company A) pay a special of £0.10, no consolidation, shares drop to £0.90 and I have £10 cash - treat as a dividend.
Company B) pay a special of £0.10 with a 9-for-10 consolidation, shares stay at £1, leaves me with 90 shares at £1 each and £10 cash - treat as a return of capital

VRD


*Capital for my personal records, normally dividends for tax purposes...

[NeilW]

I continue to believe that the developing concentration-risks to income and capital of taking such an approach

Yet after 19 years there is no evidence of the manifestation of this theoretical risk, and in that time we had the biggest financial crisis since 1929.

19 years is a retirement. You retire at 67 and you'd be 86 by then end of the period - and probably dead.

Which says that whatever risk level you are seeing falls into the same level of risk as having an operation and failing to recover from it. Would you turn down the operation?

I'd suggest HYP has proved it is 'good enough' and the approach is sufficient to render the risk mitigated for all practical purposes.

[GoSeigen]

I continue to believe that the developing concentration-risks to income and capital of taking such an approach

Yet after 19 years there is no evidence of the manifestation of this theoretical risk, and in that time we had the biggest financial crisis since 1929.

19 years is a retirement. You retire at 67 and you'd be 86 by then end of the period - and probably dead.

Which says that whatever risk level you are seeing falls into the same level of risk as having an operation and failing to recover from it. Would you turn down the operation?

I'd suggest HYP has proved it is 'good enough' and the approach is sufficient to render the risk mitigated for all practical purposes.

Loving the Wilsonian maths.


TJH has "proved" it wrong unless one believes he died some years ago!!


GS

[tjh290633]

Loving the Wilsonian maths.

TJH has "proved" it wrong unless one believes he died some years ago!!

I am still here and hope to remain on this earth for a few more years yet, but one never knows:-)

TJH

[Alaric]

The diversification argument is at least in part that if handed a sum of money equal to the current market value with an instruction to invest it for equity income at around the same annualised amount as HYP1, it would be unlikely that your replacement portfolio would rely on two Companies for approaching 50% of the income.

The argument against is perhaps that if you notionally took the market value every year and decided afresh what it should be invested in, you might be talking dealing costs of up to £ 300 ( 15 buys, 15 sells at £ 10 a time) plus Stamp Duty costs.

If a portfolio is bought once and held forever, or at least until the Executors get their hands on it, the relative success or failure is going to depend on the initial selections and how they develop over time.

[vrdiver]

I continue to believe that the developing concentration-risks to income and capital of taking such an approach

Yet after 19 years there is no evidence of the manifestation of this theoretical risk, and in that time we had the biggest financial crisis since 1929.

19 years is a retirement. You retire at 67 and you'd be 86 by then end of the period - and probably dead.

Which says that whatever risk level you are seeing falls into the same level of risk as having an operation and failing to recover from it. Would you turn down the operation?

I'd suggest HYP has proved it is 'good enough' and the approach is sufficient to render the risk mitigated for all practical purposes.

HYP1 is a great real-time experiment (as opposed to backtesting) but is effectively anecdotal. It, by itself, doesn't "prove" anything, unless you also accept that HYP would be disproved by a single failure.

It (HYP1) shows the impact of non-tinkering. If, say in a couple of years, one of the key shares in it takes a tumble, that will not disprove HYP, just highlight the risk of over-reliance on too few companies.

A few years ago we had a poster in the other place argue that investing in Lloyds was the way forward, with dividends to be reinvested in more Lloyds. It worked great, until it didn't. Currently we have another poster who has done very well with a highly concentrated portfolio. Comments suggesting diversification are dismissed. I hope it continues to go well for him, but the risk is very real, whether it materialises or not.

HYP1 carries a very real risk of sudden, significant income loss. There is nothing about it having succeeded for 19 years that mitigates that risk. Far from it, each year it continues to concentrate its winners, the impact of the risk becomes more severe. The severity of the risk is a different question - what is the likelyhood of a single company failing? I'd suggest that a look at the FTSE100 in 2000 to see which of those are still in the index today might give an indication of the chances of a company having problems over this sort of timescale - indeed, we see that very situation by looking at the smaller constituents of HYP1!

The question is whether, given 15 picks out of 100 candidates, even applying various quality filters, any HYP thus constructed would do as well as HYP1, or better, or worse? We don't know the answer to that, so most of us set limits as to how reliant we will accept being on a single company or sector.

Finally, with your operation analogy, if only one other person had had this operation, and the surgeon explained to you that it had gone really well, but if it goes wrong, it will go really wrong, would you have the operation, or opt for the alternative* that lots of people had had, with perhaps a lower "best outcome" but also a better "worst outcome". You only have one life; do you feel lucky?

VRD


*as in, spreading the single-company risk a bit more evenly.

[IanTHughes]

First of all, if I remember correctly, HYP1 was not set up as being the only income stream available to its owner. On the contrary, the original investment amount of £75,000 was arrived at because it was then the maximum amount allowable for withdrawal from a Personal Pension scheme. In other words the dividend income was not required to cover day to day living expenses, the aforesaid Pension would do that. HYP1 was to provide a pool of money for life's luxuries: a Cruise, a Caribbean holiday, a monthly supply of caviar or whatever else the owner was into. It therefore follows that even a significant drop in HYP1's dividends from one year to the next would not materially affect the owner, except perhaps being forced to holiday in Skegness that year!

Secondly, anyone suggesting that, because of the lack of diversification both by value and by income, HYP1 should immediately be re-balanced to make the income stream going forward "safer", must also accept that if such a re-balancing is now required, it would have already been required several years ago. It must also be accepted that such a re-balancing exercise, if it had been enforced from the very beginning, would have materially changed the results that we are now looking at, 19 years later. Sure, there would be no serious lack of value and income diversification, but what about the annual income now being enjoyed? Would that be lower now as a result of continually selling down the very holdings that, if left alone, would otherwise go on to create the impressive results we see today? Because there has been no re-balancing, it is now impossible to answer that question, without the introduction of hindsight bias, so I am afraid we shall just have to live with it.



Ian

[jackdaww]

Loving the Wilsonian maths.

TJH has "proved" it wrong unless one believes he died some years ago!!

I am still here and hope to remain on this earth for a few more years yet, but one never knows



TJH

===================================

and capital wont matter then ...........................

[vrdiver]

First of all, if I remember correctly, HYP1 was not set up as being the only income stream available to its owner.

No. It was offered as an alternative to an annuity, so quite feasible to be the only income stream.

the dividend income was not required to cover day to day living expenses, the aforesaid Pension would do that. HYP1 was to provide a pool of money for life's luxuries: a Cruise, a Caribbean holiday, a monthly supply of caviar or whatever else the owner was into. It therefore follows that even a significant drop in HYP1's dividends from one year to the next would not materially affect the owner, except perhaps being forced to holiday in Skegness that year!

No. Pyad made no reference to how the income should be spent. He confined himself to producing the HYP, not on controlling its owner.

Secondly, anyone suggesting that, because of the lack of diversification both by value and by income, HYP1 should immediately be re-balanced to make the income stream going forward "safer", must also accept that if such a re-balancing is now required, it would have already been required several years ago. ... <snip> ... Because there has been no re-balancing, it is now impossible to answer that question, without the introduction of hindsight bias, so I am afraid we shall just have to live with it.

I think you've missed the point somewhat. Nobody is saying that HYP1 hasn't delivered the goods. Just that it has delivered them with a fair amount of risk attached that could have been mitigated by rebalancing. You are absolutely correct that rebalancing would most likely come at a cost. Again, I don't think anybody would disagree.

In summary, HYP1 has delivered, but by taking some risks that many around here would not be comfortable with. What it has not done, in and of itself, is prove that HYP works, in the same way that surviving a barrel ride over Niagra falls doesn't prove that barrel riding is a safe form of transport.

VRD

[Dod101]

First of all, if I remember correctly, HYP1 was not set up as being the only income stream available to its owner. On the contrary, the original investment amount of £75,000 was arrived at because it was then the maximum amount allowable for withdrawal from a Personal Pension scheme. In other words the dividend income was not required to cover day to day living expenses, the aforesaid Pension would do that. HYP1 was to provide a pool of money for life's luxuries: a Cruise, a Caribbean holiday, a monthly supply of caviar or whatever else the owner was into. It therefore follows that even a significant drop in HYP1's dividends from one year to the next would not materially affect the owner, except perhaps being forced to holiday in Skegness that year!

Secondly, anyone suggesting that, because of the lack of diversification both by value and by income, HYP1 should immediately be re-balanced to make the income stream going forward "safer", must also accept that if such a re-balancing is now required, it would have already been required several years ago. It must also be accepted that such a re-balancing exercise, if it had been enforced from the very beginning, would have materially changed the results that we are now looking at, 19 years later. Sure, there would be no serious lack of value and income diversification, but what about the annual income now being enjoyed? Would that be lower now as a result of continually selling down the very holdings that, if left alone, would otherwise go on to create the impressive results we see today? Because there has been no re-balancing, it is now impossible to answer that question, without the introduction of hindsight bias, so I am afraid we shall just have to live with it.

Are you saying you remember the rules of the game of 19 years ago? I had never heard of TMF at that time and certainly had never heard of pyad.

You are introducing things that are certainly news to me, not to say that they did not exist of course...........

Dod

[StepOne]

The thing about HYP1 is that I don't imagine many people run a strict no-tinkering portfolio like it. And if there are any 'Dorises' around, they're not reading TLF.

Almost everyone reading these boards will have state pension entitlement plus company pension schemes. So having something HYP-like in addition to those is not as risky as it might seem. Personally I let my HYP picks get on with it, and I quite enjoy the fact that I never have to worry about how they are doing, or whether it's time to tinker. I've got another, trading, portfolio for things like that, and so far it's under-performing my 'risky' HYP

StepOne

[NeilW]

Loving the Wilsonian maths.


TJH has "proved" it wrong unless one believes he died some years ago!!


GS

I suspect I have not explained the assumption. Always remember that while you are earning an alternative income and 'building' you can rebalance the HYP. It's when you move into drawdown that the HYP becomes fixed.

[NeilW]

Just that it has delivered them with a fair amount of risk attached

There is no material risk attached. The mitigation has been sufficient over the time period *which included a major once in three generations global financial event*.

Therefore you are tilting at windmills. The risk you state has been shown to be immaterial by concrete example.

[Alaric]

. On the contrary, the original investment amount of £75,000 was arrived at because it was then the maximum amount allowable for withdrawal from a Personal Pension scheme.

As far as I recall there was never a cash limit on the amount that could be withdrawn without tax (25% of fund) from a Personal Pension scheme. There was however a limit that the maximum tax free cash was 150% of "final pensionable salary" from a defined benefit scheme. So you would need a final salary of £ 50,000 to get a cash payout of £ 75,000.

[vrdiver]

Just that it has delivered them with a fair amount of risk attached

There is no material risk attached. The mitigation has been sufficient over the time period *which included a major once in three generations global financial event*.

Therefore you are tilting at windmills. The risk you state has been shown to be immaterial by concrete example.

No. Simply no. That HYP1 did not suffer a catastrophic failure does not mean that all such unbalanced portfolios do not have a risk of losing a significant share. Yes, HYP1, over the last 19 years has "delivered the business", but that does not mean that the companies that have carried it thus far will continue to be successful for the next 19.

HYP1 started out with 15 equal shares. They are not equal now. QED individual shares can and do fail or "shrink". That HYP1 is now reliant on a small number of shares puts us in mini-HYP territory, where, when such concepts were discussed, it was ruled out as "too risky". Having a rump of small holdings does not balance the risk of relying on four or five shares for over 75% of your income.

There is no material risk attached. The mitigation has been sufficient

What mitigation? That's like saying you bet on rolling six sixes and won, therefore it wasn't risky! Mitigation is about protecting yourself from risk; how did HYP1 attempt to do that when the risk we are talking about is the potential failure of one of its major holdings to continue to pay dividends in the future?

The risk I am referring to is of course interruption to the income stream, as that is the main purpose of HYP. Mitigating that risk is, for example, achieved by ensuring no single failure will impact more than say, 10% of that income. HYP1 is risking over 20% income loss if either of its two major holdings has a significant problem.

VRD

[Backache]

I continue to believe that the developing concentration-risks to income and capital of taking such an approach

Yet after 19 years there is no evidence of the manifestation of this theoretical risk, and in that time we had the biggest financial crisis since 1929.

19 years is a retirement. You retire at 67 and you'd be 86 by then end of the period - and probably dead.

Which says that whatever risk level you are seeing falls into the same level of risk as having an operation and failing to recover from it. Would you turn down the operation?

I'd suggest HYP has proved it is 'good enough' and the approach is sufficient to render the risk mitigated for all practical purposes.

Life expectancy at retirement is 18.6 years for a male and a little longer for a female. As life expectancy is slightly skewed by people dying very shortly after retirement the average person lives longer than the average life expectancy.
Furthermore those with the wherewithall to invest a reasonable sum and not live entirely or nearly entirely from the state pension will have a somewhat longer than average life expectancy so the majority can expect to be still alive 19 years after retirement.

[ayshfm1]

I always multiply everything by 10 on those 75k portfolio's because then they fit my criteria, which is enough to live off from the start. HYP1 is a success it did what was intended, maybe it's been lucky but had you upscaled it by 10 you'd being having a pretty comfortable retirement.

But I don't like the concentration or the income volatility which gradually evolved.

So I drew three lessons.

15 shares wasn't enough for me, I wanted a lot more. I accept I'm pretty much certain to pick the dogs, which I think ends up impacting capital value more than income. But it should mitigate my two main worries at the macro level.
Cash buffer was necessary. I accept lower income as a consequence. But mitigates my income concern on micro scale.
IT's and pref shares would also feature, I accept I'm not a purist.

[Gengulphus]

I think HYP1 is a great experiment in how a completely 'hands-off' income-portfolio might develop, ...

Sorry, but I disagree, because HYP1 has not been run in a completely 'hands-off' way. If it had been, it would be missing its current holdings in BA Tobacco, BT Group, Dixons Carphone, GlaxoSmithKline, Persimmon and Pearson, because nothing would have been done following the cash takeovers of Gallaher, Associated British Ports, Alliance Boots (derived from original holding Boots), Blue Circle, Resolution (derived from original holding Britannic) and Scottish & Newcastle. Together, those six holdings produced £5,628.70 income out of the portfolio total of £10,557.29, and had a capital value of £76,339 out of the portfolio total of £159,682, so their absence would remove a bit over half of the portfolio's share income and a bit under half of the portfolio's share capital. It would have a large pile of cash instead of those holdings, which would go about 71% of the way towards replacing the capital (*) but only a tiny bit of the way towards replacing the income, cash interest rates being as derisory as they are.

Those are major differences from how HYP1 has actually performed, and in fact it's almost certainly an underestimate of those differences because a completely 'hands-off' version of HYP1 would also not have had the benefit of the topped-up holdings produced by reinvesting the proceeds of the numerous partial capital returns made by HYP1's holdings over the years. So the actual sizes of many of the 9 holdings it would have kept would be smaller than the real HYP1's, and the unproductive cash total larger - it would take a lot of work to establish just how big those differences are, but my gut feel is not as major as those due to cash takeovers but still quite noticeable (**). For completeness, there would also be some differences due to the fact that HYP1 has entirely voluntarily sold (***) on four occasions in order to simplify the admin: United Utilities rights in 2003 to simplify a complex rights issue, Anglo American's demerger on tiny holdings in Mondi in 2007, Banco Santander's takeover of Alliance & Leicester for shares in 2008 and GVC's takeover of Ladbrokes Coral for a mix of shares, cash and 'Contingent Value Rights' in 2018. For various reasons, those differences look quite minor - the United Utilities differences are essentially only a matter of share price fluctuations affecting the outcomes of different routes through the rights issue, the Mondi holdings would have been really small, the Alliance & Leicester holding was only about a third of its original size when it was taken over and Banco Santander's performance since October 2008 is not impressive, and the completely 'hands off' version of HYP1 wouldn't have faced the Ladbrokes Coral takeover at all because it wouldn't have had Ladbrokes Coral (it having been produced by reinvestment of the Blue Circle takeover proceeds in Hilton Group, which was later renamed Ladbrokes and later still merged to form Ladbrokes Coral).

The practical points I'm making with all that are:

A) In the long run, it isn't realistic to expect to run a HYP in a completely 'hands-off' way, because cash takeovers and returns of capital are basically one-way routes back to cash whose effect is cumulative. It's highly uncertain when they'll accumulate to enough to drastically affect the HYP's performance, but in the long run they're practically certain to do so, and HYP1's experience of five cash takeovers (plus a share takeover that was rather awkward for a demo 'paper' portfolio) in the three years 2006-2008 shows that it can happen quite quickly.

B) So in practice, a HYP needs a way to decide how to reinvest returned capital, and no way of doing that can be completely 'hands-off'. It can be quite close to completely 'hands-off' if it's a single company returning part of its capital, with its shares remaining on the market and without various complications, by using the mechanical rule of reinvesting the capital in the share that produced it - that's not completely 'hands-off' since the HYPer does have to do something, namely make the buy, but once the decision has been made to use that rule, the HYPer only needs to act on it, not make further decisions. Though complications can change that fairly easily - for instance, when Six Continents demerged Mitchells & Butlers in 2003 and renamed what was left of itself as International Hotels Group, it did a return of capital at the same time - and that did present a HYPer who wanted to follow the mechanical rule with an extra decision, namely how to split reinvesting it between the two resulting companies.

C) But that mechanical rule doesn't work for reinvesting capital returned by cash takeovers, since it's impossible to reinvest that capital in the share that produced it. Furthermore, experience of HYP1 says that cash takeovers are probably the biggest source of returned capital, and that without reinvesting it, a HYP's number of shares will probably decline over time. That's not absolutely certain, because demergers will produce increases to offset decreases caused by cash takeovers, but HYP1's record of six cash takeovers and two demergers (one of them only producing a small 'splinter' holding, which HYP1 sold) suggests probably not enough increases. So it seems likely that one will need to select new holdings from time to time, and that necessarily involves making decisions that are at least very hard to make well using a purely mechanical rule.

So I'm not criticising HYP1 for failing to be a completely 'hands off' HYP, since I don't think running such a HYP is a realistic aim - I'm just saying that HYP1 isn't one. It would be more accurate to describe it as a non-tinkering HYP, though the small number of voluntary sales it as done mean that that isn't a fully accurate description either. Perhaps calling it an admin-tinkering-only HYP would be best, since the common theme of the few voluntary sales it has done is avoiding admin to do with multi-stage corporate actions, tiny holdings, foreign holdings, etc. Note incidentally that such admin avoidance tends to be a rather more important factor in the HYPer's mind for a 'paper' demo HYP than it would be for a real-money HYP that was actually a significant part of the HYPer's financial prospects, both because the HYPer is probably going to be less willing to spend significant time and effort on it and because there won't be a broker collecting dividends, handling obligatory corporate actions and alerting voluntary corporate actions for the HYPer.

(*) Those takeovers delivered cash of roughly £13,556+£14,032+£9,328+£5,330+£3,881+£8,080 = £54,207 according to https://web.archive.org/web/20170213161 ... sort=whole. That's not necessarily totally accurate, as those figures were calculated by kool4kats in 2009 rather than by pyad at the time of the takeover - his figures tended to be a bit vague, e.g. he described the proceeds of the Associated British Ports takeover as "about £14,000" - but it should be pretty close.

(**) In particular, those due to IHG's numerous capital returns by a special dividend plus share consolidation must be mounting up!

(***) I say "entirely voluntarily" because there are also a few occasions on which it sold earlier than a near-certain takeover actually went through, just to get the admin done and dealt with. I don't have a record of exactly which occasions they were.

Gengulphus