Wizard wrote:... it could just be that the performance of HYP1 has had a large slice of luck in it. As an experiment it is merely a single data point. ...
FWIW, I agree about HYP1 probably having had a slice of good luck. HYP1 has had five really excellent income performers over the years, judging taken-over shares by the income their replacements produced in their first full year in HYP1 after the takeover:
Associated British Ports: bought 2000, first year income £204, taken over 2006, replacement BT produced £858 in 2007.
Gallaher: bought 2000, first year income £289, taken over 2007, replacement BATS produced ~£601 in 2008.
BATS: bought 2007, first year income ~£601, still owned 2019 with annual income £1,735.
Persimmon: bought 2008, expected first year income ~£540 (*), still owned 2019 with annual income £2,484.
Rio Tinto: bought 2000, first year income £178, still owned 2019 with annual income £2,547.
Note that I have not included the current 4th-highest income producer BT in that list because its income £875 in this last year is barely above its income £858 in 2007. I.e. its income performance hasn't been anything special, it's simply been coasting on the high starting position it got courtesy of Associated British Ports.
The three very successful long-term income growers out of those that it's still got produce £6,766 out of the portfolio total of £10,557, i.e. about 64% of it. For comparison, the whole portfolio was originally forecast to produce a yield of 4.8%, which on an investment of £75,000 is about £3,600 income. Or in other words, even if the other shares in HYP1 had produced no dividends at all (which is of course not the case), those three would by now be producing the originally-expected income with 3.4% compound growth - i.e. inflation-equalling or slightly inflation-beating growth. And on capital, those three shares have total capital value £72,716, which is 97.0% of the starting capital of £75,000, or if you also include Intercontinental Hotels, which has been a much better capital grower than income grower, the total comes to £92,567, or 123.4% of the starting capital.
On that 'what the really excellent performers have done even if you totally discount what everything else has done' basis, we have got a bigger sample than just the single data point of HYP1, because we've got also got 2007-or-2008 portfolio listings for HYP2, HYP3 and HYP4:
*
HYP2 certainly had a very successful long-term income grower in the form of BATS, and I think Legal & General also qualifies - the dividend for 2003 (the first company financial year from which HYP2 received all the dividends) was 5.06p, for 2018 16.42p. Hanson was
taken over for cash at a price of 1100p in August 2007, a big enough capital increase on the original price of 320p that judged by its replacement's first year income, it will almost certainly have been a third, albeit a historical one rather than a still-current one (like Associated British Ports and Gallaher in HYP1). BAA's
takeover for cash at a price of 935p in June/July 2006 and BOC's
takeover for cash at a price of 1600p in September 2006 might well make them fourth and fifth good income performers, again historical only, though at only just over twice their original prices of 465p and 795p respectively they're not quite as convincing. AMVESCAP
became Invesco in May 2007 and then
moved abroad in November 2007. I believe it is still on the New York Stock Exchange, but since pyad's demo HYPs have avoided having foreign holdings and have made exceptions to their no-tinkering policy when a holding or a demerged part of a holding has gone abroad, I think HYP2 would have sold and replaced it at that point. That
might make it yet another historical excellent income grower, but I don't have any clear idea what price it might have been sold at, so cannot tell whether it actually does. And of course, replacements for Hanson, BAA, BOC and AMVESCAP/Invesco might have included Persimmon or other shares that went on to do very well, but that's very speculative and so not to be counted in this exercise.
*
HYP3 had Gallaher, which was
replaced by BATS just a few days after that report. Gallaher was held for too short a period to have been a serious contributor to income growth for the portfolio, but BATS will have gone on to produce total dividends of 66.2p in 2007 that had more-than-tripled to 203p in 2018. It also had Legal & General, which has already been discussed for HYP2 above, and Unilever. I had a look on the Wayback Machine for an archived copy of the article in which it was selected - finding such a copy when I don't already have a link is something that I usually cannot manage, but on this occasion I've managed to track it down via a chain of links from links I do have - it's
here. It was bought in June 2006 at 1204p, on a forecast dividend of 47.9p (putting it on a forecast yield of 4.0% - which wouldn't be enough to count as high-yield on this board now, but as the article says "
Not huge but current yields in general are not that great once you've exhausted the few real biggies at the top of the FTSE 100. However it is decently above the current average for this index, which is 3.3%.". With its total dividend for 2018 having been 135.3p and its current share price of 4548p, it's undoubtedly been an excellent performer for HYP3. HYP3's
investment in ITV has also done very reasonably overall, with dividends rising to 8p in 2018 from the 3.12p historical/3.55p forecast quoted in the article, and its investment in Scottish & Newcastle might very plausibly have been reinvested in Persimmon, since HYP1's investments in Scottish & Newcastle and Resolution were invested in Persimmon and Pearson and HYP3 already had Pearson, but neither of them seems to me to be both definite and truly excellent, so I won't count them. It's also worth mentioning that HYP3 was set up as a tinkering HYP, so it's possible that further excellent performers might have arisen as a result of tinkering by pyad - or indeed that actual excellent performers would have been tinkered away by him at some point where things were looking bad for them. But that leads to a mass of possibilities that cannot be resolved without introducing far too much hindsight into the projected portfolio performance, so the only way I can reasonably project HYP3 to the present day is by assuming it was held by a HYPer who owned HYP3 until pyad last said anything about it (in
February 2008 AFAIAA) and then decided to leave it untinkered - call it NUHYP3 for Nearly Untinkered HYP3 if you like (it's nearly untinkered rather than completely so because its original selection of Rank was
replaced by William Hill while it was being constructed).
* A similar comment applies to HYP4, including having been tinkered once during its construction (to replace F&C Asset Management by BATS), so call the resulting portfolio NUHYP4. IMHO it has had truly excellent long-term performers in the form of BATS, Persimmon and Compass Group. The first two are adequately discussed above, and Compass Group's dividend of 10.8p for the year to 30/09/2007 (which had been completed when it was selected but not yet reported on) and share price of about 330p have risen to 37.7p for the year to 30/09/2018 and about 2014p currently. That makes it more than a 'six-bagger' - a stunning capital performance that puts even Persimmon's to shame...
So summarising those, I can identify the following excellent long-term income performers in each of HYP1, HYP2, NUHYP3 and NUHYP4 (the last two differing from HYP3 and HYP4 only in having been run differently to how pyad had intended, after pyad abandoned them):
HYP1: Total income £6,766 from 3 holdings worth £72,716, from BATS producing £1,735 income and worth £25,290; PSN producing £2,484 income and worth £25,981; RIO producing £2,547 income and worth £21,445.
HYP2: Total income ~£2,855 from 2 holdings worth ~£44,144, from BATS producing ~£1,733 income and worth ~£25,261; LGEN producing £1,122 income and worth ~£18,883.
NUHYP3: Total income ~£2,191 from 3 holdings worth ~£43,591, from BATS producing ~£862 income and worth ~£12,044; LGEN producing ~£749 income and worth ~£12,611; ULVR producing ~£580 income and worth ~£18,936.
NUHYP4: Total income ~£2,175 from 3 holdings worth ~£49,889, from BATS producing ~£607 income and worth ~£8,850; PSN producing ~£985 income and worth ~£10,441; CPG producing ~£583 income and worth ~£30,608.
So all three of HYP2, NUHYP3 and NUHYP4 recover around 60% of their initial capital cost (£80k for NUHYP4, £75k for the others) from those few excellent long-term income performers - not as much as HYP1's 97%, but the remaining 12 or 13 holdings only need to be worth a bit over £30k between them (i.e. about half the £60-65k invested in them) for the whole portfolio to have broken even on capital. I have little doubt that they'll have managed that - there are enough OK performers among those shares that even the undoubted major losers won't have dragged their average down to a 50%ish loss or worse. And on income, it depends how much one was looking for, but if one assumes a 5% target yield (so income averaging £250 per holding, or £4,000 for NUHYP4, £3,750 for the others), the other holdings only need to average about £70-£140 per holding to make the target after taking the income contributions of those 2-3 holdings into account. That's not as good as HYP1's top three totally smashing the target ion their own. but I've little doubt they would have managed it.
The same income target compounded by say 3% p.a. to compensate for inflation would be more challenging - very roughly £6,580 for HYP1, £5,930 for HYP2, £5,590 for NUHYP3 and £5,700 for NUHYP4. HYP1 has overshot that, HYP2 would require its remaining 13 holdings to average about £235 per holding, NUHYP3 its remaining 12 holdings about £283 per holding and NUHYP4 its remaining 13 holdings about £271 per holding. I'm not certain whether the others besides HYP1 would have achieved those targets, but none of them strike me as impossible - remember that those remaining holding will include some decent dividend growers as well as cutters, since I've only filtered out the excellent dividend-growers.
So the evidence I've seen suggests that HYP1 has enjoyed significantly better luck than HYP2, NUHYP3 and NUHYP4, probably more in terms of the degree of excellence of its excellent dividend-growers than the number of them, and at a guess that luck is mainly the luck of having been started while the market was still distorted in an overvalue-low-yield/undervalue-high-yield way by the deflating tech bubble (a suggestion made earlier in the thread). But it looks unlikely that those three will have done significantly worse than a FTSE100 tracker - the question that's open in my mind is whether they'll have done roughly equally well or significantly better. More detailed study of the holdings that can be studied (i.e. those present in them in late 2007 or early 2008, when we last had a "from the horse's mouth" report on them, and not taken over for cash since) would produce better lower bounds on how well they've done, but whether they would be high enough to resolve that question - but it would be a lot more work!
(*) Persimmon's purchase in HYP1 was on an unknown date in 2008, which complicates matters considerably about judging exactly what its actual first-year income was - though it was almost certainly much less than had been expected, possibly nothing at all. The expected first year income figure I've given is basically what the previous year's dividend had been.
Gengulphus