10 Years Strong

[Wizard]

An excellent observation. I can couch it in terms that 'masterly inactivity' led to no great catastrophe (which arguably could happen with straight equity investing, rather than via a collective tool like an OEIC, IT or ETF), but you're right, that with one business a year roughly going into forced disposal, left alone, after ~30 years a HYP may have no or very few active stocks left, no dividend income, and simply be mouldering in unproductive cash. That would include quasi-disposals such as the PNN and TSCO special dividends as they sold off portions of their businesses.

The dividends could have been handled by using the DRIP function that my ISA provider allows. And that may also have handled special dividends. But DRIP brings its own issues of possible imbalance (and by its nature, those stocks paying dividends, or the larger dividends, get reinvested into the most).

I've suspect the HYP could have performed better in income terms had we reinvested the cash along the way - as we'd have only picked new stocks with a reasonable yield, or topped-up those existing HYP stocks currently paying dividends. Capital is trickier to say, as we may have picked another dog or two.

Without the forced disposals I doubt we would have added beyond 35 stocks, the last three years would have been a somewhat mechanistic reinvestment into the stocks already in the HYP. I do have caps (similar to TJH's/Geng's from memory) around amount invested, current value, and income, ranging from the company level (5%), sector (10%), super-sector (15%) and industry (20%). Currently Financial industry is 'red' for value (25%) and income (24%) so would not be getting more invested until some balancing happened, as is Basic Resources (the two miners) with income 29%! from just two (of 30) stocks, skewed massively by their largesse in a cyclical market.

I was surprised by how much cash there was. The dividends cash accounts for around 15% of the HYP value (but then after 3.5 years, at say 4% p.a. dividend yield, it's not THAT surprising), the specials another 3%, and the forced disposal monies are 12%. So 30% cash in total.

Yes, we'll now most likely put the cash back into the HYP, to perhaps add a few more stocks back to ~33-35 (currently 30) to further improve diversification, and then top-up others to restore a little balance. We might also sell off some of the rump items that aren't generating dividends currently, like KIE, CNA and WG. While they have a bit of diversity (certainly KIE) they are relatively small sums now, aren't paying dividends currently, and probably not that HYPable.

The rebalance looks a little tricky right now, with many firms breaking stride on dividend payments for Covid (understandable). For some it was a brief pause, others rebased, and a few have not restarted paying again. With 'blocks' in financials, miners and pharma (10), plus those not currently paying (5) that's half the options gone, then a further bunch with recent cuts (Covid or otherwise) (7), that leaves only eight possible stocks (TSCO, TATE, PNN, NG, BVIC, ULVR, BA, SGE) but all-bar-two are yielding below the 4.0%, so would be dilutive, although once topped up would start to bring some of the 10 'blocks' back into play. Choices, choices.

Thanks for the very considered reply. Your "'masterly inactivity" has indeed resulted in no disaster, presumably to a great extent showing the benefit of the initial level of diversification in your portfolio. I know there are frequent debates here about how many shares are needed to achieve an adequately diversified portfolio, and 15 is often quoted as enough. In the case of a long period of activity your experience may show that more is better from the perspective of forced disposal.

I do not have the time to do what I think would be a time consuming piece of analysis, but I do wonder if the issue of forced disposal and / or significant capital returns triggering the need for action to reinvest, even in a draw down portfolio, is greater now than it was in past periods.

It seems our thoughts are aligned on the point that the inactivity probably cost income due to there being no reinvestment, but how much is impossible to say. In this particular period you could have reinvested in a high yielding share at the beginning of the three years that then shortly afterwards cut or halted its dividend for most of the period you were away. I also agree with you, a dividend reinvestment plan (I presume that is what DRIP stands for) could have caused as many problems (imbalance) as it solved (drag) and indeed given the COVID period could have been only a partial solution if the shares reinvested into early on then took action on dividends in a response to COVID.

As you say rebalancing may be tricky, and I also agree that with a large proportion of the portfolio in cash in some cases some reinvestment may allow you to look at some of the currently blocked candidates again. I tried looking at Financials as an example, I assumed four constituents and as far as I can see there is scope for candidates being blocked at different and multiple levels. For example, Admiral looks blocked at all levels, but HSBC (yielding over 4%) maybe only at the higher levels of aggregation. I had to try and come up with a view of income concentration, as this is not explicitly shown. I took the value concentration as shown, divided by the percentage of median value then multiplied by percentage of median income to arrive at an income concentration number. But something is not right, if I am correct and you Financials grouping is made up of ADM, AV., LGEN and HSBA I only make the income concentration 14%, rather than the 24% you mention. However, using the aprroach above I do arrive at 25% income concentration. So something is wrong with my attempt to reach a proxy. Are REITs in your Financials?

Finally you mention shares blocked as cutters. I think, given the rather unusual COVID period some flexibility could be taken here. IMHO it is reasonable to look differently on a share which cut and rapidly resumed at the historic dividend level (possibly even paying the missed dividends), as compared to one which merely used COVID as cover for taking action to necessarily rebase its dividend.

It would be interesting if you could provide further insight when you come to making your decisions around reinvestment.

[JohnnyCyclops]

Thanks for the very considered reply. Your "'masterly inactivity" has indeed resulted in no disaster, presumably to a great extent showing the benefit of the initial level of diversification in your portfolio. I know there are frequent debates here about how many shares are needed to achieve an adequately diversified portfolio, and 15 is often quoted as enough. In the case of a long period of activity your experience may show that more is better from the perspective of forced disposal.

I do not have the time to do what I think would be a time consuming piece of analysis, but I do wonder if the issue of forced disposal and / or significant capital returns triggering the need for action to reinvest, even in a draw down portfolio, is greater now than it was in past periods.

It seems our thoughts are aligned on the point that the inactivity probably cost income due to there being no reinvestment, but how much is impossible to say. In this particular period you could have reinvested in a high yielding share at the beginning of the three years that then shortly afterwards cut or halted its dividend for most of the period you were away. I also agree with you, a dividend reinvestment plan (I presume that is what DRIP stands for) could have caused as many problems (imbalance) as it solved (drag) and indeed given the COVID period could have been only a partial solution if the shares reinvested into early on then took action on dividends in a response to COVID.

As you say rebalancing may be tricky, and I also agree that with a large proportion of the portfolio in cash in some cases some reinvestment may allow you to look at some of the currently blocked candidates again. I tried looking at Financials as an example, I assumed four constituents and as far as I can see there is scope for candidates being blocked at different and multiple levels. For example, Admiral looks blocked at all levels, but HSBC (yielding over 4%) maybe only at the higher levels of aggregation. I had to try and come up with a view of income concentration, as this is not explicitly shown. I took the value concentration as shown, divided by the percentage of median value then multiplied by percentage of median income to arrive at an income concentration number. But something is not right, if I am correct and you Financials grouping is made up of ADM, AV., LGEN and HSBA I only make the income concentration 14%, rather than the 24% you mention. However, using the aprroach above I do arrive at 25% income concentration. So something is wrong with my attempt to reach a proxy. Are REITs in your Financials?

Finally you mention shares blocked as cutters. I think, given the rather unusual COVID period some flexibility could be taken here. IMHO it is reasonable to look differently on a share which cut and rapidly resumed at the historic dividend level (possibly even paying the missed dividends), as compared to one which merely used COVID as cover for taking action to necessarily rebase its dividend.

It would be interesting if you could provide further insight when you come to making your decisions around reinvestment.

You've spotted all the pertinent points. Thank you. Let me try and cover some in return.

Diversity
The 'classic' 15 stock HYP portfolio does provide sectorial diversity. However I opted for a 30+ model, doubling-up all sectors where I could, to achieve a measure of company diversity too. Thus, both AV/LGEN, BP/RDSB, BHP/RIO, etc. The latter is a good example that both miners had similar fortunes in economic cycles (sector risk), and both have weathered them (company risk). Whereas BA/COB saw the latter flounder and get bought out (company risk) whereas BA has continued to deliver well in HYP terms.

Classification
The above answer takes us to classification. Following the ICB classes leads to 41 sectors (or did until recent changes expanded to 45 - more below). Two sectors were Equity ITs and Nonequity ITs and I chose to exclude Tobacco (personal choice), leaving 38 sectors. I've acknowledged I was following a 'doubling up' approach, so unlikely to buy a stock in every one of 38 sectors. But, the classic HYP at 15 would likely need to leave out a few 'useful' sectors. We ended up in 23 (of 38) sectors, before the forced sales and CLLN demise.

Upon my return to the HYP/investing, I saw the ICB had updated the classification. Now 11 Industries (prev 10), splitting REITs out from Financials. 20 Super-Sectors (prev 19) and 45 Sectors (prev 41).

Our HYP
I spent a bit of time last night remapping the HYP to the new classes. There's a little bit of sleight of hand, but my moving REITs (SGRO/BLND) out of Financials it sees the latter 'improve' to the numbers you'd spotted related to the four remaining. The Insurance Super-Sector is still 'blocked' but the one Bank (HSBC) is back in play. Other Fools had commented previously about the wisdom/utility of some bundling by sector.

I've also reflected (not for the first time!) how useful (or not) the official classes are. On the upside it provides a framework, and a methodical way to analyse sector/industry exposure, and explore optional HYP candidates. On the downside, it can be rigid, and should I really be lumping PNN and CNA into the same "Gas, Water & Multiutilities" sector, or split them out a little. The new sector "Personal Care, Drug & Grocery Stores" sees ULVR lumped in with TSCO - I'm not sure it's helpful.

It turns out there's still no Automobiles & Parts HYP candidates(!), and the whole Industrials industry group still looks sparse in the FT350 (a feature of the types of companies that LSE list, I expect) - rueing the loss of BBA (SIG), CLLN and COB.

Covid Cutters
My first job back was to review the recent dividend histories. There's been a mixed bag. Some delayed then paid. Some reduced and kept paying. Others skipped a turn or two and then mostly came back with reduced payments, possibly addressing weak dividend cover even before Covid. Some have not yet restarted paying. I'll treat each on its merits.

The Benefit of Notes
As a final mention, having kept copious notes, it does help me review. In looking at the sector analysis table, of held stocks, good HYP candidates, dark horses, and no hopers, in the Leisure Goods sector I saw I'd noted Games Workshop (FT250) as a high yielder (~8%) and low market cap (£250m) so didn't buy at the time (below £1bn). That was around 2014/15. It's now £3.8bn cap, but a low yielder (~2%) despite dividend CAGR of 40% (5yr) and 20% (10Yr) - ho hum

[JohnnyCyclops]

In answer to a question on HYP yields, I've now worked through four years of data, and got to some portfolio yields.



Obviously, many ways to report yield annually (start year value, end year value, etc.) and how to deal with the fact new monies were being added, and thus new dividends being generated, up unitl early 2018.

Of the two columns given, both are on the HYP value at the start of the year. The first column includes cash held in the HYP (ISA) from accrued dividends and forced sales. It includes all dividends, even in their first year of payment.

The second column focuses on the stock holding only (not cash), and only includes dividends from 'full holdings' of stocks - this helps eliminate (in the building phase) the impact of a buy that generates no or only partial dividends in its first year.

Yes, I know unitisation would address a lot of the above, but I've not got around to updating the unitising, and recall I didn't have a happy experience with it prior to 2018, despite being numerate and Excel-savvy

SO WHAT
Basically, we've got an equity portfolio generating 4%+ returns (plus whatever the capital is doing). At a time when looking at my pension pots, L&G say they'll give me an annuity rate just under 2% if I take the pension at age 60 in around a decade PLUS L&G will swallow the entire capital.

The effects of Covid don't fully show up here. For the year ending 31 Mar 2020 I'm taking the HYP value on 1 Apr 2019 and applying the dividends through that year to derive the yields (i.e. I invest X and receive Y). The share price "hit" from Covid on most of our holdings creates a deflated valued on 1 Apr 2020 but then are set against lower dividends in the following year through to 31 Mar 2021. If I look at yields vs year-end values it's 5.24% at 31 Mar 2020 and 2.77% at 31 Mar 2021 (i.e. the Covid share price hit happens in March 2020, but the drop in dividends is in the 12 months following).

The impact of holding ~20% of the HYP value in cash as at 31 March 2021 shows up on that date's yields with the total yield of 3.37% lying around 20% below the "full holding excl cash" yield of 4.19%. Time to put that cash back to work!

HYP years from 2011 to 2013 are pretty meaningless for yields as that was the early years of building and massively skewed by partial holdings, etc. so not included here.

Ten years in, and HYP is delivering an acceptable income stream seamingly 2x better than a leading annuity provider, plus that allows us to retain the capital (which has also grown over the decade, notwithstanding CLLN!).

[Itsallaguess]

In answer to a question on HYP yields, I've now worked through four years of data, and got to some portfolio yields.



The effects of Covid don't fully show up here. For the year ending 31 Mar 2020 I'm taking the HYP value on 1 Apr 2019 and applying the dividends through that year to derive the yields (i.e. I invest X and receive Y).

The share price "hit" from Covid on most of our holdings creates a deflated valued on 1 Apr 2020 but then are set against lower dividends in the following year through to 31 Mar 2021.

If I look at yields vs year-end values it's 5.24% at 31 Mar 2020 and 2.77% at 31 Mar 2021 (i.e. the Covid share price hit happens in March 2020, but the drop in dividends is in the 12 months following).

It's an interesting table JC, but it would be a much more interesting table if we were perhaps able to see the level of dividends delivered over the same yearly periods as well, and especially over the past few years where we know the HYP went untouched - do you have that data available from the trawl you carried out to generate the above?

I'd completely understand if you might prefer not to disclose precise levels of monetary values on a public board, so a level of 'scaling' would hopefully get around that potential issue, and still leave us with an appropriate data-set to visualise against the 'yield-only'' data currently seen in the above table...

Great to see you back..

Cheers,

Itsallaguess

[JohnnyCyclops]

It's an interesting table JC, but it would be a much more interesting table if we were perhaps able to see the level of dividends delivered over the same yearly periods as well, and especially over the past few years where we know the HYP went untouched - do you have that data available from the trawl you carried out to generate the above?

I'd completely understand if you might prefer not to disclose precise levels of monetary values on a public board, so a level of 'scaling' would hopefully get around that potential issue, and still leave us with an appropriate data-set to visualise against the 'yield-only'' data currently seen in the above table...

Great to see you back..

Cheers,

Itsallaguess

Thank you. This morning I was catching up on some HYSS posts from the summer (TR vs dividend/yield) and realised my post from last night above only gives half the story (yield) not the other half (income). While 'patting myself on the back' for our 4%+ yield trumping L&Gs ~2% annuity rate, I had casually ignored that our 4% was off a lower portfolio valuation (Covid) and then saw an actual drop in income (dividends).

Without the (dreaded!) unitisation task, the best I can do for myself is as follows. No new monies were added or stocks bought after 31 Mar 2018, so that provides a baseline. No dividends or special dividends reinvested (or removed from the ISA). Interactive Investor fees have come out of the total (through depleting cash held). There were four 'forced' sales (BBA(SIG), GNK, ISAT, COB) that turned stocks into cash (not reinvested). For the dividend numbers below I've excluded most special dividends (I've left in ADM and RIO, though).



In income terms (rather than yield) we are obviously down (around 36% year-on-year) as a result of Covid. There has been a bounceback in the first four months of the 2021/22 year (Apr/Jul) but still not enough get back to 2019/20 levels (maybe a 1/3rd recovered, now down -34% whereas last year it was down -49%).

However, some of that income depletion is also a factor of carrying so much in cash right now. Cash was ~20% of the portfolio in Mar-2021 now risen to ~25% with the summer forced disposal of BBA (SIG). I see that represented in the yield numbers I gave above/yesterday.

[Itsallaguess]

This morning I was catching up on some HYSS posts from the summer (TR vs dividend/yield) and realised my post from last night above only gives half the story (yield) not the other half (income). While 'patting myself on the back' for our 4%+ yield trumping L&Gs ~2% annuity rate, I had casually ignored that our 4% was off a lower portfolio valuation (Covid) and then saw an actual drop in income (dividends).

Without the (dreaded!) unitisation task, the best I can do for myself is as follows.

No new monies were added or stocks bought after 31 Mar 2018, so that provides a baseline. No dividends or special dividends reinvested (or removed from the ISA). Interactive Investor fees have come out of the total (through depleting cash held). There were four 'forced' sales (BBA(SIG), GNK, ISAT, COB) that turned stocks into cash (not reinvested). For the dividend numbers below I've excluded most special dividends (I've left in ADM and RIO, though).



In income terms (rather than yield) we are obviously down (around 36% year-on-year) as a result of Covid. There has been a bounceback in the first four months of the 2021/22 year (Apr/Jul) but still not enough get back to 2019/20 levels (maybe a 1/3rd recovered, now down -34% whereas last year it was down -49%).

However, some of that income depletion is also a factor of carrying so much in cash right now. Cash was ~20% of the portfolio in Mar-2021 now risen to ~25% with the summer forced disposal of BBA (SIG). I see that represented in the yield numbers I gave above/yesterday.

Thanks JC - and yes, it was the primary focus on just yields in your earlier table that prompted me to ask about the underlying income, as I had a suspicion that it might have been illuminating to do so...

I posted on a separate company-specific thread a few weeks ago to show how yields on their own can often remain quite static and perhaps give an incorrect impression on underlying income -

https://www.lemonfool.co.uk/viewtopic.php?f=54&t=28087&p=434164#p434164

As can be seen from the above link, where underlying share-price rises might accompany a rising dividend over longer periods, then concentrating solely on a single yield figure that's remaining fairly steady *might* hide the fact that received income might sometimes rise quite considerably over the same period, and clearly the opposite is also possible, where a relatively steady yield figure might perhaps hide an underlying *drop* on both the capital and the income side of things, and perhaps give the impression things are better than they are on the actual, 'real-life income' side of the story...

Cheers,

Itsallaguess

[JohnnyCyclops]

Thanks JC - and yes, it was the primary focus on just yields in your earlier table that prompted me to ask about the underlying income, as I had a suspicion that it might have been illuminating to do so...

I posted on a separate company-specific thread a few weeks ago to show how yields on their own can often remain quite static and perhaps give an incorrect impression on underlying income -

https://www.lemonfool.co.uk/viewtopic.php?f=54&t=28087&p=434164#p434164

As can be seen from the above link, where underlying share-price rises might accompany a rising dividend over longer periods, then concentrating solely on a single yield figure that's remaining fairly steady *might* hide the fact that received income might sometimes rise quite considerably over the same period, and clearly the opposite is also possible, where a relatively steady yield figure might perhaps hide an underlying *drop* on both the capital and the income side of things, and perhaps give the impression things are better than they are on the actual, 'real-life income' side of the story...

Cheers,

Itsallaguess

Very much so. In totting up dividends received I'd seen quite a Covid dent. But then comparing the start of each year's value (1 Apr 20xx) to that year's subsequent income, gave the impression of a fairly static yield. Whereas, end of year values didn't, especially the 'crunch' date around March 2020 when valuations were lowest and dividends highest (but about to tumble).

I think unitisation will help too in giving a clearer picture. I've 'only' got to back-calculate the HYP value on seven different dates over the last four years

[funduffer]

I shouldn't worry too much JC, nearly every HYP reporting last financial year saw a dip of 30% to 40% in HYP income due to the pandemic.

Mine was 36.7% after a year of very little activity.

Anyway, you has a nice cash buffer to see you through!

FD

[tjh290633]

Without the (dreaded!) unitisation task, the best I can do for myself is as follows. No new monies were added or stocks bought after 31 Mar 2018, so that provides a baseline. No dividends or special dividends reinvested (or removed from the ISA). Interactive Investor fees have come out of the total (through depleting cash held). There were four 'forced' sales (BBA(SIG), GNK, ISAT, COB) that turned stocks into cash (not reinvested). For the dividend numbers below I've excluded most special dividends (I've left in ADM and RIO, though).

It's interesting to speculate about the likely outcome on the dividend front, had you reinvested those accumulated cash balances as they arose. I'm thinking that you would have gone for companies neither ceasing to pay nor reducing their dividends.

I had to introduce two new holdings, one to replace Indivior and the other to replace William Hill. I chose Primary Health Properties (PHP) in 2020 and IG Group (IGG) in 2021.

Presumably you would have gone for 4 new shares at the relevant times.

TJH

[JohnnyCyclops]

It's interesting to speculate about the likely outcome on the dividend front, had you reinvested those accumulated cash balances as they arose. I'm thinking that you would have gone for companies neither ceasing to pay nor reducing their dividends.

I had to introduce two new holdings, one to replace Indivior and the other to replace William Hill. I chose Primary Health Properties (PHP) in 2020 and IG Group (IGG) in 2021.

Presumably you would have gone for 4 new shares at the relevant times.

TJH

Yes, ideally I'd have picked four replacement stocks, chiefly to maintain levels of sector and company diversity. What those four would have been is now moot (although BBA(SIG) is only from June this year). It's likely I'll still do that (add four new) but there are also accumulated dividends to reinvest. I'm not beholden to zdding four new, and might just reinvest all the cash into the remaining 30 holdings.

[JohnnyCyclops]

Hurrah!!! I've managed to work through the unitisation steps. The main effort was back calculating HYP values on seven dates, by looking up each stock's historic share prices. Then just needed to plug those numbers into existing unitisation sheets (fortunately very few transactions (income units) or fresh cash added (accumulation units). All done now.

Quite pleased to see the strong dividend per unit grow in 2017 & 2018. I can see more clearly that the 2021 Covid 'hit' drops us back six years to 2015 in dividend terms, but hopefully this year will begin to catch that up.

The following is on an accumulation basis. Baseline was 100.0 on 03 April 2011.