Sell or not sell

[IanTHughes]

If we take HYP1 as an example of what can happen over many years though Ian, we do know *now* which shares you see as the 'best performers', as we can see that two of them are delivering nearly 50% of the HYP1 income.

If we take your view that all the market has done has been to 'expose' the best performing shares from the original HYP1 selection, I think Alaric has a point if he were to ask why not simply now sell the large number of shares delivering the other half of the income, and plough that capital into what we can clearly now see as 'the two best performers'.....

Is that something you'd countenance?

No of course not!

How would we know that those two, now historical "best performers", would continue into the future as the two "best performers"?

Hang on though Ian - the reason you've given for not rebalancing HYP1, so that it's not relying on just two shares to deliver nearly 50% of the current dividend-income is that you'd be, in your own words - 'cutting your winners'.

But now you seem to also be saying that we shouldn't sell the dog-half of the HYP1 portfolio and concentrate all the HYP1 capital into those two stellar-performers, because we don't know that they are going to 'continue to be best-performers'.

You do seem to want it both ways Ian....

Surely it can't be both ways, so which is it?

Sorry, but you please hang on a moment. In your post above, you did not ask me whether, if I owned HYP1 as it is now currently constituted, I would countenance re-balancing it now. In fact what you asked was whether I would:

1) Completely un-balance it by concentrating all capital into only those two, now historical, "best performers", or
2) Starting out from scratch, re-create such an un-balanced portfolio as HYP1 is now, over-concentrated on those two, now historical, "best performers"'.

The answer to both questions was of course no.

Furthermore the question of whether, if I owned HYP1 as it is now currently constituted, would I now countenance re-balancing it is, if you will excuse me, irrelevant.

The scenario is: someone, not me of course, has owned HYP1, for 18+ years, right from the outset, without any re-balancing process at all, such that years ago the concentration of the portfolio income became un-balanced and has now ended up so completely un-balanced as you believe it now is.

The relevant question you should be asking is: would that person, after years of inactivity, suddenly spring into action and re-balance it now? To me the answer is clear: that person would do no such thing!

As I tried to explain in my earlier posts, if you want to avoid such over-concentration of Income, you have to do so from the outset, you do not wait 18+ years and only then act! Furthermore, I whole-heartedly agree that such a process, if properly conducted from the outset, will surely result in less over-concentration of Income, not least because there will be less Income growth to create over-concentration in the first place!

Would I institute such a process from the outset, possibly damaging, even foregoing Income Growth? Doubtful!


I do hope that is now clear


Ian

[Alaric]

No of course not! How would we know that those two, now historical "best performers", would continue into the future as the two "best performers"?

I think the point is that the hypothetical review is taking place a few years into the portfolio when the issue of unbalanced income has started to emerge. The choice would be to continue the run the self identified winners so far or switch partly from them to other shares of a similar yield so as to reduce the risk of the income drying up because of a shock to a particular share or sector. So for HYP1 it's a review taking place around 2005 or 2006.

[IanTHughes]

If we take HYP1 as an example of what can happen over many years though Ian, we do know *now* which shares you see as the 'best performers', as we can see that two of them are delivering nearly 50% of the HYP1 income.

If we take your view that all the market has done has been to 'expose' the best performing shares from the original HYP1 selection, I think Alaric has a point if he were to ask why not simply now sell the large number of shares delivering the other half of the income, and plough that capital into what we can clearly now see as 'the two best performers'.....

Is that something you'd countenance?

No of course not! How would we know that those two, now historical "best performers", would continue into the future as the two "best performers"?

I think the point is that the hypothetical review is taking place a few years into the portfolio when the issue of unbalanced income has started to emerge. The choice would be to continue the run the self identified winners so far or switch partly from them to other shares of a similar yield so as to reduce the risk of the income drying up because of a shock to a particular share or sector. So for HYP1 it's a review taking place around 2005 or 2006.

So, back in 2005 or 2006, you would have concentrated all the portfolio value into the, at that time, two "best performers"?

Really? How would that "reduce the risk of the income drying up because of a shock to a particular share or sector"?


Ian
By the way, I have completed your quote of my post, so as to ensure no misunderstanding

[Alaric]

So, back in 2005 or 2006, you would have concentrated all the portfolio value into the, at that time, two "best performers"?

Someone wishing to reduce income shock risk would sell part of the holdings where the income had exceeded a risk tolerance. Someone content with the risk would have let them run. If you really wanted to dial up the risk, bet the lot on the winners !

But if when you started, you had 15 stocks or sectors, none of which represented more than 10% of the portfolio income, would you not agree that if over time you had 20% of the income from one stock or sector, your income was more at risk than when you started? One poster observed that if it had increased at a rate much greater than inflation, you might not be unduly bothered as even if you lost 20% to a dividend suspension, your necessary income would be untouched.

[tjh290633]

Something which seems to have been forgotten, is that the imbalance of both income and weight was the result of original holdings being taken over. Because the proceeds of those takeovers was put into a single share each time, the imbalance was preserved.

It is not down to the shares originally chosen but to the action taken on takeover. Had the current holdings been bought at the original equal weight, the situation would be completely different.

The argument is based on a false premise.

TJH

[IanTHughes]

So, back in 2005 or 2006, you would have concentrated all the portfolio value into the, at that time, two "best performers"?

If you really wanted to dial up the risk, bet the lot on the winners !

Which is what you stated you would have done, back in 2005 or 2006!

Based on the question originally posed by itsallaguess:

If we take your view that all the market has done has been to 'expose' the best performing shares from the original HYP1 selection, I think Alaric has a point if he were to ask why not simply now sell the large number of shares delivering the other half of the income, and plough that capital into what we can clearly now see as 'the two best performers'.....

Is that something you'd countenance?

To which I very clearly responded:

No of course not! How would we know that those two, now historical "best performers", would continue into the future as the two "best performers"?

Whereas your equally clear response was:

So for HYP1 it's a review taking place around 2005 or 2006.

So I must ask again: if as you appear to recommend, one was to concentrate all the portfolio capital into what are at that time, the two "best performers", how would that "reduce the risk of the income drying up because of a shock to a particular share or sector"?


Ian

[Alaric]

So I must ask again: if as you appear to recommend, one was to concentrate all the portfolio capital into what are at that time, the two "best performers", how would that "reduce the risk of the income drying up because of a shock to a particular share or sector"?

I was giving three options, the first of which reduces income risk, the second and third increase it. The second is the "do nothing" one and allow concentration of income to potentially increase.

It did strike me that a review in 2005 or 2006 which switched income from 20% banks, 10% tobacco, 70% other to 15% banks, 15% tobacco, 70% other would have looked prescient by 2008, whereas one that went the other way, rather less so. But any holding above nil in banks in 2008 would not have been good.

[PinkDalek]

In order to enable Gengulphus to pick up whatever Notifications he may see, it might be best to quote his name correctly:

[Gengulphus]

The point I am making is that - using HYP1 as an example - would you have waited until now before selling/trimming, or would you have started the process before that one share got to 46.5%, maybe long before? Assuming some time before, you would of course now have a more diversified pool of income. But you would also have sold a share before some of that growth in income even occurred. In this way yes, your income is less concentrated, but surely in part because you would have sold the share that would have created the income that would have resulted in the over concentration in the first place!

Therefore, part of your solution to the over-concentration of income is in fact the prevention of some of that income even appearing in the first place!

Look, you or anyone else may still say that that is a price worth paying for less worry and better sleep, fair enough. But at least recognise that such a process comes with a cost. Indeed, if such an ongoing process had been instituted for HYP1, from the outset mind, it is a certainty that the income would now be less than it actually is, maybe significantly less. Mind you, it would be beautifully balanced

No, what would have happened if HYP1 had had a policy of trimming its winners is not a certainty - to get a certainty, you would have to also say exactly what criteria it used to decide what to trim, how much by and when, and to decide what to do (and when to do it) with the sales proceeds. And for us to actually know that certainty, it would all have to be mechanically driven by criteria that depend only on historical facts that can still be determined - too much danger of hindsight influencing the outcome otherwise.

There is also a danger of hindsight influencing our judgement about the results of a trimming policy applied to HYP1 would have been - we can easily see the current big income producers, i.e. the ones that trimming would have had a negative effect on portfolio income, but the past big income producers that have fallen by the wayside are harder to see and (depending on the exact trimming criteria) trimming them might have had a positive effect on it. Rather than illustrating that immediately, I'll re-pose a question I asked as an aside here:

(***) I'll pose a 'HYP trivia' question that illustrates that point: which HYP1 company (see list here) made the biggest contribution to its major income drop in the financial crisis, from £5,040 in 2008 to £3,187 in 2009? (I have posted the answer to that in the past, certainly on TMF and I think also on TLF, so some readers may remember the answer from that - but I don't think I've posted it for quite a long time.)

The answer is a case where a judicious trimming and reinvestment of the proceeds might well have proved to have a positive long-term effect on HYP1 portfolio income - and I have archived evidence in the form of this TMF post that I was nervous about the size of the HYP1 holding at what would have turned out to be a good point to trim (warning: that link is a 'spoiler' for the trivia question!).

Don't get what I'm saying here wrong: I do think it likely that there would have been a cost to pay for a trimming policy (if it had had one) in terms of HYP1's long-term income performance being less than it has actually proved to be. But it's not a certainty, because there is at least one case where it could easily have been beneficial in the long term, and quite possibly also have provided some much shorter-term mitigation of the shock of the financial crisis's 37% year-on-year reduction to HYP1 income.

Gengulphus

[Gengulphus]

In order to enable Gengulphus to pick up whatever Notifications he may see, it might be best to quote his name correctly:

A good general principle, but as it happens, I have the "Someone quotes you in a post" notification option switched off in the interests of keeping the size of my list of not-yet-dealt-with notifications down. If the software had a "Someone quotes you in a post to a topic you haven't bookmarked" notification option, I might consider switching that on - but that's probably a bit too much of an ask!

Gengulphus

[IanTHughes]

Indeed, if such an ongoing process had been instituted for HYP1, from the outset mind, it is a certainty that the income would now be less than it actually is, maybe significantly less. Mind you, it would be beautifully balanced

No, what would have happened if HYP1 had had a policy of trimming its winners is not a certainty - to get a certainty, you would have to also say exactly what criteria it used to decide what to trim, how much by and when, and to decide what to do (and when to do it) with the sales proceeds. And for us to actually know that certainty, it would all have to be mechanically driven by criteria that depend only on historical facts that can still be determined - too much danger of hindsight influencing the outcome otherwise.
.
.
.
Don't get what I'm saying here wrong: I do think it likely that there would have been a cost to pay for a trimming policy (if it had had one) in terms of HYP1's long-term income performance being less than it has actually proved to be. But it's not a certainty, because there is at least one case where it could easily have been beneficial in the long term, and quite possibly also have provided some much shorter-term mitigation of the shock of the financial crisis's 37% year-on-year reduction to HYP1 income.

You are of course quite correct. In my defence what I had intended to say was that the income growth and total from "those trimmed holdings" would certainly have been reduced. By the time I had noticed the omission of "those trimmed holdings", it was already too late to edit the post


Ian

[Alaric]

The answer is a case where a judicious trimming and reinvestment of the proceeds might well have proved to have a positive long-term effect on HYP1 portfolio income - and I have archived evidence in the form of this TMF post that I was nervous about the size of the HYP1 holding at what would have turned out to be a good point to trim (warning: that link is a 'spoiler' for the trivia question!).

From your link

For example, if your worry is that too much of your income will be coming from one company, making you too vulnerable to that company cutting its dividend, you could specifically trigger tinkering on the proportion of income coming from each share. In a 15-share HYP with reasonably large holding sizes, for instance, "review the ordinary dividend income at the end of each year; sell 20% of any share that provided 10% or more of that income and reinvest the proceeds" might be a reasonable tinkering strategy.)

Did anyone at that time pick up HYP1's increasing reliance on tobacco for returns?

[Itsallaguess]

As I tried to explain in my earlier posts, if you want to avoid such over-concentration of income, you have to do so from the outset, you do not wait 18+ years and only then act!

Furthermore, I whole-heartedly agree that such a process, if properly conducted from the outset, will surely result in less over-concentration of Income, not least because there will be less income growth to create over-concentration in the first place!

[my bold]

How do you know that Ian?

Cheers,

Itsallaguess

[funduffer]

This debate of balancing v not balancing a HYP is a very interesting and relevant discussion to the management of a HYP! (Not sure why it isn't on the HYP-P board, but then, who cares!)

I think Alaric has put his finger on the answer - it is all about risk tolerance.

HYP1 was started with equal amounts of capital that initially produced fairly similar amounts of annual income from its constituents.

Over time (18+ years), HYP1 is hugely unbalanced (at least partly due to how actions from corporate events were re-invested, as tjh pointed out).

Now consider the opposite extreme, of a HYP where re-balancing is carried out regularly to maintain strict limits on the proportion of income generated from each constituent company. I am talking about tjh's long running HYP, of course.

Both portfolios have produced fantastic income performance (with short blips in 2007-8), showing that both balancing, or not, can be effective.

The main difference, in my eyes, is that the level of risk to income from not balancing (as in HYP1) has increased from what it was at the outset, whereas, tjh's top-slice/top-up balancing methodology has kept risk to income relatively constant.

Surely this debate is all about personal attitude to risk?

FD

[Alaric]

Surely this debate is all about personal attitude to risk?

It should be, were it not for a TMF author who seemingly had an aversion to a bit of share dealing in his suggested annuity replacement portfolio and wrote that those who did so were ""tinkerers".

If you wanted income to live off as opposed to reinvest, it might be better to balance the income at the outset, although that's a more complex process involving not buying equal weights by market value.

[TUK020]

This debate of balancing v not balancing a HYP is a very interesting and relevant discussion to the management of a HYP! (Not sure why it isn't on the HYP-P board, but then, who cares!)

I think Alaric has put his finger on the answer - it is all about risk tolerance.

HYP1 was started with equal amounts of capital that initially produced fairly similar amounts of annual income from its constituents.

Over time (18+ years), HYP1 is hugely unbalanced (at least partly due to how actions from corporate events were re-invested, as tjh pointed out).

Now consider the opposite extreme, of a HYP where re-balancing is carried out regularly to maintain strict limits on the proportion of income generated from each constituent company. I am talking about tjh's long running HYP, of course.

Both portfolios have produced fantastic income performance (with short blips in 2007-8), showing that both balancing, or not, can be effective.

The main difference, in my eyes, is that the level of risk to income from not balancing (as in HYP1) has increased from what it was at the outset, whereas, tjh's top-slice/top-up balancing methodology has kept risk to income relatively constant.

Surely this debate is all about personal attitude to risk?

FD

Very well put FD.
A disciplined rebalancing set of rules such as TJH's does indeed manage that risk. Above this, it also captures gain from market noise and mean reversion. Major of TJH's shares have not just been top sliced or topped up, but have gone through an iteration of the two as market sentiment for the individual share moves around. Aka as "ratcheting up income".

In my mind, this is one of the most important ways in which investors here have evolved the original Bland Annuity Substitute concept into something superior. Ably supported by tools supplied here as well (HYPTUSS)

[Bubblesofearth]

Did anyone at that time pick up HYP1's increasing reliance on tobacco for returns?

Let me ask you a question - would you prefer to have £10,000 in each of 30 shares OR £10,000 in each of 29 shares and £100,000 in the other one?

IMO if you continuously sell down any shares that outperform then you will approximate the former whereas if you leave your portfolio alone you will approximate the latter. Along with the difference in total value implied. Obviously not as clean cut as that but along those lines. The same argument applies to income.

You cannot simply ignore the way markets evolve and assume periodically ducking out of winners will be just as good as keeping them.

It's a shame tjh did not keep a record of how his portfolio would have evolved had he never voluntarily trimmed any shares. I say voluntarily because appreciate there have been forced corporate actions.

BoE

[pendas]

The main difference, in my eyes, is that the level of risk to income from not balancing (as in HYP1) has increased from what it was at the outset, whereas, tjh's top-slice/top-up balancing methodology has kept risk to income relatively constant.

The impression I gained, rightly or wrongly from following some of the early HYP articles on TMF was that the initial selection process of choosing high yield introduced additional risk to kick start the income from the portfolio. From that point on, income was the main consideration and if yields fell due to share price recovery, so much the better as the risk had reduced.

As I understand, TJH ratchets up the yield with a purchase following a top slice and so maintains the initial level of risk whereas some in retirement may be looking to reduce the level of risk.

I appreciate not everyone will agree with my correlation between yield and risk and I'm not suggesting the majority of income should be left in the hands of a few successful shares.

[Alaric]

Let me ask you a question - would you prefer to have £10,000 in each of 30 shares OR £10,000 in each of 29 shares and £100,000 in the other one

The idea of HYP1 was income rather than value. So the question really should be whether you were happy with £ 1000 income from each of 29 shares and £ 10,000 from the other one. If your spending requirements were such that you needed an income of £ 39,000, you would feel vulnerable to the £10,000 income being suspended

[scrumpyjack]

I don’t know if I am allowed make comments which would otherwise be considered heresy. However some shares which are not high yield when you buy them may become so (relative to your purchase price) after years of growth.

I have never ever focused on income from my investments, but simply invested in companies that I thought would be a good investment long term. The income was incidental and was sometimes important to help prove the company was actually generating the real profits to pay the dividend.

I was not interested in short term market fluctuations and tended to hold shares for decades or until they were taken over. Probably that attitude started because in my youth dividends were so ludicrously highly taxed (98%), there wasn’t much point in having them, except that they could be offset by mortgage interest.

The result now, 50 years later, is that my dividends now cover my expenditure several times over and I haven’t needed to draw a pension. I can leave that to the kids. I have to say in retrospect that the companies I bought which had a high yield at the time of purchase have often been the worst performers and after 5 or 10 years the growth company dividends overtake them and then carry on growing and in some cases the annual dividend ends up being higher than the cost price of the share.

That is not to denigrate the HYP approach because things are obviously different if you start investing aged 20 as compared to starting at 65. However I still think the fact that a share has a high yield often reflects the view of the market that that company’s profits and dividends are not going to grow strongly. That may still be a lot better than an annuity but one ought to be conscious of it.

Back to the title of this topic – my approach has always been to hang on to the winners ‘cos they often keep on winning, but only if you are comfortable doing that and can sleep at night.