SDN’s HYP Report – Jan 2016

[SDN123]

SDN’s HYP Report – Jan 2016

Hi everyone, this is my second annual HYP report, the first was on TMF.

As before the post has ended up quite long as I decided to include all of the introductory context again for the TLF platform - so please bear with me. I don't plan to include all of the introduction in future TLF updates.

All feedback welcome

Context

- I am saving towards early retirement which will happen in 5 years at the earliest.
- While I’m saving my HYP represents approximately 15% of my total investments. The rest of my investments are in cash, Buy-To-Let and a Tim Hale style passive index portfolio.
- My plan is that, in retirement, I will use a HYP as my main tool to generate income (rather than, for example, buying an annuity) eg at the point of "retiring" I plan to transfer most, or possibily all, of my other investments into the HYP.
- My investments are spread across ISA, SIPP and tax exposed accounts.
- I started investing in 2010 and bought my first HYP shares (IMT, SSE, TSCO) in March 2012.

HYP guidelines*

* Note that "guidelines" are a little looser than "rules".

This section looks like a lot of rules but most of the detail is handled by an excel spreadsheet. I update the spreadsheet with current values whenever a dividend arrives, when I make a purchase or sale and at the end of each month. I use the HYP Top Up Spreadsheet extensively. A spreadsheet update when a dividend arrives takes about 1 minute. The monthly update takes less than 10 minutes. A purchase takes longer, depending on how much money I have to invest and my free time it typically takes about an hour.

1. New share selection

- min 5 years rising dividend
- p/e <= 15
- cover >= 1.5 (depends on sector)
- yield > FSTE 100 average and ideally > portfolio median
- look at history of net debt and free cash flow (actually this inconsistently applied as my ability to read and interpret annual reports evolves).

2. Diversification (purchases and top-ups)

- Don’t top or buy a share if that causes it to break any of the following limits:
---Max cost, current value or forecast income from share >= 5% HYP total
---Max cost, current value or forecast income from sector >= 10% HYP total
---Max cost, current value or forecast income from super-sector >= 20% HYP total

- Consider trimming share > twice HYP median current value
- Consider trimming share > twice HYP median forecast dividend
- Consider trimming share with forecast yield <= 2%

3. Other top-up / purchase guidelines

- Don’t top-up a share you would not buy using new share selection guidelines
- Don’t buy REITs (they are overweight in the overall portfolio)
- Max costs 1% (including all platform commission and all taxes) – aim for much less
- No foreign (non-UK) tax regimes

4. Activity

- In general this is a LTBH HYP.

5. Dividend reinvestment

Dividend income is considered as income to the WHOLE portfolio (not just the HYP) and is reinvested to maintain overall asset allocation whenever an "economic" amount is avaliable to invest.

6. Unitisation

I started to unitise my HYP in January 2016. A consequence of the dividend reinvestment strategy is that I only calculate income units and yield per unit (not accumulation units).

Portfolio @ 3/01/17

Shares
------

Code: Select all

Share                       EPIC  Weight  F Yield
=====                       ====  ======  ======
Admiral Group               ADM     3.0%     6.8%
AMEC Foster Wheeler         AMFW    2.0%     4.4%
AstraZeneca                 AZN     2.8%     5.0%
BAE Systems                 BA.     4.2%     3.7%
British American Tobacco    BATS    3.3%     3.9%
BHP Billiton                BLT     1.7%     3.7%
BP                          BP.     4.5%     6.2%
Carillion                   CLLN    4.6%     8.1%
Cobham                      COB     3.0%     4.4%
G4S                         GFS     3.7%     4.2%
GlaxoSmithKline             GSK     3.7%     5.2%
HSBC Holdings               HSBA    4.8%     6.0%
Imperial Tobacco Group      IMT     3.1%     4.9%
Interserve                  IRV     2.8%     7.1%
Legal & General Group       LGEN    4.4%     6.1%
Lloyds Banking Group        LLOY    2.5%     5.8%
Marston's                   MARS    3.7%     5.6%
National Grid               NG.     2.8%     4.8%
Pennon Group                PNN     2.7%     4.6%
Pearson                     PSON    3.2%     6.2%
Royal Dutch Shell 'B'       RDSB    4.0%     6.3%
Rio Tinto                   RIO     2.5%     4.4%
Stagecoach Group            SGC     3.4%     5.7%
Standard Life               SL.     3.1%     5.7%
DS Smith                    SMDS    4.2%     3.7%
Scottish & Southern Energy  SSE     3.8%     6.0%
Taylor Wimpey               TW.     4.2%     9.0%
Unilever                    ULVR    3.1%     3.4%
United Utilities Group      UU.     2.7%     4.4%
Vodafone Group              VOD     2.4%     6.2%
=====                       ====  ======  ======
Count/Median                 30     3.2%     5.4%

Sectors
-------

Code: Select all

Sector                          Shares  Weight
======                          ======  ======
Aerospace and Defence                2    7.2%
Banks                                2    7.3%
Consumer staples                     1    3.1%
Education                            1    3.2%
Energy                               2    6.6%
Homebuilding                         1    4.2%
Insurance (life)                     2    7.6%
Insurance (non-life)                 1    3.0%
Mining                               2    4.2%
Oil & Gas Producers                  2    8.6%
Oil Services                         1    2.0%
Packaging                            1    4.2%
Pharmaceuticals                      2    6.5%
Security (Support Services)          1    3.7%
Outsourcing (Support Services)       2    7.4%
Telecommunications                   1    2.4%
Tobacco                              2    6.4%
Travel & Leisure                     2    7.1%
Water                                2    5.3%
======                          ======  ======
Count/Median  19                    30    5.3%

Super-Sectors
-------------

Code: Select all

Super Sector       Shares  Weight
=============      ======  ======
Financials              5   17.9%
Resources               5   14.7%
Health Care             2    6.5%
Industrials             6   22.5%
Consumer Goods          4   13.8%
Consumer Services       3   10.3%
Utilities               5   14.3%
=============      ======  ======
Count/Median  7        30   14.3%

Activity in last twelve months

- Purchases: 4 (LLOY, COB, GFS, SGC); 2015 8
- Top-ups: 10 (SSE, RDSB, COB*, LGEN, MARS, BP., TW., CLLN, IRV, SL.); 2015 6
- Sales: 2 (S32, TSCO); 2015 1

(*= rights issue)

Others stats

As at 03/01/17
- HYP Unit value: 109.4; 2015 100.0 - 9.4% increase
- FTSE-All Share (unitised): 105.1; 2015 100.0 - 5.1% increase
- RPI: 264.5; 2015 259.8 - 1.8% increase
- Unit Yield (dividends received in last 12 months / current portfolio unit value): 4.70%; 2015 4.11% - 14.4% increase
- Forecast Yield (as per HYPTUSS ): 5.5%; 2015 5.4% - 1.9% increase
- Costs (stamp duty, platform fees, commission): 0.32%; 2015 0.38% - 15.8% reduction

Commentary - small stuff

- I got bored of TSCO and sold - probably a mistake in a HYP that is currently in savings mode. Maybe because of that I am holding onto BLT despite it yielding below 2%. Maybe another mistake?
- Vodafone is top of my current top-up list but the low cover worries me.
- Persimmon is top of my current wish list for new shares but adding another cyclical share in a sector that is already covered (by TW.) worries me.

Commentary - big stuff

- Capital and actual yield outperformed my benchmarks (FTSE-AllShare and RPI).
- The increase in forecast yield outperformed RPI, but only just.
- I expect that the strong increase in actual yield is due to dividend lag effects.
- The HYP section of my overall investments lagged behind my overall portfolio in TR terms (by about 1%).
- The HYP section is also more volatile (judged by standard deviation) than the overall portfolio (by about 2%).
- The HYP income is higher than my other shares / gilts but lower than my Buy-to-Let investments.
- The rate of increase HYP income is higher than my other shares / gilts and my Buy-to-Let investments.

My conclusion are that:
--- HYP isn't the best savings vehicle for me but remains my first choice for post-retirement drawdown;
--- Keeping a small HYP section is very valuable for me to learn / practice how HYP works in reality;
--- Keeping a small HYP section is very valuable for me to play with, helping me to resit "tinkering" with the more passive parts of my portfolio.

Finally

Many thanks to all of the readers and contributors to the HYP Practical Board, your contributions area really appreciated. Special thanks to Stooz and Clariman for keeping this amazing forum alive.

SDN
Jan 2017

[YeeWo]

Unless your holding in each share is very small indeed, if the portfolio shown represents 15% of your overall Investments you're clearly well on your way to FIRE already. Well done!

- Why did you decide/settle on a 30 stock portfolio?
- How long, roughly, have you held each stock?
- I agree with most of your principles!

Good Lick for 2017.

[SDN123]

Hi YeeWo,

Unless your holding in each share is very small indeed, if the portfolio shown represents 15% of your overall Investments you're clearly well on your way to FIRE already.

I have a decent amount, a lot more than I started with but a lot less than I need

Why did you decide/settle on a 30 stock portfolio?

I have no planned limit on the number of shares. I tried to get to 10-ish sectors as quickly as I could. After that I top up existing shares if the rules allow or add new shares if they look better to me. I like the idea of two shares per sector but not enough to go outside my guidelines to make it happen.

How long, roughly, have you held each stock?

I could look up a mechanical answer to that but I think it's more instructive to say that the vast majority have been held since I bought them. Previous year's sales have been "forced" by corporate actions. This year's sales have been to follow my guidelines (TSCO - forward yield below 2%) and to get rid of a runt holding (S32).

SDN

[Breelander]

...this is my second annual HYP report, the first was on TMF

Here: https://web.archive.org/web/20170104140 ... 15127.aspx

[SDN123]

Thanks Bree

[Arborbridge]

- Consider trimming share > twice HYP median forecast dividend

SDN, thanks for the report - you are getting off to a good start.

Your suggestion above, for trimming is an interesting one - I'm not sure I've seen that before. It sounds a good idea, except that it will often mean a share has dropped a great deal. This can happen in the anticipation of a dividend cut, but one willnot necessarily take place.
Depending on the circumstances, that might be the time to buy, rather than sell!

Arb.

[SDN123]

Hi Arb,

- Consider trimming share > twice HYP median forecast dividend

...It sounds a good idea, except that it will often mean a share has dropped a great deal.

Thank-you for the comment but I don't think that you are correct in this case. A test for the yield being greater than twice the median wouldn't make sense for a trim (and would flag, depending on taste, a screaming opportunity to sell out or to top-up!)

My test is for the dividend being greater than twice the median. In my mind that is closely linked to the weight being being greater than twice the median. In either case (weight or dividend) being high I'd want to consider trimming. When both are over median the case for trim would be even stronger in my mind.

I included this rule because I don't check portfolio values every day and I want plenty of warnings if action may be needed. (And because the flag is "automatically" generated by my spreadsheet it is no extra work for me.)

Having said all of that I'm in the early stages of my HYP and neither case (high weight or high dividend) has occurred yet. And there is a good chance that am confusing myself!

Does that make sense?

Simon

[MDW1954]

My test is for the dividend being greater than twice the median. In my mind that is closely linked to the weight being being greater than twice the median. In either case (weight or dividend) being high I'd want to consider trimming. When both are over median the case for trim would be even stronger in my mind.

I have to say that I'm a bit puzzled by this. Perhaps you could provide a mocked-up example, with numbers, so that we can see the calculations involved?

Other than that, a good HYP, and your guidelines make a lot of sense -- especially the distinction between "rules" and "guidelines".

MDW1954

[tjh290633]

My test is for the dividend being greater than twice the median. In my mind that is closely linked to the weight being being greater than twice the median. In either case (weight or dividend) being high I'd want to consider trimming. When both are over median the case for trim would be even stronger in my mind.

I think that it is the terminology that is the problem here. I have the same sort of limit, but only when it comes to topping up a holding. The criterion is the contribution to dividend income, in other words the dividend from a holding divided by the total dividend income for the year.

My limit is 5%, which is actually just over twice the median of 2.56%. If topping up a share would take it over that limit, then I don't top it up. The current situation, down to the median holding, is:

Rank   EPIC   % Income
   1   GSK       4.82%
   2   TW.       4.64%
   3   ADM       4.61%
   4   BP.       4.30%
   5   RDSB      4.15%
   6   CLLN      4.13%
   7   PSON      4.08%
   8   MKS       3.96%
   9   SSE       3.92%
  10   LGEN      3.78%
  11   VOD       3.62%
  12   MARS      3.50%
  13   AV.       3.30%
  14   NG.       3.22%
  15   AZN       3.14%
  16   BLND      2.96%
  17   IMI       2.77%
  18   UU.       2.66%
  19   IMB       2.56%

All other factors being equal, shares from position 5 down would be eligible for a 20% top up.

TJH

[SDN123]

MDW:

Perhaps you could provide a mocked-up example, with numbers, so that we can see the calculations involved?

TJH:

I think that it is the terminology that is the problem here.

It is terminology issue I think. In creating the worked example below I realise that I could have said that the guideline was "consider trimming if the income is great than twice the median" which might have helped. Anyway, hopefully the example below clarifies what I am talking about. The example is for a fictitious five share £50,000 portfolio.

Code: Select all

Share   Value    Weight  F Yield  F Dividend  Income
        (GBP)    (%)     (%)      (GBP)       (%)
======  ======   ======  ======   ======      ======
AA      £17,500     35%       6%      £1,050     47%
BB       £7,500     15%       5%        £375     17%
CC      £10,000     20%       2%        £200      9%
DD      £10,000     20%       4%        £400     18%
EE       £5,000     10%       4%        £200      9%
======  ======   ======  ======   ======      ======
Median  £10,000     20%       4%        £375     17%

Share AA is less than twice the median in terms of % weight.
Share AA's yield is less than twice the median in terms of forward yield.
Share AA's dividend (and therefore % income) is more than twice the median in terms of total portfolio dividend (or % income).

I'm suggesting that in this case I would consider the portfolio "imbalanced" and I'd want to consider trimming share AA. This is different from not topping up AA (which would be blocked by another rule as the income is so high).

In the real world, rather than a contrived example, I expect that the % income getting this imbalanced will be closely related to the % weight also getting similarly imbalanced. I feel that if this happened I'd become nervous that one share was providing too much income and I'd want to spread the capital a bit differently. But this is only theory because it hasn't actually happened to me yet. I realise that I could well be using the convenience of spreadsheets to over-complicate things. I'd be very interested to hear other's views.

Hope that helps,

Simon

[MDW1954]

Ah! Income. All is clear. And thank you for the mocked-up example.

That said, in my own 32-share HYP, I have three shares contributing over 10% of the income (and one, a mildly disturbing 16%).

They were all bought opportunistically when prices were depressed, and yields high.

I believe that the Forecast Yield column in your example is obscuring this. In my (roughly equivalent) spreadsheet, I calculate "yield on bought cost" per share. As you can imagine, some are very juicy indeed -- RDSB almost 8%, for instance.

MDW1954

[Itsallaguess]

In my (roughly equivalent) spreadsheet, I calculate "yield on bought cost" per share.

As you can imagine, some are very juicy indeed -- RDSB almost 8%, for instance.

Can I ask why you do this?

Every time this has come up in the past, regarding HYP shares, the majority of discussions around the topic have come to the conclusion that it's a really distorting metric, and one that's got no real-world use other than as a 'comfort-blanket' of some sort, and that 'yield on current price' is always the better metric to use, as it allows proper comparison with other available options in the real world.

Thinking that a HYP share is 'delivering a yield on bought cost' of 12%, for instance, when in reality it may well be delivering a 'yield on current price' of 3% (as an example), might perhaps put you in the position of regarding the '12%' share in a much more favourable light than perhaps knowing that in reality it's actually delivering less yield than the majority of the other HYP shares.

The historical price paid for a HYP share is useless other than for record-purposes, as soon as you've actually paid it. After that, the only really important figure to a HYP owner, in my opinion, is what the current yield is, which is of course a metric based against current price and the current dividend paid.

If there is a good reason that you use 'yield on bought cost' as a metric at all, then could you please explain what it offers you?

Cheers,

Itsallaguess

[tjh290633]

I'm suggesting that in this case I would consider the portfolio "imbalanced" and I'd want to consider trimming share AA. This is different from not topping up AA (which would be blocked by another rule as the income is so high).

In the real world, rather than a contrived example, I expect that the % income getting this imbalanced will be closely related to the % weight also getting similarly imbalanced. I feel that if this happened I'd become nervous that one share was providing too much income and I'd want to spread the capital a bit differently. But this is only theory because it hasn't actually happened to me yet. I realise that I could well be using the convenience of spreadsheets to over-complicate things. I'd be very interested to hear other's views.

Hope that helps,

Simon

I think that, in your contrived example, the situation could only arise over several years, during several of which it would be out of balance. The weight can probably only get overheavy if you bought more, despite the warning signal. If it was down to share price rise, with a concomitant dividend rise (which is not impossible), then it would be due for trimming back on weight grounds alone. I don't think that I have ever experienced this type of situation, although I have seen the income from a share rise above my 5% limit because of dividend increases.

Looking back at 12 months ago, my highest contributor to income was SSE at 4.70%, followed by BP. at 4.69% and RDSB at 4.50%. Now those three rank 9th, 4th and 5th respectively, at 3.62%, 4.30% and 4.15% respectively. The then median was 2.70%, compared with 2.56% now. Top of the midden are:

Rank   EPIC   % Income
   1   GSK       4.82%
   2   TW.       4.64%
   3   ADM       4.61%
   4   BP.       4.30%
   5   RDSB      4.15%
   6   CLLN      4.13%
   7   PSON      4.08%
   8   MKS       3.96%
   9   SSE       3.92%
  10   LGEN      3.78%

What has affected the rankings? GSK has been topped up twice, TW. has been topped up once and increased its dividend, ADM increased its dividend, CLLN, PSON, MKS and LGEN have all been topped up as well and increased their dividends from the previous calendar year. SSE has not been topped up, but has had a small increase in dividend of 1.13%.

It is certainly good to look at hypothetical situations, but best to concentrate on the actual results and respond as and when required.

TJH

[funduffer]

If there is a good reason that you use 'yield on bought cost' as a metric at all, then could you please explain what it offers you?

I can think of a reason.

Say you bought a HYP as well as an annuity at retirement (maybe different amounts). The annuity rate is 4% flat and the HYP yield is 4% on purchase.

In the annuity case the capital has gone forever, and for the HYP its still there but not of any real importance (Let's say it is all going to be inherited by some of the family).

After a few years, you might want to see how much better your HYP has done on income compared to the annuity. The HYP yield based on bought cost would tell you how much better, or worse, your HYP income was doing relative to the annuity rate of 4%.

Just one way of doing this, I am sure there are others.

FD

[Itsallaguess]

If there is a good reason that you use 'yield on bought cost' as a metric at all, then could you please explain what it offers you?

I can think of a reason.

Say you bought a HYP as well as an annuity at retirement (maybe different amounts). The annuity rate is 4% flat and the HYP yield is 4% on purchase.

In the annuity case the capital has gone forever, and for the HYP its still there but not of any real importance (Let's say it is all going to be inherited by some of the family).

After a few years, you might want to see how much better your HYP has done on income compared to the annuity. The HYP yield based on bought cost would tell you how much better, or worse, your HYP income was doing relative to the annuity rate of 4%.

Just one way of doing this, I am sure there are others.

FD

Thanks FD, I hadn't considered the 'whole HYP comparison with an annuity' angle, and can see where you're coming from regarding perhaps wanting to do this.

That said, it still doesn't really answer the question as to why people would want to consider the 'yield on bought cost' of individual HYP holdings though...

Cheers,

Itsallaguess

[SDN123]

TJH:

It is certainly good to look at hypothetical situations, but best to concentrate on the actual results and respond as and when required.

I completely agree, thanks for the data and experience to say this theoretical situation is unlikely to happen. One motivation for documenting my HYP and gathering figures is that one day I may have enough year-on-year data objectively to help others in the same way that you (and others in here) do so regularly and generously.

Simon

[77ss]

My test is for the dividend being greater than twice the median. In my mind that is closely linked to the weight being being greater than twice the median. In either case (weight or dividend) being high I'd want to consider trimming. When both are over median the case for trim would be even stronger in my mind.

My limit is 5%, which is actually just over twice the median of 2.56%.

I too have limits - but like SDN123, I take capital weight into account as well as dividend weight. It doesn't really matter much to me whether the figures are 4% and 6% or 6% and 4%.

I am reluctant to go over a total of 10 for the two weightings, although not inflexible. I currently have 3 holdings breaching this figure (LGEN, RDSB and SSE). I don't currently intend to do anything about LGEN or RDSB - both the result of opportunistic top-ups last year. I shall almost certainly be trimming SSE (5.9% and 7.8% by capital and dividend income) in the coming tax year.

My median for dividend weight is 2.55%, so my limits are very similar to the twice median constraint - not that I had ever thought of them in that way!

[Gengulphus]

Thanks FD, I hadn't considered the 'whole HYP comparison with an annuity' angle, and can see where you're coming from regarding perhaps wanting to do this.

That said, it still doesn't really answer the question as to why people would want to consider the 'yield on bought cost' of individual HYP holdings though...

Actually, the issue isn't so much a whole-portfolio-vs-individual-holding one, as an evaluating-past-decisions-vs-making-decisions-now one. 'Yield on bought cost' is a plausible measure of how one's past decisions have worked out, at either whole-portfolio or individual-holding level - not necessarily the best such measure, but I certainly wouldn't dismiss it out of hand. It is however just about virtually useless for the purpose of making decisions now.

As a somewhat extreme example, in 1991 I acquired 105 shares in Scottish Hydro-Electric when it was privatised, at a total cost of 240p each, and later got 5 extra shares for no extra cost as a bonus under the privatisation arrangements. Those 110 shares therefore cost me a total of £252.

I still own those shares - they've gone through a merger and a couple of name changes with no change to their number, becoming 110 shares in SSE. Last year, they paid me dividends totalling 89.4p each, for total dividends received of £98.34. So my historical 'yield on bought cost' for that purchase is £98.34/£252 = 39.0%.

That does tell me something about that past decision to invest in the 1991 privatisation - which is basically that it's been a very successful investment decision from a growing-income point of view. It's not necessarily the best such measure - a year-by-year income record would tell me more, for instance - but as a single-number measure, it isn't bad.

But it tells me virtually nothing about whether buying (or selling) SSE is currently a good idea. To see that, imagine that the current share price were ten times what it actually is - i.e. £153.90 rather than £15.39. My 'yield on bought cost' would still be 39.0% - the current share price simply doesn't affect the calculation of 'yield on bought cost' at all, and so it doesn't tell me anything about whether the shares are currently cheap or expensive. Nor does it tell me anything about what my income prospects from a purchase now are (nor how much future income I might be giving up if I were to sell).

I say "virtually nothing" rather than "absolutely nothing" because it does suggest reasonably strongly that the company's management over the last quarter-century have known what they're doing! But what I'm actually interested in for a purchase (or sale) now is whether the company's current management know what they're doing - and on that, it's no more than a very vague hint...

Gengulphus

[Itsallaguess]

Thanks FD, I hadn't considered the 'whole HYP comparison with an annuity' angle, and can see where you're coming from regarding perhaps wanting to do this.

That said, it still doesn't really answer the question as to why people would want to consider the 'yield on bought cost' of individual HYP holdings though...

Actually, the issue isn't so much a whole-portfolio-vs-individual-holding one, as an evaluating-past-decisions-vs-making-decisions-now one. 'Yield on bought cost' is a plausible measure of how one's past decisions have worked out, at either whole-portfolio or individual-holding level - not necessarily the best such measure, but I certainly wouldn't dismiss it out of hand. It is however just about virtually useless for the purpose of making decisions now.

Gengulphus

I'd probably go beyond saying it's 'virtually useless' for the purpose of making decisions and actually say that it's 'downright dangerous' for the purpose of making decisions, if that's ever what people actually think they can usefully use the metric for....

What I mean by that is that I'm not often convinced that the people using 'yield on bought cost' that we come across now and then are using it for anything other than some sort of comfort-blanket, and using it as something to perhaps put off what might be quite proper and correct decisions about particular holdings, were they to view those holdings using much more appropriate metrics....

I'll agree that within someone's personal investment-sphere, they could use the metric to allow some 'what-if' scenarios to be played out, but as soon as people start bringing the metric into open discussion in terms of wanting to compare their holdings with anyone else's, then as you say, it becomes useless, and I'd suggest misleading, as a historical price paid is simply that - a historical price....

Cheers,

Itsallaguess

[MDW1954]

If there is a good reason that you use 'yield on bought cost' as a metric at all, then could you please explain what it offers you?

For my overall HYP, I calculate yield on total new money invested, yield on bought cost, and yield on current prices (the latter being the TTM historic yield, not forecast yield).

For individual shares within the HYP, I calculate yield on bought cost, and yield on present price.

Both sets of calculations are performed just once per year, when I update the HYP spreadsheet with the year's dividends and latest values etc. Purchasing decisions within the year are based on forecast yield, and some rough guidelines re: sectoral concentration. The annual HYP spreadsheet contains a tab which calculates sectoral concentration.

I think the point that you are making is that yield on bought cost provides no indication of the opportunity cost of capital, as yield on current prices does. You are quite correct; it doesn't.

My view is that this opportunity is not always available. Take RDSB: whatever the forecast yield, the opportunity to swap the capital into another global London-headquartered high-yielding oil giant is fairly limited. Ditto GSK, AZN, HSBC, BLT etc etc. It is easy to take that opportunity by swapping across sectors, less easy by swapping within sectors. This is where the opportunity cost of capital breaks down, IMV: it is easy to identify what to sell by looking at forecast yield, but less easy to identify what to buy.

In this context, I find yield on bought cost gives me some insight into how well (or badly) I made the initial capital deployment decision, although that insight can be clouded or distorted by top-ups subsequent to the initial purchase. It is an additional data point; it costs nothing to calculate; the calculation is performed automatically; and I am not obligated to take any decision based on it.

Please also note that I am not arguing that my approach is in any way superior, or recommending it. I am simply answering the question you asked. I am also not arguing that yield on bought cost is in any way a substitute for forecast yield; they are very different things.

MDW1954