P is for Portfolio

[moorfield]

A Friday/Weekend diversion for you "builders".

Several regulars here will know I place great store in my overall portfolio income (and yield) and measure regularly against a long term retirement target.

I wanted to share these wise words from the Monevator website (if I'm allowed to), and pose: how many questions can you answer ?


http://monevator.com/

Breaking down your investment goal

The key components of any investing goal are:

Vision – For example, “I want to retire at 55.”

Target – What is the number in pounds and pence that you need to achieve?

Time horizon – How many years can you take to hit the target?

Contribution level – How much can you invest? This may be a lump sum or a regular amount, such as £250 a month.

Expected rate of return – What growth rate do you need for your contribution to mushroom into your magic number, given your time frame? You’ll need to come up with a credible expected return for your portfolio – and come to terms with the fact that expected returns are not guaranteed.

[tjh290633]

That makes the fundamental error of asking how much capital you need, rather than how much income you need.

I've always maintained that the best way is to build a flow of income and reinvest it until it is time to draw it. Others take the contrary view, but I prefer my way.

TJH

[moorfield]

That makes the fundamental error of asking how much capital you need, rather than how much income you need.

Not necessarily. Those questions can also be framed here in terms of future income needed. Fixing an expected rate of growth and a time to draw it enables one to discount (estimate) to present value a future income, which is how I measure progress.

[tjh290633]

That makes the fundamental error of asking how much capital you need, rather than how much income you need.

Not necessarily. Those questions can also be framed here in terms of future income needed. Fixing an expected rate of growth and a time to draw it enables one to discount (estimate) to present value a future income, which is how I measure progress.

Yes, but it does not target a future income. As you will be aware, dividends are much less volatile than capital values. Far better to follow the growth in income and to direct the monthly contribution towards buying more income.

TJH

[Itsallaguess]

I wanted to share these wise words from the Monevator website (if I'm allowed to), and pose: how many questions can you answer ?

http://monevator.com/

Breaking down your investment goal

The key components of any investing goal are:

Vision – For example, “I want to retire at 55.”

Target – What is the number in pounds and pence that you need to achieve?

Time horizon – How many years can you take to hit the target?

Contribution level – How much can you invest? This may be a lump sum or a regular amount, such as £250 a month.

Expected rate of return – What growth rate do you need for your contribution to mushroom into your magic number, given your time frame? You’ll need to come up with a credible expected return for your portfolio – and come to terms with the fact that expected returns are not guaranteed.

Being a bit of a spreadsheet junkie, I've probably spent more time on things like this than I really should do, but I always find it interesting to both model my assumptions, and then test those assumptions with some stress-tests, regarding things like natural-dividend-increases over time, additional capital (and the subsequent dividends from that additional capital, given that I'm still working and have no current need to draw income from my portfolio..), and suchlike.

Let's face it, there's not a lot on telly these days....

Anyhow, one of the first things I ever wanted to model was to see just what situation I would need to be in, whilst still working and able to add extra capital each year to my investments, that would deliver the first yearly increase in dividend-income of £1000 over the previous year, and importantly this needed to be done in a scenario that didn't assume huge yields from my investments.

Here's where I ended up with one of my imagined calculations -

Current Portfolio income = £10,222

Portfolio Yield = 4.5%

Additional capital invested each year = £12,000

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Next year additional dividend from re-investing the portfolio income from this year = £10,222 * 4.5% = £460

Next year additional dividend from the additional capital invested this year = £12,000 * 4.5% = £540

Growth in portfolio income (excluding any calculation of any 'natural dividend increases' at all) = £460 + £540 = £1000.

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------

So I worked out that if I could get to a point where my portfolio income was £10,222 and I could also still re-invest that sum, as well as invest additional capital of £12,000 each year, from still being in paid employment, then (with a fair wind) I should see my total portfolio income rise each year by around £1000, and that's ignoring any additional benefits regarding naturally-rising dividends from within the already-invested portfolio components.

I can't remember just when I did the above calculation, but it was very early on in my income-investment journey, possibly around 18 years ago now, and I'll admit that the above scenario, whilst very interesting, seemed completely unachievable at that time, but what it did do was to instil a sense of drive, purpose, and discipline into my investment approach that I've managed to keep up over subsequent years.

It's surprising, over an investment lifetime, how situations that would have scared the living daylights out of you in those much earlier years, seem to gradually work their ways towards describing achievable goals.

Keep plugging away over many years, allowing time itself to do the heavy lifting, and just continue putting one investment-foot in front of the other, and eventually what might have once seemed like a scenario completely out of reach, might one day start to look a little bit less so....

Cheers,

Itsallaguess

[ReformedCharacter]

That makes the fundamental error of asking how much capital you need, rather than how much income you need.

Not necessarily. Those questions can also be framed here in terms of future income needed. Fixing an expected rate of growth and a time to draw it enables one to discount (estimate) to present value a future income, which is how I measure progress.

I don't dispute what you say at all but I feel far less confident buying investments which are intended to grow capital rather than provide income. Perhaps that reflects my own psychology but also perhaps that capital growth is less predictable than income growth from reinvesting dividends. I'm happy to accept that this debatable. Doubtless not everyone can do as well as tjh but to my mind his performance over a fairly long period of time (as I recall) provides good evidence of the benefits of income investment. I would certainly be happy to do as well (and am grateful for his input on this board).

RC

[moorfield]

That makes the fundamental error of asking how much capital you need, rather than how much income you need.

Not necessarily. Those questions can also be framed here in terms of future income needed. Fixing an expected rate of growth and a time to draw it enables one to discount (estimate) to present value a future income, which is how I measure progress.

Yes, but it does not target a future income. As you will be aware, dividends are much less volatile than capital values. Far better to follow the growth in income and to direct the monthly contribution towards buying more income.

Those variables can be swapped around of course, ie. fixing a target future income and time to draw it implies a rate of growth from the present day.

[monabri]

I think we can reasonably say that any reinvested dividends might be invested at 4 to 5% but how much the compound annual growth rate in dividends (CAGR) will be is difficult to estimate..but is by far the most important factor.

I decided to use the recent 5 year data for CAGR for each company weighted by the holding in the company and then to summate over the portfolio as a whole...which is all well and good until the likes of WPP suddenly decide to freeze their dividend increases after stellar increases in recent years ( but at least I can tweak the spreadsheet).

[Itsallaguess]

I think we can reasonably say that any reinvested dividends might be invested at 4 to 5% but how much the compound annual growth rate in dividends (CAGR) will be is difficult to estimate..but is by far the most important factor.

I decided to use the recent 5 year data for CAGR for each company weighted by the holding in the company and then to summate over the portfolio as a whole...which is all well and good until the likes of WPP suddenly decide to freeze their dividend increases after stellar increases in recent years ( but at least I can tweak the spreadsheet).

I do agree that internal-CAGR is a very important factor, but I also agree that it's difficult to incorporate very well into these types of calculations, and certainly so at a component-by-component level...

So given these issues, and also given the important consideration that you've also highlighted, where components will stumble during periods of portfolio-ownership, and often to different degrees and for different lengths of time, I think a fairly neat and simple thing to do is to perhaps assume that all of those 'un-calculated influences' might tend to cancel themselves out over long periods of time, and then just ignore them for the sake of simplicity.....

Cheers,

Itsallaguess

[tjh290633]

I think we can reasonably say that any reinvested dividends might be invested at 4 to 5% but how much the compound annual growth rate in dividends (CAGR) will be is difficult to estimate..but is by far the most important factor.

I decided to use the recent 5 year data for CAGR for each company weighted by the holding in the company and then to summate over the portfolio as a whole...which is all well and good until the likes of WPP suddenly decide to freeze their dividend increases after stellar increases in recent years ( but at least I can tweak the spreadsheet).

You might like an actual example. In June 1969 I took out a unit-linked endowment, which invested £5 each month into the then Prudential Unit Trust, the starting price was the equivalent of 88p. Dividends in the next 12 months came to 2.81p/unit, with a resultant initial yield of 3.2%.

By December 1980, the dividends equalled my monthly contributions (actually a bit more because I received tax relief on my premium plus the policy fee, so net I only paid about £4.80 a month). At this point the unit price had risen to 175p, and the distribution was 6.82p/unit. Unit price virtually doubled, distribution more than doubled.

When the policy matured in December 1997, the unit price had risen to 1032p and the last distributions were 21.73p, which were increases of 1073% and 673% respectively, the yield now being 2.11%. My IRR over the whole period was 12.5%. In terms of the monthly premiums, the dividends were now just over 6 times the amount paid.

It's one example from actual life, and dividend taxation changed along the way, as did LAPR, I believe. However the 12 years to have dividends equal to the contributions is significant.

TJH

[Arborbridge]

No one doubts the power of compounding, but the point at issue for me would be whether a better route is via investments which do not pay out at all, but "inwardly" compound for capital growth. OT for this board, but something people in that position should investigate. One can get an idea at to the possible answer by checking the TR of average equity income versus growth funds, from time to time, and see which come out on top regularly.

HYP is certainly one way to go, but it was intended as an income strategy: it may not be the best pot building strategy.


Arb.

[Itsallaguess]

HYP is certainly one way to go, but it was intended as an income strategy: it may not be the best pot building strategy.

I've absolutely no doubt that you're right there Arb - but alternative strategies that would require me to swap approaches when I come to require an income-strategy would (for me anyway...) necessarily have to provide some level of performance-improvement just to overcome the superb simplicity of a full-trip income-approach (with income re-investment during working years) having an in-built 'switch' that simply diverts funds when the time comes to want them. One lifetime investment approach, and then just choose the time to flip the switch.....

I think that simplicity, and the subsequent removal of a potentially quite different accumulation-approach, should be recognised as a good benefit, and one that for me personally would need a really good justification to give up...

I don't think anyone ever really tries to suggest that HYP is 'the best' one-shot strategy (even if some detractors often seem to think that this is the case...) - it's just that it tends to do the job in a very simple way, and that, in itself, might well be considered such an advantage that it might well turn out to be 'the best' strategy for some people who really value it's distinct merits...

Cheers,

Itsallaguess

[TUK020]

I think we can reasonably say that any reinvested dividends might be invested at 4 to 5% but how much the compound annual growth rate in dividends (CAGR) will be is difficult to estimate..but is by far the most important factor.

I decided to use the recent 5 year data for CAGR for each company weighted by the holding in the company and then to summate over the portfolio as a whole...which is all well and good until the likes of WPP suddenly decide to freeze their dividend increases after stellar increases in recent years ( but at least I can tweak the spreadsheet).

You might like an actual example. In June 1969 I took out a unit-linked endowment, which invested £5 each month into the then Prudential Unit Trust, the starting price was the equivalent of 88p. Dividends in the next 12 months came to 2.81p/unit, with a resultant initial yield of 3.2%.

By December 1980, the dividends equalled my monthly contributions (actually a bit more because I received tax relief on my premium plus the policy fee, so net I only paid about £4.80 a month). At this point the unit price had risen to 175p, and the distribution was 6.82p/unit. Unit price virtually doubled, distribution more than doubled.

When the policy matured in December 1997, the unit price had risen to 1032p and the last distributions were 21.73p, which were increases of 1073% and 673% respectively, the yield now being 2.11%. My IRR over the whole period was 12.5%. In terms of the monthly premiums, the dividends were now just over 6 times the amount paid.

It's one example from actual life, and dividend taxation changed along the way, as did LAPR, I believe. However the 12 years to have dividends equal to the contributions is significant.

TJH

Terry,
The trouble with the period chosen is that I am not sure if the anecdote illuminates the power of compounding, or if it illuminates the effects of inflation
tuk020

[tjh290633]

I think we can reasonably say that any reinvested dividends might be invested at 4 to 5% but how much the compound annual growth rate in dividends (CAGR) will be is difficult to estimate..but is by far the most important factor.

I decided to use the recent 5 year data for CAGR for each company weighted by the holding in the company and then to summate over the portfolio as a whole...which is all well and good until the likes of WPP suddenly decide to freeze their dividend increases after stellar increases in recent years ( but at least I can tweak the spreadsheet).

You might like an actual example. In June 1969 I took out a unit-linked endowment, which invested £5 each month into the then Prudential Unit Trust, the starting price was the equivalent of 88p. Dividends in the next 12 months came to 2.81p/unit, with a resultant initial yield of 3.2%.

By December 1980, the dividends equalled my monthly contributions (actually a bit more because I received tax relief on my premium plus the policy fee, so net I only paid about £4.80 a month). At this point the unit price had risen to 175p, and the distribution was 6.82p/unit. Unit price virtually doubled, distribution more than doubled.

When the policy matured in December 1997, the unit price had risen to 1032p and the last distributions were 21.73p, which were increases of 1073% and 673% respectively, the yield now being 2.11%. My IRR over the whole period was 12.5%. In terms of the monthly premiums, the dividends were now just over 6 times the amount paid.

It's one example from actual life, and dividend taxation changed along the way, as did LAPR, I believe. However the 12 years to have dividends equal to the contributions is significant.

TJH

Terry,
The trouble with the period chosen is that I am not sure if the anecdote illuminates the power of compounding, or if it illuminates the effects of inflation
tuk020

You take any period of 29 years and inflation is bound to be a factor.

TJH

[Arborbridge]

Judge for yourselves: RPI June 1969 = 17.46. RPI Dec 1997 = 160. A factor of over 9x


Arb.

[Alaric]

Judge for yourselves: RPI June 1969 = 17.46. RPI Dec 1997 = 160. A factor

One of my personal benchmarks for relating prices in the 1960s, in other words before decimalisation is to relate the price in shillings then to the price in pounds today. A 20 times multiplier in other words. Beer at under 3/- a pint comes out cheaper, whilst LPs at 37/6 look expensive.

[tjh290633]

Judge for yourselves: RPI June 1969 = 17.46. RPI Dec 1997 = 160. A factor

One of my personal benchmarks for relating prices in the 1960s, in other words before decimalisation is to relate the price in shillings then to the price in pounds today. A 20 times multiplier in other words. Beer at under 3/- a pint comes out cheaper, whilst LPs at 37/6 look expensive.

Petrol was 4/-(£0.20) per gallon, now £5.80p, so 29 times.

Looking at the RPI from 1969 to 1987, that's a rate of increase of 8.1%. For the unit price it is 9.0% and for the dividends it is 7.4%, but the compounding gave a return of 12.5%. Effectively a real rate of return of 4.4% over the 29 years.

TJH