Portfolio Unitization - Results Interpretation

[PrefInvestor]

Hi All,

Browsing the HYP pages of this site yesterday I see that many people seem to be unitizing their portfolios so that they can compare how they are doing. I googled this today and found it quite interesting. If you are unaware of this process it involves presenting your portfolio valuation rather as a fund manager does ie as a number of units at a certain unit price. For a more detailed explanation of how to go about applying this to your portfolio see below:-

https://monevator.com/how-to-unitize-your-portfolio/

Anyway I had a go this morning as I was interested to see what it would produce as I have weekly portfolio data going back to 2012 along with full details of all deposits and withdrawals.

My personal result seems to be that I now have about 3 times the number of units at just over double the original unit price from when I started in 2012. I am interpreting this as saying that the increase in the number of units reflects the fact that I have added quite a lot of money to the portfolio over the years, which I certainly have. The doubling of the unit price I am interpreting as being a positive sign that my investments have in fact made money, clearly a reduction in the unit price would represent having made a loss.

Not quite sure what else to make of it right now. My own adhoc analysis is that while I haven’t doubled my money, I am perhaps ~75% up in capital terms on what I started with plus my deposits.

Appreciate any informed comments on how to interpret the unitization results.

ATB

Pref

[tjh290633]

Are you working on an income unit basis or the accumulation unit basis? Are you reinvesting dividends? If so, how?

Your conclusions are near enough correct.

Just to give you a comparison, my unitisation dates from 1987 when PEPs began. Initially I set both types of unit at £1.00. Today the income unit is £6.21 and the accumulation unit is £26.04. the dividend per income unit has risen from 2.83p in 87-88 to 29.21p in the current year, with a few yet to be declared.

I calculate the units created from reinvested dividends at the end of each month, likewise the units cancelled by cash withdrawn.

TJH

[PrefInvestor]

Well TJH, thanks for the response. I am working on an accumulation basis (I think). I have just followed the instructions contained in the monevator article for which there was a link in my first post.

My investment spreadsheet has my portfolio value for every week since 2012 and each row also records if I added or withdrew any money and the amount. I have used these figures to calculate the unit prices for any week where there was no deposit or withdrawal. If there was a deposit/withdrawal then I have adjusted the number of units by the figure corresponding to the amount of that deposit/withdrawal using the unit price at the time.

I have just added two extra columns (Units & Unit Prices) to my weekly record of portfolio performance and thats all I have done thus far.

I DO re-invest dividends but not automatically, rather I save them up till they reach about £3,000 and then invest in some investment of my choice. I have ignored dividends in what I have done unitization wise as thats what the monevator article said to do. I do have all of the dates and amounts in my spreadsheet as it happens, but as I say I've not used them for anything.

Having done it though I'm just not immediately sure what any of it really means and how I can use it !!.

ATB

Pref

[OZYU]

Well TJH, thanks for the response. I am working on an accumulation basis (I think). I have just followed the instructions contained in the monevator article for which there was a link in my first post.

My investment spreadsheet has my portfolio value for every week since 2012 and each row also records if I added or withdrew any money and the amount. I have used these figures to calculate the unit prices for any week where there was no deposit or withdrawal. If there was a deposit/withdrawal then I have adjusted the number of units by the figure corresponding to the amount of that deposit/withdrawal using the unit price at the time.

I have just added two extra columns (Units & Unit Prices) to my weekly record of portfolio performance and thats all I have done thus far.

I DO re-invest dividends but not automatically, rather I save them up till they reach about £3,000 and then invest in some investment of my choice. I have ignored dividends in what I have done unitization wise as thats what the monevator article said to do. I do have all of the dates and amounts in my spreadsheet as it happens, but as I say I've not used them for anything.

Having done it though I'm just not immediately sure what any of it really means and how I can use it !!.

ATB

Pref

Take a look at your PMs for addtitional help on the subject. The monevator article only scratched the surface imho.

I strongly advise you, since you are just starting out unitising to calculate both Inc and Acc units, since it is no additional bother at all on a properly set up spreadsheet. There are important complementary aspects to each.

Ozyu

[kiloran]

This might be useful: http://lemonfoolfinancialsoftware.weebl ... folio.html

I put this together a few years ago in Motley Fool days, to try and summarise some of the discussions at the time, and it was never really finished. The Motley Fool links are no longer valid and I don't have equivalents on the Wayback Machine.

--kiloran

[tjh290633]

Hello, pref.

I agree that calculating as income units will be worthwhile for the future. You don't have to go back to square one, as you can start from now if you prefer, just using your current accumulation unit number and value as the start point.

No compulsion, just a suggestion.

Sorry about the confusion over the locked thread. It seemed to do it of its own accord. I unlocked it twice.

TJH

[nmdhqbc]

Having done it though I'm just not immediately sure what any of it really means and how I can use it !!.

Well this is a good start to the unitization being useful. Before you could not separate accurately what growth of your portfolio was from additional cash vs investment returns...

My personal result seems to be that I now have about 3 times the number of units at just over double the original unit price from when I started in 2012. I am interpreting this as saying that the increase in the number of units reflects the fact that I have added quite a lot of money to the portfolio over the years, which I certainly have. The doubling of the unit price I am interpreting as being a positive sign that my investments have in fact made money, clearly a reduction in the unit price would represent having made a loss.

You can do year on year comparisons of the unit price to see the returns made each year. Does rely on having the portfolio valued at the same date each year I guess.

You can work out annualised returns (in excel perhaps)...
B=start price
C=end price
D=number of years
annualised returns=(C/B)^(1/D)-1 (this expresses it as a decimal you can get excel to format it to a %)

[PrefInvestor]

Hi All, Thanks for the inputs. I will read kiloran’s link and Ozyus PM when I get some time. Up until now I have used the info on my spreadsheet to analyse my YTD capital performance by calculating the capital increase and reducing it by the amount of new money invested (or the reverse if I made withdrawals). Very easy to do with the data on my spreadsheet, that’s why it is the way it is. Income I can analyse separately as I have a complete record of all dividend payments received. These forms of analysis have been sufficient for me to date. Thought I’d try out unitization - but to date I still don’t really see exactly what it’s telling me except in the most general of terms.

ATB

Pref

[BusyBumbleBee]

I calculate the units created from reinvested dividends at the end of each month, likewise the units cancelled by cash withdrawn.TJH

Surely the dividends belong to the existing units? So no new units should be created when the dividends are paid. In fact when the individual shares (or whatever) go ex dividend the existing units have the entitlement to that forthcoming dividend otherwise being a fund manager would be even more profitable than it is now!

This is buried in the monevator article which in my view is very poorly written. See

What about dividends, spin-offs, dealing fees, stamp duty, rights issues and so on?
None of this matters, so long as all the incoming money (from sources such as dividends and spin-offs) and expenses (such as dealing fees and stamp duty) are retained or paid from within the portfolio.

The number of units you own doesn’t change because you were paid a dividend – no more than if one of your shares went up by 20p

[mc2fool]

The doubling of the unit price I am interpreting as being a positive sign that my investments have in fact made money, clearly a reduction in the unit price would represent having made a loss.

Well, no, not necessarily.

The advantage of unitisation is that it removes the effect of cash flows, and so measures the performance of your portfolio and you as a "fund manager" of it -- and the disadvantage of unitisation is that it removes the effect of cash flows, and so doesn't measure the performance of your investment and you as an investor.

Example: you start off your investing career with £10K in the bank that you plan to invest and make your first step with £1K, unitising to 1000 units at £1.

Some time later your invested portfolio has doubled, so is now 1000 units at £2, worth £2K, and you decide you're a genius so pile in the other £9K, "buying" 4,500 more units, giving a total of 5,500 units at £2, worth £11K.

Then, by the end of the year your investments have reversed somewhat such that you have 5,500 units at £1.20, worth £6,600.

So, across the year your units have gone up in value by 20%, a good result for the performance of your "fund" and of you as a fund manager of it, but you have lost £3,400 as a result of your poor timing decision as an investor in "buying" the additional units in your "fund".

To measure the "investor" performance you need to get the discounted cash flow rate of return, which takes cash flows (in and out) into account. This is most easily done using the Excel XIRR function. Look it up and ask if you're confused.

Both unitisation and XIRR are useful tools, but they tell you separate things and neither alone tells you the whole story.

[EssDeeAitch]

To measure the "investor" performance you need to get the discounted cash flow rate of return, which takes cash flows (in and out) into account. This is most easily done using the Excel XIRR function. Look it up and ask if you're confused.

Both unitisation and XIRR are useful tools, but they tell you separate things and neither alone tells you the whole story.

Got it! Did the XIRR and as you say, it is a useful measure.

[PrefInvestor]

Yes i see that most people posting their portfolios here include an XIRR figure for each investment. I believe this gives some kind of annualised rate of return figure ?. Personally I am quite happy with capital gain, dividends received, total gain (capital + dividends), yield for the coming year and break-even figure (across all purchases and including dividends received) for each stock - seems to tell me everything that I am interested in......?. Clearly others think not.......

ATB

Pref

[EssDeeAitch]

Yes i see that most people posting their portfolios here include an XIRR figure for each investment. I believe this gives some kind of annualised rate of return figure ?. Personally I am quite happy with capital gain, dividends received, total gain (capital + dividends), yield for the coming year and break-even figure (across all purchases and including dividends received) for each stock - seems to tell me everything that I am interested in......?. Clearly others think not.......

ATB

Pref

Well your missing something elemental here. There can NEVER be enough statistics

[tjh290633]

I calculate the units created from reinvested dividends at the end of each month, likewise the units cancelled by cash withdrawn.TJH

Surely the dividends belong to the existing units? So no new units should be created when the dividends are paid. In fact when the individual shares (or whatever) go ex dividend the existing units have the entitlement to that forthcoming dividend otherwise being a fund manager would be even more profitable than it is now!

This is buried in the monevator article which in my view is very poorly written. See

What about dividends, spin-offs, dealing fees, stamp duty, rights issues and so on?
None of this matters, so long as all the incoming money (from sources such as dividends and spin-offs) and expenses (such as dealing fees and stamp duty) are retained or paid from within the portfolio.

The number of units you own doesn’t change because you were paid a dividend – no more than if one of your shares went up by 20p

That is correct for Accumulation units where the dividends roll up inside the units. It is not correct for income units, where the dividends may be withdrawn as cash or reinvested into whichever share the owner wishes.

TJH

[PrefInvestor]

Hi EssDeeAitch, Well in general terms I am up for including any metric in my investment spreadsheet if it has some obvious use. Now I’ve tried thinking about what adding an XIRR column to my investments spreadsheet would do for me. As I understand it what it would give me is an annualised gain figure over the whole time that I’ve held the investment.

So considering possible values for such a field:-
1. If it’s a really low number and less than the yield then that would tell me that I must have been making a capital loss for some of the time – but then my capital gain/loss figure would tell me that.
2. Similarly if it’s a really big figure and greater than the yield – then ditto.
3. It’s going to tell me nothing about the future of the investment, nothing can do that !.

So I don’t immediately see any use for it. My existing set of fields tell me everything I think I need to know that such a figure might tell me.

Obviously lots of people here feel differently. I’m not sure why. Perhaps they will explain.

I also note that calculating it wouldn’t be any fun either. You need to have a contiguous set of cells which hold the date values and amounts of all transactions (buys, sells and dividends). To do that I’d probably need to write a VBA function and collect together the various data items in local variables before calling the XIRR function. And considering that I just spent a whole load of time removing all the VBA from my spreadsheet last time I revamped it (because it stopped undo from working reliably) I am not keen to start putting it back in to compute this metric.

So unless someone can explain some really compelling reason why I need it then no XIRR for me I think…..

ATB

Pref

[EssDeeAitch]

Hi EssDeeAitch, Well in general terms I am up for including any metric in my investment spreadsheet if it has some obvious use. Now I’ve tried thinking about what adding an XIRR column to my investments spreadsheet would do for me. As I understand it what it would give me is an annualised gain figure over the whole time that I’ve held the investment.

So considering possible values for such a field:-
1. If it’s a really low number and less than the yield then that would tell me that I must have been making a capital loss for some of the time – but then my capital gain/loss figure would tell me that.
2. Similarly if it’s a really big figure and greater than the yield – then ditto.
3. It’s going to tell me nothing about the future of the investment, nothing can do that !.

So I don’t immediately see any use for it. My existing set of fields tell me everything I think I need to know that such a figure might tell me.

Obviously lots of people here feel differently. I’m not sure why. Perhaps they will explain.

I also note that calculating it wouldn’t be any fun either. You need to have a contiguous set of cells which hold the date values and amounts of all transactions (buys, sells and dividends). To do that I’d probably need to write a VBA function and collect together the various data items in local variables before calling the XIRR function. And considering that I just spent a whole load of time removing all the VBA from my spreadsheet last time I revamped it (because it stopped undo from working reliably) I am not keen to start putting it back in to compute this metric.

So unless someone can explain some really compelling reason why I need it then no XIRR for me I think…..

ATB

Pref

I can only apply this (and any) measure since August last year when I started self-investing so cannot really comment on how useful XIRR will be in the long term. But on the basis that its easier to start with something than backtrack, I think it's good to include it.

Perhaps others can comment on the points you raise.

BR

[nmdhqbc]

So unless someone can explain some really compelling reason why I need it then no XIRR for me I think…..

Your method of working out your yearly returns is just not accurate. If you can't be bothered to go to the effort of getting accurate numbers then that's fair enough. I could not be bothered for a long time and then my life changed and I now have more time to fill so I started doing it just over a year ago.

[OZYU]

Hi EssDeeAitch, Well in general terms I am up for including any metric in my investment spreadsheet if it has some obvious use. Now I’ve tried thinking about what adding an XIRR column to my investments spreadsheet would do for me. As I understand it what it would give me is an annualised gain figure over the whole time that I’ve held the investment.

So considering possible values for such a field:-
1. If it’s a really low number and less than the yield then that would tell me that I must have been making a capital loss for some of the time – but then my capital gain/loss figure would tell me that.
2. Similarly if it’s a really big figure and greater than the yield – then ditto.
3. It’s going to tell me nothing about the future of the investment, nothing can do that !.

So I don’t immediately see any use for it. My existing set of fields tell me everything I think I need to know that such a figure might tell me.

Obviously lots of people here feel differently. I’m not sure why. Perhaps they will explain.

I also note that calculating it wouldn’t be any fun either. You need to have a contiguous set of cells which hold the date values and amounts of all transactions (buys, sells and dividends). To do that I’d probably need to write a VBA function and collect together the various data items in local variables before calling the XIRR function. And considering that I just spent a whole load of time removing all the VBA from my spreadsheet last time I revamped it (because it stopped undo from working reliably) I am not keen to start putting it back in to compute this metric.

So unless someone can explain some really compelling reason why I need it then no XIRR for me I think…..

ATB

Pref

I can only apply this (and any) measure since August last year when I started self-investing so cannot really comment on how useful XIRR will be in the long term. But on the basis that its easier to start with something than backtrack, I think it's good to include it.

Perhaps others can comment on the points you raise.

BR

I can't see what all the fuss is about. ‘A set of cells with date, transaction, etc..’ is, put in a table, the core of any portfolio system. Another table is the ‘prices table’. From that all my unitisation, XIRR is automatically computed without the slightest amount of extra work. Tables, pivots and slicers make, in Excel, the whole thing a doddle. From that everything can be automatically extracted and plotted. It requires proper design to start with. It requires at most a few minutes data entry when portfolio activity occurs. I spent my working lifetime designing systems, and in my experience, messy spreadsheets produced by spreadsheet champions were legion, but properly analysed and designed systems were few. A portfolio system, compared with industrial automation/robotics/AI systems we designed, is an elementary application. I run a cloud based system in use by 21 family and friends, we have settled on two versions, a slick one and a more complete one which deals with options, CFDs etc.... It has not required any new design once implemented.


Ozyu

[PrefInvestor]

HI OZYU, Well FWIW I also used to work in IT and I am generally pretty handy when it comes to spreadsheets. My spreadsheet is based on Excel tables but the design and implementation has incrementally grown over the 7 years that I have been using it through progressive feature additions. For the most part the data is held as you would in a relational database but I confess that I know that with 20:20 hindsight I have a major flaw in the data model in that I ought to have a single date ordered transaction table containing all portfolio events such as buys, sells, deposits, withdrawals, dividend payments as this would make a lot of things much easier. That would making doing an XIRR calculation very easy. But unfortunately I dont have that, I have an Investments table (holding all my buys and sell), a dividends table (holding all my dividends), a stocks table (which combines everything into a single co-ordinated view). So I have a 7 year old growing legacy system which serves my needs very well day to day but isnt always the easiest thing to change, and starting again from scratch isnt a project that appeals.

None of which changes my point that I dont really know what I would do with an XIRR field even if I had one, so to go to any significant effort to put it in doesnt immediately seem worthwhile.

ATB

Pref

[tjh290633]

I'm with OZYU. All you need is a column with the cash flow for the portfolio or the individual share. This aligns with the dates for the various events, which are usually in the left hand column. Money in is negative, money out is positive and the final row in the table has today's date and the current value as a positive entry. Dividends are money in. If reinvested in the portfolio, a corresponding entry cancels them out.

For individual shares, dividends may be reinvested elsewhere or withdrawn. Hence they are positive and any money used to buy more shares is negative.

XIRR takes account of serial purchases and sales, and the effects of dividends very simply. It gives an accurate measure of Total Return. It can also be used for individual years or longer periods if wished but short periods of less than a year are inadvisable.

TJH