The trouble with unitising one’s portfolio is there’s nowhere to hide. Unitising lets you track the effects of adding money, which helps avoid the easiest gotcha in fooling yourself on returns. The Beardstown Ladies Investment Club effect. The hard earned cash you lob into the pot makes your portfolio go up, but it’s not profit, or ROI, or anything like that.
Unlike starting with a one off lump sum from which you draw nothing, evaluating performance gets a lot more complicated if you draw a yearly stipend from your stash. It gets a lot more complicated if you’re one of the ordinary mugs who has to actually, y’know, earn the money they are putting into their future freedom fund, paying it in year by year as they go.
The UK version of the Motley Fool used to have the greatest description of how to unitise and worked example called Stockpicking – Are You As Good As You Think? by G.A.Chester which was still visible to freebie members, but all that was lost when they reorganised the website(19 January update – see Neil’s comment for the original text – he had saved this). Sadly G.A. Chester seems to serve up endless spammy clickbait articles these days, what the hell happened to you, man? Stockpicking was an article pure genius, putting across a tough concept in actionable bitesize steps.
Monevator has a description here but for some reason I really struggle to follow that, although I recognise the moving parts when I analyse my spreadsheet written to implement GA Chester’s more ermine-friendly narrative. I tested the spreadsheet against Chester’s example. Pity that gem of wisdom is lost to linkrot.
Unitising is quite a grief-stricken and error-prone process because it involves going through the spreadsheet and entering the current price of holdings I own at the January sampling datum point. After 10 years, particularly with some occasional muppetry I have a few dead lines of stocks I have no holdings in, but it’s easy to miss the odd line where I do have holdings. It fails safe in that if I don’t enter the price of a holding I own, it says the value of that line is 0 which makes the unit price lower, which is an incentive to go back and catch all of ’em on the grounds I can’t be that crap, surely? There’s also a error-checking catch line that tots all the holdings up, it’s kinda nice if it matches Iweb’s view of my world. Obviously fans of Cloud Services like Money Dashboard will have this easier, though you still need to do the annual spreadsheetery to unitise. Money Dashboard claims to be
a secure cloud-based open banking website that enables you to replicate and then track all the spending categories you set up in MSE’s Budget Planner
Colour me a cynical sonofagun but I am of the firm opinion that secure and cloud-based do not belong in the same sentence. See Equifax for a worked example.
The Ermine portfolio unit value is down 5% this January to last January. It’s also changed nature, more gold and I have taken 20k out as cash, though I may stick that back in to Charles Stanley, which is a Flexible ISA, and pull it out again halfway through April. And I may contribute something to Iweb this year, though I can’t make the full 20k.
Now that’s not dreadful. What would I have been doing otherwise – I’d be in VWRL a la Lars Kroijer.
I get divi from VWRL, which is about 2%, I guess there’s a .25% platform fee too. So instead of all that tracking, I could have had one lot of VWRL and been about the same.
What about VGLS100? That was about -5.36% in acc units. Much of a muchness and not worth the Sturm und Drang. In general, a little bit shit. Where Eagles Fear to Perch did better than me last year for instance, congratulations that man!
Defence, not offence is the word at the moment
Now I did shift much more defensively, there’s a lot of gold, there are some government bonds in there. I am probably suffering the deadweight drag of the gold not earning an income. Well, that’s my excuse. I shifted more defensively for several reasons. It is not quite determinate when the best time to take my main pension is, there is a balance between the actuarial reduction because I am not 60 and what appears to be high CETVs which incidentally seem to reduce the actuarial reduction, for reasons I don’t understand.
So I have to keep on pinging the pension modeller. I might need some of that cash if the modeller says delay a bit, and money you might need in the next five years has no business being in the stock market. Particularly when said stock markets are at high valuations. I did much of the switching mid last year, but all that gold and the cash is pretty much a passenger now. I am not one of you young finance workers getting a savings rate of 50% into your SIPPs, I might have a negative savings rate this year.
I’m also trying to keep some of this year and last year’s ISA allowance, because I will draw a pension commencement lump sum from my main pension. And there is some hazard of a Corbyn led government in the future. As a retiree I won’t have a particularly spectacular income1 so I will probably be safe from his ministrations, but an ISA allowance of £20000 is way above what the vast majority of the population could even dream of saving. The argument that letting the rich shelter such a large yearly amount from tax does have some cogency, so I want the possibility of getting that PCLS into the ISA within the next year or two. Whether £20,000 will have any useful value in the Brexit Brave New World of buccaneering brio will remain to be seen.
- by the standards of my professional self or indeed the general UK PF scene – even the employed Ermine was way down in the ranks of finance whizz-kids well represented on the UK PF scene now. It wil be fine and more than my early retired self, but I don’t expect to be a tall poppy in Corbyn’s sights. Hopefully Corbyn won’t have the Blairite ambitions of siring a baby-boom through pronatal giveaways as we had in a tough period midway through my career, where every other bugger seemed to be getting the breaks. ↩
Simple Living In Somerset 
Couldn’t agree more re: the Money Dashboard and similar apps…. As one man whom I greatly admire would put it, I fart in their general direction. Yah, dahlink, of course. All I need to complete happiness is to give away a bit more of my personal info to some data mining collective so they can profit from it.
LikeLike
“but an ISA allowance of £20000 is way above what the vast majority of the population could even dream of saving.” A lifetime allowance was applied as a deterrent to some for further investment in pensions, have you ever considered the possibility that a Corbyn government could do similar to ISAs – thus increasing the tax take for those with substantial longer term portfolios, while still offering a wide annual allowance for the average person?
LikeLike
I’m pretty sure he will, should he get that far. But I worked in engineering, not finance, so my ISA will never reach such heady heights, in the same way as my pension is not in hazard of the LTA. So that sort of risk is not my circus, not my monkeys. I’m not saying Corbyn couldn’t bugger my life up in other ways, though at the moment it’s hard to see him making a bigger cod’s of it than Call Me Dave with his blasted referendum, the twazzock. But that’s not setting the bar that high, really.
LikeLiked by 1 person
Dear Simple Living in Somerset, re Money Dashboard I have to say you really are coming across as a cynical sonofagun. 🙂 I can only assume both you and the other gentleman who has commented have no personal experience of using it. I’ve used it for over four years now and never had a problem. I would call it the single most useful financial tool I have ever come across, one that has helped me turn myself around. You were fretting on Monevator about what might happen if a toerag stole your phone. I had a toerag steal my laptop and there were no subsequent MD-related problems (or bank account problems of any kind). This was also some years ago so I think an issue would have emerged by now if it was going to.
Incidentally, I’d also say that about Money Saving Expert, which you’ve had a go at in the past, so I may as well mention that here in the same breath. MSE has helped me enormously. I think you had the impression it was all about persuading people to buy stuff. It has millions of users and its forums cover a vast array of money-related topics and approaches, mostly run by the users themselves. I’ve been a member for over 10 years now and have never found it anything but helpful and eminently sensible, and if anything have found it to be about saving as much as you can, spending as little as possible and getting rid of unnecessary stuff.
I’m not having a go at you per se – yours is one of the few financial blogs I do regularly read. I’m just mentioning that I’ve had a lot of experience of both MD and MSE and the negativity towards both seems to me to be misplaced.
LikeLiked by 1 person
Absolutely fair cop on MD, I haven’t used it because I use an old program Quicken to do the same job, which resides on my own computer. And doesn’t download stuff from my bank, when I buy stuff I enter it manually. I did use something similar to MD many years ago, called Egg Money Manager. But when I read an article in The Register that this basically stored the passwords in an ActiveX control I came to the sad conclusion that convenience wasn’t really worth the risk to my paranoid mind. Quicken is okay and does the job.
But as for the charge of being mean to MSE, not guilty IMO. I have generally only admiration for them. Obviously I’m cynical on some of the products they say are a great deal, some of these shouldn’t be in any FIRE aspirant’s shopping trolley, pretty much anything in the deals/vouchers section f’rinstance. But if you are going to buy that sort of thing then overspending the least is a less bad way to go.
So no, MSE are a fine crew in the Ermine viewpoint!
LikeLiked by 1 person
I use the Monevator way to unitise and I too am down by 5% at the end of 2018. I don’t compare my investing ‘skills’ to any benchmark but compared to other years where I did better, my performance is rather rubbish!
I can only dream of maxing my ISA limit (having only done it once before with the help of my redundancy money and when it was at £15k) but I’d still think it a shame if this were reduced as it’s something to aim for. Let’s hope they just cap it at £20k.
LikeLike
It’s good to be in good company on this! TMF has a list of historical ISA allowances and it was interesting to plug in the 1999 value of £7000 into the BoE’s inflation calculator. It’s worth about £11500 today, so the limit has roughly doubled in real terms.
ISTR I managed to also max the ISA once in those heady dotcom days. A foolish Ermine and his money were easily parted…
LikeLiked by 1 person
Thanks for the hat-tip. I’m with you on MD and any other software where the data doesn’t reside solely on my own drives. It must be an age thing, the young are far too trusting …
LikeLike
@Ermine – Good provocation as usual. Like you I am a Quicken user (2004!) and it remains useful. But it unquestionably would not be reinvented in the cloud-based Open Banking world.
Re unitising, I think it is much easier than you imply. Try the following:
– take your existing portfolio, which is worth say £260k.
– unitise it by saying it represents 100k units. I insist, 100k on the nail. You have just ‘unitised your portfolio’. Congrats!
– as of today each unit is worth £2.60. [£260k/100k]
– now take a portfolio topup of, say, £4k. Every time you either top up or withdraw you need to keep a log and do the very simple calculation into what the latest unit price is, and thus how many units this is buying (in a topup) or selling (in a withdrawal). This is the additional step required when you are tracking a unitised portfolio.
– For the sake of argument at the point you do this £4k topup, your portfolio has dropped in value to £258k. This means that just before your topup your unit price is £2.58. £4k buys 1550.4 units. So you now have 101550.4 units and your portfolio is worth £262k. Your price per unit has not changed and is still £2.58.
– note that dividends received don’t alter your number of units so long as they stay in the portfolio. If you withdraw them then you need to work out how many units you have just withdrawn; if, for instance, you take out £130 of dividends, then this represents a reduction in units of about 50.
– at the end of the year, imagine you haven’t done any more topups/withdrawals, and the portfolio is now worth £271k. Your 101550 units are each now worth £2.669. This represents a 2.65% increase in the unit price. This is your portfolio’s underlying investment return.
Sorry for the grandmothers egg sucking lessons, but hopefully we can avoid scaring readers off from one of the easiest tricks in the tracking book.
LikeLiked by 1 person
Now THAT is a description I can understand, and gels with my recollections of GACs article. And matches what happens in the spreadsheet if I do the year before last’s figures by hand 😉 Thank you!
My copy of Quicken is also 2004, I think this was the last UK variant. It wouldn’t take much innovation to make Money Dashboard safer. Allow people in their online banking to create a read-only password to give aggregators visibility of their account. It wouldn’t help with the privacy and datamining issues.
Moneydance looks like a fairly close modern version of Quicken for the UK market as well as the US. It has the capability to d/l from banks and make online payments but you don’t have to use that.
LikeLike
The problem with unitising on the fly is that, strictly speaking, every injection of new capital requires a complete portfolio valuation at that particular time to establish the new unit price, something I only do at the end of each month because if you also hold bonds etc it’s not such a simple calculation and takes a bit of digging around various web accounts to pull in all the components.
If anyone’s having problems with following the written descriptions of unitisation, maybe just consider it in simple formulaic terms.
Unit price change factor = (M2-X+Y)/M1, where M2 and M1 are the portfolio valuations at period ends (say monthly), X is money added during the period and Y is money withdrawn.
Let’s say the valuation at the end of the first month, M1, is £260k based on 100,000 units and the next month’s valuation M2 is £271k, using the figures from above. At sometime during the month (it doesn’t really matter when) you’ve added £4k of new cash but have made no withdrawals.
(M2-X+Y)/M1 = £267k/£260k = 1.0269 x the initial unit price of £2.60 gives a current unit price of £2.6694. The number of units now becomes somewhat irrelevant because it’s always just the latest portfolio valuation divided by the current unit price, but in this example would be 101,520.9. Monthly portfolio return is (£2.6694-£2.60)/£2.60 = +2.7% (a very good month !)
Now say you withdraw £10k in the third month, and the total portfolio valuation at the month end, M3, happens to be £259k.
(M3-X+Y)/M2 = £269k/£271k (assuming no additions in this particular month) = 0.9926 x the M2 unit price of £2.6694 = £2.6497. Number of units is £259k/£2.6497 = 97,746.9. So you’ve effectively ‘sold’ 3,774 units at £2.6497 (i.e. the £10k withdrawn).
This latest monthly return is (£2.6497-£2.6694)/£2.6694 = -0.7% and over the two-month period is (£2.6497-£2.60)/£2.60 = +1.9%
If you keep all your portfolio valuations, money added and money withdrawn within monthly columns on a spreadsheet, as I do, this process is very simple to automate.
LikeLiked by 1 person
I follow the same sort of process in a spreadsheet as FIREvLONDON. I do it once a month (too frequently really, but it fits in with a whole bunch of end of month/beginning of month assorted admin and takes no more than 15 mins). Current valuation before in and outs (OK it’s inaccurate, but transaction rate is low for me); therefore current value of a unit. Money in and out at this unit price, therefore new number of units and total portfolio value.
I then calculate some other interesting numbers like net cash in or out since I started the spreadsheet, which shows that I have only added a trivial amount in net cash over the last few years an and that most of the growth in value has come from reinvested dividends and fund growth.
On a separate sheet, I use the XIRR function to calculate the rate of return. For this I use actual dates for withdrawals and additions and it gives (supposedly) an accurate rate of return. Again it is pretty easy as transaction rate is low. I calculate rates of return for tax years (don’t really know why it just seemed convenient).
I do this across 4 stocks and shares accounts, 3 bank accounts and some NS&I holdings and it is pretty quick and easy.
Interestingly my unit price dropped 0.5% over the calendar year (too much cash and cash-like investments), but the rate of return this financial year to date is 2.76% ‘cos the unit price had dropped to a low point in April 2018.
I am sure there are better ways of doing these things, but this is good enough for me and at least give me some idea what is going on.
LikeLiked by 1 person
I am starting to look like the idle one only doing this annually. Having said that, I am not tremendously active. Sure, I do some market timing in trying to buy when people are having a good moan about the stock/index in question. But having bought it, I leave it be. Monthly rounding up would be too much. Interesting comment thread on Monevator’s unitisation post mentioned above, in that XIRR is a better measure of if you are any good at market timing.
LikeLike
@Ermine – that Monevator thread was the one that set me off on tracking my portfolio by unitising and calculating rate of return.
LikeLiked by 1 person
Doesn’t such a unitisation approach have problems with ACC funds, as you are rolling up their dividends invisibly. I know you need to extract this information for tax returns if unsheltered, so only have INC there.
I have a spreadsheet with current values, and historic information for CGT porpoises, which I store in configuration management so it has snapshots whenever I think to commit it.
I don’t try to work out changes year on year as its too hard. I just look at the first 2 digits of the total and smile/groan if they change.
I don’t use automated unit price lookups, but if I did, I’d create a fake portfolio in the cloud rather than reveal true numbers of units.
LikeLike
> Doesn’t such a unitisation approach have problems with ACC funds, as you are rolling up their dividends invisibly.
I’m running on the assumption that dividends increase the unit price of acc funds over and above the underlying asset price. F’rinstance if you plot a UK tracker, eg L&G UK Index if you plot the acc and the inc funds prince they slowly split apart in favour of the acc fund, which I guess is the cumulative effect of the dividends.
As such your portfolio will get more valuable because of the increasing price of the units held within it, even if the number of units is the same. In the same way as mine gets more valuable because of the divi cash explicitly added in it, though the value of the inc units track the SP of the asset year on year in a way that the acc units don’t due to the rolled up divi.
I guess if you used the divi to buy the inc units your number of inc units would drift up in the same proportion as the unit price of the acc units drifts up, less your greater transaction fees.
I only hold equities in an ISA nowadays but I still use inc funds, and use the dividends to give more flexibility in rebalancing
LikeLike
Acc funds should work seamlessly as long as you update the latest value of the holdings. Inc funds are the same provided the divi goes into the portfolio and not out into your current account. I tend to prefer inc funds and then reinvest them which, if I reinvest into the same fund, is the same as Acc.
LikeLiked by 2 people
I know how it works, I just suspect it complicates your accounting . The total return is fine, but I suspect you want to know the like-for-like unit cost changes over a time period, which is being distorted.
LikeLike
Should I hang my head in shame because I’m far too lazy to track my investment returns.
I do a net worth calculation once a year, but didn’t bother this year because I guessed it would be down. So I’m just chucking all of money into VWRL until I get to my net worth goal.
LikeLiked by 1 person
Should I hang my head in shame
Monevator started his article on unitisation:
Nobody’s going to screw up buying VWRL 😉 Okay, they might in theory but then we’ve all got serious problems that are probably larger than a retrenchment in networth.
LikeLike
Here is the detail from the G A Chester page:
There’s only one reason to invest in individual stocks: to earn a better return than you’d get by investing in the whole market. If you can’t choose stocks that consistently outperform a passive index tracker, your stockpicking approach will not only be a waste of time, but will also seriously damage your wealth.
The only way to know if you’re any good as a stockpicker is to measure your performance.
Unitisation
Most investors don’t make a one-off, lump-sum investment. We tend to add money to our portfolios over time, whether regularly or irregularly; and, occasionally, we may withdraw money, for example, to pay a child’s school fees or fund a holiday.
To get over the difficulty of measuring the performance of a portfolio that has money going in and coming out, we can adopt the principle of ‘unitisation’, which is used by unit trusts/OEICs.
Concrete examples are generally easier to grasp than abstract theory, so I’m going to use the Family Firms Portfolio to show you how unitisation works in practice.
The Family Firms Portfolio is being constructed over a period of time, using The Motley Fool ShareBuilder service. We are making regular investments of £150 a month.
First investment — 7 July
The first time you put money into your portfolio, you decide on an arbitrary unit value. You then convert your initial cash injection into units.
In the case of the Family Firms Portfolio, let’s say we decide our unit value is £1. As our initial investment is £150, the portfolio consists of 150 units (£150/£1).
Second investment — 5 August
Each time we inject new money into the portfolio, we look at its value just before doing so.
Ahead of our second family firms investment the portfolio value was £178. We divide the value by the number of units to get the current unit value: in this case, £178 divided by 150 units, giving a unit value of £1.19. We then divide the new cash being injected (£150) by the unit value (£1.19) to give the number of new units being created: 126.
So our portfolio now consists of 276 units (150 + 126).
Third investment — 7 September
We repeat the process with each new injection of cash.
Ahead of our third investment the value of the portfolio was £348. Dividing £348 by the number of units (276) gives a unit value of £1.26. Again we divide our new cash injection of £150 by the unit value — this month £1.26 — giving us 119 new units.
The portfolio now consists of 395 units (276 + 119).
Current value — 17 September
We can see how the Family Firms Portfolio has performed in its short life to date.
The current value of the portfolio is £502, which, divided by the number of units (395), gives a unit value of £1.27. So, the unit value has increased 27% (£1.00 to £1.27) since we made our first investment on 7 July.
Selling
If money is permanently removed from the portfolio, we follow the same procedure, but subtract the units. So, if on 17 September we decided to take £254 out of the portfolio, we would divide that amount by the unit value of £1.27, which would tell us that £254 is the equivalent of 200 units. As we are ‘subtracting’ the cash from the portfolio, we subtract the units.
The portfolio now consists of 195 units (395-200) and is valued at £248 (£502-£254); and the unit value remains the same — as it should — at £1.27 (£248/195 units).
Simplicity
Going back and unitising an existing portfolio can be a lot of work if it was started a long time ago. But you can unitise it from today, simply by looking at the current total monetary value, assigning an arbitrary unit value, and calculating a base number of units for going forward.
It doesn’t take much time to manually update a unitised portfolio — and even less time if it’s set up in a spreadsheet. Remember, the number of units only changes when new money is added or money is permanently withdrawn.
All trades within the portfolio, including the reinvestment of dividends and temporary holdings of cash, have no impact whatsoever on the number of units.
Benchmarking
Once your portfolio is unitised, you are in a position to make a like-for-like comparison of your performance with a suitable index, index tracker or actively managed fund.
In the case of the Family Firms Portfolio, our ‘investment universe’ is UK-listed stocks. If we weren’t stockpicking the money would probably be going into a FTSE All-Share tracker, so it’s appropriate to use such a tracker as our benchmark.
Also, because we are planning to reinvest dividends, it makes sense to benchmark against the ‘accumulation’ version of the tracker; if we were routinely taking dividends out, the ‘income’ version would be the choice.
We can pit our portfolio against the HSBC FTSE All-Share Index fund, for example, which is just one of a number of suitable All-Share trackers:
Unit value
at 7 Jul Unit value
at 17 Sep Gain/loss
Portfolio £1.00 £1.27 +27%
Tracker £2.84 £3.23 +14%
I hasten to add that this is merely an illustration of what unitisation allows you to do. A couple of months is way too short a period to draw any conclusions about a portfolio’s performance.
However, if after 5 or 10 years you find you’ve been handsomely outperforming your benchmark, you may just be cut out for stockpicking.
Equally, if you’ve miserably underperformed over such a period, it may be time to start thinking about passive investment.
You won’t know either way, though, unless you measure your performance. Get unitising!
LikeLiked by 1 person
Thank you! Much appreciated. I found this the easiest to walk through building the spreadsheet. I don’t run the same level of detail, I simply do it one year to the next on the value of the entire portfolio + cash. That’s effectively the Ermine investment trust as a whole.
LikeLike