HYP 2019 - the results are in

[Arborbridge]

My results are distorted by the take over of Greene King because this reduced the number of units near the year end. If I take the units at face value I get an increase in income of 12.7%. If I take the units prior to the GNK sale the increase is 6.6%. After GNK there were no other sales or purchases in the year.

My results are:

Income per unit (daily calculation method) 6.6%

Income unit price: up 12.8%

Arb.

So, if you intend reinvesting the Greene King proceeds, you will get the opposite distortion in the current years results, won't you?

Doesn't seem to be achieving the purpose of unitising as I understood it.

Adrian

It's always been true that if you make large changes the result will not initially show a faithful result. For example, if I add loads of units today, there will be no dividends accruing from those units for months, possibly half a year. The reverse is true if and when one sells. I roughly circumvented this for years by taking the current income and dividing it by the number of units which existed three months prior (I mentioned this in one of my reports) which was a handy compromise as some of those units would begin banging off dividends in that time - bearing in mind I was buying most cum-dividend. It all comes out in the wash, and one can see one's income progress accurately enough, which is the purpose of the exercise.

However, a few years back someone here suggested that I should set up a daily*calculation in my spreadsheet as it is a more accurate method and I relulctantly did so. I have not regrets: once set up, it's easy to do with speed filling etc. so that's what I'm doing now, but by tradition I usually have quoted the old "year end dividend divided by year end units" since that is the longest record I have.

Interestingly, "the year end income divided by units existing three-months prior" or alternatively "end of quarter.........." gives virtually the same answer to a decimal point as the daily calculation, so my initial guess about three months being a reasonable delay is about right.

The important thing to note is that we are mostly interested in change and provided one keeps a consistency about how one is calculating, I believe it is fit for my purpose. Other people may use slightly different methods - possibly TJH will be along soon to comment on what he does in this regard.


*this is a sloppy term. It is "on the day" not literally every day. The calculation is done on the day each event occurs, whether it's a change in the number of units or recording a dividend.


Arb.

Incidentally, if you think there are some inaccuracies in my attempt to unitise, how much greater are those of not unitising!
My own income is up about 16% this year but that isn't owing to the success of my existing capital - I am not getting 16% more bang for the buck! Most of it is from re-investment or new capital, but what use is that to anyone else? I cannot tell from that whether my investments are efficient or not compared with any other investment method, only unitising can tell me that.

[tjh290633]

I assume that cash from dividends or other sources is reinvested at the end of each month, if the source is within the portfolio. Cash input from outside, or withdrawn, is accounted for on that day.

It's not unusual to have up to about 10 dividends arising in a single month. I haven't counted them, but the annual total from 35 holdings, some of which pay quarterly, is probably more than 80.

TJH

[Arborbridge]

I assume that cash from dividends or other sources is reinvested at the end of each month, if the source is within the portfolio. Cash input from outside, or withdrawn, is accounted for on that day.

It's not unusual to have up to about 10 dividends arising in a single month. I haven't counted them, but the annual total from 35 holdings, some of which pay quarterly, is probably more than 80.

TJH

It therefore follows (as in my case with Greene King) that if a large change of units happens from one month end to another, theres can be an unusual change in the income per unit value.

I think it all comes out in the wash, in time. But these days, as I mentioned, my preference is to account for each dividend on the day it arrives and with the number of units on that day. That seems the most accurate with surprisingly little extra effort.

Thanks for your confirmation about how you do it in your case.

[Hypster]

A bit odd that income per unit went up so much, and unit value so little.

To be honest, I thought the same! It's entirely possible that I'm handling income units incorrectly. I'll go over some of the posts I used to inform my methodology to make sure I did it right.

Just to update the board, after forensically going over the data I found an error in 2018's valuation, which overinflated 2018's unit value.

The correct increase in income unit value was 15%

[Arborbridge]

A bit odd that income per unit went up so much, and unit value so little.

To be honest, I thought the same! It's entirely possible that I'm handling income units incorrectly. I'll go over some of the posts I used to inform my methodology to make sure I did it right.

Just to update the board, after forensically going over the data I found an error in 2018's valuation, which overinflated 2018's unit value.

The correct increase in income unit value was 15%

Thanks for coming back on that.