Leases and IFRS16

[Snoopy]

I can't recall a change in accounting rules that has caused more disruption to financial statements than IFRS16. The principle behind IFRS16 is to bring leased assets onto the company balance sheet. A lease becomes a 'right to use asset' on one side of the balance sheet offset by a 'lease payable liability' on the other. The 'right to use asset' is then amortised each year as annual lease payments are made, with a consummate reduction in 'lease payment liability'. In practice things are not quite as simple as this :-(.

The 'right to use asset' and the 'lease payment liability' are nominally of equal value, but they do not appear that way when listed on the company balance sheet. Each lease payable is measured as the 'present value' of all future lease payments, discounted back to the present at the incremental borrowing rate of the company. Lease costs are then recognised through 'lease interest expense' over the life of the lease.

SNOOPY

[Snoopy]

Lease costs are then recognised through 'lease interest expense' over the life of the lease.

I have a problem when assessing company solvency, which is a core part of any investment strategy. One way to assess whether a bank is generating enough income to pay its funders (think banks) is to look at the ratio: EBIT/I

'I' in the denominator is the net interest paid. However it is not clear to me that 'lease interest expense' is part of this. Before IFRS16, one of the most common forms of 'lease interest expense' was called 'rent'. 'Rent' was typically a business expense shoved in with the power bill and the wages bill in an 'operating expense bucket'. Post IFRS16, what was a 'day to day' expense has been pulled out of the overall 'operating expense bucket' and turned into a balance sheet item that did not exist before. Banks did not consider 'lease interest expense' as part of the overall company 'interest bill' because that 'lease interest expense' was generally owed to a different third party. Yet rent was taken account of by the banks indirectly, because rent reduces EBIT, the numerator of our 'ability to fund' assessment ratio. But under IFRS16, the treatment of 'rent'/'lease interest expense' as part of the debt assessment process is not clear to me.

None of this is about 'me' though, it is all about the banks. This suggests two alternative treatments for EBIT/I in the IFRS16 era.

1/ If 'lease interest payments' (sic, I can't help thinking of them as rent) have already been removed when calculating EBIT, then you should not include any lease interest payments in 'I' , the denominator of our ratio. To do so would mean you are accounting for the effect of rent twice, by:

a/ reducing earnings with rent AND
b/ converting rent into higher interest payment that must be serviced.

all at the same time. However

2/ If you remove 'rent' as a factor when calculating EBIT, because it is now classed under IFRS16 as a 'finance interest expense', then you will get a higher EBIT than before the change to IFRS16. In that situation, it would be appropriate to regard rent, now reported as a finance interest expense, as part of the denominator 'I'. Although doing it this way gives the company higher EBIT earnings, those higher EBIT earnings have a higher interest expense to cover. In this version of EBIT/I, the effect of 'rent' moves to the numerator.

The question I have is, under IFRS16, do I use method 1 or method 2 to calculate EBIT/I? Or do I have a choice which method I use?

SNOOPY

[Snoopy]

The question I have is, under IFRS16, do I use method 1 or method 2 to calculate EBIT/I? Or do I have a choice which method I use?

Sometimes the easiest way to make sense of these 'theoretical questions' is to do an example. I will do the calculation of "EBIT/I" for Chorus over the last reported year, FY2020.

AR2020 p27 tells us that :

1/ 'Other interest expense' includes $21m of 'lease interest', $5m of amortisation and a $1m restructuring expense due to interest rate swaps. That works out to a total of $27m.
2/ Now, if we go back to p24 of AR2020 and look under the 'expenditure commentary' we can see an entry 'Other $27m'. which ties in.
3/ Back a couple of pages further to p22 of AR2020. Right under the main 'Management Commentary' header you can see how the earnings are calculated. The first step is to take the 'operating revenue' and remove the 'operating expenses'. These are the same 'operating expenses' we have just looked at in a more detailed way on page 24.

These three steps show me that 'lease interest' has already been subtracted from profits. This means we should use 'Method 1' from my previous post when calculating EBIT/I.

EBIT is easy to find, it is listed as $246m on p22 of AR2020.

The slightly more tricky thing is figuring out the 'I' bit.

On page 27 of AR2020 we can see a 'Total Finance Expense' of $185m. But this is not the figure we use.

1/ Right at the top of the page we see 'Finance Income' of $12m that we have to offset against out finance expense.
2/ We must subtract from the 'net interest total' the $29m of 'CIP securities notional interest', because this is an accounting construction that is never actually paid (this is all explained on the Chorus thread, but for the purposes of this exercise please trust me on this point).
3/ Look further up the column and you will see the 'Other interest expense' of $27m that we have been discussing. That $27m has already been used in calculating EBIT. So we have to remove that from the interest bill as well , because if we did not we would, in effect, be counting it twice.
This means the 'Total Net Finance Expense' for our purposes is:

(-$12m + $185m) - $29m - $27m = $117m

So the obvious calculation of EBIT/I for Chorus for FY2020 is: $246m/$117m = 2.10 (using method 1)

SNOOPY

[Snoopy]

The calculation of EBIT/I for Chorus for FY2020 is: $246m/$117m = 2.10 (using method 1)

Now let's look at the alternative calculation method 2, where 'lease interest' (as part of 'Other Interest') is not taken into account when calculating EBIT.

EBIT now changes to $246m + $27m = $273m

With this iteration we have chosen not to take the lease interest into account as earnings. So instead we must include 'Other interest' as part of the total interest bill due by not subtracting it. This means the 'Total Net Finance Expense' for our purposes is now:

-$12m + $185m - $29m = $144m

So the calculation of EBIT/I for Chorus for FY2020 becomes: $273m/$144m = 1.89 (using method 2)

That isn't grossly different, except there is a 'rule of thumb' that says an EBIT/I ratio above 2 is passable, while anything below that is dodgy. So is 'method 1' or 'method 2' the better way of calculating this ratio? I don't know the answer. My solution is to curse IFRS16 and go to bed.

SNOOPY

[Ferg]

Now let's look at the alternative calculation method 2, where 'lease interest' (as part of 'Other Interest') is not taken into account when calculating EBIT.

EBIT now changes to $246m + $27m = $273m

With this iteration we have chosen not to take the lease interest into account as earnings.

Hello Snoopy

Why are you adding the lease interest to EBIT? I saw your explanation in the earlier post, but it does not make sense. I believe you are tying yourself in knots when the numbers we need are in the cashflow statement.*

How does one find the post I put up for you in another thread explaining IFR16?

I view the 2 options as being:
1) take the numbers as is, or
2) adjust the numbers back to the old method

Under method 1 you would use the published numbers - which I freely admit are "corked" due to the inclusion of lease interest and you lose the ability to compare to prior periods which did not have IFRS16.

Under method 2 we would adjust EBIT down (not up) for the rent payments (ie the total lease payments outflow from the cashflow statement, less GST = $23/1.12 = $20) and adjust EBIT up by the depreciation component of leased assets (i.e. $14), and would adjust "I" down for the lease elements only (i.e. $21).

EBIT: $246 - 20 + 14 = $240
I: $173 - $21 = $151
EBIT/I = 240 / 152 = 1.58

*Also, any differences between the P&L and cashflow will be due to accrual accounting.

FERG

[Snoopy]

How does one find the post I put up for you in another thread explaining IFR16?

As you have put it yourself, albeit not quite in these words: Seeking any rational explanation on the conceptualization and implementation of IFRS16 is a task that is, almost by definition, impossible.

However, to answer your search question, go to 'Search', then 'Advanced Search'. A box will pop up where you can put in a 'term to be searched for' (I put in 'IFRS16') then the name of the post author (I put in 'Ferg'). That search will then pull up a link to all posts that you have authored that mention IFRS16.

That lead me to your post 13827 on the Heartland thread.

https://www.sharetrader.co.nz/showth...l=1#post848304

I don't know if that was the one you were looking for?

SNOOPY

[Snoopy]

Hello Snoopy
Why are you adding the lease interest to EBIT? I saw your explanation in the earlier post, but it does not make sense.

To contextualize this discussion, we are talking about the FY2020 results of Chorus and I am trying to decide whether to add lease interest payments to EBIT or not. The post you quote (post 4) is my 'method 2' calculation where I do add the lease interest to EBIT. The post you do not quote is my 'method 1' calculation (post 3) where I do not add the lease interest to EBIT. It sounds like you favour my 'method 1' (post 3) calculation, of my two options.

Although we are talking about EBIT, I am considering how that figure is useful in evaluating banking covenants - in particular EBIT/I.

I believe you are tying yourself in knots when the numbers we need are in the cashflow statement.*

I view the 2 options as being:
1) take the numbers as is,

Under method 1 you would use the published numbers - which I freely admit are "corked" due to the inclusion of lease interest and you lose the ability to compare to prior periods which did not have IFRS16.

OK under method 1, I just used the published figure of EBIT which is $246m (AR2020 p22). The contentious factor in doing this is that what used to be classed as 'rent' has now been moved to a 'finance expense' under 'Other interest expense' header of $27m (AR2020 p27). In the text further down that same page we find that $27m figure includes $21m of 'lease interest' (formerly called rent). I notice that in your own adjusted 'method 2' calculation (quoted further down this page), you have adjusted for this. But you did not adjust for the rest of the 'Other interest expense' (again refer to AR2020 p27) where $5m of 'amortisation arising from the difference between fair value and proceeds received from interest rate swaps' and $1m of 'one off expense in FY2020 for restructuring forward interest rate swaps'. Of course that $5m and $1m are not directly related to IFRS16, which may be why you did not mention them. But in the context of the EBIT/I banking covenant, I considered those one off charges that any bank would look through, which is why I added those back to the EBIT as well (my post 4).

However with the aim of keeping this discussion 'on topic' to IFRS16, I will forgo those other adjustments and go with your figure of $21m

or 2) adjust the numbers back to the old method

Under method 2 we would adjust EBIT down (not up) for the rent payments (ie the total lease payments outflow from the cashflow statement, less GST = $23/1.15 = $20)

I think you made a typo in that GST calculation (now corrected). When I corrected the GST rate from 12% to 15% it worked, giving the $20m figure you quoted.

and adjust EBIT up by the depreciation component of leased assets (i.e. $14),

OK I see that $14m 'right of use asset' depreciation, listed in the Depreciation & Amortisation break down in AR2020 on p26

and would adjust "I" down for the lease elements only (i.e. $21).

"lease interest of $21m" in the text in AR2020 p27, - gotcha

EBIT: $246 - 20 + 14 = $240
I: $173 - $21 = $152
EBIT/I = 240 / 152 = 1.58

OK this is where you and I have rather different adjusted EBIT calculations.

I hadn't considered depreciation before. Because as we were talking about EBIT, I had considered the 'D'epreciation' part of the EBITDA to EBIT calculation already dealt with. But now I see you are correct Ferg, because the 'depreciation of the lease liability' did not exist under the old accounting rules. So to make a comparison with previous years, you have to increase the EBIT by any depreciation that was not present and in fact did not even exist at all under the old accounting rules.

Under the new accounting rules, what were 'rent payments' (an operating cost) have turned into 'lease interest liabilities' (a finance cost). Under the new rules of EBIT, all financing costs are yet to be subtracted before NPAT is calculated. Under the old accounting rules the equivalent of 'lease interest liabilities' (the old 'rent') would have already come out of the EBIT figure. This means in consideration of interest payments only, that to convert the EBIT you see published in the annual accounts under IFRS16, back to the lower EBIT figure we would have seen under the old accounting rules, we have to subtract the 'lease interest liability' from the EBIT figure quoted. So it looks like you are right again Ferg. (note: Typo on your interest calculation corrected).

What slightly puzzles me is why you have chosen to reduce the interest bill by the 'lease interest liability' ($21m), yet reduced the EBIT figure by an interest figure derived from the cashflow statement ($20m). Shouldn't these two numbers be exactly the same?

*Also, any differences between the P&L and cashflow will be due to accrual accounting.

FERG

You refer to the difference between $21m of "lease interest" (AR2020 p27) and the 'net outflow from leases' of $23m in the cashflow statement (AR2020 p38) ?

SNOOPY

[Snoopy]

Time to eat a bit of 'humble pie' this evening and admit when I did my EBIT/I covenant calculation (post 4 on this thread) - where I adjusted for the effect of any 'lease interest deductions' (that are partially reflective of the old 'rent' charge under the old pre-IFRS16 accounting standard)- I got it wrong.

The 'wrong' calculation I have quoted below.....

Now let's look at the alternative calculation method 2, where 'lease interest' (as part of 'Other Interest') is not taken into account when calculating EBIT.

EBIT now changes to $246m + $27m = $273m

With this iteration we have chosen not to take the lease interest into account as earnings. So instead we must include 'Other interest' as part of the total interest bill due by not subtracting it. This means the 'Total Net Finance Expense' for our purposes is now:

-$12m + $185m - $29m = $144m

So the calculation of EBIT/I for Chorus for FY2020 becomes: $273m/$144m = 1.89 (using method 2)

That isn't grossly different, except there is a 'rule of thumb' that says an EBIT/I ratio above 2 is passable, while anything below that is dodgy. So is 'method 1' or 'method 2' the better way of calculating this ratio? I don't know the answer. My solution is to curse IFRS16 and go to bed.

.....while the changes I wish to introduce are summarized below (thanks to Ferg for pointing this stuff below out to me).

OK this is where you and I have rather different adjusted EBIT calculations.

I hadn't considered depreciation before. Because as we were talking about EBIT, I had considered the 'D'epreciation' part of the EBITDA to EBIT calculation already dealt with. But now I see you are correct Ferg, because the 'depreciation of the lease liability' did not exist under the old accounting rules. So to make a comparison with previous years, you have to increase the EBIT by any depreciation that was not present and in fact did not even exist at all under the old accounting rules.

Under the new accounting rules, what were 'rent payments' (an operating cost) have turned into <added> 'lease liabilities' (an operating cost) and </added> 'lease interest liabilities' (a finance cost). Under the new rules of EBIT, all financing costs are yet to be subtracted before NPAT is calculated. Under the old accounting rules the equivalent of 'lease interest liabilities', (<added> part of </added> the old 'rent') would have already come out of the EBIT figure. This means in consideration of interest payments only, that to convert the EBIT you see published in the annual accounts under IFRS16, back to the lower EBIT figure we would have seen under the old accounting rules, we have to subtract the 'lease interest liability' from the EBIT figure quoted. So it looks like you are right again Ferg.

So let's go through this 'step by step', adjusting EBIT back to what it was in the 'old days' (pre IFRS16 reporting standards), to make sure I don't 'stuff up' again.

Declared EBIT	= $246m	(AR2020 p22)
add 'Right of Use Asset' Depreciation	+ $14m	(AR2020 p26) (Non existent asset under old regime, so not there to depreciate)
subtract 'Rent Paid'	- $20m	(AR2020 p38, $23m exGST) (post IFRS16 EBIT is not adjusted for the old 'rent' concept)
equalsAdjusted EBIT	= $240m	(Calculated)

Previously I had also added back a couple of one off interest charges

add 'Amortisation Loss'	+ $5m	(AR2020 p47) (One off Swap reset reversed: difference between fair value and proceeds realised)
add 'Restructuring Loss'	+ $1m	(AR2020 p47) (One off cost to restructure interest rate swaps reversed)

However, these are interest charges, not directly related to Covid-19. While they do produce a higher normalised profit at NPAT, they do not produce a higher profit at EBIT level. So I think, it is best to leave these changes out of my EBIT adjustment.

The interest calculation is therefore as follows:

Declared Interest Expense	= $185m	(AR2020 p47)
subtract Interest Income	- $12m	(AR2020 p27)
subtract 'Amortisation Loss'	- $5m	(AR2020 p47) (One off Swap reset reversed: difference between fair value and proceeds realised)
subtract 'Restructuring Loss'	- $1m	(AR2020 p47) (One off cost to restructure interest rate swaps reversed)
subtract 'CIP Notional Interest'	- $29m	(AR2020 p47) (This is an artificial accounting construct based on defined CIP construction and is never paid in practice)
equals Adjusted Net Interest Bill	= $138m	(Calculated)

So EBIT/I = $240m/$138m = 1.74

SNOOPY

[Snoopy]

The calculation of EBIT/I for Chorus for FY2020 is: $246m/$117m = 2.10 (using method 1)

The half year results for FY2021 are available. So let's see what EBIT/I did for the six months after EOFY2020. This isn't straightforward, because the disclosure at half year result time is less full than in the whole of year accounts.

EBIT = $114m vs $246m for the FY2020 full year (p5 HYR2021),
Net Finance Expense = $0m - $77m = -$77m vs $12m - $185m = $173m for the FY2020 full year. (p5 HYR2021)

A problem now arises because there is no breakdown of the finance expenses given for the half year. Given the comparative figures are given for the previous full year in the half year report, we have to assume there is no difference in the way the calculations have been made at HY2021 when compared with FY2020. That means if we work out our metric using the raw figures in HYR2021:

EBIT/I = $114m / $77m = 1.48

then we are 'double counting' the effect of 'lease interest payments'. Firstly because they have reduced EBIT in the numerator. Secondly because they have increased I in the denominator. You should do one or the other, but not both. My contention then is that there is insufficient information disclosed by Chorus to allow the calculation of EBIT/I over the half year period.

If we instead focus on the twelve month period comprising 2HY2020 and HY2021, then the calculation changes to this:

EBIT/I = (($246m-$134m) + $114m)/ (($185m - $95m) + $77m) = 1.35

This metric is also wrong, because it suffers from the same 'double counting' problem I have just described. If we look at the published information from third parties:

Morningstar:

https://www.morningstar.com/stocks/xnze/cnu/financials

lists the available EBIT/I for Chorus as 1.38 (close to my 2HY2020 + HY2021 figure in this post)

Simply Wall Street:

https://simplywall.st/stocks/nz/tele...nsidered-risky

lists the available EBIT/I for Chorus as 1.5 (close to my HY2021 figure in this post)

I submit that both of the 'Morningstar' and 'Simply Wall Street' EBIT/I figures are wrong, because they have both 'double counted' lease interest. I can't tell you what the correct figure is because there is insufficient disclosure by Chorus to allow us to work that out. Blame IFRS16 for this mess.

SNOOPY

[winner69]

Snoops - what’s the “Net outflow from leases” in Cash Flow Statement under Financing?

As an aside even Operating Cash Flow doesn’t include whst you and I call rent