CNU - Chorus

[Snoopy]

Note the value of $707m for UFB also includes UFB2 in the past 2 years per the notes. I too am curious why the CIP debt does not equal CIP equity - a quick search of comprehensive income and reserves did not reveal anything obvious.

The above $707m number is pulled from Note 7 on 'Crown Funding' in AR2020. I think we are agreed that despite the title, this section on 'Crown Funding' does not contain any CIP (Crown Infrastructure Partners) money on loan to Chorus. The CIP funding leaks across into 'Crown Funding' and ultimately returns again to CIP funding. Yet the presented picture is muddled by the UFB1 and UFB2/UFB2 roll outs, and separate CIP funding for both (with different conditions) overlapping.

Consequently I thought it might be worth going back to FY2012, which was the very first year of CIP (or CFH as it was called then) funding for the second call of such funding taken on board by Chorus. Under Note 24 on "Post Balance date Events' we learn:

"Chorus issued a call notice on 17th August 2012 to CFH with an aggregate issue price of $13m. The component of the cash received will be allocated as follows. CFH debt securities $2m, CFH Equity Securities $1m and Crown Funding $10m."

Now we know that 'CIP debt' and 'CIP equity' (Preference Shares) are issued in equal capital amounts. So why even at this very early stage of UFB roll out are they so badly out of whack on relative value? Before I answer that question, I need to address why so much of the cash drawn down has gone straight to Crown Funding.

Back in 2012, interest rates were 5 percentage points higher than today. So I am going to take the implicit averaged discount rate you calculated for CIP funding of 5.88% and add another 5% to that making a total of 10.88%. The projected weighted average payback time from a 2012 perspective was 20 years. This means that $1,000 of interest free loan paid back in 20 years forward from the FY2012 draw down date has a present value today of:

$1,000 / 1.1088^20 = $127

I think that means that for each $1,000 drawn down from the CFH lending regime, $127 remains as capital and $873 (87.3%) becomes a 'notional interest' benefit that is transferred to 'crown funding' (for now). Thus the actual ratio of money transferred to 'crown funding' of $10m/$13m = 76.9% using the numbers referred to in my quoted text from AR2012 therefore comes as no surprise.

Returning to the question of the CFH debt being valued at only half that of the CFH equity. Despite the nominal dollar values of each being the same 'at creation', the equity portion carries a heavy associated payment burden reaching a peak of probably 7% of $959m = $67m per year (see my post 2646). Even though that 'annual bill' doesn't reach its peak until 2036 (24 years after this study point), the present day value of all those future interest bills must start to add up.

This is my explanation of why 'CFH Debt' and 'CFH Equity' are issued in a ratio of 1:1 but valued in a ratio of 1:2 . At least that is how it was back in FY2012!

SNOOPY

[Snoopy]

My turn for a couple of (hopefully easy) questions.

I'm not currently an investor in CNU. Is this another dividend yield stock? Or is there room for some sort of growth over and above inflation?

I am approaching Chorus as a 'yield share', with any growth that does amount being a bonus. Chorus is in a perplexing position in certain areas. If customers convert to fibre in those areas where someone else has built the fibre network other than Chorus, then Chorus lose their existing copper broadband customers entirely. Thus growth in fibre broadband on a national basis has some negative aspects for Chorus.

Where fibre has been laid, copper broadband comes out of price controls from 01/01/2022. So we might see Chorus dropping their copper broadband prices in Christchurch, for example, in an effort to slow down the loss of copper broadband customers to Enable's fibre network. Who knows, they may even win some Enable fibre customers back onto Chorus copper, as Chorus sweat some more value from their old school copper network! Looking further out, the ability to bring small niche retailers onto the Chorus network in a cost effective manner may open up new business opportunities that even Chorus cannot imagine. The ability to offer short term broadband, without using a mobile network, to a festival or something, at a cost effective price is coming. And there do seem to be potential applications for the new WiFi 6 standard devices, used in conjunction with fibre. These are three growth initiatives for the future. But how successful or significant they will be, no-one knows at this point.

If you look at my post 2658, I am picking a fair value for Chorus of $8.36 by 2023, using a capitalised future dividend approach. i am working on a 35cps annual dividend - my broker says 33cps. But to be fair, that is ignoring the very large debt repayment stream that will begin from 2025. In my post 2648, my corrected calculation shows you may need to take 67c off that target price for proposed debt repayments, which gives a fair value of $7.69. The share closed at $6.55 on Friday, some 15% below my $7.69 fair value figure. While not a supreme bargain, I would say that this is discounted enough for potential new shareholders to at least take a closer look.

SNOOPY

[Aaron]

Many thanks Snoopy & Ferg

I have skim read your notes but will have to come back to properly understand them.

Accounting standards seem to just make things harder to understand for the lay person. I guess the rules are required as CEOs try to deceive the investing public but they don't seem to help.

Why would you put a notional interest portion on an interest free loan?

Would it be easier to just expense marketing costs as you pay them? zspoon mentions they are trying to match expenses with income but is this any more accurate with estimates regarding the effectiveness of the advertising campaigns and how to spread the expense to match the income being very arbitrary. Something for the auditors to argue over I suppose to generate good fees.

Again Snoopy and Ferg appreciate the time taken to explain what is being presented in the annual report.

[Snoopy]

It is my understanding that Chorus is obliged to maintain their copper network for any customer who does not have access to fibre, i.e. as long as the fibre rollout is not completed there will be parts of the copper network Chorus must maintain - and I suppose that these remaining parts of the copper network will as well generate revenue.

I assume as well that any price control is only taken off from copper for customers who have the option to go fibre ...

From slide 25 of the November 17th presentation, reinforcing BlackPeter's points.

Copper –where fibre is not available:

1/ Copper remains regulated and TSO(*) applies
2/ Copper pricing capped at 2019 levels with CPI adjustments
3/ Commission required to review pricing framework no later than 31 December 2025

(*) TSO means 'Telecommunications Services Obligation': A universal service obligation under which Chorus must maintain certain coverage and service on the copper network.

SNOOPY

[Snoopy]

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

PH "Andrew, I knew you would come around to my way of thinking. Must get back to the office. I am working on a new standard I am calling IFRS 16. We are really going to muck things around by dragging all leases onto every company's balance sheet!"

AC "Hah hah hah hah! Pull the other one Pointy Head! You will never get away with that!

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

As history has subsequently revealed, the 'Pointy Heads' got their way with IFRS 16.

Here is what Chorus said about it in AR2017 p34:

"NZ IFRS 16 introduces a single lessee accounting model and requires a lessee to recognise assets and liabilities for all leases with a term of more than twelve months, unless the underlying asset is of low value. Accounting by lessors is unchanged under IFRS 16. A lessor continues to classify its leases as operating leases or finance leases, and to account for those types of leases differently. Management are in the process of finalizing the assets subject to lease agreements and the impact of these on the balance sheet and income statement. Management's process to date highlights the that the impact is expected to be material (greater than $20m), given Chorus's asset intensive nature. The agreements identified to date relate to poles, buildings, easements and IT equipment."

One year later, in AR2018 p37 we get this:

"Chorus has early adopted NZ IFRS 16 with a date of initial application of 1 July 2017, and has not restated comparative information.."

The 'Statement of Changes in Equity' (AR2018 p33) confirmed the $20m equity 'hit' predicted in AR2017 (at least it wasn't greater than $20m).

What is of particular interest to me is the effect of IFRS 16 on EBITDA, because that impacts banking covenants. While some effort has been made to show the effect of IFRS on assets (AR2018 Note 2 p40 $226m carrying value of 'Right of Use Assets') and liabilities (AR2018 Note 5 p45, An incremental $84m of 'Lease Liabilities'), I can't find any reported quantifiable analysis on the effect that adopting IFRS 16 has had on earnings,

I know that IFRS 16 introduces new interest costs to the business in the form of a 'lease interest' expense. This will boost EBITDA for a given NPAT as the interest total 'I' to be added back in increases. In the case of Chorus, lease interest over FY2018 totalled $18m (AR2018 p45).

Chorus have declared that EBITDA from FY2017 to FY2018 rose from $652m to $653m, (AR2018 p18) a rise of $1m. However if we adjust for IFRS 16 then EBITDA drops from $652m to $635m, a sizeable drop of $17m.

Slide 32 of the 17th November 2020 presentation shows the 'Net Senior Debt' / EBITDA ratio should be less than 4.75 times.

Yet, before the early adoption of IFRS 16 was announced on 1st July 2017, Chorus and their bankers had already announced revised banking syndicate arrangements on 25th May 2017.

"The facility has also been repriced to reflect current market rates and the covenants have been revised from 4.0 to 4.75 times debt to EBITDA and 3.0 to 2.75 times interest coverage, to better align with Chorus’ rating thresholds."

This means that within an historical period of two months, Chorus has let shareholders know that the amount of debt their company can hold can increase, in two jumps. Why the banks regard Chorus as suddenly more trustworthy in holding debt I am not sure. I guess the banks are just saying that as a 'big infrastructure company' with a 'good credit rating', we should have given you this much leash before?

SNOOPY

[Snoopy]

What is of particular interest to me is the effect of IFRS 16 on EBITDA, because that impacts banking covenants.

I know that IFRS 16 introduces new interest costs to the business in the form of a 'lease interest' expense. This will boost EBITDA for a given NPAT as the interest total 'I' to be added back in increases. In the case of Chorus, lease interest over FY2018 totalled $18m (AR2018 p45).

Chorus have declared that EBITDA from FY2017 to FY2018 rose from $652m to $653m, (AR2018 p18) a rise of $1m. However if we adjust for IFRS 16 then EBITDA drops from $652m to $635m, a sizeable drop of $17m.

Slide 32 of the 17th November 2020 presentation shows the 'Net Senior Debt' / EBITDA ratio should be less than 4.75 times.

Yet, before the early adoption of IFRS 16 was announced on 1st July 2017, Chorus and their bankers had already announced revised banking syndicate arrangements on 25th May 2017.

"The facility has also been repriced to reflect current market rates and the covenants have been revised from 4.0 to 4.75 times debt to EBITDA and 3.0 to 2.75 times interest coverage, to better align with Chorus’ rating thresholds."

Moving on to the numerator of banking covenant, I imagined it would be a straightforward process to simply pluck the relevant figures out of the balance sheet. However, after I had done this, I noticed Chorus themselves had done the calculation on the same 30th June 2020 balance date, and come up with a significantly different answer (Slide 41, November 17th 2020 Presentation).

	Net Debt Nov 2020 Presentation (Chorus)	Net Debt FY2020 Balance Sheet (Snoopy)
Borrowings	$2,234m	$2,327m
add PV of CIP debt securities (senior)	$183m	$461m
add Net Leases Payable	$263m	$263m
less Cash	$0m	$0m
equals Total Net Debt	$2,680m	$3,051m

The differences in the two calculations are outlined below:

Borrowings

Chorus Interpretation: (Refer Slide 42, November 2020 Presentation):

$30m (Long term bank facilities) + $5m (Overdraft) + $400m (NZ Bond) + $500m (NZ Bond) + $1,299m (European Medium Term Notes) = $2,234m (Total)

Snoopy Interpretation: (Refer Note 4 AR2020 and Balance Sheet for Overdraft):

$30m (Syndicated banking facilities) + $5m (Current Liability Overdraft)+ $400m (NZ Bond) + $500m (NZ Bond) +$1,410m (EMTN) -$18m (facility fees) = $2,327m (Total)

The EMTN Total by Chorus of $1,299m is described as "$NZ equivalent at hedged rates". Why would the figures in the balance sheet not be at $NZ equivalent hedged rates"?

Why do Chorus not count the 'facility fee', (an annual payment on account of their banking facilities) as part of their banking liabilities?

You can take it from those questions that I am backing my own interpretation of debt from the balance sheet, over that espoused by Chorus.


CIP Debt Facilities

Chorus have used the face value of the debt facilities issued (Slide 42, November 2020 Presentation) and used an 8.5% annual discount factor (AR2020 p50) to get their 'present value' of CIP debt:

$85m / 1.085^5 + $86m / 1.085^10 + $128m / 1.085^13 + $163m / 1.085^16 = $183m

The undiscounted value of that CIP debt is: $85m+$86m+$128m+$163m = $462m, equal to .the total value of the UFB1 debt. (Slide 42, November 2020 presentation), However, this may be a co-incidence, as why would the UFB2/UFB2+ funding be left out of the Chorus debt picture?

The total of $461m on the balance sheet represents both 'CIP equity' and 'CIP debt'. The 'CIP equity' is actually preference shares, which in my view are more closely classed as a form of debt (it is listed as a liability in the balance sheet after all). Nevertheless, this is a grey area and Chorus is not wrong to leave out 'CIP equity' as part of their debt load. But I would class the Chorus view as a favourable interpretation of the company's debt situation.

SNOOPY

[Snoopy]

Moving on to the numerator of banking covenant,

	Net Debt Nov 2020 Presentation (Chorus)	Net Debt FY2020 Balance Sheet (Snoopy)
Borrowings	$2,234m	$2,327m
add PV of CIP debt securities (senior)	$183m	$461m
add Net Leases Payable	$263m	$263m
less Cash	$0m	$0m
equals Total Net Debt	$2,680m	$3,051m

	Chorus View	Snoopy View
Total Net Debt	$2,680m	$3,051m
divided by EBITDA	$648m	$648m
equals 'Net Selected Debt' / EBITDA	4.14	4.71

The number the banks are on the lookout for is anything over 4.75. Chorus would have you believe that they are doing really well. Anything under 4.2 ( Moody's ) and 4.25 ( S&P ) are the trigger ratings for a credit upgrade. I, however, consider that Chorus are within a whisker of breaking this banking covenant. Chorus won't go broke though. But it is very possible that in the future at least some preference shares could be converted to Chorus ordinary shares. That would be eps dilutive for existing shareholders, and not welcome!

SNOOPY

[Snoopy]

"The facility has also been repriced to reflect current market rates and the covenants have been revised from 3.0 to 2.75 times interest coverage, to better align with Chorus’ rating thresholds."

	Snoopy View
EBIT	$246m
divided by Net Interest Expense	$114m
equals 'Interest Coverage'	2.2

Notes

1/ For the interest bill I am using in this calculation (Ref AR2020 p47):

($185m -$12m) - $29m - $27m - $3m = $114m

I have removed from the net interest bill:

(i) $29m of CIP securities 'notional interest'
(ii) $21m of 'lease expense' interest and $5m amortisation from a swaps reset and $1m to restructure interest rate swaps.
(iii) $3m from the ineffective portion of a cashflow hedge.

This calc is telling me the 'Interest Coverage Ratio' is not greater than 2.75 and so is broken. Not good news!

SNOOPY

[Ferg]

The EMTN Total by Chorus of $1,299m is described as "$NZ equivalent at hedged rates". Why would the figures in the balance sheet not be at $NZ equivalent hedged rates"?

My guess is this is either CNU and the bankers giving the finger to IFRS and/or the debt covenants use a specific definition of debt that is not in accordance with IFRS. CNU give a reconciliation of the $1,299m (being $514m+$785m) and the $1,410m (being $527m+$883m) on the bottom of page 46 AR2020. My understanding (and I am not expert in this area) is that as interest rates change, then the fair value of the debt changes for IFRS accounting. However, in my experience with senior debt and covenants, it is unlikely long term covenant calculations of debt will bounce around based on short term fluctuations in interest rates, hence the differing values.

Why do Chorus not count the 'facility fee', (an annual payment on account of their banking facilities) as part of their banking liabilities?

Are you referring to the deduction of $18m per your calculation? If so, it is correct to not include that deduction as part of debt, as Chorus has done (in other words they are correct). Again that is an IFRS construct of spreading the cost of raising the debt over the term of loan and should be thought of as a prepayment. It doesn't reduce the debt obligation, rather it is an item on the Balance where the money has already been spent but it is to be expensed.

You can take it from those questions that I am backing my own interpretation of debt from the balance sheet, over that espoused by Chorus.

I respectfully caution against such an approach given debt covenants, especially of this magnitude, are set in stone after what are usually protracted and bloody negotiations between lawyers and analysts for both parties.

Chorus have used the face value of the debt facilities issued (Slide 42, November 2020 Presentation) and used an 8.5% annual discount factor (AR2020 p50) to get their 'present value' of CIP debt:

$85m / 1.085^15 + $86m / 1.085^10 + $128m / 1.085^13 + $163m / 1.085^16 = $183m

The undiscounted value of that CIP debt is: $85m+$86m+$128m+$163m = $462m, equal to .the total value of the UFB1 debt. (Slide 42, November 2020 presentation), However, this may be a co-incidence, as why would the UFB2/UFB2+ funding be left out of the Chorus debt picture?

Ah the penny drops. Previously I mentioned senior and subordinated CIP debt was muddying the waters but I think you have unpicked it. I believe UFB1 is the senior debt, UFB2 is the subordinated debt. The difference between the raw values you have above and those per the PV calculation we did earlier (being $85.3m + $104.7m + $166.7m + $210.2m) will be for UFB2. These will be the subordinated portions of the CIP debt (as opposed to senior).

Covenants usually only apply to senior debts and/or debts held by the senior banks. The subordinated debts are either lower ranking or held by banks that are not party to the contracted covenant, hence the reason they are excluded from the covenant calculations. Two words missing from the term "covenant calculation" will be "senior debt covenant calculation". Keep in mind senior refers to debts that are ranked higher than others, and sometimes banks who view themselves as more senior to others, i.e. usually local banks versus overseas banks (historically from Scotland).

The total of $461m on the balance sheet represents both 'CIP equity' and 'CIP debt'. The 'CIP Equity' is actually preference shares, which in my view are more closely classed as a form of debt (it is listed as a liability in the balance sheet after all). Nevertheless, this is a grey area and Chorus is not wrong to leave out 'CIP equity' as part of their debt load. But I would class the Chorus view as a favourable interpretation of the company's debt situation.

SNOOPY

I would not include the equity component, despite being classified under term liabilities for 2 reasons:

• it is debt that will revert to equity, so it is equity but just not right now, and
• more importantly, covenant calculations are very strict and not usually open to interpretation.

I cannot stress that last point enough - all I can point to is my experience in this area with some large NZ organisations with similar amounts of debt. I highly recommend you do not use your own version of debt for these calculations. Not only should you trust their process, but the bankers will be scrutinising these covenants every quarter. There is no (or very little) room for "favourable interpretations". I suggest you run with their values because the difference will be legitimate and the calculations are set in stone.

Edit: I see you have another post on senior debt coverage etc and debt ratings. I don't disagree with what you are doing there, the part I disagree with is recalculating their senior debt covenants.

[Snoopy]

My guess is this is either CNU and the bankers giving the finger to IFRS and/or the debt covenants use a specific definition of debt that is not in accordance with IFRS. CNU give a reconciliation of the $1,299m (being $514m+$785m) and the $1,410m (being $527m+$883m) on the bottom of page 46 AR2020. My understanding (and I am not expert in this area) is that as interest rates change, then the fair value of the debt changes for IFRS accounting. However, in my experience with senior debt and covenants, it is unlikely long term covenant calculations of debt will bounce around based on short term fluctuations in interest rates, hence the differing values.

Talking about the EMTN notes, my reading of what happens to the valuation of any debt on CNUs books is that the IFRS valuation of overseas denominated overseas debt is changing with the exchange rate, If the debt valuations changed with prevailing interest rates as you allege, then you would expect the New Zealand denominated debt would change in value from year to year as well. But if you look at the book valuation of the Fixed rate NZD bonds (AR2020 p46,p47) you will see that there is no change in the value of these from year to year.

Returning to EMTN and the bottom of page 59 (AR2020) we learn:

"Chorus designated the EMTN and cross currency interest rate swaps into three part hedging relationships for each issue: A fair value hedge of EUR benchmark interest rates, a cash flow hedge of margin and a cash flow hedge on the principal exchange."

Moving back to the bottom of page 46, I interpret the "impact of fair value hedge" of $5m and $12m to relate to the relative change of interest rates between what happened in NZ and Europe over the year. The 'cash flow hedge of margin' I presume must refer to the margin between NZ interest rates and EMTN interest rates at the point the loan was entered into? The annual payment to cover that difference when translated into NZ dollars would go up and down. To stabilize that you would have to hedge the exchange rate variations in that annual payment. Then once the loan finally expires, Chorus would have to pay back a European currency loan in New Zealand dollars. The estimate of that would change annually, requiring another hedge, the $8m and $86m on the bottom of page 46.

Having said all of that, the fact that these changes are not taken into the balance sheet suggests to me that by the time the loan has run its course they will all unwind and reverse. Or as you put in Ferg, the long term covenant calculations of debt do not change.

SNOOPY