Fibre Broadband Shares

[Snoopy]

The Average Revenue Per User (ARPU) broadband trend for NBN and Chorus (as published by the respective companies) is as follows:

	NBN	Chorus
FY2014	$A37
FY2015	$A40
FY2016	$A43
FY2017	$A43
FY2018	$A44
FY2019	$A44	$NZ47.50
FY2020	$A45	$NZ48.12

Having embarrassed myself on the Chorus thread by being unable to calculate the ARPU figures for Chorus, I will now continue my embarrassment by trying to do the same for NBN here,

From NBN AR2020 p133:
Total Broadband Revenue for FY2020 was: $2,979m + $666m = $3,645m

From NBN AR2020 p7:
Average number of 'premises activated' over FY2020 was 0.5x (7.3m + 5.5m) = 6.4m

=> monthly ARPU = (1/12) x $3,645m / 6.4m = $47

That is a bit higher than the $45 figure quoted. But on re-reading the text, that $45 ARPU is for residential connections only (AR2020 p7). If the business rate was higher, that could explain why my 'average calculation' was pushed up. Of course no ARPU figure is quoted for all users including business.

I also see in the fine print that the connection figures include those customers, retail and business, 'ready to connect' as well as those 'actually connected'. If you are looking for a way to obfuscate your calculation methods, NBN is certainly pulling out all stops!

SNOOPY

[Snoopy]

If NBN are relying on ever increasing ARPU to pay off their massive debts, yet technological developments have - to some extent- caught up with functional demand, what is the incentive for customers to pay more and more for higher bitstream rates? And if ARPU does not continue to increase, what does that mean for the longer term capital position of network companies (hint: IMO it isn't looking good)?

Based on the Australian telecommunications network as it existed in 2009, NBN was expecting that some 70% of those existing customers, after the NBN was installed outside their gate, would eventually migrate to NBN. The remainder would likely eventually 'go fixed wireless', (not on a system run by NBN), or not have any fixed line at all. Because NBN is publicly owned, projections of EBITDA well out into the future have been made.

https://www.nbnco.com.au/content/dam...iew-report.pdf

According to graph 2-3 on page 39 of the above reference, projected EBITDA from FY2021 onwards are as follows (I have interpolated the numbers from the graph).

Financial Year	Projected EBITDA	Actual EBITDA	Projected CAPEX	Actual CAPEX
2019	-$1,200m	$600m-$1,900m= -$1,300m	-$7,000m	-$5,905m
2020	-$600m	$1,800m-$2,400m= -$600m	-$7,000m	-$5,038m
Final Network Build Cost EOFY2020				-$42,512m
2021	$300m		-$6,600m
2022	$1,100m		-$6,400m
2023	$1,900m		-$6,000m
2024	$3,000m		-$4,000m
Final Network Build Cost EOFY2024			-$55,900m
2025	$4,400m		-$800m
2026	$4,800m		-$1,200m
2027	$5,200m		-$2,000m
2028	$5,400m		-$2,200m

Notes

1/ Network build costs and the sum of capital expenditure from FY2010 and do not include associated interest borrowing costs.



The above figures are from the revised outlook where completion of the core network was not envisaged until 2024.
Considering these projections were made back in 2013, the EBITDA figures are surprisingly well on track.

SNOOPY

[Snoopy]

The 'Return on Equity' for 'Enable' will certainly improve going forwards, as more connections are made to the existing fibre network. It will also improve as the overall value of the network depreciates and the accompanying ongoing capital expenditure reduces. At EOFY2020 (30th June 2020), the customer connection rate at 'Enable' had risen to 63%. That compares favourably to 'Chorus' which, at the same date, had a fibre network connection rate of 60%.

Enable are set to catch up with Chorus in offering 'hyperfibre' to the residents and businesses of Christchurch. From p14 of the current edition of the 'Christchurch Star'.

https://www.yumpu.com/en/document/re...r-july-22-2021

"Next month Enable is launching a range of hyperfibre products including 2Gbps, 4Gbps and 8Gbps, all with symmetrical download and upload speeds."

"Most households would still fall short of using 2Gbps of bandwidth even if every family member is mainlining 4k Netflix at the same time, while on-line gaming and watching a Zoom call in the background."

"But Enable (like Chorus and other fibre operators) is in an arms race against Spark and Vodaphone in their push to promote fixed-wireless broadband, which uses a mobile network to deliver fast internet to home or small business - and the fibre companies want to keep bragging rights about offering the fastest service."

"Spark and Vodaphone argue their cheapest unlimited data broadband plans start from just $50 a month with same day install."

"5G rollouts are allowing increasing amounts of bandwidth that, depending on location, can handily outpace entry level 100Mbps UFB fibre plans."

I understand Enable and the others need to compete on speed, because ultimately you can't compete on portability with a mobile network. I understand medium to large business may benefit from Hyperfibre. However, I would suggest that Mom and Dad users may be perfectly well served by fixed mobile broadband, especially with 5G fixed mobile technology coming on stream. I guess there is always the status symbol of having the fastest broadband, just like some prefer to own the fastest car, (notwithstanding the fact that NZ roads won't let you exploit it). But while others can see you have the fastest car, no-one is going to know that you have the fastest broadband. So where is the status in that?

SNOOPY

[Snoopy]

Does Vital fit into this group somewhere?

Hi Lola,

You have piqued my interest in Vital now, and in particular their 'wired network' business in Wellington. I see back before the nationwide fibre roll out, Teamtalk, as Vital was called then, put in a bid with Fulton Hogan to build fibre broadband throughout Wellington. Ultimately that partnership didn't get the gig and Chorus did the job. Then Spark had a go at taking out Teamtalk back in 2017. But they walked away after the 'Vital' Farmside rural division got divested to Vodaphone. So I guess Spark didn't see the value in taking over what was left?

There were some sobering words on Vital's Wellington 'wired network' in the Vital Chief Executives address at the last AGM:

"There was a clear gap on what we could offer verses what customers could get elsewhere with UFB. Our previous network design limited us as to what we could offer in terms of different types of services and associated pricing plans. Our services could not be compared to UFB offerings, they were too expensive and had limited capability. being only offered at 100Mbps , 1Gbps and some limited 10Gbps ports. I am pleased to announce that we now have our new products and pricing plans for our wholesale partners ready to launch, following board approval earlier today. We have moved to meet the market and we now have 74 different wholesale products compared to our legacy 14. Our old pricing and products were a major impediment to new business."

That all sounds very well, but it still looks like Teamtalk are only just ahead of the Chorus freight train, with Chorus rolling out 8Gbps hyperfibre. So Vital have outlaid heaps of capital to upgrade their Wellington network, but will now be charging less money for their services, the services that haven't been usurped by Chorus, than before. Have I got that right?

SNOOPY

[Snoopy]

There were some words on Vital's Wellington 'wired network' in the Vital Chief Executives address at the last AGM:

"We can now offer services utilising our new PE nodes with MPLS as the underlying technology which will provide greater scale and flexibility. Our services will be MEF compliant, UFB comparable along with faster delivery times and priced to compete."

Now I don't mind admitting I have no idea what the CE has just said in that sentence. So using Google to take that sentence apart 'byte by byte'....

1/ PE stands for 'Provider Edge' that I take to mean an exit gateway to the fixed Wellington network. A 'PE router' is a router between one network service provider's area and areas administered by other network providers. So the 'PE node' is where Vital's wholesale customers connect.

2/ MPLS stands for 'Multi Protocol Label Switching'. MPLS is a routing technique in telecommunications networks that directs data from one node to the next based on assigned short path labels rather than long network addresses, thus avoiding complex 'look ups' in a routing table and speeding traffic flows. MPLS can be used to deliver many different categories of traffic.

3/ MEF stands for 'Metro Ethernet Forum', a global alliance of more than 220 organizations including telecommunications service providers, cable TV Multi System Operators, network equipment/software manufacturers, semiconductors vendors and testing organizations. Vital are just embracing common industry standards by saying this.

4/ UFB Comparable means Ultra Fast Broadband comparable which is an interesting thing to say because it implies even the new subterranean Vital city network might not be using the current best practice 'dark fibre' that Chorus have laid as competition?

Phew! I don't see any enduring competitive advantage with any of this though. Is there anything in the above that Chorus cannot replicate?

SNOOPY

[Snoopy]

I have always been a bit debt averse with my investments. That's why I embrace the concept of figuring out how long a company would take to repay all its debts if it diverted all of its profits toward that goal (I have coined the term MDRT or 'Minimum Debt Repayment Time' for this). I am aware that 'certainty of cashflow' means it is fiscally efficient to leverage up a company's balance sheet. Yet, I regard an MDRT figure above 10 years as a sign of a heavily indebted company. Chorus, is on an MDRT figure of 44.7 years. I find that eye watering. I tell myself that extreme certainty of cashflow means extreme levels of debt are sustainable. A counterpoint view to this is that Enable (MDRT of 26 years) have stated that FY2020 is their year of 'peak debt' and they will target debt reduction from here on in. Call me too conservative if you like, and maybe it is my mindset that is wrong. But that debt position at Chorus worries me.

To answer my question first, it is how much cashflow that you need to service your debt that matters. The way that the banks answer this question is to look at:

'Net Debt' / EBITDA

This is a measure of a company's ability to deal with the absolute quantum of its debt. Chorus's banking syndicate says that they want this figure to be below 4.75. So let's take four companies that own fibre broadband circuits and see how this banking covenant stacks up for the FY2020 financial year. The companies are spread across the page in order of descending size.

Broadband Network Company	National Broadband Network (Aus)	Chorus	Enable	Vital
Total Net Debt	$20,619m	$2,680m	$278.426m	$17.619m
EBITDA	$1,789m	$648m	$52.585m	$8.136m
'Total Net Debt'/EBITDA	11.5	4.14	5.29	2.17

Notes

1/ Total net debt at NBN: $19,548m +1,000m + $415m - $344m = $20,619m
2/ Total net debt at Chorus: $2,234m + $183m + $263m = $2,680m.
3/ Total net debt at Enable: $294.4m - $10m -$5.974m = $278.426m.
4/ Total net debt at Vital: $14.000m + $5.477m - $1.858m = $17.619m
5/ EBITDA for NBN is: -$5,239m (NPAT) + $1,460m (finance costs) + $3,154m (D&A) + $2,414m (Subscriber Costs) = $1,789m. 'Subscriber Costs' are a special payment made to incumbent telcos to turn off their legacy networks once the new national broadband network is in place. Subscriber costs are one off payments.
6/ EBITDA for Vital is: $3.515m +$4.621m = $8.136m


The high depreciation costs have largely turned around the insufficient profit worries that I had. Depreciation is not a cash item and that makes the resultant incremental cashflow sufficient to deal with the heavy debt burden of all four companies. I had thought that Chorus was particularly heavily indebted, butting up close to the banking syndicate standard of 4.75 . However, it is clear that neither Enable nor NBN would even be close to meeting this banking standard. Both are fully government owned, so I guess the safety net of a government bail out has made the difference to allowing them to continue in business under their current financial structures. Chorus claim that debt has peaked at EOFY2020, even if the HY2021 result suggests debt is not yet on target to reduce. Long time readers of this thread will know that the Chorus CIP preference shares and debt, which is part of the above Chorus total debt, has been discounted in accordance with accounting standards too. Curiously, the preference shares on the books at Enable are allowed to be counted as equity in contrast to what has happened at Chorus. So I would argue the net debt to EBITDA ratio for both Enable and Chorus, as I have calculated above, are lower than they could be than under alternative legal accounting presentations.

Now I know that Enable and NBN are both less able to service their debts than Chorus, does this make me a bit more warm and fuzzy about Chorus's debt position? Not really! Maybe I need to go into one of those military run 'accounting camps' to be re-educated?

SNOOPY

[Snoopy]

Interest Cover = EBIT / 'Net Interest'

This is a measure of the importance of 'earned by work' company income, in comparison with that company's "lender's interest income" (an outgoing). It represents how many times a company can pay its bank interest obligations using its earnings. Chorus's banking syndicate have previously said that they want this figure to be greater than 2.75: From the 25th May 2017 Chorus announcement:

"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."

I have looked at all subsequent renewal of banking facilities announcements for Chorus. There is no mention of the interest coverage requirement being relaxed further.

According to investopedia:
"Generally, an interest coverage ratio of at least "2" is considered the minimum acceptable goal for a company that has solid, consistent revenues."

So let's take four companies that own fibre broadband infrastructure (note that the other half of Vital Ltd. is a nationwide two way radio service network), and see how this banking covenant stacks up for the FY2020 financial year. The four companies in the table below are spread across the page in order of descending size.

Broadband Network Company	National Broadband Network (Aus)	Chorus	Enable	Vital
EBIT	-$1,364m (1)	$246m	$28.399m	$3.515m
Net Interest Bill	$631m (2)	$117m (3)	$13.382m (4)	$1.001m (5)
EBIT / 'Net Interest'	NM	2.10	2.12	3.51

Notes

1/ EBIT for NBN is: -$5,238m (NPBT) + $1,460m (finance costs) + $2,414m (Subscriber Costs) = -$1,364m. 'Subscriber Costs' are a special payment made to incumbent telcos to turn off their legacy networks once the new national broadband network is in place. Subscriber costs are 'one off' payments that will cease when the initial nationwide roll out of NBN is finished.
2/ Total net interest at NBN: -$1,460m + $829m = -$631m
3/ Total net interest at Chorus: +$12m - $185m + $29m + $27m = -$117m (Ref AR2020 p47)
4/ Total net interest at Enable: +$0.081m - $13.499m + $0.036m = -$13.382m
5/ Total net interest at Vital: +$0.100m - $2.593m + $1.492m = -$1.001m

In all 'net interest' calculations, I have added back any 'lease interest'. This is because 'lease interest' has already been dealt with, in that it has reduced EBIT - lease interest is an expense - (equivalent to how rent was treated in pre-IFRS16 days). If I were to leave 'lease interest' as part of the overall interest bill as well, then that would be double counting its effect. Also 'lease interest' is generally owed to the owner of leased property, not a bank.

So there are the raw numbers for how these four protagonists stack up together. But what does it all mean?

SNOOPY

[Snoopy]

So there are the raw numbers for how these four protagonists stack up together. But what does it all mean?

NBN

NBN looks like an 'outlier case'. Given the extreme debt position at NBN, I was rather surprised to read in the annual report (NBN AR2020 p11)

"In order to lend additional support to our future growth ambitions, NBN Co this year completed its inaugural borrowing from private debt markets with the addition of credit facilities totalling $6.1 billion."

Looking at the EBIT/I statistic, some would say this is showing that NBN is insolvent (the negative EBIT figure, in the context of the the ceiling of government loan support being reached). It is only the depreciation and amortisation (not incorporated within the EBIT/I statistic) from what are 'long lived assets' that will provide any hope of positive cashflow for NBN in the near term. Just as well NBN is fully Australian Federal Government owned then! It would be interesting to know what "group of banks" has put up the new $6.1billion loan facility. To me, they look like fools. And it would be good to know which Australian banks an investor should not buy shares in! I note from p57 of the annual report that the NBN company has not yet drawn down on these facilities. So perhaps it is just a 'look friendly' to the government measure 'at this stage' for the banks?

Chorus

Interest coverage at Chorus and Enable meets the Investopedia standard of 2. But those figures of 2.10 and 2,12 respectively are close enough to that '2' target to demand scrutiny of the respective business models going forwards. All EBIT/I statistics are snapshots of a past time period. So what is really important, even if the EBIT/I figures aren't great (like Chorus and Enable at EOFY2020), is that these statistics keep getting better.

My post 2735 on the Chorus thread predicts interest rate reductions should lead to $7.5m in annual interest rate savings for Chorus from FY2022 onwards in the denominator. Yet EBIT has already fallen to $114m in HY2021 from $134m in HY2020, a drop of 15% (or 13% if we leave out $3m of proceeds from a HY2020 legal settlement). Annualising those changes I get:

EBIT/I = ($114m x2)/($117m-$7.5m) = 2.08

That is still OK, albeit moving in the wrong direction. But there are no more interest rate savings on the horizon until 2028. And there is no guarantee that the downward trend in EBIT over the last five years ($267m, $313m, $266m, $243m, $246m) will not continue. I would describe the Interest Coverage Ratio position at Chorus to be 'tight'.

Enable

Enable, fully owned by the Christchurch City Council, doesn't provide the same amount of continuous disclosure as a publicly listed company. We do know that as at 30th June 2021, they were providing broadband to 132,000 homes and businesses, up from 117,690 on 30th June 2020. Last year I estimate the ARPU to be:

$67,729,000 / (0.5x(101,271 + 117,690)) = $619 per year (Or $51.55 per user per month).

If we assume the average number of new customers over a year for FY2021 was:

0.5 x (132,000 - 117,690) = 7155

Then that equates to incremental revenue of: $51.55 x 7155 = $370,000 over FY2021

With no legacy network to wind down, like Chorus has, and with the underlying infrastructure complete (unlike Chorus), incremental revenue gains should flow straight through to profits. Funding for Enable is via the business holding company of the council, 'Christchurch City Holdings Limited'. Interest is charged at Christchurch City Council borrowing rates, plus 1%. The most recent local government funding authority presentation:

https://www.lgfa.co.nz/sites/lgfa.co...esentation.pdf

shows eight year bond rates at 2%. So if the Christchurch City Council's financing committee has been prudent, borrowing rates at Enable should be locked in below 3% until 2028. Low borrowing rates and increasing EBIT should see the 'Interest Coverage Ratio' at Enable rise over time. And that is a good thing.

Vital

Finally Vital is seeing a revenue squeeze as they readjust their pricing (to match the likes of Chorus) . Vital is coming off a period of heavy infrastructure investment and the jury is still out on whether profit margins will recover to the level of even three years ago. However, the EBIT/I 'Interest coverage ratio' is still the most favourable of the four protagonists, even under post Covid-19 depressed profitability. Debt is funded by the BNZ and is relatively short dated (as at 30th June 2020 the facility would expire in December 2022). So Vital is far more exposed to market rises in interest rates compared to the others. Averaging the outstanding loan balance over the year, the average interest rate paid by Vital over FY2020 was:

$1.001m / (0.5x($14.000m + $12.500m)) = 7.6%

That seems very high for an infrastructure company, even a small one. Furthermore the short dated nature of the funding would suggest the interest rate payable may increase, ahead of the other three. While meritorious that the interest coverage ratio at Vital is the best of this comparative group, the less certain revenue path forwards combined with the higher risk of interest rate rises probably demands a superior 'interest coverage ratio'.

SNOOPY

[Waltzing]

your a machine

[Marilyn Munroe]

Some are of the opinion 5G wireless broadband is a fiber killer.

Here is a link to a YouTube video which examines the 5G rollout in China. It suggests the 5G reality is less than the hype.

https://www.youtube.com/watch?v=4tE24j5WFw4

Boop boop de do
Marilyn