Borrowings and Potential Interest Bill Savings (FY2020 perspective)
Quote:
Originally Posted by
Snoopy
Borrowings
Chorus Interpretation: (Refer Side 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)
I have previously talked about the maturity and reissuing of bonds as an interest saving measure. But there is another way of looking at what has happened. You could think of this bond refinancing - at a much lower interest rate- as freeing up more shareholder cash that could fund increased borrowing. Why would Chorus suddenly want to borrow a whole lot more money, especially as they are on the cusp of completing their UFB build program? The answer is, they already have borrowed more, in the form of interest free 'CIP debt' and dividend deferred 'CIP preference shares', for each of which, a brutal repayment (or refinancing?) schedule is looming ever closer.
The $400m NZ Bond referred to above are the CNU010 4.12% bonds that matured in May 2021. These bonds were effectively replaced by the CNU030 and CNU040 bonds as outlined in the quoted text below.
The other NZ bond mentioned in the quote above is the $500m CNU020 issue with a coupon rate of 4.35%. That originated on 23rd November 2018 and does not mature until 2028. That means there is little chance of interest rate reset relief from that bond money until 2028.
The next chance for 'interest rate relief' will come from the maturity of the EMTN (Euro Medium Term Note) 500 million bond in 2023, currently valued at $NZ883m in the EOFY2020 statements ($NZ776m in the FY2017 statements). This bond, taken out on 18th October 2016 carries a minuscule nominal interest rate of just 1.125%. However the interest payments are only partially hedged to the NZD - based on a bond capital of $NZD500m or about 300m euro (3/5 of the actual value of the underlying bond). That means any decline in the NZD to Euro exchange rate will result in a higher interest rate being paid in NZD terms.
On 18/10/2016 NZD1 = 0.6513 Euro. Today we are sitting at NZD1 = 0.59 Euro
This means the effective interest rate that Chorus is paying on this bond today is:
(3/5)*1.125% + (2/5) (0.6513/0.59)*1.125% = 1.172%
In dollar terms that translates to an incremental annual payment of $776m x (0.01172-0.01125) = $0.365m. In this context, no auditor would count that as significant. Nor can I see any significant interest payment savings when this bond is refinanced in 2023.
The 300 Euro 7 year EMTN bond that matures in December 2026, was only taken out 5th December 2019. IMO this is unlikely to be refinanced at lower rates than the 0.875% coupon rate achieved.
What does all this mean? We are stuck with the just matured CNU010 bond as providing the only meaningful borrowing cost saving before 2028.
Quote:
Originally Posted by
Snoopy
In December 2020 Chorus issued two new bonds:
1/ CNU030 7 year bonds: $200m with a coupon rate of 1.98%
2/ CNU040 10 year bonds: $200m with a coupon rate of 2.51%
At the time it was made known that these two bonds would replace the $400m of CNU010 4.12% bonds, due to mature on Thursday 6th May. It is clear there will be an ongoing saving in interest charges with the replacement of $400m of CNU010 bonds with $200m of CNU030 bonds and $200m of CNU040 bonds.
Previous Annual Interest Charge: $400m x 0.0412 =
$16.48m
Forward Annual Interest Charge 1: |
$200m x 0.0198 |
= $3.96m |
plus Forward Annual Interest Charge 2: |
$200m x 0.0251 |
= $5.02m |
equals Total Forward Interest Charge : |
|
$8.98m |
This should produce an annualised saving of: $16.48m - $8.98m = +$7.5m
Working backwards, that $7.5m in 'interest saved' could be used to support a new loan, taken out at 2.5% of:
$7.5m / 0.025 = $300m
SNOOPY
The first CIP payoff period
Quote:
Originally Posted by
Snoopy
Working backwards, that $7.5m in 'interest saved' could be used to support a new loan, taken out at 2.5% of:
$7.5m / 0.025 = $300m
Quote:
Originally Posted by
Snoopy
I notice in the 'UBS Australia virtual conference' on 17th November 2020, the Chorus CIF debt repayment schedule has been updated to include repayments from the UFB2 roll out (slide 43).
Repayment Year |
2025 |
2030 |
2033 |
2036 |
Total |
Percentage to Repay |
15% |
18.5% |
29.4% |
37.1% |
100% |
Dollars to Repay |
$85.3m |
$104.7m |
$166.7m |
$210.2m |
$566.9m |
Discount Rate Divisor: 5.5% discount rate (FY2020 perspective) |
1.307 |
1.708 |
2.006 |
2.355 |
|
PV Dollars to Repay (FY2020 perspective) |
$65m |
$61m |
$83m |
$89m |
$298m |
Note
At EOFY2020 there were 444.492m Chorus shares in issue. So to work the future CFH capital repayments into today's Present value of Chorus shares, you have to reduce the value of Chorus shares by:
$298m / 444.492m = 67cps
This is the amount that must be taken off the value of CNU based on earnings potential to adjust for the debt repayment schedule.
Looking at the 'capacity to support a larger loan' verses the 'rolling mounds of debt' that are appearing on the horizon, it looks like the first $85.3m mound of debt could easily be serviced from current cashflows ( $85.3m << $300m ). However I have come to the conclusion that the real debt load to worry about in the future are the CIP preference shares. What looked like fair market interest rates when the CIP Preference Share terms were set up in 2012 ( the 180 day NZ bank bill rate plus 6% ) are now looking positively usurious. I can't find a 180 day rate any more. But if we take the current 90 day rate at the all time low of 0.25% we are still talking a 6.25% interest rate. 6.25% is monstrous for a utility.
The preference share capital repayment schedule, once again taken from 17th November 2020 Chorus presentation slide 43, to avoid interest payments (otherwise known as preference share dividends) is as below:
Repayment Year |
2025 |
2030 |
2033 |
2036 |
Total |
Percentage to Repay |
15% |
18.5% |
29.4% |
37.1% |
100% |
Cumulative Dollars to Repay |
$85.3m |
$197.1m |
$377.7m |
$766.4m |
$766.4m |
Discount Rate Divisor: 5.5% discount rate (FY2020 perspective) |
1.307 |
1.708 |
2.006 |
2.355 |
|
PV Dollars to Repay (FY2020 perspective) |
$65m |
$115m |
$188m |
$325m |
|
It looks to me as though by 2025, it would be realistic to assume that Chorus could initiate enough new borrowing to eliminate both the 2025 CIP debt repayment obligation of $85.3m and the 2025 CIP equity repayment obligation of $85.3m. ( $85..3m + $85.3m = $170.6m << $300m ). The subsequent following 'CIP debt' and 'CIP equity' repayment obligation is not until 2030. In internet terms that is a lifetime away (just think how far the internet has come in terms of network capability over the last ten years). Is it even meaningful to think that far ahead if you are running a broadband network company? Maybe the board and senior management all consider that they will be retired by then, so they don't have to worry about it?
Nevertheless I have noticed an interesting sharemarket phenomenon. Every time Chorus announces to the market that more CIP securities have been taken on as the fibre broadband rollout continues, the CNU share price goes down a few cents. CIP securities, be they debt or preference shares, ultimately represent debt that must be repaid. So maybe Mr Market isn't as cavalier about Chorus's ever ballooning debt as the board is?
SNOOPY
MAR end of May 2021 update
Quote:
Originally Posted by
Snoopy
On 6th April Chorus announced that the indicative maximum allowable revenue range to $680m to $710m. I will take the mid point of that range - $695m - to illustrate my point. Revenue from 'fibre broadband' and 'fibre premium' (AR2020 p52) totalled over FY2020:
$393m + $73m = $466m
This implies a profit maximising growth rate over the ensuing four years of:
$466m(1+g)^4= $695m => g=10.5%
The implied compounding annual growth rate for fibre revenue should be up to a maximum of 10.5%.
There has been an update on Chorus's 'Maximum Allowable Revenue' (for fibre broadband) position for the three years to 2024 since I last discussed this on 3rd May. The 27th May 'stock exchange announcement' states:
"The Commerce Commission has released its draft decision on Chorus’ price-quality determination for the first regulatory period for fibre (RP1, 2022-2024). The draft decision references an annual revenue range of $689 million to $786 million, including pass-through costs, during RP1 and remains broadly consistent with Chorus’ forecast fibre revenues for the period."
This is in response to interim correspondence from Chorus to the Commerce Commission on 17th May, where Chorus was talking about an MAR of $720m to $820m, So while the Commerce Commission has come up in the maximum dollar ceiling from their 3rd May MAR suggestion, the result is still well below (by $30m) the revenue cap that Chorus is asking for. However slide 29 on the 17th November 2020 presentation, the page titled 'Regulated Asset Base Implementation' has given me pause for thought.
That slide states the 'Anchor and Price Capped Products' include:
1/ Voice
2/ 100/20 Mbps fibre
3/ Direct fibre
Then the footnote says 'symmetrical wash up for unders and overs". I don't understand what that footnote means. All I can think of is that if a fibre plan was offered 60/60 Mbps with the reduced maximum download speed, offset by an increased maximum upload speed, then that would come under the same rules. Because the reduced download performance has been 'symmetrically offset' by an improved upload speed?
What does appear indisputable is that the MAR applies to a minimum 'bottom line of service'. However, if Chorus chooses to offer a higher level broadband plan, there is no mention of this coming under the Maximum Allowable Revenue cap. According to the Q4 Operating Statistics released to the market on 12th July 2021, 19% of all fibre broadband connections are 1Gbps as of EOFY2021. Wholesale prices went up last year for the 100Mbps plan from $46 to $47 per month. But at the same time the price of the 1Gbps plan fell from $60/month to $56/month. The wording on the slide I referenced above would suggest that the 1Gbps plan is not subject to the MAR price cap. But whether:
1/ The whole amount is not subject to the MAR price cap ($56/month) OR
2/ The difference between the 1 Gbps price and the 100Mbps price ($56/month - $47/month = $9/month) is not subject to the MAR price cap OR
3/ I am totally wrong and all the wholesale fibre broadband charges are included in the MAR price cap (as I originally surmised)
I do not know. Can anyone out there provide any clarity on how these 'superior' broadband plans will interact with the Commerce Commission set MAR price cap for fibre broadband?
SNOOPY
Capex: Total vs Sustaining (FY2020 Perspective)
Quote:
Originally Posted by
Snoopy
Contrary to popular belief, the roll out of fibre has only been partially and transiently funded by the government. Note 13 of AR2019 suggests that the total budget to roll out UFB1 (which covers the major towns and cities) followed by UFB2 and UFB2+ will cost $2.3b to $2.4b by the end of FY2022. Of that $548m to $568m is the cost of UFB2 and UFB 2+ going out to smaller towns (note 13 AR2020). By simple subtraction then, the cost of rolling out UFB1 alone must be between $1.732b and $1.852b. The total government funding for Chorus's share of UFB1 is $959m (Note 5 AR2012), or about half the cost of building the network.
Capital spending will not stop after FY2022. But it will be a step change down from the hundreds of millions of dollars of recent years.
Operational Year |
FY2016 |
FY2017 |
FY2018 |
FY2019 |
FY2020 |
FY2021 |
FY2022 |
FY2023 |
FY2024 |
Capital Expenditure Fibre |
$486m |
$503m |
$620m |
$664m |
$548m (3) |
$560m-$590m (1) |
$399.9m (2) |
$333.2 (2) |
$295.9m (2) |
Notes
1/ Forecast from HY2021 presentation: Slide 20
2/ These are the forecast 'Fibre Fixed Line Access Services' FFLAS 'Capital expenditure Proposals' as outlined in the 17th December 2020 market release on the 'Price Quality Expenditure Proposal Overview', Slide 10
3/ For FY2020, $186m of the total Capex for the year of $663m was defined as 'sustaining' (Full Year Result presentation, 24-08-2020, slide 25)
Potentially we have a very significant drop in CAPEX coming through, notwithstanding the fact that UFB2 and UFB2+ and the Rural Broadband Initiative (in partnership with Vodaphone) are still rolling out. FY2023 and FY2024 look like they will become comparatively sweet years for Chorus, before a very heavy debt repayment schedule disrupts things.
The difference between total capital expenditure and sustaining capital expenditure is set to become critical in determining the future dividend returns from Chorus shareholders. From the 17th November 2020 Presentation, Slide 33:
"From FY22 we will transition to a dividend policy based on a pay-out range of free cash flow
▪ free cash flow will be defined as net cash flows from operating activities minus sustaining capex."
The following table is compiled from information in:
a/ Slide 34 of the 17th November 2020 presentation.
b/ AR2020 p28
Capex over 2020 |
Fibre Layer 2 Sustaining |
$31m |
|
|
Fibre Products & Services Sustaining |
$14m |
|
|
Fibre 'Other Fibre Connections' Sustaining |
$20m |
|
|
Fibre Customer Retention Sustaining |
$7m |
|
|
Fibre UFB Communal Development |
|
$170m |
|
Fibre Connections & Layer 2 Development |
|
$251m |
|
Other Fibre Connections & Growth Development |
|
$42m |
|
Fibre Customer Retention Development |
|
$13m |
|
Fibre Total |
|
|
$548m |
Copper Network Sustaining |
$31m |
|
|
Copper Layer 2 Sustaining |
$7m |
|
|
Copper Customer Retention Sustaining |
$16m |
|
|
Copper Connections Sustaining |
$1m |
|
|
Copper Development |
|
$0m |
|
Copper Total |
|
|
$55m |
Common IT Sustaining |
$43m |
|
|
Common Building & Engineering Services Sustaining |
$17m |
|
|
Common Development |
|
$0m |
|
Common Total |
|
|
$60m |
Sustaining Total |
$186m |
|
|
Development Total |
|
$476m |
|
Overall Total |
|
|
$663m |
In light of the new revised dividend policy, a useful question to ask is: What sort of dividend would shareholders expected if such a policy had been operating over FY2020? (the last year for which full published results are available). Over FY2020 operating cashflow was $476m which equates to:
$476m/ 444.492m = $1.07 per share
From that we have to remove 'Sustaining Capex'
$186m / 444.492m = 42cps
I have argued in post 2766 that we must remove a 9.9cps CIP debt repayment allowance from the total free cashflow available as well. This means the total free cashflow available was:
107cps - 42cps -9.9cps = 55cps
A policy of distributing 80% of free cashflow available would see a dividend of 44cps.
What actually happened? The interim dividend paid was 10cps which together with a final dividend of 14cps, made a total dividend of 24cps relating to the FY2020 financial year. On paper then, we are looking at the prospect of a very substantial dividend hike going forwards. Or are we...?
SNOOPY
Directors buying Chorus Shares: Be Careful!
I had noted that in April two Chorus directors notified the market, that they had bought Chorus shares 'on market':
1/ On 31-03-2021 director Sue Bailey bought 5,000 shares increasing her holding to 25,000. At the price paid for this acquisition, $6.94 per share, Ms Bailey's total shareholding was worth $173,500, up from $138,800 before the purchase.
2/ On 07-04-2021 director Prue Flacks bought 18,900 shares increasing her holding to 43,334. At the price paid for this acquisition, $6.57 per share, Ms Flack's total shareholding was worth $284,708.38, up from $160,531.58 before the purchase.
Generally directors buying shares on the market of a company they govern is a positive sign for investors: A signal for outsiders that they should consider buying into the company. However a not so well publicised change in the director shareholding rules should cause potential shareholders in Chorus to pause. From p76 AR2020 there is a new minimum shareholding policy for directors:
"Chorus' Minimum Shareholding Policy sets the expectation on Director's to hold, at minimum, shares equal in value to one year's director base fee (after tax). If not held at date of appointment (or policy), the policy expects Directors to accumulate this holding over the first three years in office."
The 'after tax' thing is a little odd as I would have thought independent directors are independent contractors and are not employees of Chorus. That would imply they would not have any tax deducted from their directors fees. Furthermore different directors are liable to have varying tax obligations depending on how they manage their financial affairs. So there would be no way to know how much tax the individual directors end up paying. Putting this aside, I see the 'base fee' for a director is $114,000.
Prue Flacks has been in the job for more that three years and her holding is worth well over 'base director fee' level. So it would be fair to see Prue's purchases as an endorsement of Chorus. However the same is not true for Sue Bailey, who only became a director in October 2019. Sue has until October 2022 to build up her shareholding, although it might not be a good look to be seen to be 'dragging the chain'. So I would argue the reason for Sue's share purchase in this instance might have been to avoid any embarrassment at a falling share price taking her Chorus shareholding below the new minimum holding requirement.
The reason I mention this now is that should the Chorus share price fall significantly, and you see directors buying, then first thoughts might be the directors, who have intimate knowledge of the business consider the prevailing share price offers value. However, with this new minimum shareholding rule, there might be another reason: Company rules demanding that they buy!
I am aware of some overseas companies having minimum director shareholding levels. But this is the first time I have noticed such a policy in New Zealand. Do any other NZ listed companies do this?
SNOOPY