The imputation credit illusion
Quote:
Originally Posted by
Snoopy
Spark certainly is one for the 'dividend hounds'.
Gross Dividend Calculations
FY2020 P2:
12.5c (Ordinary, 75% imputed) = 9.375c (FI) + 3.125c (NI) = 9.375c/0.72 +3.125c = 13.021c +3.125c = 16.15c (gross dividend)
FY2021 P1, FY2021 P2, FY2022 P1:
12.5c (Ordinary, 100% imputed) = 12.5c (FI) = 12.5c/0.72 = 17.36c = 17.36c (gross dividend)
In AR2020 on p94 we learn:
"Spark has a nil imputation credit account as at 30 June 2020"
This is not a surprise. The dividend paid in 'Part 2' of FY2020 (referenced above) was not fully imputed. That is consistent with the imputation credit account running out. However the fact that the imputation credit account was empty at SOFY2021 means we can run a little 'imputation credit check'.
The total dividend paid out during the FY2021 year was 25cps 'fully imputed'. 'Fully imputed' means 'after tax is paid at the company rate of 28%'. So how much tax was paid by Spark to allow the dividend to be fully imputed?
Answer: 25cps / (1-0.28) = 34.72cps (gross dividend) => Tax paid = 34.72cps - 25cps = 9.72cps
At EOFY2021 there were 1,867m Spark shares on issue. If we assume that there were 1,867m on issue at the time of both 12.5cps dividend payments to shareholders (actually there likely weren't because there were 30m employee performance share options cashed in over the year) then the amount of dividend paid out in dollars over the FY2021 year was:
1,867m x $0.0972 = $181m (the possible maximum error in this calculation, should none of the 30m performance rights have been exercised before any of the dividend was paid is -$2m)
The cashflow statement for FY2021 shows payment of income tax at $188m. Anyway you look at things, that is more than enough tax to pay a fully imputed dividend (only a $181m payment is required). However, if I look at the income tax expense information for FY2021 (AR2021 p100) then the current year tax expense was $172m. So it looks to me as though Spark has paid $188m-$172m= $16m more tax than was due. $181m - $172m = $9m of that went towards making the dividend fully imputed, even though technically the dividend was not fully imputed. It only was allowed to became fully imputed when Spark handed more money to the tax man than was actually due. If the odd one off tax adjustments were taken out of the FY2021 Spark result, then the underlying tax bill for the year was even less, only $155m. $155m is $181m - $155m = $26m short of the actual underlying tax bill paid by Spark. So:
1/ IF $181m was the actual tax paid to give 100% imputed dividend AND
2/ $155m was the normalised requirement for tax to pay THEN
the underlying dividend imputation rate at Spark over FY2021 was only $155m/$181m = 85.64%. This is more in line with earlier imputed tax rates of 80.6% and 75% paid over FY2019 and earlier. This means I would not be surprised if for FY2022 - if the dividend rate is kept at 25cps - then the Spark dividend for FY2022 will no longer be fully imputed. The only reason the dividends paid in the year FY2021 were fully imputed was that Spark paid the IRD in advance more tax than they had to.
There follows a very strange statement on p101 of AR2021 regarding the EOFY imputation credit position.
"Spark has a negative $18 million imputation credit account balance as at 30 June 2021, due to the timing of dividend and tax payments (30 June 2020: nil)."
A negative imputation credit balance means that the company has somehow paid out more imputation credits than it had earned. But this is impossible, because if a company hasn't paid the required amount of tax, then it cannot pay a dividend with full imputation credits. Yet in direct contradiction to this, the dividend paid over the second half of FY2021 was fully imputed! How is this possible?
A rather confused SNOOPY
SPK vs CNU 'Head to Head' [FY2021 Perspective]
OK I am first to admit that this is not an 'apples with apples' comparison. Both companies are operating in the telecommunications space in New Zealand, but that is where the similarities end. Spark is primarily a 'retailer' that also happens to own their own 'mobile network' and is on the drive to become a wider player in the digital space. Chorus OTOH is strictly a fibre broadband network company that operates entirely in the wholesale space.
Nevertheless both are competitors in the sense that they compete for my investment dollar. So how do the investment fundamentals stack up when the key metrics are tabulated side by side?
FY2021 |
Spark |
Chorus |
No. Shares |
1,867m |
447.025m |
Share Price (10-10-2021) |
$4.83 |
$6.85 |
Declared eps |
20.6c |
10.5c |
Normalised eps |
20.1c |
11.5c |
Normalised PE |
24.0 |
59.6 |
Normalised NPAT Margin |
10.5% |
5.4% |
Normalised ROE |
25.0% |
5.4% |
Net Bank Debt |
$1,403m - $72m |
$2,373m - $53m (1) |
Declared NPAT |
$384m |
$47m |
Min. Debt Repayment Time |
3.5 years |
49 years |
Snoopy's Fair Share Price Valuation (2) |
$5.58 |
$8.36 |
Market Discount to Fair Value |
-13% |
-18% |
Notes
1/ Bank debt for Chorus excludes crown funding
2/ For this comparison my 'Fair Share Price Valuation' is based on my 'capitalised dividend valuation' model (Post 2658 on Chorus Thread, Post 1820 on Spark Thread). Being a monopoly wholesale provider, I have judged a fair yield in today's ultra low interest rate market for Chorus shares to be 5%. For Spark being in a different competitive market place (albeit starting from a strong incumbent position) I have judged a fair yield for Spark shares to be 6%.
3/ For normalised earnings calculations see post 1808 (Spark thread) and 2792 (Chorus thread)
Lined up so starkly like that, when I was looking recently for a space for my 'telecommunicatioons dollar ' to go, I think you can see why I pushed it the way of Spark. It came down to a better earnings ability (higher ROE) and much more conservative debt position (the MDRT figure for Chorus of 49 is eye watering). I actually hold both shares. But my 'commitment' to each, in dollar terms, is now 4:1 in favour of Spark.
My fair value calculations, both yield based, both have a whisker of unease attached to them. Actual dividends paid out from Spark have been above core earnings in recent years. So being able to keep paying those dividends in the future will require some of those Spark 'business development plans' to come to fruition. LIkewise Chorus is transitioning to a new 'cashflow based' dividend policy. I was already projecting sharply lower imputation credit percentages going forwards, a reality which has since been confirmed by Chorus. But it does seem the Chorus capital spend is tracking higher than expected, and they are facing more competition from 'disruptive technology' in the form of 'fixed mobile broadband' from Spark, Vodafone and 2 degrees. So I may have to rethink that 'acceptable monopoly yield' that I have attributed to Chorus
SNOOPY