Husky Energy has announced:
that 40,800 Cumulative Redeemable Preferred Shares, Series 5 (Series 5 Shares) were tendered for conversion, which is less than the one million shares required to give effect to conversions into Cumulative Redeemable Preferred Shares, Series 6 (Series 6 Shares). As a result, none of the Series 5 Shares will be converted into Series 6 Shares on March 31, 2020.
HSE.PR.E is a FixedReset, 4.50%+357, that commenced trading 2015-3-12 after being announced 2015-3-4. It will reset at 4.591% effective 2020-3-31. I made a preliminary recommendation not to convert. The issue is tracked by HIMIPref™ and has been assigned to the FixedReset (Discount) subindex.
Husky 7 year bond is yielding 8.8%.
Think about that.
Anyone interested in hse.pr.e now?
25 x 3.57% / 11.9 = 7.9% now
Husky 2037(!) bond closed at 5.74% on Friday, 2029s at 4.77%. HSE.PR.E (second “cheapest” by my calcs, Gs are better) yields 9.6%, resets at 8.3%. Seniority spread is extreme on Husky prefs. All of the Husky prefs are ripe for a recalibration IMO.
“Husky 7 year bond is yielding 8.8%. Think about that.”
Skeptical, would be interested to know where you are you getting your numbers from? I am assuming you are referring to the Cda$ 3.6% 10 March 2027.
However, the US$ 04/15/2022 3.95% is trading around 101.5 and has a YTM of some 3.4%. And Stusclues has already posted with respect to the longer 2029s and 2037s.
What I refer is the worse case scenario assuming 5 year Canadian gov bond is zero after 2025.
Currently it is just reset to 4.591% for 2020 to 2025：
25 x 4.591% / 11.9 = 9.6%
Then 2025, if not called, it will at least 7.9%
If it is called at 25 after 2025, the potential gain is （25/11.9）-1= 110%..
If it is never called, you stay at 7.9% at least
Currently it is just reset to 4.591% for 2020 to 2025：
25 x 4.591% / 11.9 = 9.6%
Then 2025, if not called, it will at least 7.9%
On9nobody, welcome to the blog!
I recommend that, for easier comparison between issues, you compare long-term yields based on identical assumptions – that is, a single number for yield calculated with specified assumptions that are the same for all comparative calculations.
There is a FixedReset yield calculator available through this blog, linked in the right-hand navigation panel under the heading “Calculators”. Its use is described in detail in the post What Is The Yield Of HSE.PR.A?
I am using the assumption to calculate the worse case, 0% on Canada bond + 3.57% and never get called and it does not default
25 x 3.57% / 11.9 = 7.9% annually
Is this correct?
Is this correct?
No. This calculation does not account for the 9.6% you cited as the yield for the next five years.
Just questioning ” 0% on Canada Bond ” as worst case scenario…
Who would have ever thought negative interest rates are a possibility. We are currently very close to 0% and other countries have gone negative. Therefor it is a possibility, but it is remote. On top of that, the total lack of repayment of principal is an even great “worst case”.
On the flip side, we may see high interest rates as well given all the debt countries are generating…
Here’s a bit more detail on why I feel the regrettably popular method of describing yield as ‘X until this date and Y afterwards’ should be deprecated.
It’s not internally consistent; any model (and “yield” is a model!) should first be examined to determine whether or not it’s internally consistent; if not, then you don’t even have to bother checking whether it actually predicts anything useful because the answer is no.
So, for instance, say we have a share that will yield ‘7% until 2024 and 5% afterwards’, while Straight Perpetuals are yielding 6%. How would one use this information to trade? A yield-maximizing algorithm might well indicate that one should buy it now with the intent of selling it in 2024 to buy the Straights.
But the realized yield will depend on the future realized price of the security, and why should this future price be equal to the current price? It seems much more reasonable to assume that the future price will be less than the current one, so that the realized yield to 2024 will be equal to the projected yield after 2024.
To put this another way, one can model the resetting issue as always paying a base rate of 5%, now and forever, and in addition paying a ‘top-up’ dividend of 2% for the next four years. Therefore, one should logically calculate the total amount of the ‘top up’ dividend, discount to Present Value if you’re being particular, and using this value to adjust the price of the issue to an effective value. Thus, you say to yourself, I am not paying $20 for this security, I am paying $19 for the security and $1 for the top-up.
Don’t anybody work out the details of the prices in this example, I’m just pulling the numbers out of the air. I will leave proper calculations as an exercise for the student.
Looking at it this way has the immediate benefit of forecasting a decline in price; every time a dividend is paid until the reset, the top-up amount will decline until it disappears. The base price of $19.00 will be constant, given unchanging market conditions.
The disadvantage of this method is that it still uses too many numbers to calculate ‘yield as defined’ – the “Expected Future Current Yield”, the Top-Up Amount and the date of the reset. This will make comparison of yields between different issues extremely difficult.
It is much better to use a single number for yield – the estimated return per unit of investment given certain assumptions. And the only assumptions necessary are those that have already been made in calculating the ‘three-number yield’.
So I will strongly suggest and reiterate my recommendation: calculate single-number yields and use those for valuation and discussion purposes.