The September, 2019, edition of PrefLetter has been released and is now available for purchase as the “Previous edition”. Those who subscribe for a full year receive the “Previous edition” as a bonus.
PrefLetter may now be purchased by all Canadian residents.
Until further notice, the “Previous Edition” will refer to the PrefLetter, 2019, issue, while the “Next Edition” will be the October, 2019, issue, scheduled to be prepared as of the close October 11, 2019, and eMailed to subscribers prior to market-opening on October 15 (the day after Thanksgiving).
PrefLetter is intended for long term investors seeking issues to buy-and-hold. At least one recommendation from each of the major preferred share sectors is included and discussed.
Note: My verbosity has grown by such leaps and bounds that it is no longer possible to deliver PrefLetter as an eMail attachment – it’s just too big for my software! Instead, I have sent passwords – click on the link in your eMail and your copy will download.
Note: The PrefLetter website has a Subscriber Download Feature. If you have not received your copy, try it!
Note: PrefLetter eMails sometimes runs afoul of spam filters. If you have not received your copy within fifteen minutes of a release notice such as this one, please double check your (company’s) spam filtering policy and your spam repository – there are some hints in the post Sympatico Spam Filters out of Control. If it’s not there, contact me and I’ll get you your copy … somehow!
Note: There have been scattered complaints regarding inability to open PrefLetter in Acrobat Reader, despite my practice of including myself on the subscription list and immediately checking the copy received. I have had the occasional difficulty reading US Government documents, which I was able to resolve by downloading and installing the latest version of Adobe Reader. Also, note that so far, all complaints have been from users of Yahoo Mail. Try saving it to disk first, before attempting to open it.
Note: There have been other scattered complaints that double-clicking on the links in the “PrefLetter Download” email results in a message that the password has already been used. I have been able to reproduce this problem in my own eMail software … the problem is double-clicking. What happens is the first click opens the link and the second click finds that the password has already been used and refuses to work properly. So the moral of the story is: Don’t be a dick! Single Click!
Note: Assiduous Reader DG informs me:
In case you have any other Apple users: you need to install a free App from the apple store called “FileApp”. It comes with it’s own tutorial and allows you to download and save a PDF file.
However, Assiduous Reader Adrian informs me in the comments to the January 2015 release:
Some nitpicking for DG:
FileApp costs $1.19 in the Apple Store.
But Adrian2 now advises:
Well, as of now, FileApp is free (again?).
FixedReset Prospectuses Are Imprecise!
September 13th, 2019As we all know, FixedResets will reset their dividend every five years based on the Government of Canada Five Year yield (“GOC-5 rate” or “GOC-5 yield”) and therefore the prospectus for each issue needs to include information regarding exactly how that yield is determined.
The prospectus for ALA.PR.G (chosen because I can link to it!) contains typical language with respect to this process:
I am not aware of any material differences in the definitions between prospectuses.
So this sounds pretty good, right? The GOC-5 yield will be calculated by an independent third party with no ambiguity and complete verifiability, right? Wrong.
As noted in the post Reset Calculation Oddity for 2019-9-30 / 2019-10-1, the following four issues had the GOC-5 rate underlying their dividends recalculated by their issuers on September 3:
The AltaGas screenshot shows they made a slight mistake: the time of the screenshot is 10:00:18, so they missed their proper time by 18 seconds, although they could argue that the prospectus only uses four significant figures and therefore their calculation is completely OK. However, each of the other screenshots shows a genuine effort being made to determine just what exactly the GOC-5 rate was at 10:00:00.00000 and each methodology resulted in a different answer.
Four companies, four identically specified calculations, four different answers.
I will be the first to agree that the variance is minor: the spread between the highest and lowest measurement is only 1.5bp and that’s not a lot. On a typical issue size of $250-million, that comes to $37,500 annually or $187,500 over the full five years. On a per-share basis, a 1.5bp yield difference comes to $0.00375 p.a., slightly less than two cents over the full five years.
But that’s not the point. First, the prospectus should specify the yield to be used in a completely precise manner. To quote again from the representative language of the ALA.PR.G prospectus:
What’s the point of being so horrifyingly precise about the rounding of the Annual Fixed Dividend Rate when the underlying figure is nowhere near that precisely measured?
In addition, once this becomes widely known, what’s to prevent a company from determining the GOC-5 yield in as many ways as their Bloomberg users can invent and choosing the lowest answer?
Clearly, the Bloomberg methodology is not adequate for the task of determining a precise, public, third-party figure and the procedure needs to be changed. The first alternative that leaps to mind is the Bank of Canada’s bond yield reporting:
Yes, it’s not quite the same thing and yes, there might be a perceived problem if the benchmark changes near the time of calculation (typically, new benchmarks will trade to yield less than the ‘off the run’ issues they supersede). I don’t care. I want something precise, public (certainly more public than a subscription to a Bloomberg terminal!) and prepared by an independent third party. If somebody has a better idea, let’s hear it.
Posted in Issue Comments, Regulation | No Comments »