The Bank of Nova Scotia has announced:
its intention to redeem all outstanding Non-cumulative 5-Year Rate Reset Preferred Shares Series 30 (“Series 30 Shares”) and Non-cumulative Floating Rate Preferred Shares Series 31 (“Series 31 Shares”) of Scotiabank on April 26, 2020 at a price equal to $25.00 per share (the “Redemption Price”), together with all declared and unpaid dividends to the date fixed for redemption. Formal notice will be issued to the shareholders in accordance with the share conditions.
The redemption has been approved by the Office of the Superintendent of Financial Institutions and will be financed out of the general funds of Scotiabank. This redemption is part of the Bank’s ongoing management of its Tier 1 capital.
On February 25, 2020, the Board of Directors of Scotiabank declared quarterly dividends of $0.113750 per Series 30 Share and $0.166480 per Series 31 Share. These will be the final dividends on the Series 30 Shares and Series 31 Shares, respectively, and will be paid on the date of redemption, April 26, 2020, to shareholders of record at the close of business on April 7, 2020, as previously announced. After April 26, 2020, the Series 30 Shares and the Series 31 Shares will cease to be entitled to dividends.
BNS.PR.Y was issued as a FixedReset, 3.85%+100, NVCC non-compliant, that commenced trading 2010-4-12 after being announced 2010-3-25. Extension was announced in March, 2015 and the issue issue reset to 1.82% effective 2015-4-26. I recommended against conversion, but 42% converted to the FloatingReset, BNS.PR.D.
BNS.PR.D is a FloatingReset, 3-Month Bills + 100, NVCC non-compliant, that resulted from a 42% conversion from BNS.PR.Y.
It will be noted that there has been (relatively!) heavy trading in these issues lately, with about 40,000 shares of BNS.PR.Y trading this week with a VWAP of under $24.00, together with about 55,000 shares of BNS.PR.D trading with a VWAP of under 23.00.
I’ll bet a nickel that Scotia made this announcement a few days in advance of original intentions, just to put an end to this nonsense.
I find this announcement at a time like this extremely odd. Not only did they have the chance to extend it with a relatively low dividend payment given where 5 year goc is, but also in a time of uncertainty. It is weird that they would basically give away free money.
They redeemed these because they are not an NVCC compliant issue, which means that post-2022, these shares would no longer be counted is Scotia’s Tier 1 capital (since the next redemption date would have been in 2025). I am not 100% sure that they absolutely had to redeem (James can confirm), but I suspect that they did not want to get the watchdogs at OSFI annoyed. RY and BMO still have some of these non_NVCC prefs outstanding and, these will almost certainly be called as well (BMO’s are reset prefs, RY are straight prefs so will probably not get redeemed until the last minute given that it would be impossible to issue new prefs to replace these in the current market). Good investment opp, these, if you don’t believe that these banks will default due to the COVID crisis.
Received today by email from RBCDS – seems you have a competitor to Prefletter……….. who may not understand the NVCC tag
“The virus outbreak has wreaked havoc on the equity markets but the fixed income markets are where the shockingly poor performance is. Credit spreads over government bonds have quick jumped faster than we have ever seen in the past (from a low level).
Spreads on high yield bonds went from ~350bps to almost 1,000bps in a little over a week. The last time we seen credit spreads this high was Feb 2016 and that was two years in the making.
This abrupt move as cause massive dislocations in prices in preferred shares and all types of bonds. But the best opportunity is in a small group of the best preferred shares in Canada. The example below is a Royal Bank 5-year rate reset preferred share paying 5.50% set to reset in August 2021 (but will get called as it is expensive financing for the bank). It has a fixed reset spread 480bps over the 5-year government bond and is trading around $22.25. In August 2021 this share will get called at $25. That means you have a capital gain of $2.75 + $1.72 in dividends. That works out to a 1.25 year total return of 20%.
This is a rare opportunity that should be taken advantage of…..”
Complete fail to copy the graphic into my post above – sorry all – its RY.PR.T that RBCDS was boosting
Pugwash, I think you meant (NVCC-compliant) RY.PR.R.
Re. the RBCDS comment: why are they so sure that RY.PR.R will be called in August 2021? For that to happen you have to make a number of assumptions, the main one being that 2021 market conditions will be such that the bank could refinance in August 2021 at a materially lower coupon. As a side note, when it was first issued in 2016, the 5-year rate as a tad over 1% and the market still “demanded” a 5.5% coupon. If it is, say, .5% in August 2021, it will be cheaper for the bank at 5.3%.
And can you exclude the possibility that RY.PR.R is still trading below par in 2021? If it is, is it reasonable to assume the bank would redeem RY.PR.R at $ 25 and then be able to replace it with a refinancing that is materially cheaper [ and therefore worse for the investor] but still get that issue out at par? I don’t think so.
Not only did they have the chance to extend it with a relatively low dividend payment given where 5 year goc is, but also in a time of uncertainty.
I should have discussed this more deeply. The problem is that these are NVCC non-compliant, which means that after the 2022-1-31 OSFI deadline, they will no longer be able to count them as Tier 1 Capital (equivalent to equity, up to a certain amount in total).
It has long been assumed that all NVCC non-compliant bank issues will be redeemed on or before – or maybe, in certain circumstances, a little after – this deadline since, from the issuers’ perspective their nature will have changed from “cheap equity” to “expensive debt”.
Are they expensive debt? They pay GOC-5 + 100bp on their par value, which means that as of now they can expect to pay 1.78% on the principal. However, this will be paid out of after-tax profit, so the interest equivalent rate is 1.3-times this figure or 2.31%.
That’s not particularly expensive debt, especially when it can be held in perpetuity, but there are other considerations.
Bank sub-debt is similar to preferred shares in concept, in that it is a regulatory requirement to have so much “Tier 2” capital and it’s cheaper to achieve this level with sub-debt than with equity. See A Vale of Tiers for discussion of these capital categories.
Sub-debt is routinely issued with a five year fixed rate that changes to floating if not redeemed at that time. These bonds are then sold to customers as having such-and-such a spread over five-year bonds; I refer to the five-year date as a “pretend-maturity”.
In late 2008, Deutsche Bank refused to honour the pretend-maturity on some of its sub-debt, largely because the rate on that issue was so low relative to where they could obtain equivalent financing at that time.
The investment world was very, very upset as discussed on December 19, 2008.
The conclusion was drawn that the bank didn’t redeem because they didn’t have any money; Deutsche’s securities didn’t do very well in the aftermath, which of course complicated the financing problems that they were experiencing (along with many other banks!). So Canadian banks continued to honour their own pretend-maturities, despite the large economic incentive to leave the issues outstanding – see, for instance National Bank Honours Sub-Debt Pretend-Maturity and Bank Sub-Debt Redemptions.
We are in a different world now, and people are not so concerned about imminent bankruptcy of the entire banking sector. But the above illustrates the potentially severe consequences of not adhering to very firm market expectations.
The banking business is based on confidence; the banks’ marketting techniques means that not honoring pretend-maturities will shake that confidence, therefore they have to redeem these things at the expected time even when it is not economic to do so.
That is what I believe is happening with this redemption; I also believe that this is why they announced their intention prior to the 30-day notice period, as they saw that the trading prices of these issues was indicating that confidence was not as total as they thought it should be.
It’s Public Relations. First, holders who panicked and sold at $23 or less would have been somewhat irritated when redemption was announce a week later; second, they don’t want people speculating, even for a day, that they might not have the money to redeem; and, bound up in all this, they want people to think of the banking industry in general (and Scotia in particular) as being nice and predictable.
I am not 100% sure that they absolutely had to redeem (James can confirm), but I suspect that they did not want to get the watchdogs at OSFI annoyed.
The only consequence of leaving them outstanding that was written down was that they couldn’t count them as Tier 1 Capital.
It is possible that OSFI would have been annoyed. When announcing the NVCC rules, they invented a new concept called “the legitimate expectations of the parties to such capital instruments” which, when translated into street language, means “contracts and prospectuses mean nothing so don’t bother reading them” (it’s a polite street and expletives are rare).
And, lo and behold, the banks agreed that contracts mean nothing and did not exercise their redemption rights on some very expensive debt, which cost them a fair bit of money.
still have some of these non_NVCC prefs outstanding
Remaining issues are BNS.PR.Z, BMO.PR.Q, RY.PR.A, RY.PR.C, RY.PR.E, RY.PR.F and RY.PR.G.
An issue that could go either way, according to whoever gets to define which expectations are legitimate and which ones aren’t, is RY.PR.W. There were some CIBC issues that were similar.
The example below is a Royal Bank 5-year rate reset preferred share paying 5.50% set to reset in August 2021 (but will get called as it is expensive financing for the bank). It has a fixed reset spread 480bps over the 5-year government bond and is trading around $22.25. In August 2021 this share will get called at $25.
It’s very bold of the analyst to assert so firmly that they will be called. My questions is that when he calls them “expensive financing”, what is he comparing them to?
It can’t be to the current preferred share market, because they’re trading below par at the moment, indicating that a new issue coming out at par tomorrow would have to have a higher coupon and reset.
If he means the preferred share market as it is expected to be in August, 2021, then he’s making a market timing call. If the prediction of a higher market at that time is accepted (even if only for the sake of an argument) then it is virtually certain that applying those assumptions to the preferred share universe as a whole will uncover many issues that will have a higher total return.
If he means expensive relative to replacing the capital with retained earnings, then I’d like to see more argument on that, because the banks like to see a Return on Equity in the 12% range.
If he means expensive relative to replacing the capital with Alternative Tier 1 Bonds (“AT1”, as us cool guys say), such as Scotia’s issue discussed on October 10, 2017, then that’s another market call.
So I would like to know more about what he means by “expensive”.
The issue, by the way, is RY.PR.R, the highest spread NVCC FixedReset from the major banks (NA.PR.X is higher).
Pugwash, I think you meant (NVCC-compliant) RY.PR.R. … you have to make a number of assumptions, the main one being that 2021 market conditions will be such that the bank could refinance in August 2021 at a materially lower coupon.
Ah, heck, peet beat me to it. You snooze, you lose! Or, in this particular case, “You work on the PrefLetter Supplement, you lose!”.
James, “You snooze, you lose!”
At least you don’t get your statistics wrong 🙂 … I have to correct my earlier side note: the 5-year when this issue was announced was .64%, not “a tad over 1%”.