PrefBlog

OK, so this is a pretty long series of posts! And I will admit, a lot of the motivation behind these posts is just to get everything clear in my own mind and look at some actual numbers … and some actual prospectus language! It really just counts as background material for an article to be written some day, if I can convince an editor that there is unsatiated demand for articles dealing with bankruptcies of Big 5 Canadian Banks!

There are two more comparitive tables I want to look at:

It is interesting to note that CIBC is now over the 25% limit for Preferreds in Tier 1 Capital (although they can apply the excess to Tier 2 Capital, which is Sub-Debt territory). If anybody needed any more indication that that provided by a quick look at the dividend paid per annum, then this is an indication that they intend to redeem CM.PR.C as soon as possible. Note that the “Issuance” figure for CIBC is a net figure : two issuances (CM.PR.I and the pending one) and one redemption (CM.PR.B).

It is also interesting that Royal has used up about half its available room. Given that their Tier 1 Ratio at year-end significantly lagged the competition, I suspect that there will be more opportunistic issuances in addition to RY.PR.D and RY.PR.E.

It’s hard to guess just what else might happen. Given that the banks like to make an ENORMOUS return on equity (BMO’s 2006 Annual Report states that the target is 17-19%; they achieved 19.2%; largely due, I suspect, to fees paid by me personally) and only have to pay a 4.5% dividend on perps, it would seem that issuing perps is a no-brainer, especially in this environment where one CAN issue perps and consensus is that interest rates are historically low due to excess global capital. But we shall see! It would be nice to get some more TD issues, especially.

In Part 2 of this series, I claimed that Preferred Shares are, at best, pari passu with Innovative Tier 1 Capital. Therefore, to look at the credit quality of the bank preferreds, we have to look at the Equity / Assets ratio, since only common equity is subordinate to prefs. It would be nice if Preferred Shareholders could stick it to the Innovative Tier 1 holders in the events of problems; but it would be nice too if it rained lemonade; we can’t have everything!

The ratios reported in the second table above are important because:

• The Equity / RWA ratio supports preferred shares and Innovative Tier 1 Capital
• The Tier 1 Ratio supports the banks’ subordinated debt, which is included in their Tier 2 Capital
• The Capital Ratio is what supports deposits

If we’re going to bring things like sub-debt and deposits into this discussion, we have to get an idea of what they are! So, sorry, but I’m going to have to do some more quotation here, this time from the Trust Indenture for the TD Bank 4.317% MTNs Due Jan 18, 2016.

The Indenture provides that an Event of Default in respect of this Debenture shall occur only if the Bank becomes insolvent or bankrupt or resolves to wind-up or liquidate or is ordered woundup or liquidated. If an Event of Default has occurred and is continuing, the Trustee may, in its discretion, and shall upon the request of Holders of not less than one-quarter in principal amount of the Debentures, declare the principal of and interest on all outstanding Debentures to be immediately due and payable. There is no right of acceleration in the case of a default in the payment of interest or a default in the performance of any other covenant of the Bank in the Indenture.

…

Unless previously redeemed, purchased for cancellation, converted or exchanged as provided herein, the principal amount of the Debentures shall be due and payable on January 18, 2016.

After Jan. 18, 2011, these notes pay BAs+100bp. The notes have a Canada Call prior to Jan 18, 2011 and a call at par afterwards. The market treats these things as five-year notes and they are currently quoted at 34bp back of the Canada 4%/2010.

As noted by the Alberta Securities Commission

Deposit Notes are deposits issued by a financial institution and are excluded from the definition of “security” pursuant to sub-paragraph 1(ggg)(v)(B) of the Act.

Accordingly, prospectuses are awfully hard to come by, so I can’t do much quoting! I’ll restrict myself to noting that there is a CIBC Deposit Note, for instance, with a coupon of 4.75% maturing December 22, 2014 and quoted at 41bp back of Canadas 5%/2014.

It is difficult to come up with comparitives for the various levels of debt but we can try! Consider:

The terms of the Royal Sub-Debt are similar to those for the TD Sub-Debt quoted above. 

I can’t get a comparitive Deposit Note. At the RBC online rate page, a GIC paying semiannually and due in 7 years, purchased in an amount in excess of $1,000,000, pays … 3.4%. Which is, more or less, 75bp THROUGH Canadas. Anybody who buys one of these is a fool, since a million bucks should be enough to get rates that … maybe aren’t completely wholesale, but will be a point or so better than what’s being offered on-line! Which is what I’ve estimated for a deposit note of similar term, based on comparables.

One may quibble about the propriety of quoting the prefs in the above table at a spread to the Canada 2013s. Prefs are generally sold to investors by stating the spread to long Canadas … but I take the view that if it’s OK to trade Tier 1 Debt against the shorter Canadas, then it must be OK to trade the prefs against them, too! 

And one may really quibble about my inclusion of equities in the table! However, they are the end-point of the capital continuum and I’ve estimated an equity risk premium of 380bp based on Richard Guay. Another good number is 400bp, which has been discussed by Campbell, Diamond & Shoven. Stick in any number you like! You can only be proved wrong in the long run, and in the long run we’re all dead anyway.

To be perfectly frank, the relative levels for the market-trading debt don’t make a lot of sense to me.

• Sub-Debt holders aren’t getting much more than Depositors. Despite giving up a whole bunch of credit protection. This just seems wrong.
• Tier 1 Investors are only getting 14bp more than Sub-Debt. As shown in the second table of this post, Equity-to-RWA for Royal is 7.56%, vs. a Tier-1-Ratio of 9.6%. That’s a significant loss of credit protection for the Tier-1 holders, and they don’t seem to be getting paid too much for it.
• Taxable Preferred Shareholders are getting distressed yields for high quality credit. Look at it! They’re getting more now than Tier-1-Debt holders will get if in a distressed situation!
• Tier 1 Holders have better rate protection : in the event that Royal Bank becomes distressed AND inflation rises, we can assume that the BA rate will at least meet inflation and pay 150bp extra. Pref holders do not have this degree of protection. Note, however, that should deflation be the operative word, the link to
  BAs will be greatly regretted.
• Taxable Preferred Shareholders are capturing a lot of the equity risk premium, with better protection than equity holders. Pref shareholders are getting 63% of the credit protection accorded to depositors (ignoring Federal Insurance, figure determined by comparing the Equity-to-RWA and Total-Capital ratios).

The fly in the ointment of this analysis is, as always, inflation. A good bout of inflation will be very harmful to the interests of perpetual preferred shareholders and it never pays to forget that. On the other hand, one shouldn’t let Fear rule Greed too harshly … I suggest that Real-Return-Bonds are an appropriate complement to a perpetual preferred portfolio.

What I find really surprising is the extraordinarily small premium that Canadian bond buyers are demanding for Innovative Tier 1 instruments. I am advised that spreads are wider elsewhere in the world. However, Tier 1 instruments are included in the Index (which was determined by … um … Scotia Capital, a subsidiary of the Bank of Nova Scotia) and there is a tendency in Canada to worship the banks anyway … so, “Why not? says Joe Bank-Employed-Portfolio-Manager. “I can pick up 15bp over sub-debt and still have an index-included bond.”

Innovative Tier 1 Capital may be a good instrument … but I don’t think they’re bonds. It’s only a bond if I can bankrupt the issuer for being a day late and a dollar short! And the more I look at them, the less convinced I am that they pay enough of a premium to be worth the chunk of equity-risk that is being taken.

Sub-Debt is bonds. The indenture contains the words “due and payable”. I like the words “due and payable”, they give me a warm fuzzy feeling. When I don’t hear them, I want more money than just straight fixed-income returns.

Perpetual Fixed Rate Preferreds aren’t bonds either, but they pick up a big chunk of the equity-risk-premium while retaining a great many bond-like features. One must be wary of inflation and adjust the remainder of one’s fixed income portfolio accordingly – but represent a great deal for taxable investors at today’s prices.

Update 2007-02-06: Added some commentary for the first table; corrected a typo in the last bulletted list.

Update 2007-02-07: With respect to inflation and the perpetual holding of prefs vs. Innovative Tier 1 (I wish there was a better word for this. Hybrids? I know some people say hybrids.) … can we use the real/nominal Canada spread as a proxy for requirements? I’ll have to think about this a little more … but the prefs pay 4.50% and distressed hybrids will pay Inflation + Real BAs + 150bp … I don’t think anybody will be all that happy about holding either security in a distress scenario, but this series is all about choosing between them.

Update 2007-02-07-A: In this post, I have expressed some surprise that the spread on Hybrids to Sub-Debt is so narrow. However, in Hold your Hybrids (RBC Capital Markets, November 10, 2006), Altaf Nanji takes the view that a spread of 15bp to sub-debt is just fine. It takes two to make a market!

Next Generation Hybrid Securities (Wall Street Lawyer, May 2006) provides a good overview of US practice.

Moodys has announced that as a result of comments received on their RFC Rating Preferred Stock and Hybrid Securities,

that, except for hybrid securities with meaningful mandatory deferral triggers, all preferred stock and hybrid securities will continue to be rated according to existing notching guidelines with no rating distinction made among cumulative, non-cash cumulative and non-cumulative obligations.

Update 2007-02-14: I really should link to Preferreds & Tier 1 Capital (Part 2), just to facilitate navigation.

In Preferreds & Tier 1 Capital (Part 1) we had a quick look at some balance sheets and at the OSFI specifications for preferred shares’ qualifications for Tier 1.

I was interested enough in the banks’ capital structures to extend the table:

So there’s a wide variation in reliance upon preferred shares, to say the least! To continue this investigation, we should look at the ranking of preferred shares relative to Innovative Tier 1 Capital … are they junior, senior, or parri passu?

For the sake of an example, I looked at the prospectus for the RY.PR.E new issue:

Rights on Liquidation

In the event of our liquidation, dissolution or winding-up, the holders of the Series AE Preferred Shares will be entitled to receive $25.00 per share, together with all dividends declared and unpaid to the date of payment, before any amount may be paid or any or our assets distributed to the registered holders of any shares ranking junior to the Series AE Preferred Shares. The holders of the Series AE Preferred Shares will not be entitled to share in any further distribution of our assets.

Restrictions on Dividends and Retirement of Shares

So long as any of the Series AE Preferred Shares are outstanding, we will not, without the approval of the holders of the Series AE Preferred Shares:

• pay any dividends on any second preferred shares, any common shares or any other shares ranking junior to the Series AE Preferred Shares (other than stock dividends in any shares ranking junior to the Series AE Preferred Shares); or
• redeem, purchase or otherwise retire any second preferred shares, any common shares or any other shares ranking junior to the Series AE Preferred Shares (except out of the net cash proceeds of a substantially concurrent issue of shares ranking junior to the Series AE Preferred Shares); or
• redeem, purchase or otherwise retire less than all the Series AE Preferred Shares; or
• except pursuant to any purchase obligation, sinking fund, retraction privilege or mandatory redemption provision attaching to any series of preferred shares, redeem, purchase, or otherwise retire any other shares ranking on a parity with the Series AE Preferred Shares;

unless all dividends up to and including the dividend payment date for the last completed period for which dividends are payable have been declared and paid, or set apart for payment, in respect of each series of cumulative first preferred shares then issued and outstanding and all other cumulative shares ranking on a parity with the first preferred shares and we have paid, or set apart for payment, all declared dividends in respect of each series of non-cumulative first preferred shares (including the Series AE Preferred Shares) then issued and outstanding, and on all other non-cumulative shares ranking on a parity with the first preferred shares. See “Bank Act Restrictions” in the prospectus.

Now lets look at a recent issue of Innovative Tier 1 Capital : RBC TruCS – Series 2015. These are sold as, and trade as, bonds with a maturity in 2015, with a spread to regular bonds due to the fact that non-repayment in 2015 is an unpleasant event for the bank, but is not actually a default:

On each Regular Distribution Date following December 31, 2015, the Indicated Distribution per RBC TruCS — Series 2015 will be determined by multiplying $1,000 by one half of the sum of the Bankers’ Acceptance Rate (as herein defined) for the Distribution Period (as herein defined) immediately preceding such Distribution Date plus 150 basis points.
…
Pursuant to the terms of a Bank Share Exchange Trust Agreement between the Bank, the Exchange Trustee (as defined herein) and the Trust (the ‘‘Bank Share Exchange Agreement’’), the Bank has covenanted for the benefit of holders of RBC TruCS — Series 2015 (the ‘‘Dividend Stopper Undertakings’’) that if, on any Regular Distribution Date, the Trust fails to pay the Indicated Distribution in full on the RBC TruCS — Series 2015, the Bank will not declare dividends of any kind on any preferred shares or common shares of the Bank (the ‘‘Bank Common Shares’’ and collectively with preferred shares, the ‘‘Dividend Restricted Shares’’) until the month commencing immediately after the third Dividend Declaration Month (as defined herein) following the Trust’s failure to pay the Indicated Distribution unless the Trust first pays such Indicated Distribution (or the unpaid portion thereof) to holders of RBC TruCS — Series 2015. It is in the interest of the Bank to ensure, to the extent within its control, that the Trust pay the Indicated Distribution on the RBC TruCS — Series 2015 on each Regular Distribution Date so as to avoid triggering the Dividend Stopper Undertakings.
…
Each RBC TruCS — Series 2015 will be exchanged automatically (the ‘‘Automatic Exchange’’), without the consent of the holder, for 40 newly issued non-cumulative, perpetual First Preferred Shares, Series Z of the Bank (‘‘Bank Preferred Shares Series Z’’) if: (i) an application for a winding-up order in respect of the Bank pursuant to the Winding-Up and Restructuring Act (Canada) is filed by the Attorney General of Canada or a winding-up order in respect of the Bank pursuant to that Act is granted by a court; (ii) the Superintendent of Financial Institutions (Canada) (the ‘‘Superintendent’’) advises the Bank in writing that the Superintendent has taken control of the Bank or its assets pursuant to the Bank Act (Canada) (the ‘‘Bank Act’’); (iii) the Superintendent advises the Bank in writing that the Superintendent is of the opinion that the Bank has a riskbased Tier 1 Capital ratio of less than 5.0% or a risk-based Total Capital Ratio of less than 8.0%; (iv) the Board of Directors advises the Superintendent in writing that the Bank has a risk-based Tier 1 Capital ratio of less than 5.0% or a risk-based Total Capital Ratio of less than 8.0%; or (v) the Superintendent directs the Bank pursuant to the Bank Act to increase its capital or provide additional liquidity and the Bank elects to cause the Automatic Exchange as a consequence of the issuance of such direction or the Bank does not comply with such direction to the satisfaction of the Superintendent within the time specified therein (each, a ‘‘Loss Absorption Event’’).

The Bank Preferred Shares Series Z will pay semi-annual, non-cumulative per share cash dividends, as and when declared by the Board of Directors on the last day of June and December in each year (subject to adjustment on the first such payment date if the Bank Preferred Shares Series Z have been issued and outstanding for less than six months), equal to $0.60625.

The RBC TruCS — Series 2010, the RBC TruCS — Series 2011 and the RBC TruCS — Series 2015 rank pari passu on the distribution of the property of the Trust in the event of a termination of the Trust (together with the Bank as sole holder of the Special Trust Securities) and rank pari passu in respect of the Indicated Distributions payable on each series of RBC TruCS.

So the Series Z preferreds carry an indicative dividend of 4.85% of par, paid semi-annually (not quarterly!). That’s a reasonable rate, considering the date of issue. Certainly nothing extraordinary.

Now here’s something that will sound very familiar!

The Bank Preferred Shares Series Z will not be redeemable prior to December 31, 2010. On and after December 31, 2010, but subject to the provisions of the Bank Act and the prior approval of the Superintendent and the provisions described below under ‘‘Description of the Bank Preferred Shares Series Z — Restrictions on Dividends and Retirement of Shares’’, the Bank may redeem at any time all, or from time to time any part, of the outstanding Bank Preferred Shares Series Z, at the Bank’s option without the consent of the holder, by the payment of an amount in cash for each such share so redeemed equal to (i) $26.00 per share if redeemed on or prior to December 31, 2011; (ii) $25.75 per share if redeemed after December 31, 2011 and on or prior to December 31, 2012; (iii) $25.50 per share if redeemed after December 31, 2012 and on or prior to December 31, 2013; (iv) $25.25 per share if redeemed after December 31, 2013 and on or prior to December 31, 2014; or (v) $25.00 per share if redeemed after December 31, 2014, plus, in each case, all declared and unpaid dividends up to but excluding the date fixed for redemption.

And in addition there is language describing the Series Z that is substantially identical to the “Restrictions on Dividends and Retirement of Shares” for the Series E Preferreds, above.

Now let’s look at the redemption provisions for the TruCS:

Upon the occurrence of a Tax Event or a Capital Disqualification Event, in each case prior to December 31, 2010, the RBC TruCS — Series 2015 will be redeemable by the Trust at its option in whole (but not in part) without the consent of the holders thereof, upon at least 30 and not more than 90 days’ prior written notice by the Trustee and with Superintendent Approval for a cash amount per RBC TruCS — Series 2015 equal to the Early Redemption Price, being the greater of: (i) the Redemption Price; and (ii) a price per RBC TruCS — Series 2015 calculated to provide an annual yield thereon to December 31, 2015 equal to the Government of Canada Yield plus 0.195% determined on the Business Day immediately preceding the date on which the Trust has given notice of the redemption of the RBC TruCS — Series 2015 as a consequence of the exercise of the Trust Special Event Redemption Right plus the Unpaid Indicated Distribution (the ‘‘RBC TruCS — Series 2015 Canada Yield Price’’).
…
On December 31, 2010 and on any Distribution Date thereafter, the Trust may, at its option, redeem the RBC TruCS — Series 2015 in whole (but not in part) without the consent of the holders thereof, upon at least 30 and not more than 60 days’ prior written notice and with Superintendent Approval, for a cash amount per RBC TruCS — Series 2015 equal to: (i) the Early Redemption Price if the redemption occurs prior to December 31, 2015; or (ii) the Redemption Price if the redemption occurs on or after December 31, 2015.

The “Redemption Price” is par value. These are good provisions – or, at least, relatively good provisions, if a buyer has to give up some call rights! Very often there are provisions in Eurobonds and related instruments that a change in tax law can lead to redemption at par. This is contrary to the interests of the holder, since it might mean he gives up capital gain if interest rates have fallen. However, this prospectus specifies that the worst redemption price will be at a spread to Canadas – and a profitable spread compared to issue price at that. I’m OK with that provision.

What does all this boil down to? After all, you must suspect that after all this quotation I’m going to get to the point eventually, right?

• The bank has some incentive to redeem the TruCS on the intended date of December 31, 2015. After this date, the interest payable on the TruCS changes to BAs + 150bp. Given current conditions, this is far more than the rate at which the bank could otherwise raise money, but this will not necessarily always be the case:
  • The credit quality of the bank may have deteriorated to the point at which BAs + 150 bp is a pretty good deal. After all, prime is now 6.00% at RBC … BAs + 170 bp.
  • The curve could get steeper. These bonds are perpetual. A 0-30 term spread of 150bp is by no means unheard of.
  • I note that an RBC Floating Rate Note maturing in 2083 is quoted at 97bp over BAs.
• Income is better protected in the TruCS than in the Preferreds – if they stop paying interest on the TruCS, dividends on the prefs can’t be paid for about a year.
• Default on Principal has the same protection on TruCS as on preferreds. This assumes that default on principal will be preceeded by an automatic conversion of the TruCS to Series Z 1st preferreds, which are pari passu with regular preferreds.

And what does all this mean in terms of investment policy? Ah, for that you’ll have to wait for Part 3!