10 Responses to “September 17, 2008”

1. Annette says:

   September 18, 2008 at 12:35 am

   Dear Mr. Hymas,

   I love your blog and thank you for all the time you put into it.

   Me too I did not like the “new” Fixed Reset prefs when they came on the Canadian market a few months ago. They reminded me the step-up extendible notes RY had introduced (I think) and keeps issuing (at least until recently). I lost money with those not fully appreciating what their fine prints meant.

   With such a preset mindset, I was prompt thinking that the “new” Fixed Reset prefs were simply just another trick whereby banks could issue prefs paying less dividends than what new issues of plain perps would otherwise have to pay.

   I was thus very much comforted reading your well reasoned analysis on Fixed Reset vs. Perps with its fabulous punchy end sentence reading: “It is the pretense that borrowers can access long term funds from borrowers assuming short-term risks that, after all, caused the credit crisis in the first place”.

   As much as I truly hate to say it (I don’t own any fixed reset so far), I think I was wrong and that your beautiful punchy line is, unfortunately, a “limpy comparison”.

   Unlike with SIVs, ABCPs and other investments vehicles I might not even know the name, one has at least a fair chance of finding out what they are made of and make his own mind as to the value of the underlying assets (the issuer of the Fixed Reset). Unlike with SIVs, ABCPs and other instruments which may hide sub-prime mortages or other bad assets mixed together and shared amongst the world (as if sharing together a complicated mix of good and bad assets make the whole thing better), there is a market for Fixed Reset prefs such that there is at least always the possibility of finding a buyer for it (at loss or with a gain obviously). The problem with ABCPs, SIVs, etc. is that one technically ends up with an instrument secured by say, for instance, a fraction of multiple mortages in various locations throughout the U.S., some of which might have already been in default when you bought the product and all of them without any proper mechanism in place to foreclose these mortages (I would be prepared to bet with you that there are people in the U.S perfectly able to pay their mortgage but which simply stop making their payments without anyone trying to collect the loan from them). This is, in my humble and insufficiently educated opinion, what caused the first round of the credit crisis.

   Other than its potential perpetual term, the fixed reset prefs or their underlying rationale have nothing in common with what caused the credit crisis in the first place. Unfortunately, your beautiful line is mixing apples with oranges.

   In addition to providing some protection for its holder against inflation and the resulting drop in value in order to bring up the yield of the plain perpetual, I now have the strong feeling that the Fixed Reset Pref is providing some protection in the case of a continued credit crisis to both the issuer and its buyer. Please correct me if I am wrong but the current confidence crisis cannot explain why a plain perp from a same issuer is doing worse nowadays than the fixed reset product from the same issuer (taking into account, obviously, all the other characteristics of each product). My fear with my current well diversied portfolio but mostly constituted of plain perps portfolio is the next issues of perps, if any, will have to yield even more than the 5.6, 5.8 and 6.0% of the most recent plain perps from banks and that this situation might be for quite a long time before the lost confidence returns to the financial market. This will bring the price of my current perps even lower.

   With the Fixed Reset, I would feel, I think, having at least the comfort that the share will either be bought out in 5 years or its yield will be fixed at a price closer to the then prevailing market. No doubts, the next plain perps are likely to be an excellent deal if they are to be issued before the current crisis is resolved. However, in the meantime, the one with a portfolio made of Fixed Reset must sleep better than I do these days as transpires from the time I am posting you this comment.

   Good nite!

2. jiHymas says:

   September 18, 2008 at 1:15 am

   Wow! Thank you for your comment! We do have a few areas of disagreement; I will briefly state them here and continue to ponder your points.

   there is a market for Fixed Reset prefs such that there is at least always the possibility of finding a buyer for it (at loss or with a gain obviously).

   Well, you are saying “possibility” rather than certainty; but I will say that markets in anything can disappear instantly. Markets for ABCP and SIV Commercial paper dried up regardless of the quality of the underlying assets; but the closest parallel would be the Auction Rate Securities. Perfectly good credits – like the State of New York, the State of California, student loans with a 95% guarantee – have their auctions failing, with no obligation to redeem.

   I don’t think there is much chance of the Fixed-Reset market drying up, but there’s no reason why it can’t happen.

   However, exactly the same risk applies to straight perpetuals, so as far as that’s concerned, it’s a tie.

   But my main point with the short-term/long-term line was to point out that Fixed-Resets are paying what are, essentially, five-year rates for money; but the credit risk is perpetual. It is this mismatch, between the terms of the rate being paid and the risk being assumed, that makes me nervous.

   Please correct me if I am wrong but the current confidence crisis cannot explain why a plain perp from a same issuer is doing worse nowadays than the fixed reset product from the same issuer (taking into account, obviously, all the other characteristics of each product).

   Quite right, and this was the point I was attempting to highlight in the post. If the recent increase in PerpetualDiscount yields is all about credit, Fixed-Resets should be doing equally poorly since they bear exactly the same credit risk. They differ only in their term risk, which is not as yet a feature of the fixed-income environment.

   What I am saying is, the market appears to be behaving irrationally with respect to these two classes of preferreds.

   With the Fixed Reset, I would feel, I think, having at least the comfort that the share will either be bought out in 5 years or its yield will be fixed at a price closer to the then prevailing market.

   Well … not necessarily! Should the flight to quality still be going full strength at reset time, the yield of the fixed-reset will be determined by the yield on Canadas, not by the yield on Corporates. Nothing makes it impossible for the five-year GOC yield to be 1% at that time (highly unlikely, but possible!) A reset to GOC+200bp won’t be all that wonderful a deal.

   Fixed-Resets do have their points, but they yield about 100bp less than straights, and are trading as if there’s a guarantee of a price of 25.00 (one way or another, by redemption or reset) in five years. But there is no such guarantee. The 100bp extra that I will make on straights if the market stays stable is the fee I charge for the risk that the issuers credit will deteriorate significantly over the next five years.

3. prefwatcher says:

   September 18, 2008 at 8:24 am

   I note your comment on the decline of the BCE prefs “Financing Jitters? see main text” but I cant find any further reference to these prefs. I think the prefs are falling in sympathy with the common but to my mind they should be rising since if the deal falls through they will no longer be behind a mountain of junk bonds.

4. jiHymas says:

   September 18, 2008 at 9:08 am

   Sorry! It was Sept. 16‘s main text that had a note about BCE.

   The BCE rating of Pfd-2(low) reflects the pre-deal financial position. If they were rated behind the Junk Mountain, they’s be more like Pfd-4, if that.

   The major worry if the deal falls throught is that the common stock will crater and the company will have to do something shareholder-friendly / creditor unfriendly to rescue it. The first plan, from before the Teachers’ bid, was a big, big share buyback.

   And their business has deteriorated since then.

5. Annette says:

   September 18, 2008 at 11:07 pm

   Thks Mr. Hymas for your detailed reply despite all what is going on at the moment. Just a few more clarifications and questions though:

   1. I did not suggest that marketability of straight perps is any different than the marketability of resets.

   2. Where we disagree the most is when you seem to consider similar the secondary market for prefs (the stock exchange) with the secondary market (if any) for ABCP and SIV commercial papers. The absence of a secondary market to value ABCPs was raised on multiple occasions as a reason for the difficulties of the Canadian Banks to assess the amount of the reserves to be taken and for the delays getting to a solution (not over yet if leave to appeal is granted by the SCC) further to the Montreal agreement. An auction process is not the stock market. Its time constraints and more restricted number of participants cannot fetch a fair market value as close to reality as a public stock market. Off course the market for a bad stock will dry out whichever vehicle used to try to sell them but buying a pref does not relieve its holder from its duty “to do his homeworks” as you say.

   3. The consideration for the lower dividend of the reset is the formula coming with it. One is then in fact buying the worse of either the five year reset formula or having your shares bought back by the issuer at par. This can be good or bad depending upon your crystal ball but the outcome of your purchase of a reset seems to me more predictable with some protection of your capital in the event interest rates and/or market spreads were to increase in the next five years. Insofar as the risk of insolvency of the issuer is concerned, I agree with you that both are exposed to the same risk but why does this prevent a “return” differential for the other risks / characteristics both types of prefs differ?

   4. Before committing myself investing in those, I would like to know from you what calculations, if any, make you say that the lesser dividend yield at issuance of say the most recent resets is not a sufficient consideration for the extra / different characteristics of the reset?

   4. I would also like someone to explain why all those resets are all exchangeable for a future “pair” yet to be issued. I just don’t understand the reasons for that. Why did they make it so complicated?

   Thks again!

6. jiHymas says:

   September 19, 2008 at 1:49 am

   1. No, that was just me saying that marketability considerations cannot be used to differentiate straights and Fixed-Resets. I was mainly trying to respond to your point differentiating between the secondary markets for preferreds and structured commercial paper – there is no guarantee that there will be a bid on any marketable instrument tomorrow.

   2. But really, I was just responding to your point that there is a market for Fixed Reset prefs such that there is at least always the possibility of finding a buyer for it (at loss or with a gain obviously). and saying that there would not necessarily be a bid for Fixed-Resets or anything else.

   3. I certainly agree that the reset provision for this batch of fixed-resets is worth something … I stated early on that if there was a 20-year no-call period, I think they’d be great things! A big objection is the short term prior to the issuer’s first call, not just the low initial dividend.

   4a. According to the Bank of Canada, the 5-30 yield spread is 105bp. That’s as interest, while the preferred share differential is 100+bp as dividend. The Canada bond spread also does not include an allowance for the extension risk that exists with the two preferred classes – that 5-year return of capital is guaranteed.

   I can’t really give you a fully satisfactory answer. The risk profiles of the two classes are sufficiently different that any model will provide a number thoroughly dependent upon the assumptions made and the time period selected for the data; most of my analytical work consists of more apples-apples comparisons.

   Based on the Canada curve, the current structure and the old stand-by “gut feel”, I suspect the indifference point is somewhere in the 30-50bp range.

   4b. I think it was just to offer a fully floating alternative for those who wanted it. I suspect that the structure was designed to be as close as possible to the BCE structure, while still being allowed as Tier 1 capital by OSFI (which would not allow – I am told – a straight copy-paste of that structure).

7. Louis says:

   September 19, 2008 at 7:13 pm

   Thanks James and Annette for the exchange as it helps me to trying to make my mind over the resets.

   To me it all boils down to point 4a. Having read over James’ excellent analysis on Resets (except perhaps, if I may tease James a bit and make Annette happy, its last sentence), my own and humble personal feeling after looking at his “Chart no 1” showing the historical spreads between 5year Canadas and the after income tax average equivalent yield of straight perps is that I would seriously consider buying at par or below par or may subscribe on a next issue (at the next quarter results I guess) if the reset formula provides for a minimum reset of +200 bps or more over the 5 year Canadas. At the moment, I only see the BNS.PR.P, NA.PR.M & CM.PR.K as having reset rates at more than +200 bps. Out of those, only CM.PR.K with its quite tempting +5.35% dividend for the first five year is below par ($24.97) and is resetable at 5 year Canadas + 218 bps. My only concern is that its issuer is the CIBC…

8. jiHymas says:

   September 20, 2008 at 11:07 am

   In the end, you have to go with your comfort level. Whenever you worry about a particular position in your portfolio, that’s a sign you own too much.

   I will note two things, however:

   (i) If you are seeking to protect your portfolio against inflation, perpetual preferreds are rather an odd choice of vehicle. I suggest that common stock in a resources company would complement a PerpetualDiscount portfolio in a more efficient manner than a portfolio comprised exclusively of Fixed-Resets.

   (ii) The yield give-up for the CM PerpetualDiscount/Fixed-Reset swap is larger than with any of the other issuers.

9. Annette says:

   September 20, 2008 at 4:53 pm

   (i) … but, if I may, common stock in a ressources company is not quite the same level of risk as a Canadian bank perp (whether it is a straight or a fixed reset) even in a period of inflation?

   (ii) It just shows that the CIBC expects to catch up with the other Canadian banks in terms of quality / reliability. Keeping away from the Enrons and high US financials exposure should do it unless there is something (or someone) in its structure making the CIBC more enclined to fall into traps than the other Canadian banks.

   Mind you. I don’t own any fixed resets as of yet, they have hurted my portfolio of straight perps every time they have been issued and I find their structure overcomplicated (and, therefore, suspect).

10. jiHymas says:

    September 20, 2008 at 5:39 pm

    (i) Quite right, and this is a good thing. A balanced portfolio will have a wide variety of elements in it, each element presenting its own profile of risk/reward. The classic example is a two-stock portfolio with shares in a suntan-oil factory and an umbrella manufacturer. Rain or shine, you’ve got one winner.

    (ii) I don’t understand your reasoning. Given that there is no difference in credit between FixedResets and Straights from the same issuer, I would expect the yield give-up between the two to reflect solely the structural difference. The fact that the give-ups are so different is one thing that leads me to suspect that these things are not being rationally priced (as opposed to being rationally priced in a way I don’t understand, which is always a possibility!).

    The structure-spread is 85bp for BNS (5.73% PerpetualDiscount Yield, less 4.88% FixedReset yield) and 158bp for CM (6.75% – 5.17%). I don’t get it.