There was a discussion at Financial Webring Forum in which one of the participants expressed a view that recent market declines proves that buying perpetuals at a premium to their ultimate redemption price is a bad idea.
So, I thought I’d provide a brief review of the arguments in favour of premiumPerpetuals as a sub-class of preferreds. The advantages are two-fold:
(i) A premium to redemption price implies that a redemption will be favourable to the issuer. Thus, an investor in such issues has at least some chance of having his capital returned in a timely manner without having to accept the usually very low yields offered on retractibles. This also mitigates the credit risk inherent in lending a company money “forever”.
(ii) A premium to redemption price, to the extent that it implies a coupon that is higher than required for new issues, provides a buffer against adverse changes in interest rates.
A quick understanding of the second point can be achieved by considering two imaginary issues. The numbers in this example are not mathematically precise – should a reader wish mathematical precision he is more than welcome to work them out!
The first issue is callable in 4 years at $25 and is currently priced at $26.00 with a coupon of 6% (that is, the annual dividend is $1.50). The holder would be very happy to hold it forever at this coupon, but must account for an expected capital loss when the issue is called of $1. The amortization of this premium amounts to $0.25 annually, thus his net income is $1.50-$0.25 = $1.25 which, taken as a fraction of $25.00 (I told you the numbers wouldn’t work out precisely!) is a net yield of 5% p.a., or, to put it another way, 6% coupon – 1% annual capital loss = 5% net income.
The second issue has a coupon of 4% and it trades at $20.00. The holder cannot rely on it being called at $25 (why would the issuer call an issue with a 4% coupon to refinance at 5%?), so the net yield is 5%.
Thus, in terms of expected annual yield, these issues are basically the same. Suppose, however, that yields increase massively … from 5% to 6%. The first issue, with its coupon of 6% will be trading at $25 … only $1 of capital has been lost and that was a dollar the holder was expecting to lose anyway! Lose it in 4 years, lose it now … it’s not the biggest deal in the world.
The second issue, with a coupon of 4%, implying an annual dividend of $1.00, will fall in price to about $16.70 (in order to keep the yield for new purchasers equal to the new market rate of 6%) – a capital loss of $3.30, or about 16.5% of the capital invested.
Thus, the premium issue provides greater protection against interest rate increases.
This doesn’t come for free, however: if interest rates fall, the discount issue will rise in price, just as dramatically. The premium issue will also rise in price, but not nearly so much – a rational buyer will still be expecting the same call at $25.00 at the same time … that will have even greater likelihood, as the issuer will cut costs even more on refinancing.
Additionally, premium issues will generally trade so that their yield-to-expected-call is a little less than discounts. Ain’t nuthin’ free in this world! Each investor has to assess the risk/reward trade-offs of each investment for himself, and come up with a diversified portfolio that meets his needs. You can’t just buy a premium issue simply because it has a premium … the yield give up might be too much. Even now, with all the recent carnage in the market, GWO.PR.F is bid at 26.95, with a yield-to-worst of 2.75% based on a call 2008-10-30 at $26. Obviously, buyers are hoping that it won’t be called at that time, that it will take longer … but even if it lasts until a $25.00 call on 2012-10-30, it will only have yielded 4.22%! One can find many high quality perpetuals yielding far in excess of 4.22% without any pious hoping that the treasurer won’t notice the fat dividend. The market obviously disagrees with me, but I say that issue is just too damn expensive.
In my article How Long is Forever?, I made the point that yield spreads between perpetuals and retractibles can be so large that rates would have to be extremely high in the future for total return over the period to favour the retractible. In Perpetual Hockey Sticks, I look at the trade-off between “expected yield” and “protection from future higher interest” in more detail. Enjoy!
Note: We can get some idea of the differences by looking at the NAV of the Claymore Preferred ETF. This peaked at 20.05 on 4/12 and the bottom (so far!) is yesterday, 6/8, at 19.08. In that time, the HIMI PerpetualPremium index has had a total return of -4.35%, the PerpetualDiscount index has returned -10.12%.
Excellent Preferred Letter!. At this point, with many prefs falling like daggers, it is an extermely valuable guide for discerning the wheat from the chaff. One question though: can any case be made for those perpetual discounts such as SLF C which have a good solid credit and an attractive YTW? I understand your point about the aatractiveness of perpetual premiums (and, for that matter, retractables) in a possibly rising rate enviroment, but can anything good be said about deep discounts?
One other question, if I may. Why have the recently issued various 5.25 split issues (the copernican cibc, bns, etc.) not been affected at all by the rcent draft down? Low liquidity or the retraction feature?
Thanks
The nice thing about deep discounts is that they can (i) provide a nice high income level (I would expect yields on discounts to exceed yields on premiums, as shown in my Canadian Moneysaver article; the fact that this is not the case for all issues right now is just another example of transient (probably transient!) preferred share market inefficiency and (ii) they provide huge leverage in the case of interest rate declines, which, for some investors, will be an attractive speculation.
The issue you mentioned, SLF.PR.C, is now quoted at 21.75-00. If we buy some at 22.00, that implies (a) a yield of 5.06% for the forseeable future as a dividend, which is an interest-equivalent of over 7%, and (b) a potential capital gain of $3 per share IF interest rates go back down. Such a capital gain will not be realized by the premiums, since they’ll just be called at the currently expected time at the currently expected price.
Mind you, a buyer of this issue is also taking on long-term credit exposure to Sunlife, so watch the credits and diversify!
I’m glad you liked the PrefLetter. Tell your friends! Tell your neighbors! Take an ad out in the paper!
The split shares have been affected. However both they and the operating retractibles have had the blow cushioned by the retraction feature; in addition, since many of the shares to which you refer are priced in the $10 range, you may be perceiving the 1%-or-so loss in the past month as a rounding error rather than a loss. Note that most of this 1%-or-so loss has occurred in June; the May Returns did leave them virtually unaffected, I agree.